Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-770/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,63 @@
+---
+base_model: Qwen/Qwen3.5-9B
+library_name: peft
+model_name: v21_semantic_boundaries
+tags:
+- base_model:adapter:Qwen/Qwen3.5-9B
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for v21_semantic_boundaries
+
+This model is a fine-tuned version of [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+ 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- PEFT 0.18.1
+- TRL: 0.24.0
+- Transformers: 5.3.0
+- Pytorch: 2.10.0
+- Datasets: 4.6.1
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForConditionalGeneration",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.5-9B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dc94603f0c6e625c9716499f91c5acf38deff4596e0ed2ca5fc3a4f5126806c
+size 173188512

chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-500/README.md

@@ -0,0 +1,210 @@
+---
+base_model: Qwen/Qwen3.5-9B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3.5-9B
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

checkpoint-500/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForConditionalGeneration",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.5-9B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7302f5a435c3f2311801ee617b710886c58a92e7197b37676d81ea4261f5f7b9
+size 173188512

checkpoint-500/chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91d2126200d3f2000e846fc2c5efb0503b7dc1e60da6cc626874b080c19aeaf4
+size 88617029

checkpoint-500/processor_config.json

@@ -0,0 +1,63 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

checkpoint-500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d6d8fafcd1ee268414be5acf0366296af5b03d60871978712eac1979cb42d65
+size 14645

checkpoint-500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc7746ec87c19743016407a58d3acdb0930ab84f523fb5edda75bbf4563c6928
+size 1465

checkpoint-500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:263dcc3feef53d558a357d52941ff3649fa9cd9a3199934376f2bbf615d49ffc
+size 19989610

checkpoint-500/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": false,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

checkpoint-500/trainer_state.json

@@ -0,0 +1,3534 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.5992779783393503,
+  "eval_steps": 770,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007220216606498195,
+      "grad_norm": 2.68963885307312,
+      "learning_rate": 0.0,
+      "loss": 1.880106806755066,
+      "step": 1
+    },
+    {
+      "epoch": 0.01444043321299639,
+      "grad_norm": 2.3691370487213135,
+      "learning_rate": 8e-05,
+      "loss": 1.882678508758545,
+      "step": 2
+    },
+    {
+      "epoch": 0.021660649819494584,
+      "grad_norm": 3.7024030685424805,
+      "learning_rate": 0.00016,
+      "loss": 1.9426246881484985,
+      "step": 3
+    },
+    {
+      "epoch": 0.02888086642599278,
+      "grad_norm": 0.7646510601043701,
+      "learning_rate": 0.00024,
+      "loss": 1.6261038780212402,
+      "step": 4
+    },
+    {
+      "epoch": 0.036101083032490974,
+      "grad_norm": 0.6628475189208984,
+      "learning_rate": 0.00032,
+      "loss": 1.478842854499817,
+      "step": 5
+    },
+    {
+      "epoch": 0.04332129963898917,
+      "grad_norm": 0.7141556143760681,
+      "learning_rate": 0.0004,
+      "loss": 1.1376413106918335,
+      "step": 6
+    },
+    {
+      "epoch": 0.05054151624548736,
+      "grad_norm": 0.7212001085281372,
+      "learning_rate": 0.0003999983135391464,
+      "loss": 1.0410869121551514,
+      "step": 7
+    },
+    {
+      "epoch": 0.05776173285198556,
+      "grad_norm": 0.7515780925750732,
+      "learning_rate": 0.0003999932541850271,
+      "loss": 0.8246882557868958,
+      "step": 8
+    },
+    {
+      "epoch": 0.06498194945848375,
+      "grad_norm": 2.474442481994629,
+      "learning_rate": 0.0003999848220229661,
+      "loss": 0.7787771821022034,
+      "step": 9
+    },
+    {
+      "epoch": 0.07220216606498195,
+      "grad_norm": 1.2009592056274414,
+      "learning_rate": 0.0003999730171951686,
+      "loss": 0.6830324530601501,
+      "step": 10
+    },
+    {
+      "epoch": 0.07942238267148015,
+      "grad_norm": 0.7545379996299744,
+      "learning_rate": 0.00039995783990071825,
+      "loss": 0.7438161969184875,
+      "step": 11
+    },
+    {
+      "epoch": 0.08664259927797834,
+      "grad_norm": 0.9728625416755676,
+      "learning_rate": 0.0003999392903955744,
+      "loss": 0.6851149797439575,
+      "step": 12
+    },
+    {
+      "epoch": 0.09386281588447654,
+      "grad_norm": 0.5509404540061951,
+      "learning_rate": 0.00039991736899256693,
+      "loss": 0.5853980183601379,
+      "step": 13
+    },
+    {
+      "epoch": 0.10108303249097472,
+      "grad_norm": 0.715829074382782,
+      "learning_rate": 0.0003998920760613919,
+      "loss": 0.7855823040008545,
+      "step": 14
+    },
+    {
+      "epoch": 0.10830324909747292,
+      "grad_norm": 0.6777167320251465,
+      "learning_rate": 0.0003998634120286046,
+      "loss": 0.5202292799949646,
+      "step": 15
+    },
+    {
+      "epoch": 0.11552346570397112,
+      "grad_norm": 0.5348321199417114,
+      "learning_rate": 0.0003998313773776128,
+      "loss": 0.6527559757232666,
+      "step": 16
+    },
+    {
+      "epoch": 0.12274368231046931,
+      "grad_norm": 0.4939132332801819,
+      "learning_rate": 0.0003997959726486683,
+      "loss": 0.584251880645752,
+      "step": 17
+    },
+    {
+      "epoch": 0.1299638989169675,
+      "grad_norm": 0.47802454233169556,
+      "learning_rate": 0.00039975719843885805,
+      "loss": 0.5235528349876404,
+      "step": 18
+    },
+    {
+      "epoch": 0.1371841155234657,
+      "grad_norm": 0.44072362780570984,
+      "learning_rate": 0.0003997150554020938,
+      "loss": 0.5184933543205261,
+      "step": 19
+    },
+    {
+      "epoch": 0.1444043321299639,
+      "grad_norm": 0.4251253306865692,
+      "learning_rate": 0.00039966954424910147,
+      "loss": 0.6167617440223694,
+      "step": 20
+    },
+    {
+      "epoch": 0.15162454873646208,
+      "grad_norm": 0.5198166370391846,
+      "learning_rate": 0.00039962066574740886,
+      "loss": 0.7412333488464355,
+      "step": 21
+    },
+    {
+      "epoch": 0.1588447653429603,
+      "grad_norm": 0.44667476415634155,
+      "learning_rate": 0.00039956842072133266,
+      "loss": 0.6361726522445679,
+      "step": 22
+    },
+    {
+      "epoch": 0.16606498194945848,
+      "grad_norm": 0.40787550806999207,
+      "learning_rate": 0.0003995128100519649,
+      "loss": 0.49531102180480957,
+      "step": 23
+    },
+    {
+      "epoch": 0.17328519855595667,
+      "grad_norm": 0.4681910276412964,
+      "learning_rate": 0.00039945383467715766,
+      "loss": 0.584551990032196,
+      "step": 24
+    },
+    {
+      "epoch": 0.18050541516245489,
+      "grad_norm": 0.41188451647758484,
+      "learning_rate": 0.00039939149559150753,
+      "loss": 0.6003429889678955,
+      "step": 25
+    },
+    {
+      "epoch": 0.18772563176895307,
+      "grad_norm": 0.43262341618537903,
+      "learning_rate": 0.0003993257938463389,
+      "loss": 0.5725842118263245,
+      "step": 26
+    },
+    {
+      "epoch": 0.19494584837545126,
+      "grad_norm": 0.45091742277145386,
+      "learning_rate": 0.0003992567305496859,
+      "loss": 0.6373569965362549,
+      "step": 27
+    },
+    {
+      "epoch": 0.20216606498194944,
+      "grad_norm": 0.3687536120414734,
+      "learning_rate": 0.00039918430686627406,
+      "loss": 0.4958152174949646,
+      "step": 28
+    },
+    {
+      "epoch": 0.20938628158844766,
+      "grad_norm": 0.3994904160499573,
+      "learning_rate": 0.00039910852401750036,
+      "loss": 0.6146813631057739,
+      "step": 29
+    },
+    {
+      "epoch": 0.21660649819494585,
+      "grad_norm": 0.41312268376350403,
+      "learning_rate": 0.0003990293832814129,
+      "loss": 0.5313689708709717,
+      "step": 30
+    },
+    {
+      "epoch": 0.22382671480144403,
+      "grad_norm": 0.37877994775772095,
+      "learning_rate": 0.0003989468859926893,
+      "loss": 0.5694329738616943,
+      "step": 31
+    },
+    {
+      "epoch": 0.23104693140794225,
+      "grad_norm": 0.32251644134521484,
+      "learning_rate": 0.0003988610335426139,
+      "loss": 0.4476504921913147,
+      "step": 32
+    },
+    {
+      "epoch": 0.23826714801444043,
+      "grad_norm": 0.40205422043800354,
+      "learning_rate": 0.0003987718273790548,
+      "loss": 0.4696302115917206,
+      "step": 33
+    },
+    {
+      "epoch": 0.24548736462093862,
+      "grad_norm": 0.3664834201335907,
+      "learning_rate": 0.0003986792690064389,
+      "loss": 0.5304725170135498,
+      "step": 34
+    },
+    {
+      "epoch": 0.2527075812274368,
+      "grad_norm": 0.5150560736656189,
+      "learning_rate": 0.000398583359985727,
+      "loss": 0.5886795520782471,
+      "step": 35
+    },
+    {
+      "epoch": 0.259927797833935,
+      "grad_norm": 0.37544649839401245,
+      "learning_rate": 0.0003984841019343872,
+      "loss": 0.6374161839485168,
+      "step": 36
+    },
+    {
+      "epoch": 0.26714801444043323,
+      "grad_norm": 0.3426123857498169,
+      "learning_rate": 0.0003983814965263676,
+      "loss": 0.507249116897583,
+      "step": 37
+    },
+    {
+      "epoch": 0.2743682310469314,
+      "grad_norm": 0.36909183859825134,
+      "learning_rate": 0.00039827554549206836,
+      "loss": 0.5216564536094666,
+      "step": 38
+    },
+    {
+      "epoch": 0.2815884476534296,
+      "grad_norm": 0.3864560127258301,
+      "learning_rate": 0.00039816625061831206,
+      "loss": 0.4607892632484436,
+      "step": 39
+    },
+    {
+      "epoch": 0.2888086642599278,
+      "grad_norm": 0.3553301692008972,
+      "learning_rate": 0.0003980536137483141,
+      "loss": 0.6021184921264648,
+      "step": 40
+    },
+    {
+      "epoch": 0.296028880866426,
+      "grad_norm": 0.3300604224205017,
+      "learning_rate": 0.0003979376367816511,
+      "loss": 0.49985307455062866,
+      "step": 41
+    },
+    {
+      "epoch": 0.30324909747292417,
+      "grad_norm": 0.3894491493701935,
+      "learning_rate": 0.0003978183216742292,
+      "loss": 0.5403987169265747,
+      "step": 42
+    },
+    {
+      "epoch": 0.3104693140794224,
+      "grad_norm": 0.32339245080947876,
+      "learning_rate": 0.0003976956704382511,
+      "loss": 0.5262795090675354,
+      "step": 43
+    },
+    {
+      "epoch": 0.3176895306859206,
+      "grad_norm": 0.3121689260005951,
+      "learning_rate": 0.00039756968514218166,
+      "loss": 0.4988410174846649,
+      "step": 44
+    },
+    {
+      "epoch": 0.3249097472924188,
+      "grad_norm": 0.37562957406044006,
+      "learning_rate": 0.0003974403679107138,
+      "loss": 0.5539376735687256,
+      "step": 45
+    },
+    {
+      "epoch": 0.33212996389891697,
+      "grad_norm": 0.3560684323310852,
+      "learning_rate": 0.0003973077209247319,
+      "loss": 0.4829791784286499,
+      "step": 46
+    },
+    {
+      "epoch": 0.33935018050541516,
+      "grad_norm": 0.3543923497200012,
+      "learning_rate": 0.0003971717464212755,
+      "loss": 0.4733068346977234,
+      "step": 47
+    },
+    {
+      "epoch": 0.34657039711191334,
+      "grad_norm": 0.3253995180130005,
+      "learning_rate": 0.0003970324466935013,
+      "loss": 0.4318773150444031,
+      "step": 48
+    },
+    {
+      "epoch": 0.35379061371841153,
+      "grad_norm": 0.3587929904460907,
+      "learning_rate": 0.0003968898240906447,
+      "loss": 0.6348429322242737,
+      "step": 49
+    },
+    {
+      "epoch": 0.36101083032490977,
+      "grad_norm": 0.3772837519645691,
+      "learning_rate": 0.0003967438810179802,
+      "loss": 0.5192855596542358,
+      "step": 50
+    },
+    {
+      "epoch": 0.36823104693140796,
+      "grad_norm": 0.3878256380558014,
+      "learning_rate": 0.0003965946199367804,
+      "loss": 0.4653000831604004,
+      "step": 51
+    },
+    {
+      "epoch": 0.37545126353790614,
+      "grad_norm": 0.306704044342041,
+      "learning_rate": 0.00039644204336427506,
+      "loss": 0.46054646372795105,
+      "step": 52
+    },
+    {
+      "epoch": 0.38267148014440433,
+      "grad_norm": 0.3290722072124481,
+      "learning_rate": 0.0003962861538736084,
+      "loss": 0.4637136459350586,
+      "step": 53
+    },
+    {
+      "epoch": 0.3898916967509025,
+      "grad_norm": 0.3877376914024353,
+      "learning_rate": 0.00039612695409379557,
+      "loss": 0.6008591651916504,
+      "step": 54
+    },
+    {
+      "epoch": 0.3971119133574007,
+      "grad_norm": 0.3738860785961151,
+      "learning_rate": 0.00039596444670967864,
+      "loss": 0.49255216121673584,
+      "step": 55
+    },
+    {
+      "epoch": 0.4043321299638989,
+      "grad_norm": 0.43176719546318054,
+      "learning_rate": 0.0003957986344618809,
+      "loss": 0.6328399777412415,
+      "step": 56
+    },
+    {
+      "epoch": 0.41155234657039713,
+      "grad_norm": 0.34965041279792786,
+      "learning_rate": 0.00039562952014676116,
+      "loss": 0.5274545550346375,
+      "step": 57
+    },
+    {
+      "epoch": 0.4187725631768953,
+      "grad_norm": 0.3525114059448242,
+      "learning_rate": 0.000395457106616366,
+      "loss": 0.5174035429954529,
+      "step": 58
+    },
+    {
+      "epoch": 0.4259927797833935,
+      "grad_norm": 0.36805179715156555,
+      "learning_rate": 0.0003952813967783822,
+      "loss": 0.5132086277008057,
+      "step": 59
+    },
+    {
+      "epoch": 0.4332129963898917,
+      "grad_norm": 0.3320341110229492,
+      "learning_rate": 0.00039510239359608735,
+      "loss": 0.47466331720352173,
+      "step": 60
+    },
+    {
+      "epoch": 0.4404332129963899,
+      "grad_norm": 0.36727970838546753,
+      "learning_rate": 0.0003949201000883001,
+      "loss": 0.5845056772232056,
+      "step": 61
+    },
+    {
+      "epoch": 0.44765342960288806,
+      "grad_norm": 0.31494706869125366,
+      "learning_rate": 0.000394734519329329,
+      "loss": 0.5123739242553711,
+      "step": 62
+    },
+    {
+      "epoch": 0.4548736462093863,
+      "grad_norm": 0.31959858536720276,
+      "learning_rate": 0.00039454565444892094,
+      "loss": 0.3888605237007141,
+      "step": 63
+    },
+    {
+      "epoch": 0.4620938628158845,
+      "grad_norm": 0.4212755262851715,
+      "learning_rate": 0.0003943535086322083,
+      "loss": 0.6065781712532043,
+      "step": 64
+    },
+    {
+      "epoch": 0.4693140794223827,
+      "grad_norm": 0.2938922643661499,
+      "learning_rate": 0.0003941580851196549,
+      "loss": 0.47488927841186523,
+      "step": 65
+    },
+    {
+      "epoch": 0.47653429602888087,
+      "grad_norm": 0.27896228432655334,
+      "learning_rate": 0.00039395938720700196,
+      "loss": 0.36559349298477173,
+      "step": 66
+    },
+    {
+      "epoch": 0.48375451263537905,
+      "grad_norm": 0.34284013509750366,
+      "learning_rate": 0.0003937574182452118,
+      "loss": 0.49944937229156494,
+      "step": 67
+    },
+    {
+      "epoch": 0.49097472924187724,
+      "grad_norm": 0.3411049544811249,
+      "learning_rate": 0.00039355218164041194,
+      "loss": 0.5018436312675476,
+      "step": 68
+    },
+    {
+      "epoch": 0.4981949458483754,
+      "grad_norm": 0.35431182384490967,
+      "learning_rate": 0.0003933436808538375,
+      "loss": 0.42643067240715027,
+      "step": 69
+    },
+    {
+      "epoch": 0.5054151624548736,
+      "grad_norm": 0.3351145386695862,
+      "learning_rate": 0.00039313191940177256,
+      "loss": 0.5355833768844604,
+      "step": 70
+    },
+    {
+      "epoch": 0.5126353790613718,
+      "grad_norm": 0.3065779507160187,
+      "learning_rate": 0.00039291690085549105,
+      "loss": 0.4329970180988312,
+      "step": 71
+    },
+    {
+      "epoch": 0.51985559566787,
+      "grad_norm": 0.5208595395088196,
+      "learning_rate": 0.00039269862884119664,
+      "loss": 0.5850666165351868,
+      "step": 72
+    },
+    {
+      "epoch": 0.5270758122743683,
+      "grad_norm": 0.3183553218841553,
+      "learning_rate": 0.00039247710703996144,
+      "loss": 0.5500473380088806,
+      "step": 73
+    },
+    {
+      "epoch": 0.5342960288808665,
+      "grad_norm": 0.32788676023483276,
+      "learning_rate": 0.0003922523391876638,
+      "loss": 0.5200982093811035,
+      "step": 74
+    },
+    {
+      "epoch": 0.5415162454873647,
+      "grad_norm": 0.2959986925125122,
+      "learning_rate": 0.00039202432907492566,
+      "loss": 0.4983074963092804,
+      "step": 75
+    },
+    {
+      "epoch": 0.5487364620938628,
+      "grad_norm": 0.32182440161705017,
+      "learning_rate": 0.0003917930805470483,
+      "loss": 0.44362854957580566,
+      "step": 76
+    },
+    {
+      "epoch": 0.555956678700361,
+      "grad_norm": 0.2816471755504608,
+      "learning_rate": 0.00039155859750394754,
+      "loss": 0.44515836238861084,
+      "step": 77
+    },
+    {
+      "epoch": 0.5631768953068592,
+      "grad_norm": 0.290783554315567,
+      "learning_rate": 0.0003913208839000882,
+      "loss": 0.46542418003082275,
+      "step": 78
+    },
+    {
+      "epoch": 0.5703971119133574,
+      "grad_norm": 0.366773396730423,
+      "learning_rate": 0.0003910799437444171,
+      "loss": 0.5370551943778992,
+      "step": 79
+    },
+    {
+      "epoch": 0.5776173285198556,
+      "grad_norm": 0.32066911458969116,
+      "learning_rate": 0.0003908357811002957,
+      "loss": 0.5146384239196777,
+      "step": 80
+    },
+    {
+      "epoch": 0.5848375451263538,
+      "grad_norm": 0.3420974016189575,
+      "learning_rate": 0.00039058840008543135,
+      "loss": 0.6218519806861877,
+      "step": 81
+    },
+    {
+      "epoch": 0.592057761732852,
+      "grad_norm": 0.3156997263431549,
+      "learning_rate": 0.000390337804871808,
+      "loss": 0.549630343914032,
+      "step": 82
+    },
+    {
+      "epoch": 0.5992779783393501,
+      "grad_norm": 0.3494798541069031,
+      "learning_rate": 0.00039008399968561597,
+      "loss": 0.5499352216720581,
+      "step": 83
+    },
+    {
+      "epoch": 0.6064981949458483,
+      "grad_norm": 0.3652164936065674,
+      "learning_rate": 0.0003898269888071803,
+      "loss": 0.5562577843666077,
+      "step": 84
+    },
+    {
+      "epoch": 0.6137184115523465,
+      "grad_norm": 0.34751319885253906,
+      "learning_rate": 0.0003895667765708887,
+      "loss": 0.5954532027244568,
+      "step": 85
+    },
+    {
+      "epoch": 0.6209386281588448,
+      "grad_norm": 0.28338274359703064,
+      "learning_rate": 0.0003893033673651189,
+      "loss": 0.4894581735134125,
+      "step": 86
+    },
+    {
+      "epoch": 0.628158844765343,
+      "grad_norm": 0.29019951820373535,
+      "learning_rate": 0.00038903676563216394,
+      "loss": 0.4635826349258423,
+      "step": 87
+    },
+    {
+      "epoch": 0.6353790613718412,
+      "grad_norm": 0.2782176434993744,
+      "learning_rate": 0.00038876697586815763,
+      "loss": 0.4121658205986023,
+      "step": 88
+    },
+    {
+      "epoch": 0.6425992779783394,
+      "grad_norm": 0.27983078360557556,
+      "learning_rate": 0.0003884940026229988,
+      "loss": 0.40188780426979065,
+      "step": 89
+    },
+    {
+      "epoch": 0.6498194945848376,
+      "grad_norm": 0.2786664366722107,
+      "learning_rate": 0.0003882178505002744,
+      "loss": 0.44132208824157715,
+      "step": 90
+    },
+    {
+      "epoch": 0.6570397111913358,
+      "grad_norm": 0.3355483412742615,
+      "learning_rate": 0.0003879385241571817,
+      "loss": 0.512315034866333,
+      "step": 91
+    },
+    {
+      "epoch": 0.6642599277978339,
+      "grad_norm": 0.29554808139801025,
+      "learning_rate": 0.0003876560283044503,
+      "loss": 0.4540112018585205,
+      "step": 92
+    },
+    {
+      "epoch": 0.6714801444043321,
+      "grad_norm": 0.3431720435619354,
+      "learning_rate": 0.00038737036770626214,
+      "loss": 0.45164141058921814,
+      "step": 93
+    },
+    {
+      "epoch": 0.6787003610108303,
+      "grad_norm": 0.37090009450912476,
+      "learning_rate": 0.0003870815471801714,
+      "loss": 0.4878736734390259,
+      "step": 94
+    },
+    {
+      "epoch": 0.6859205776173285,
+      "grad_norm": 0.35553082823753357,
+      "learning_rate": 0.0003867895715970231,
+      "loss": 0.5573399662971497,
+      "step": 95
+    },
+    {
+      "epoch": 0.6931407942238267,
+      "grad_norm": 0.30914071202278137,
+      "learning_rate": 0.00038649444588087115,
+      "loss": 0.5417919754981995,
+      "step": 96
+    },
+    {
+      "epoch": 0.7003610108303249,
+      "grad_norm": 0.27384090423583984,
+      "learning_rate": 0.0003861961750088954,
+      "loss": 0.4475242793560028,
+      "step": 97
+    },
+    {
+      "epoch": 0.7075812274368231,
+      "grad_norm": 0.2919006049633026,
+      "learning_rate": 0.0003858947640113171,
+      "loss": 0.44341611862182617,
+      "step": 98
+    },
+    {
+      "epoch": 0.7148014440433214,
+      "grad_norm": 0.33408623933792114,
+      "learning_rate": 0.000385590217971315,
+      "loss": 0.5056685209274292,
+      "step": 99
+    },
+    {
+      "epoch": 0.7220216606498195,
+      "grad_norm": 0.3498390018939972,
+      "learning_rate": 0.0003852825420249386,
+      "loss": 0.45975226163864136,
+      "step": 100
+    },
+    {
+      "epoch": 0.7292418772563177,
+      "grad_norm": 0.32381942868232727,
+      "learning_rate": 0.0003849717413610224,
+      "loss": 0.43437302112579346,
+      "step": 101
+    },
+    {
+      "epoch": 0.7364620938628159,
+      "grad_norm": 0.30022308230400085,
+      "learning_rate": 0.0003846578212210979,
+      "loss": 0.41799411177635193,
+      "step": 102
+    },
+    {
+      "epoch": 0.7436823104693141,
+      "grad_norm": 0.31061094999313354,
+      "learning_rate": 0.0003843407868993054,
+      "loss": 0.5682204365730286,
+      "step": 103
+    },
+    {
+      "epoch": 0.7509025270758123,
+      "grad_norm": 0.31062883138656616,
+      "learning_rate": 0.0003840206437423047,
+      "loss": 0.3520132303237915,
+      "step": 104
+    },
+    {
+      "epoch": 0.7581227436823105,
+      "grad_norm": 0.36826908588409424,
+      "learning_rate": 0.00038369739714918464,
+      "loss": 0.546819269657135,
+      "step": 105
+    },
+    {
+      "epoch": 0.7653429602888087,
+      "grad_norm": 0.35304203629493713,
+      "learning_rate": 0.00038337105257137263,
+      "loss": 0.5633726716041565,
+      "step": 106
+    },
+    {
+      "epoch": 0.7725631768953068,
+      "grad_norm": 0.2879215478897095,
+      "learning_rate": 0.00038304161551254207,
+      "loss": 0.42212507128715515,
+      "step": 107
+    },
+    {
+      "epoch": 0.779783393501805,
+      "grad_norm": 0.32802048325538635,
+      "learning_rate": 0.0003827090915285202,
+      "loss": 0.4973101019859314,
+      "step": 108
+    },
+    {
+      "epoch": 0.7870036101083032,
+      "grad_norm": 0.29462599754333496,
+      "learning_rate": 0.0003823734862271937,
+      "loss": 0.4543662667274475,
+      "step": 109
+    },
+    {
+      "epoch": 0.7942238267148014,
+      "grad_norm": 0.34288832545280457,
+      "learning_rate": 0.0003820348052684146,
+      "loss": 0.5599743127822876,
+      "step": 110
+    },
+    {
+      "epoch": 0.8014440433212996,
+      "grad_norm": 0.2871699333190918,
+      "learning_rate": 0.00038169305436390477,
+      "loss": 0.5132560133934021,
+      "step": 111
+    },
+    {
+      "epoch": 0.8086642599277978,
+      "grad_norm": 0.3307580053806305,
+      "learning_rate": 0.00038134823927715933,
+      "loss": 0.5319223403930664,
+      "step": 112
+    },
+    {
+      "epoch": 0.8158844765342961,
+      "grad_norm": 0.3292873799800873,
+      "learning_rate": 0.0003810003658233498,
+      "loss": 0.5933513045310974,
+      "step": 113
+    },
+    {
+      "epoch": 0.8231046931407943,
+      "grad_norm": 0.34675127267837524,
+      "learning_rate": 0.0003806494398692258,
+      "loss": 0.6370081901550293,
+      "step": 114
+    },
+    {
+      "epoch": 0.8303249097472925,
+      "grad_norm": 0.31837624311447144,
+      "learning_rate": 0.00038029546733301615,
+      "loss": 0.44530773162841797,
+      "step": 115
+    },
+    {
+      "epoch": 0.8375451263537906,
+      "grad_norm": 0.3014152944087982,
+      "learning_rate": 0.0003799384541843291,
+      "loss": 0.5310161113739014,
+      "step": 116
+    },
+    {
+      "epoch": 0.8447653429602888,
+      "grad_norm": 0.30693644285202026,
+      "learning_rate": 0.00037957840644405164,
+      "loss": 0.5011177659034729,
+      "step": 117
+    },
+    {
+      "epoch": 0.851985559566787,
+      "grad_norm": 0.2840467095375061,
+      "learning_rate": 0.000379215330184248,
+      "loss": 0.37618446350097656,
+      "step": 118
+    },
+    {
+      "epoch": 0.8592057761732852,
+      "grad_norm": 0.3380180597305298,
+      "learning_rate": 0.00037884923152805715,
+      "loss": 0.5536396503448486,
+      "step": 119
+    },
+    {
+      "epoch": 0.8664259927797834,
+      "grad_norm": 0.30026137828826904,
+      "learning_rate": 0.00037848011664958956,
+      "loss": 0.5487097501754761,
+      "step": 120
+    },
+    {
+      "epoch": 0.8736462093862816,
+      "grad_norm": 0.3990669250488281,
+      "learning_rate": 0.00037810799177382325,
+      "loss": 0.34712734818458557,
+      "step": 121
+    },
+    {
+      "epoch": 0.8808664259927798,
+      "grad_norm": 0.3168427646160126,
+      "learning_rate": 0.0003777328631764986,
+      "loss": 0.44405245780944824,
+      "step": 122
+    },
+    {
+      "epoch": 0.8880866425992779,
+      "grad_norm": 0.2761262059211731,
+      "learning_rate": 0.0003773547371840124,
+      "loss": 0.4112606942653656,
+      "step": 123
+    },
+    {
+      "epoch": 0.8953068592057761,
+      "grad_norm": 0.27056002616882324,
+      "learning_rate": 0.00037697362017331164,
+      "loss": 0.45218032598495483,
+      "step": 124
+    },
+    {
+      "epoch": 0.9025270758122743,
+      "grad_norm": 0.2833767831325531,
+      "learning_rate": 0.00037658951857178543,
+      "loss": 0.4833817481994629,
+      "step": 125
+    },
+    {
+      "epoch": 0.9097472924187726,
+      "grad_norm": 0.30818742513656616,
+      "learning_rate": 0.0003762024388571569,
+      "loss": 0.4847618341445923,
+      "step": 126
+    },
+    {
+      "epoch": 0.9169675090252708,
+      "grad_norm": 0.32447490096092224,
+      "learning_rate": 0.00037581238755737405,
+      "loss": 0.5068320035934448,
+      "step": 127
+    },
+    {
+      "epoch": 0.924187725631769,
+      "grad_norm": 0.307781845331192,
+      "learning_rate": 0.0003754193712504992,
+      "loss": 0.455619752407074,
+      "step": 128
+    },
+    {
+      "epoch": 0.9314079422382672,
+      "grad_norm": 0.29909971356391907,
+      "learning_rate": 0.00037502339656459853,
+      "loss": 0.5618020296096802,
+      "step": 129
+    },
+    {
+      "epoch": 0.9386281588447654,
+      "grad_norm": 0.33699020743370056,
+      "learning_rate": 0.0003746244701776302,
+      "loss": 0.50435870885849,
+      "step": 130
+    },
+    {
+      "epoch": 0.9458483754512635,
+      "grad_norm": 0.2932925820350647,
+      "learning_rate": 0.0003742225988173315,
+      "loss": 0.503471314907074,
+      "step": 131
+    },
+    {
+      "epoch": 0.9530685920577617,
+      "grad_norm": 0.27647075057029724,
+      "learning_rate": 0.00037381778926110567,
+      "loss": 0.4979417324066162,
+      "step": 132
+    },
+    {
+      "epoch": 0.9602888086642599,
+      "grad_norm": 0.23902934789657593,
+      "learning_rate": 0.00037341004833590734,
+      "loss": 0.4127359390258789,
+      "step": 133
+    },
+    {
+      "epoch": 0.9675090252707581,
+      "grad_norm": 0.3308359980583191,
+      "learning_rate": 0.0003729993829181276,
+      "loss": 0.6199836730957031,
+      "step": 134
+    },
+    {
+      "epoch": 0.9747292418772563,
+      "grad_norm": 0.30997487902641296,
+      "learning_rate": 0.00037258579993347795,
+      "loss": 0.5043972134590149,
+      "step": 135
+    },
+    {
+      "epoch": 0.9819494584837545,
+      "grad_norm": 0.5225651264190674,
+      "learning_rate": 0.00037216930635687357,
+      "loss": 0.5078832507133484,
+      "step": 136
+    },
+    {
+      "epoch": 0.9891696750902527,
+      "grad_norm": 0.2910468280315399,
+      "learning_rate": 0.00037174990921231557,
+      "loss": 0.5478032827377319,
+      "step": 137
+    },
+    {
+      "epoch": 0.9963898916967509,
+      "grad_norm": 0.3448226749897003,
+      "learning_rate": 0.00037132761557277257,
+      "loss": 0.5461761355400085,
+      "step": 138
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5731388926506042,
+      "learning_rate": 0.00037090243256006155,
+      "loss": 0.39478886127471924,
+      "step": 139
+    },
+    {
+      "epoch": 1.0072202166064983,
+      "grad_norm": 0.3017331659793854,
+      "learning_rate": 0.0003704743673447276,
+      "loss": 0.3729703426361084,
+      "step": 140
+    },
+    {
+      "epoch": 1.0144404332129964,
+      "grad_norm": 0.29272574186325073,
+      "learning_rate": 0.0003700434271459229,
+      "loss": 0.45498228073120117,
+      "step": 141
+    },
+    {
+      "epoch": 1.0216606498194947,
+      "grad_norm": 0.2818097770214081,
+      "learning_rate": 0.0003696096192312852,
+      "loss": 0.37411680817604065,
+      "step": 142
+    },
+    {
+      "epoch": 1.0288808664259927,
+      "grad_norm": 0.31325557827949524,
+      "learning_rate": 0.00036917295091681526,
+      "loss": 0.4299224019050598,
+      "step": 143
+    },
+    {
+      "epoch": 1.036101083032491,
+      "grad_norm": 0.2971343994140625,
+      "learning_rate": 0.0003687334295667533,
+      "loss": 0.4103966951370239,
+      "step": 144
+    },
+    {
+      "epoch": 1.0433212996389891,
+      "grad_norm": 0.32004889845848083,
+      "learning_rate": 0.00036829106259345465,
+      "loss": 0.49494248628616333,
+      "step": 145
+    },
+    {
+      "epoch": 1.0505415162454874,
+      "grad_norm": 0.29688122868537903,
+      "learning_rate": 0.00036784585745726535,
+      "loss": 0.4550933241844177,
+      "step": 146
+    },
+    {
+      "epoch": 1.0577617328519855,
+      "grad_norm": 0.3322184681892395,
+      "learning_rate": 0.00036739782166639556,
+      "loss": 0.46367114782333374,
+      "step": 147
+    },
+    {
+      "epoch": 1.0649819494584838,
+      "grad_norm": 0.267844021320343,
+      "learning_rate": 0.00036694696277679367,
+      "loss": 0.33760106563568115,
+      "step": 148
+    },
+    {
+      "epoch": 1.0722021660649819,
+      "grad_norm": 0.3240787386894226,
+      "learning_rate": 0.0003664932883920182,
+      "loss": 0.43709027767181396,
+      "step": 149
+    },
+    {
+      "epoch": 1.0794223826714802,
+      "grad_norm": 0.29441356658935547,
+      "learning_rate": 0.00036603680616311015,
+      "loss": 0.4205693304538727,
+      "step": 150
+    },
+    {
+      "epoch": 1.0866425992779782,
+      "grad_norm": 0.33195292949676514,
+      "learning_rate": 0.00036557752378846357,
+      "loss": 0.44472751021385193,
+      "step": 151
+    },
+    {
+      "epoch": 1.0938628158844765,
+      "grad_norm": 0.3113086223602295,
+      "learning_rate": 0.00036511544901369597,
+      "loss": 0.3894999921321869,
+      "step": 152
+    },
+    {
+      "epoch": 1.1010830324909748,
+      "grad_norm": 0.30926311016082764,
+      "learning_rate": 0.00036465058963151744,
+      "loss": 0.33893927931785583,
+      "step": 153
+    },
+    {
+      "epoch": 1.108303249097473,
+      "grad_norm": 0.31725767254829407,
+      "learning_rate": 0.00036418295348159967,
+      "loss": 0.4384872019290924,
+      "step": 154
+    },
+    {
+      "epoch": 1.1155234657039712,
+      "grad_norm": 0.3411625027656555,
+      "learning_rate": 0.00036371254845044297,
+      "loss": 0.4130849838256836,
+      "step": 155
+    },
+    {
+      "epoch": 1.1227436823104693,
+      "grad_norm": 0.3470022976398468,
+      "learning_rate": 0.0003632393824712444,
+      "loss": 0.4436883330345154,
+      "step": 156
+    },
+    {
+      "epoch": 1.1299638989169676,
+      "grad_norm": 0.31315702199935913,
+      "learning_rate": 0.00036276346352376266,
+      "loss": 0.39922523498535156,
+      "step": 157
+    },
+    {
+      "epoch": 1.1371841155234657,
+      "grad_norm": 0.36140701174736023,
+      "learning_rate": 0.00036228479963418465,
+      "loss": 0.44266432523727417,
+      "step": 158
+    },
+    {
+      "epoch": 1.144404332129964,
+      "grad_norm": 0.37621912360191345,
+      "learning_rate": 0.0003618033988749895,
+      "loss": 0.3703776001930237,
+      "step": 159
+    },
+    {
+      "epoch": 1.151624548736462,
+      "grad_norm": 0.40457242727279663,
+      "learning_rate": 0.0003613192693648125,
+      "loss": 0.48410341143608093,
+      "step": 160
+    },
+    {
+      "epoch": 1.1588447653429603,
+      "grad_norm": 0.35907864570617676,
+      "learning_rate": 0.00036083241926830833,
+      "loss": 0.3249608874320984,
+      "step": 161
+    },
+    {
+      "epoch": 1.1660649819494584,
+      "grad_norm": 0.3328985869884491,
+      "learning_rate": 0.00036034285679601336,
+      "loss": 0.3642440140247345,
+      "step": 162
+    },
+    {
+      "epoch": 1.1732851985559567,
+      "grad_norm": 0.3639651834964752,
+      "learning_rate": 0.00035985059020420695,
+      "loss": 0.4125223457813263,
+      "step": 163
+    },
+    {
+      "epoch": 1.1805054151624548,
+      "grad_norm": 0.3122009038925171,
+      "learning_rate": 0.0003593556277947725,
+      "loss": 0.26333218812942505,
+      "step": 164
+    },
+    {
+      "epoch": 1.187725631768953,
+      "grad_norm": 0.3917737901210785,
+      "learning_rate": 0.0003588579779150572,
+      "loss": 0.41683149337768555,
+      "step": 165
+    },
+    {
+      "epoch": 1.1949458483754514,
+      "grad_norm": 0.3892008066177368,
+      "learning_rate": 0.00035835764895773164,
+      "loss": 0.34965166449546814,
+      "step": 166
+    },
+    {
+      "epoch": 1.2021660649819494,
+      "grad_norm": 0.32206663489341736,
+      "learning_rate": 0.0003578546493606477,
+      "loss": 0.3289134204387665,
+      "step": 167
+    },
+    {
+      "epoch": 1.2093862815884477,
+      "grad_norm": 0.373098760843277,
+      "learning_rate": 0.0003573489876066967,
+      "loss": 0.37044239044189453,
+      "step": 168
+    },
+    {
+      "epoch": 1.2166064981949458,
+      "grad_norm": 0.4301343262195587,
+      "learning_rate": 0.00035684067222366614,
+      "loss": 0.46339157223701477,
+      "step": 169
+    },
+    {
+      "epoch": 1.2238267148014441,
+      "grad_norm": 0.3458724915981293,
+      "learning_rate": 0.00035632971178409603,
+      "loss": 0.41736456751823425,
+      "step": 170
+    },
+    {
+      "epoch": 1.2310469314079422,
+      "grad_norm": 0.45238223671913147,
+      "learning_rate": 0.0003558161149051341,
+      "loss": 0.4182761311531067,
+      "step": 171
+    },
+    {
+      "epoch": 1.2382671480144405,
+      "grad_norm": 0.32411810755729675,
+      "learning_rate": 0.0003552998902483907,
+      "loss": 0.30536049604415894,
+      "step": 172
+    },
+    {
+      "epoch": 1.2454873646209386,
+      "grad_norm": 0.40648043155670166,
+      "learning_rate": 0.0003547810465197926,
+      "loss": 0.45940107107162476,
+      "step": 173
+    },
+    {
+      "epoch": 1.2527075812274369,
+      "grad_norm": 0.34474775195121765,
+      "learning_rate": 0.00035425959246943614,
+      "loss": 0.4267578423023224,
+      "step": 174
+    },
+    {
+      "epoch": 1.259927797833935,
+      "grad_norm": 0.33473461866378784,
+      "learning_rate": 0.00035373553689143977,
+      "loss": 0.31750980019569397,
+      "step": 175
+    },
+    {
+      "epoch": 1.2671480144404332,
+      "grad_norm": 0.29968079924583435,
+      "learning_rate": 0.0003532088886237956,
+      "loss": 0.3412264883518219,
+      "step": 176
+    },
+    {
+      "epoch": 1.2743682310469313,
+      "grad_norm": 0.3362952172756195,
+      "learning_rate": 0.00035267965654822063,
+      "loss": 0.347223162651062,
+      "step": 177
+    },
+    {
+      "epoch": 1.2815884476534296,
+      "grad_norm": 0.38441580533981323,
+      "learning_rate": 0.0003521478495900065,
+      "loss": 0.4132072925567627,
+      "step": 178
+    },
+    {
+      "epoch": 1.288808664259928,
+      "grad_norm": 0.3259715437889099,
+      "learning_rate": 0.00035161347671786947,
+      "loss": 0.3259287476539612,
+      "step": 179
+    },
+    {
+      "epoch": 1.296028880866426,
+      "grad_norm": 0.30564218759536743,
+      "learning_rate": 0.00035107654694379876,
+      "loss": 0.33284130692481995,
+      "step": 180
+    },
+    {
+      "epoch": 1.303249097472924,
+      "grad_norm": 0.3464779853820801,
+      "learning_rate": 0.0003505370693229049,
+      "loss": 0.3613452613353729,
+      "step": 181
+    },
+    {
+      "epoch": 1.3104693140794224,
+      "grad_norm": 0.4113422632217407,
+      "learning_rate": 0.0003499950529532668,
+      "loss": 0.49952858686447144,
+      "step": 182
+    },
+    {
+      "epoch": 1.3176895306859207,
+      "grad_norm": 0.3503414988517761,
+      "learning_rate": 0.0003494505069757782,
+      "loss": 0.43928611278533936,
+      "step": 183
+    },
+    {
+      "epoch": 1.3249097472924187,
+      "grad_norm": 0.3231402337551117,
+      "learning_rate": 0.000348903440573994,
+      "loss": 0.35302072763442993,
+      "step": 184
+    },
+    {
+      "epoch": 1.332129963898917,
+      "grad_norm": 0.3849830627441406,
+      "learning_rate": 0.00034835386297397486,
+      "loss": 0.42382097244262695,
+      "step": 185
+    },
+    {
+      "epoch": 1.339350180505415,
+      "grad_norm": 0.35116714239120483,
+      "learning_rate": 0.00034780178344413185,
+      "loss": 0.3274853229522705,
+      "step": 186
+    },
+    {
+      "epoch": 1.3465703971119134,
+      "grad_norm": 0.3941578269004822,
+      "learning_rate": 0.0003472472112950701,
+      "loss": 0.34617680311203003,
+      "step": 187
+    },
+    {
+      "epoch": 1.3537906137184115,
+      "grad_norm": 0.32104700803756714,
+      "learning_rate": 0.00034669015587943184,
+      "loss": 0.39410364627838135,
+      "step": 188
+    },
+    {
+      "epoch": 1.3610108303249098,
+      "grad_norm": 0.42431211471557617,
+      "learning_rate": 0.00034613062659173867,
+      "loss": 0.49896761775016785,
+      "step": 189
+    },
+    {
+      "epoch": 1.3682310469314078,
+      "grad_norm": 0.4387979209423065,
+      "learning_rate": 0.00034556863286823284,
+      "loss": 0.4435005784034729,
+      "step": 190
+    },
+    {
+      "epoch": 1.3754512635379061,
+      "grad_norm": 0.3983216881752014,
+      "learning_rate": 0.00034500418418671855,
+      "loss": 0.44415876269340515,
+      "step": 191
+    },
+    {
+      "epoch": 1.3826714801444044,
+      "grad_norm": 0.30740320682525635,
+      "learning_rate": 0.00034443729006640184,
+      "loss": 0.36664414405822754,
+      "step": 192
+    },
+    {
+      "epoch": 1.3898916967509025,
+      "grad_norm": 0.35397860407829285,
+      "learning_rate": 0.0003438679600677302,
+      "loss": 0.36680418252944946,
+      "step": 193
+    },
+    {
+      "epoch": 1.3971119133574006,
+      "grad_norm": 0.3522591292858124,
+      "learning_rate": 0.0003432962037922312,
+      "loss": 0.3908054232597351,
+      "step": 194
+    },
+    {
+      "epoch": 1.404332129963899,
+      "grad_norm": 0.36350661516189575,
+      "learning_rate": 0.0003427220308823505,
+      "loss": 0.34363672137260437,
+      "step": 195
+    },
+    {
+      "epoch": 1.4115523465703972,
+      "grad_norm": 0.3392064869403839,
+      "learning_rate": 0.00034214545102128946,
+      "loss": 0.40172988176345825,
+      "step": 196
+    },
+    {
+      "epoch": 1.4187725631768953,
+      "grad_norm": 0.3826808035373688,
+      "learning_rate": 0.0003415664739328418,
+      "loss": 0.4292600452899933,
+      "step": 197
+    },
+    {
+      "epoch": 1.4259927797833936,
+      "grad_norm": 0.4102179706096649,
+      "learning_rate": 0.00034098510938122947,
+      "loss": 0.40238481760025024,
+      "step": 198
+    },
+    {
+      "epoch": 1.4332129963898916,
+      "grad_norm": 0.38712596893310547,
+      "learning_rate": 0.00034040136717093796,
+      "loss": 0.3980088233947754,
+      "step": 199
+    },
+    {
+      "epoch": 1.44043321299639,
+      "grad_norm": 0.35182803869247437,
+      "learning_rate": 0.0003398152571465512,
+      "loss": 0.3569096028804779,
+      "step": 200
+    },
+    {
+      "epoch": 1.447653429602888,
+      "grad_norm": 0.4129750728607178,
+      "learning_rate": 0.0003392267891925854,
+      "loss": 0.4729324281215668,
+      "step": 201
+    },
+    {
+      "epoch": 1.4548736462093863,
+      "grad_norm": 0.3427615463733673,
+      "learning_rate": 0.000338635973233322,
+      "loss": 0.38466876745224,
+      "step": 202
+    },
+    {
+      "epoch": 1.4620938628158844,
+      "grad_norm": 0.29937732219696045,
+      "learning_rate": 0.0003380428192326411,
+      "loss": 0.24431845545768738,
+      "step": 203
+    },
+    {
+      "epoch": 1.4693140794223827,
+      "grad_norm": 0.35563215613365173,
+      "learning_rate": 0.00033744733719385253,
+      "loss": 0.34825772047042847,
+      "step": 204
+    },
+    {
+      "epoch": 1.476534296028881,
+      "grad_norm": 0.3395077586174011,
+      "learning_rate": 0.0003368495371595279,
+      "loss": 0.37704023718833923,
+      "step": 205
+    },
+    {
+      "epoch": 1.483754512635379,
+      "grad_norm": 0.32358789443969727,
+      "learning_rate": 0.0003362494292113308,
+      "loss": 0.29866379499435425,
+      "step": 206
+    },
+    {
+      "epoch": 1.4909747292418771,
+      "grad_norm": 0.44219517707824707,
+      "learning_rate": 0.00033564702346984684,
+      "loss": 0.5036117434501648,
+      "step": 207
+    },
+    {
+      "epoch": 1.4981949458483754,
+      "grad_norm": 0.358549565076828,
+      "learning_rate": 0.00033504233009441287,
+      "loss": 0.42227745056152344,
+      "step": 208
+    },
+    {
+      "epoch": 1.5054151624548737,
+      "grad_norm": 0.3784799575805664,
+      "learning_rate": 0.0003344353592829461,
+      "loss": 0.4267178773880005,
+      "step": 209
+    },
+    {
+      "epoch": 1.5126353790613718,
+      "grad_norm": 0.3619599938392639,
+      "learning_rate": 0.00033382612127177166,
+      "loss": 0.4075395166873932,
+      "step": 210
+    },
+    {
+      "epoch": 1.5198555956678699,
+      "grad_norm": 0.44484809041023254,
+      "learning_rate": 0.0003332146263354501,
+      "loss": 0.5016494989395142,
+      "step": 211
+    },
+    {
+      "epoch": 1.5270758122743682,
+      "grad_norm": 0.34379443526268005,
+      "learning_rate": 0.00033260088478660407,
+      "loss": 0.317508339881897,
+      "step": 212
+    },
+    {
+      "epoch": 1.5342960288808665,
+      "grad_norm": 0.4008065164089203,
+      "learning_rate": 0.0003319849069757446,
+      "loss": 0.3974913954734802,
+      "step": 213
+    },
+    {
+      "epoch": 1.5415162454873648,
+      "grad_norm": 0.4006946384906769,
+      "learning_rate": 0.00033136670329109624,
+      "loss": 0.5162532925605774,
+      "step": 214
+    },
+    {
+      "epoch": 1.5487364620938628,
+      "grad_norm": 0.32255908846855164,
+      "learning_rate": 0.0003307462841584223,
+      "loss": 0.37669965624809265,
+      "step": 215
+    },
+    {
+      "epoch": 1.555956678700361,
+      "grad_norm": 0.7169090509414673,
+      "learning_rate": 0.0003301236600408484,
+      "loss": 0.4477896988391876,
+      "step": 216
+    },
+    {
+      "epoch": 1.5631768953068592,
+      "grad_norm": 0.3694959580898285,
+      "learning_rate": 0.00032949884143868675,
+      "loss": 0.45744967460632324,
+      "step": 217
+    },
+    {
+      "epoch": 1.5703971119133575,
+      "grad_norm": 0.3709769546985626,
+      "learning_rate": 0.0003288718388892583,
+      "loss": 0.4952532947063446,
+      "step": 218
+    },
+    {
+      "epoch": 1.5776173285198556,
+      "grad_norm": 0.29159069061279297,
+      "learning_rate": 0.00032824266296671567,
+      "loss": 0.28994375467300415,
+      "step": 219
+    },
+    {
+      "epoch": 1.5848375451263537,
+      "grad_norm": 0.3600374460220337,
+      "learning_rate": 0.0003276113242818645,
+      "loss": 0.43825557827949524,
+      "step": 220
+    },
+    {
+      "epoch": 1.592057761732852,
+      "grad_norm": 0.2843645215034485,
+      "learning_rate": 0.0003269778334819846,
+      "loss": 0.32124435901641846,
+      "step": 221
+    },
+    {
+      "epoch": 1.5992779783393503,
+      "grad_norm": 0.3688318431377411,
+      "learning_rate": 0.0003263422012506502,
+      "loss": 0.4742942750453949,
+      "step": 222
+    },
+    {
+      "epoch": 1.6064981949458483,
+      "grad_norm": 0.3213002383708954,
+      "learning_rate": 0.00032570443830755015,
+      "loss": 0.42100387811660767,
+      "step": 223
+    },
+    {
+      "epoch": 1.6137184115523464,
+      "grad_norm": 0.3240319490432739,
+      "learning_rate": 0.0003250645554083068,
+      "loss": 0.29403799772262573,
+      "step": 224
+    },
+    {
+      "epoch": 1.6209386281588447,
+      "grad_norm": 0.3569975197315216,
+      "learning_rate": 0.0003244225633442948,
+      "loss": 0.4555712938308716,
+      "step": 225
+    },
+    {
+      "epoch": 1.628158844765343,
+      "grad_norm": 0.43396008014678955,
+      "learning_rate": 0.00032377847294245895,
+      "loss": 0.4877527356147766,
+      "step": 226
+    },
+    {
+      "epoch": 1.6353790613718413,
+      "grad_norm": 0.3694853186607361,
+      "learning_rate": 0.0003231322950651316,
+      "loss": 0.41299888491630554,
+      "step": 227
+    },
+    {
+      "epoch": 1.6425992779783394,
+      "grad_norm": 0.3358525335788727,
+      "learning_rate": 0.00032248404060985,
+      "loss": 0.4551587998867035,
+      "step": 228
+    },
+    {
+      "epoch": 1.6498194945848375,
+      "grad_norm": 0.38492903113365173,
+      "learning_rate": 0.0003218337205091715,
+      "loss": 0.4316423535346985,
+      "step": 229
+    },
+    {
+      "epoch": 1.6570397111913358,
+      "grad_norm": 0.34664952754974365,
+      "learning_rate": 0.0003211813457304902,
+      "loss": 0.39314523339271545,
+      "step": 230
+    },
+    {
+      "epoch": 1.664259927797834,
+      "grad_norm": 0.3254722058773041,
+      "learning_rate": 0.00032052692727585133,
+      "loss": 0.3327226936817169,
+      "step": 231
+    },
+    {
+      "epoch": 1.6714801444043321,
+      "grad_norm": 0.4386709928512573,
+      "learning_rate": 0.0003198704761817658,
+      "loss": 0.4558185935020447,
+      "step": 232
+    },
+    {
+      "epoch": 1.6787003610108302,
+      "grad_norm": 0.5447010397911072,
+      "learning_rate": 0.00031921200351902446,
+      "loss": 0.48867422342300415,
+      "step": 233
+    },
+    {
+      "epoch": 1.6859205776173285,
+      "grad_norm": 0.2976516783237457,
+      "learning_rate": 0.000318551520392511,
+      "loss": 0.3376544117927551,
+      "step": 234
+    },
+    {
+      "epoch": 1.6931407942238268,
+      "grad_norm": 0.3854595124721527,
+      "learning_rate": 0.00031788903794101477,
+      "loss": 0.45083147287368774,
+      "step": 235
+    },
+    {
+      "epoch": 1.7003610108303249,
+      "grad_norm": 0.37051668763160706,
+      "learning_rate": 0.00031722456733704297,
+      "loss": 0.4483460485935211,
+      "step": 236
+    },
+    {
+      "epoch": 1.707581227436823,
+      "grad_norm": 0.42628687620162964,
+      "learning_rate": 0.0003165581197866322,
+      "loss": 0.5173429250717163,
+      "step": 237
+    },
+    {
+      "epoch": 1.7148014440433212,
+      "grad_norm": 0.44184717535972595,
+      "learning_rate": 0.0003158897065291596,
+      "loss": 0.45995786786079407,
+      "step": 238
+    },
+    {
+      "epoch": 1.7220216606498195,
+      "grad_norm": 0.3687286376953125,
+      "learning_rate": 0.00031521933883715293,
+      "loss": 0.33321040868759155,
+      "step": 239
+    },
+    {
+      "epoch": 1.7292418772563178,
+      "grad_norm": 0.3514050841331482,
+      "learning_rate": 0.00031454702801610103,
+      "loss": 0.3461155891418457,
+      "step": 240
+    },
+    {
+      "epoch": 1.736462093862816,
+      "grad_norm": 0.4148181676864624,
+      "learning_rate": 0.0003138727854042627,
+      "loss": 0.4323638677597046,
+      "step": 241
+    },
+    {
+      "epoch": 1.743682310469314,
+      "grad_norm": 0.44285446405410767,
+      "learning_rate": 0.0003131966223724756,
+      "loss": 0.4780181646347046,
+      "step": 242
+    },
+    {
+      "epoch": 1.7509025270758123,
+      "grad_norm": 0.3196166157722473,
+      "learning_rate": 0.00031251855032396464,
+      "loss": 0.3380797505378723,
+      "step": 243
+    },
+    {
+      "epoch": 1.7581227436823106,
+      "grad_norm": 0.48847395181655884,
+      "learning_rate": 0.0003118385806941494,
+      "loss": 0.4878506064414978,
+      "step": 244
+    },
+    {
+      "epoch": 1.7653429602888087,
+      "grad_norm": 0.398876816034317,
+      "learning_rate": 0.0003111567249504515,
+      "loss": 0.4532281458377838,
+      "step": 245
+    },
+    {
+      "epoch": 1.7725631768953067,
+      "grad_norm": 0.34218019247055054,
+      "learning_rate": 0.0003104729945921012,
+      "loss": 0.3738666772842407,
+      "step": 246
+    },
+    {
+      "epoch": 1.779783393501805,
+      "grad_norm": 0.4071042835712433,
+      "learning_rate": 0.0003097874011499433,
+      "loss": 0.4437788724899292,
+      "step": 247
+    },
+    {
+      "epoch": 1.7870036101083033,
+      "grad_norm": 0.35893604159355164,
+      "learning_rate": 0.0003090999561862428,
+      "loss": 0.35174691677093506,
+      "step": 248
+    },
+    {
+      "epoch": 1.7942238267148014,
+      "grad_norm": 0.34495967626571655,
+      "learning_rate": 0.0003084106712944899,
+      "loss": 0.34072697162628174,
+      "step": 249
+    },
+    {
+      "epoch": 1.8014440433212995,
+      "grad_norm": 0.4653945565223694,
+      "learning_rate": 0.0003077195580992045,
+      "loss": 0.45023709535598755,
+      "step": 250
+    },
+    {
+      "epoch": 1.8086642599277978,
+      "grad_norm": 0.42124322056770325,
+      "learning_rate": 0.0003070266282557401,
+      "loss": 0.4846741557121277,
+      "step": 251
+    },
+    {
+      "epoch": 1.815884476534296,
+      "grad_norm": 0.38208404183387756,
+      "learning_rate": 0.0003063318934500872,
+      "loss": 0.3900390565395355,
+      "step": 252
+    },
+    {
+      "epoch": 1.8231046931407944,
+      "grad_norm": 0.3911641538143158,
+      "learning_rate": 0.0003056353653986764,
+      "loss": 0.3795917332172394,
+      "step": 253
+    },
+    {
+      "epoch": 1.8303249097472925,
+      "grad_norm": 0.34185168147087097,
+      "learning_rate": 0.00030493705584818065,
+      "loss": 0.39876672625541687,
+      "step": 254
+    },
+    {
+      "epoch": 1.8375451263537905,
+      "grad_norm": 0.4229717552661896,
+      "learning_rate": 0.000304236976575317,
+      "loss": 0.357430100440979,
+      "step": 255
+    },
+    {
+      "epoch": 1.8447653429602888,
+      "grad_norm": 0.35156992077827454,
+      "learning_rate": 0.0003035351393866485,
+      "loss": 0.37869274616241455,
+      "step": 256
+    },
+    {
+      "epoch": 1.8519855595667871,
+      "grad_norm": 0.3945099711418152,
+      "learning_rate": 0.0003028315561183845,
+      "loss": 0.45211949944496155,
+      "step": 257
+    },
+    {
+      "epoch": 1.8592057761732852,
+      "grad_norm": 0.31473544239997864,
+      "learning_rate": 0.0003021262386361814,
+      "loss": 0.32624444365501404,
+      "step": 258
+    },
+    {
+      "epoch": 1.8664259927797833,
+      "grad_norm": 0.4487520158290863,
+      "learning_rate": 0.00030141919883494247,
+      "loss": 0.48553428053855896,
+      "step": 259
+    },
+    {
+      "epoch": 1.8736462093862816,
+      "grad_norm": 0.3847423791885376,
+      "learning_rate": 0.00030071044863861713,
+      "loss": 0.457511842250824,
+      "step": 260
+    },
+    {
+      "epoch": 1.8808664259927799,
+      "grad_norm": 0.3822494447231293,
+      "learning_rate": 0.00030000000000000003,
+      "loss": 0.38700124621391296,
+      "step": 261
+    },
+    {
+      "epoch": 1.888086642599278,
+      "grad_norm": 0.37934961915016174,
+      "learning_rate": 0.0002992878649005293,
+      "loss": 0.4150720238685608,
+      "step": 262
+    },
+    {
+      "epoch": 1.895306859205776,
+      "grad_norm": 0.37126392126083374,
+      "learning_rate": 0.00029857405535008467,
+      "loss": 0.3389851450920105,
+      "step": 263
+    },
+    {
+      "epoch": 1.9025270758122743,
+      "grad_norm": 0.3264472484588623,
+      "learning_rate": 0.0002978585833867847,
+      "loss": 0.327554315328598,
+      "step": 264
+    },
+    {
+      "epoch": 1.9097472924187726,
+      "grad_norm": 0.3889877200126648,
+      "learning_rate": 0.00029714146107678413,
+      "loss": 0.4393918812274933,
+      "step": 265
+    },
+    {
+      "epoch": 1.916967509025271,
+      "grad_norm": 0.4926806092262268,
+      "learning_rate": 0.0002964227005140698,
+      "loss": 0.5468560457229614,
+      "step": 266
+    },
+    {
+      "epoch": 1.924187725631769,
+      "grad_norm": 0.4259951114654541,
+      "learning_rate": 0.0002957023138202573,
+      "loss": 0.39216533303260803,
+      "step": 267
+    },
+    {
+      "epoch": 1.931407942238267,
+      "grad_norm": 0.39791861176490784,
+      "learning_rate": 0.00029498031314438626,
+      "loss": 0.37104716897010803,
+      "step": 268
+    },
+    {
+      "epoch": 1.9386281588447654,
+      "grad_norm": 0.3682475686073303,
+      "learning_rate": 0.0002942567106627155,
+      "loss": 0.41986221075057983,
+      "step": 269
+    },
+    {
+      "epoch": 1.9458483754512637,
+      "grad_norm": 0.3230769634246826,
+      "learning_rate": 0.00029353151857851735,
+      "loss": 0.35548532009124756,
+      "step": 270
+    },
+    {
+      "epoch": 1.9530685920577617,
+      "grad_norm": 0.4607864320278168,
+      "learning_rate": 0.0002928047491218727,
+      "loss": 0.4246275722980499,
+      "step": 271
+    },
+    {
+      "epoch": 1.9602888086642598,
+      "grad_norm": 0.4302678108215332,
+      "learning_rate": 0.0002920764145494638,
+      "loss": 0.42147913575172424,
+      "step": 272
+    },
+    {
+      "epoch": 1.967509025270758,
+      "grad_norm": 0.36212071776390076,
+      "learning_rate": 0.00029134652714436814,
+      "loss": 0.4659682512283325,
+      "step": 273
+    },
+    {
+      "epoch": 1.9747292418772564,
+      "grad_norm": 0.31535375118255615,
+      "learning_rate": 0.00029061509921585107,
+      "loss": 0.38229790329933167,
+      "step": 274
+    },
+    {
+      "epoch": 1.9819494584837545,
+      "grad_norm": 0.3685518801212311,
+      "learning_rate": 0.0002898821430991582,
+      "loss": 0.46467846632003784,
+      "step": 275
+    },
+    {
+      "epoch": 1.9891696750902526,
+      "grad_norm": 0.3998563885688782,
+      "learning_rate": 0.0002891476711553077,
+      "loss": 0.42970407009124756,
+      "step": 276
+    },
+    {
+      "epoch": 1.9963898916967509,
+      "grad_norm": 0.35800614953041077,
+      "learning_rate": 0.00028841169577088115,
+      "loss": 0.4408795237541199,
+      "step": 277
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.5006654858589172,
+      "learning_rate": 0.0002876742293578155,
+      "loss": 0.38447684049606323,
+      "step": 278
+    },
+    {
+      "epoch": 2.0072202166064983,
+      "grad_norm": 0.31330448389053345,
+      "learning_rate": 0.00028693528435319305,
+      "loss": 0.30373793840408325,
+      "step": 279
+    },
+    {
+      "epoch": 2.0144404332129966,
+      "grad_norm": 0.29946643114089966,
+      "learning_rate": 0.0002861948732190319,
+      "loss": 0.26699960231781006,
+      "step": 280
+    },
+    {
+      "epoch": 2.0216606498194944,
+      "grad_norm": 0.26591944694519043,
+      "learning_rate": 0.0002854530084420762,
+      "loss": 0.23807109892368317,
+      "step": 281
+    },
+    {
+      "epoch": 2.0288808664259927,
+      "grad_norm": 0.34681931138038635,
+      "learning_rate": 0.00028470970253358487,
+      "loss": 0.32200539112091064,
+      "step": 282
+    },
+    {
+      "epoch": 2.036101083032491,
+      "grad_norm": 0.3542478382587433,
+      "learning_rate": 0.0002839649680291211,
+      "loss": 0.3349723815917969,
+      "step": 283
+    },
+    {
+      "epoch": 2.0433212996389893,
+      "grad_norm": 0.3007175624370575,
+      "learning_rate": 0.0002832188174883408,
+      "loss": 0.2172810435295105,
+      "step": 284
+    },
+    {
+      "epoch": 2.050541516245487,
+      "grad_norm": 0.381719708442688,
+      "learning_rate": 0.00028247126349478075,
+      "loss": 0.3058876693248749,
+      "step": 285
+    },
+    {
+      "epoch": 2.0577617328519855,
+      "grad_norm": 0.4056151211261749,
+      "learning_rate": 0.0002817223186556463,
+      "loss": 0.25402823090553284,
+      "step": 286
+    },
+    {
+      "epoch": 2.064981949458484,
+      "grad_norm": 0.5069922804832458,
+      "learning_rate": 0.00028097199560159913,
+      "loss": 0.2375117838382721,
+      "step": 287
+    },
+    {
+      "epoch": 2.072202166064982,
+      "grad_norm": 0.574663519859314,
+      "learning_rate": 0.00028022030698654374,
+      "loss": 0.3134710490703583,
+      "step": 288
+    },
+    {
+      "epoch": 2.07942238267148,
+      "grad_norm": 0.5239531993865967,
+      "learning_rate": 0.00027946726548741443,
+      "loss": 0.20423674583435059,
+      "step": 289
+    },
+    {
+      "epoch": 2.0866425992779782,
+      "grad_norm": 0.5845417976379395,
+      "learning_rate": 0.0002787128838039612,
+      "loss": 0.3297584354877472,
+      "step": 290
+    },
+    {
+      "epoch": 2.0938628158844765,
+      "grad_norm": 0.47120949625968933,
+      "learning_rate": 0.00027795717465853585,
+      "loss": 0.32672739028930664,
+      "step": 291
+    },
+    {
+      "epoch": 2.101083032490975,
+      "grad_norm": 0.40136632323265076,
+      "learning_rate": 0.00027720015079587743,
+      "loss": 0.23871661722660065,
+      "step": 292
+    },
+    {
+      "epoch": 2.108303249097473,
+      "grad_norm": 0.5156406164169312,
+      "learning_rate": 0.00027644182498289677,
+      "loss": 0.23360338807106018,
+      "step": 293
+    },
+    {
+      "epoch": 2.115523465703971,
+      "grad_norm": 0.4302510917186737,
+      "learning_rate": 0.0002756822100084621,
+      "loss": 0.31100887060165405,
+      "step": 294
+    },
+    {
+      "epoch": 2.1227436823104693,
+      "grad_norm": 0.3673456609249115,
+      "learning_rate": 0.00027492131868318246,
+      "loss": 0.21802300214767456,
+      "step": 295
+    },
+    {
+      "epoch": 2.1299638989169676,
+      "grad_norm": 0.32138535380363464,
+      "learning_rate": 0.00027415916383919216,
+      "loss": 0.19008119404315948,
+      "step": 296
+    },
+    {
+      "epoch": 2.137184115523466,
+      "grad_norm": 0.3203730285167694,
+      "learning_rate": 0.00027339575832993444,
+      "loss": 0.1889929622411728,
+      "step": 297
+    },
+    {
+      "epoch": 2.1444043321299637,
+      "grad_norm": 0.4612288773059845,
+      "learning_rate": 0.0002726311150299443,
+      "loss": 0.22323200106620789,
+      "step": 298
+    },
+    {
+      "epoch": 2.151624548736462,
+      "grad_norm": 0.36709311604499817,
+      "learning_rate": 0.00027186524683463155,
+      "loss": 0.20466932654380798,
+      "step": 299
+    },
+    {
+      "epoch": 2.1588447653429603,
+      "grad_norm": 0.4253864586353302,
+      "learning_rate": 0.0002710981666600636,
+      "loss": 0.2431459128856659,
+      "step": 300
+    },
+    {
+      "epoch": 2.1660649819494586,
+      "grad_norm": 0.4018993675708771,
+      "learning_rate": 0.00027032988744274736,
+      "loss": 0.2751820385456085,
+      "step": 301
+    },
+    {
+      "epoch": 2.1732851985559565,
+      "grad_norm": 0.4428861737251282,
+      "learning_rate": 0.000269560422139411,
+      "loss": 0.269814133644104,
+      "step": 302
+    },
+    {
+      "epoch": 2.1805054151624548,
+      "grad_norm": 0.47122061252593994,
+      "learning_rate": 0.00026878978372678564,
+      "loss": 0.22829411923885345,
+      "step": 303
+    },
+    {
+      "epoch": 2.187725631768953,
+      "grad_norm": 0.4426339268684387,
+      "learning_rate": 0.00026801798520138646,
+      "loss": 0.2637779414653778,
+      "step": 304
+    },
+    {
+      "epoch": 2.1949458483754514,
+      "grad_norm": 0.4073690176010132,
+      "learning_rate": 0.0002672450395792935,
+      "loss": 0.1904142051935196,
+      "step": 305
+    },
+    {
+      "epoch": 2.2021660649819497,
+      "grad_norm": 0.4896828234195709,
+      "learning_rate": 0.00026647095989593193,
+      "loss": 0.34650635719299316,
+      "step": 306
+    },
+    {
+      "epoch": 2.2093862815884475,
+      "grad_norm": 0.48900970816612244,
+      "learning_rate": 0.0002656957592058528,
+      "loss": 0.2542106807231903,
+      "step": 307
+    },
+    {
+      "epoch": 2.216606498194946,
+      "grad_norm": 0.4600171446800232,
+      "learning_rate": 0.0002649194505825121,
+      "loss": 0.21008341014385223,
+      "step": 308
+    },
+    {
+      "epoch": 2.223826714801444,
+      "grad_norm": 0.43380653858184814,
+      "learning_rate": 0.00026414204711805103,
+      "loss": 0.2683791518211365,
+      "step": 309
+    },
+    {
+      "epoch": 2.2310469314079424,
+      "grad_norm": 0.5792773365974426,
+      "learning_rate": 0.0002633635619230746,
+      "loss": 0.33160915970802307,
+      "step": 310
+    },
+    {
+      "epoch": 2.2382671480144403,
+      "grad_norm": 0.5626965165138245,
+      "learning_rate": 0.0002625840081264309,
+      "loss": 0.26700398325920105,
+      "step": 311
+    },
+    {
+      "epoch": 2.2454873646209386,
+      "grad_norm": 0.4952887296676636,
+      "learning_rate": 0.00026180339887498953,
+      "loss": 0.2689879536628723,
+      "step": 312
+    },
+    {
+      "epoch": 2.252707581227437,
+      "grad_norm": 0.4501405656337738,
+      "learning_rate": 0.0002610217473334199,
+      "loss": 0.26021096110343933,
+      "step": 313
+    },
+    {
+      "epoch": 2.259927797833935,
+      "grad_norm": 0.36552777886390686,
+      "learning_rate": 0.00026023906668396933,
+      "loss": 0.26156651973724365,
+      "step": 314
+    },
+    {
+      "epoch": 2.2671480144404335,
+      "grad_norm": 0.3733731806278229,
+      "learning_rate": 0.00025945537012624054,
+      "loss": 0.21406950056552887,
+      "step": 315
+    },
+    {
+      "epoch": 2.2743682310469313,
+      "grad_norm": 0.4240867495536804,
+      "learning_rate": 0.00025867067087696923,
+      "loss": 0.32105109095573425,
+      "step": 316
+    },
+    {
+      "epoch": 2.2815884476534296,
+      "grad_norm": 0.4198870360851288,
+      "learning_rate": 0.000257884982169801,
+      "loss": 0.3047807216644287,
+      "step": 317
+    },
+    {
+      "epoch": 2.288808664259928,
+      "grad_norm": 0.4444144666194916,
+      "learning_rate": 0.00025709831725506845,
+      "loss": 0.23268437385559082,
+      "step": 318
+    },
+    {
+      "epoch": 2.2960288808664258,
+      "grad_norm": 0.41258376836776733,
+      "learning_rate": 0.00025631068939956726,
+      "loss": 0.26838576793670654,
+      "step": 319
+    },
+    {
+      "epoch": 2.303249097472924,
+      "grad_norm": 0.5179259777069092,
+      "learning_rate": 0.00025552211188633293,
+      "loss": 0.2882261574268341,
+      "step": 320
+    },
+    {
+      "epoch": 2.3104693140794224,
+      "grad_norm": 0.4838998019695282,
+      "learning_rate": 0.0002547325980144166,
+      "loss": 0.30366814136505127,
+      "step": 321
+    },
+    {
+      "epoch": 2.3176895306859207,
+      "grad_norm": 0.4862940311431885,
+      "learning_rate": 0.0002539421610986605,
+      "loss": 0.29283270239830017,
+      "step": 322
+    },
+    {
+      "epoch": 2.324909747292419,
+      "grad_norm": 0.441685289144516,
+      "learning_rate": 0.000253150814469474,
+      "loss": 0.22393643856048584,
+      "step": 323
+    },
+    {
+      "epoch": 2.332129963898917,
+      "grad_norm": 0.39844486117362976,
+      "learning_rate": 0.0002523585714726081,
+      "loss": 0.17744283378124237,
+      "step": 324
+    },
+    {
+      "epoch": 2.339350180505415,
+      "grad_norm": 0.4977899491786957,
+      "learning_rate": 0.0002515654454689307,
+      "loss": 0.24069495499134064,
+      "step": 325
+    },
+    {
+      "epoch": 2.3465703971119134,
+      "grad_norm": 0.4101859927177429,
+      "learning_rate": 0.0002507714498342016,
+      "loss": 0.25044453144073486,
+      "step": 326
+    },
+    {
+      "epoch": 2.3537906137184117,
+      "grad_norm": 0.3521929085254669,
+      "learning_rate": 0.0002499765979588462,
+      "loss": 0.20651675760746002,
+      "step": 327
+    },
+    {
+      "epoch": 2.3610108303249095,
+      "grad_norm": 0.5269883871078491,
+      "learning_rate": 0.00024918090324773024,
+      "loss": 0.301647424697876,
+      "step": 328
+    },
+    {
+      "epoch": 2.368231046931408,
+      "grad_norm": 0.47672051191329956,
+      "learning_rate": 0.0002483843791199335,
+      "loss": 0.23506474494934082,
+      "step": 329
+    },
+    {
+      "epoch": 2.375451263537906,
+      "grad_norm": 0.5546545386314392,
+      "learning_rate": 0.00024758703900852375,
+      "loss": 0.3066970705986023,
+      "step": 330
+    },
+    {
+      "epoch": 2.3826714801444044,
+      "grad_norm": 0.3792511820793152,
+      "learning_rate": 0.0002467888963603298,
+      "loss": 0.20687991380691528,
+      "step": 331
+    },
+    {
+      "epoch": 2.3898916967509027,
+      "grad_norm": 0.4949963092803955,
+      "learning_rate": 0.00024598996463571474,
+      "loss": 0.2865826189517975,
+      "step": 332
+    },
+    {
+      "epoch": 2.3971119133574006,
+      "grad_norm": 0.4250241219997406,
+      "learning_rate": 0.00024519025730834954,
+      "loss": 0.25499409437179565,
+      "step": 333
+    },
+    {
+      "epoch": 2.404332129963899,
+      "grad_norm": 0.30932316184043884,
+      "learning_rate": 0.00024438978786498524,
+      "loss": 0.18888786435127258,
+      "step": 334
+    },
+    {
+      "epoch": 2.411552346570397,
+      "grad_norm": 0.5183025002479553,
+      "learning_rate": 0.00024358856980522556,
+      "loss": 0.2950358986854553,
+      "step": 335
+    },
+    {
+      "epoch": 2.4187725631768955,
+      "grad_norm": 0.46398693323135376,
+      "learning_rate": 0.00024278661664129948,
+      "loss": 0.28177422285079956,
+      "step": 336
+    },
+    {
+      "epoch": 2.4259927797833933,
+      "grad_norm": 0.4938473701477051,
+      "learning_rate": 0.00024198394189783317,
+      "loss": 0.31616371870040894,
+      "step": 337
+    },
+    {
+      "epoch": 2.4332129963898916,
+      "grad_norm": 0.4439496099948883,
+      "learning_rate": 0.00024118055911162198,
+      "loss": 0.2702808380126953,
+      "step": 338
+    },
+    {
+      "epoch": 2.44043321299639,
+      "grad_norm": 0.45705899596214294,
+      "learning_rate": 0.00024037648183140205,
+      "loss": 0.33506399393081665,
+      "step": 339
+    },
+    {
+      "epoch": 2.4476534296028882,
+      "grad_norm": 0.4123782217502594,
+      "learning_rate": 0.00023957172361762199,
+      "loss": 0.2569669783115387,
+      "step": 340
+    },
+    {
+      "epoch": 2.4548736462093865,
+      "grad_norm": 0.5178057551383972,
+      "learning_rate": 0.00023876629804221402,
+      "loss": 0.2357858568429947,
+      "step": 341
+    },
+    {
+      "epoch": 2.4620938628158844,
+      "grad_norm": 0.5206875205039978,
+      "learning_rate": 0.0002379602186883652,
+      "loss": 0.30847087502479553,
+      "step": 342
+    },
+    {
+      "epoch": 2.4693140794223827,
+      "grad_norm": 0.4264805018901825,
+      "learning_rate": 0.00023715349915028823,
+      "loss": 0.27529531717300415,
+      "step": 343
+    },
+    {
+      "epoch": 2.476534296028881,
+      "grad_norm": 0.5162002444267273,
+      "learning_rate": 0.00023634615303299233,
+      "loss": 0.3068305253982544,
+      "step": 344
+    },
+    {
+      "epoch": 2.483754512635379,
+      "grad_norm": 0.45099541544914246,
+      "learning_rate": 0.00023553819395205378,
+      "loss": 0.2681335508823395,
+      "step": 345
+    },
+    {
+      "epoch": 2.490974729241877,
+      "grad_norm": 0.49566006660461426,
+      "learning_rate": 0.00023472963553338613,
+      "loss": 0.2626950144767761,
+      "step": 346
+    },
+    {
+      "epoch": 2.4981949458483754,
+      "grad_norm": 0.4181882441043854,
+      "learning_rate": 0.00023392049141301055,
+      "loss": 0.2935839891433716,
+      "step": 347
+    },
+    {
+      "epoch": 2.5054151624548737,
+      "grad_norm": 0.4298352897167206,
+      "learning_rate": 0.000233110775236826,
+      "loss": 0.2565048038959503,
+      "step": 348
+    },
+    {
+      "epoch": 2.512635379061372,
+      "grad_norm": 0.46044519543647766,
+      "learning_rate": 0.00023230050066037872,
+      "loss": 0.2740374207496643,
+      "step": 349
+    },
+    {
+      "epoch": 2.51985559566787,
+      "grad_norm": 0.3774242401123047,
+      "learning_rate": 0.00023148968134863233,
+      "loss": 0.23688335716724396,
+      "step": 350
+    },
+    {
+      "epoch": 2.527075812274368,
+      "grad_norm": 0.4179275631904602,
+      "learning_rate": 0.00023067833097573714,
+      "loss": 0.2116960883140564,
+      "step": 351
+    },
+    {
+      "epoch": 2.5342960288808665,
+      "grad_norm": 0.5179049968719482,
+      "learning_rate": 0.0002298664632247994,
+      "loss": 0.2446313500404358,
+      "step": 352
+    },
+    {
+      "epoch": 2.5415162454873648,
+      "grad_norm": 0.44997915625572205,
+      "learning_rate": 0.000229054091787651,
+      "loss": 0.3158394694328308,
+      "step": 353
+    },
+    {
+      "epoch": 2.5487364620938626,
+      "grad_norm": 0.41355574131011963,
+      "learning_rate": 0.0002282412303646183,
+      "loss": 0.2720823884010315,
+      "step": 354
+    },
+    {
+      "epoch": 2.555956678700361,
+      "grad_norm": 0.36182165145874023,
+      "learning_rate": 0.00022742789266429095,
+      "loss": 0.22511911392211914,
+      "step": 355
+    },
+    {
+      "epoch": 2.563176895306859,
+      "grad_norm": 0.43841102719306946,
+      "learning_rate": 0.00022661409240329076,
+      "loss": 0.21212677657604218,
+      "step": 356
+    },
+    {
+      "epoch": 2.5703971119133575,
+      "grad_norm": 0.4933500289916992,
+      "learning_rate": 0.0002257998433060407,
+      "loss": 0.3581770956516266,
+      "step": 357
+    },
+    {
+      "epoch": 2.577617328519856,
+      "grad_norm": 0.4325965344905853,
+      "learning_rate": 0.00022498515910453296,
+      "loss": 0.3252311646938324,
+      "step": 358
+    },
+    {
+      "epoch": 2.5848375451263537,
+      "grad_norm": 0.45968931913375854,
+      "learning_rate": 0.00022417005353809772,
+      "loss": 0.3097744286060333,
+      "step": 359
+    },
+    {
+      "epoch": 2.592057761732852,
+      "grad_norm": 0.43083667755126953,
+      "learning_rate": 0.00022335454035317124,
+      "loss": 0.25429341197013855,
+      "step": 360
+    },
+    {
+      "epoch": 2.5992779783393503,
+      "grad_norm": 0.4504421353340149,
+      "learning_rate": 0.00022253863330306425,
+      "loss": 0.315784215927124,
+      "step": 361
+    },
+    {
+      "epoch": 2.606498194945848,
+      "grad_norm": 0.44943565130233765,
+      "learning_rate": 0.0002217223461477296,
+      "loss": 0.22878926992416382,
+      "step": 362
+    },
+    {
+      "epoch": 2.6137184115523464,
+      "grad_norm": 0.4279639720916748,
+      "learning_rate": 0.00022090569265353072,
+      "loss": 0.2623154819011688,
+      "step": 363
+    },
+    {
+      "epoch": 2.6209386281588447,
+      "grad_norm": 0.4907088279724121,
+      "learning_rate": 0.00022008868659300905,
+      "loss": 0.317205548286438,
+      "step": 364
+    },
+    {
+      "epoch": 2.628158844765343,
+      "grad_norm": 0.49302950501441956,
+      "learning_rate": 0.000219271341744652,
+      "loss": 0.30752891302108765,
+      "step": 365
+    },
+    {
+      "epoch": 2.6353790613718413,
+      "grad_norm": 0.38831761479377747,
+      "learning_rate": 0.00021845367189266042,
+      "loss": 0.27115532755851746,
+      "step": 366
+    },
+    {
+      "epoch": 2.6425992779783396,
+      "grad_norm": 0.47725188732147217,
+      "learning_rate": 0.00021763569082671635,
+      "loss": 0.20296484231948853,
+      "step": 367
+    },
+    {
+      "epoch": 2.6498194945848375,
+      "grad_norm": 0.4198381006717682,
+      "learning_rate": 0.00021681741234175027,
+      "loss": 0.22836127877235413,
+      "step": 368
+    },
+    {
+      "epoch": 2.6570397111913358,
+      "grad_norm": 0.5218875408172607,
+      "learning_rate": 0.00021599885023770835,
+      "loss": 0.29816269874572754,
+      "step": 369
+    },
+    {
+      "epoch": 2.664259927797834,
+      "grad_norm": 0.4321244955062866,
+      "learning_rate": 0.00021518001831932022,
+      "loss": 0.21631377935409546,
+      "step": 370
+    },
+    {
+      "epoch": 2.671480144404332,
+      "grad_norm": 0.4036301374435425,
+      "learning_rate": 0.00021436093039586552,
+      "loss": 0.19903570413589478,
+      "step": 371
+    },
+    {
+      "epoch": 2.67870036101083,
+      "grad_norm": 0.543828010559082,
+      "learning_rate": 0.00021354160028094153,
+      "loss": 0.2922933101654053,
+      "step": 372
+    },
+    {
+      "epoch": 2.6859205776173285,
+      "grad_norm": 0.5146738290786743,
+      "learning_rate": 0.00021272204179222982,
+      "loss": 0.2749359607696533,
+      "step": 373
+    },
+    {
+      "epoch": 2.693140794223827,
+      "grad_norm": 0.47690504789352417,
+      "learning_rate": 0.00021190226875126352,
+      "loss": 0.296896755695343,
+      "step": 374
+    },
+    {
+      "epoch": 2.700361010830325,
+      "grad_norm": 0.545520007610321,
+      "learning_rate": 0.000211082294983194,
+      "loss": 0.27578210830688477,
+      "step": 375
+    },
+    {
+      "epoch": 2.707581227436823,
+      "grad_norm": 0.42757925391197205,
+      "learning_rate": 0.00021026213431655792,
+      "loss": 0.2315436154603958,
+      "step": 376
+    },
+    {
+      "epoch": 2.7148014440433212,
+      "grad_norm": 0.5044121146202087,
+      "learning_rate": 0.0002094418005830439,
+      "loss": 0.34600791335105896,
+      "step": 377
+    },
+    {
+      "epoch": 2.7220216606498195,
+      "grad_norm": 0.5155430436134338,
+      "learning_rate": 0.00020862130761725917,
+      "loss": 0.3247469663619995,
+      "step": 378
+    },
+    {
+      "epoch": 2.729241877256318,
+      "grad_norm": 0.5196051001548767,
+      "learning_rate": 0.00020780066925649634,
+      "loss": 0.3277815878391266,
+      "step": 379
+    },
+    {
+      "epoch": 2.7364620938628157,
+      "grad_norm": 0.44720444083213806,
+      "learning_rate": 0.00020697989934050025,
+      "loss": 0.25156158208847046,
+      "step": 380
+    },
+    {
+      "epoch": 2.743682310469314,
+      "grad_norm": 0.43984708189964294,
+      "learning_rate": 0.00020615901171123412,
+      "loss": 0.28022634983062744,
+      "step": 381
+    },
+    {
+      "epoch": 2.7509025270758123,
+      "grad_norm": 0.47703853249549866,
+      "learning_rate": 0.00020533802021264648,
+      "loss": 0.2780344486236572,
+      "step": 382
+    },
+    {
+      "epoch": 2.7581227436823106,
+      "grad_norm": 0.43746253848075867,
+      "learning_rate": 0.00020451693869043763,
+      "loss": 0.23613542318344116,
+      "step": 383
+    },
+    {
+      "epoch": 2.765342960288809,
+      "grad_norm": 0.41468992829322815,
+      "learning_rate": 0.00020369578099182597,
+      "loss": 0.3185104429721832,
+      "step": 384
+    },
+    {
+      "epoch": 2.7725631768953067,
+      "grad_norm": 0.5149108171463013,
+      "learning_rate": 0.0002028745609653147,
+      "loss": 0.2768135070800781,
+      "step": 385
+    },
+    {
+      "epoch": 2.779783393501805,
+      "grad_norm": 0.47348129749298096,
+      "learning_rate": 0.00020205329246045798,
+      "loss": 0.31962212920188904,
+      "step": 386
+    },
+    {
+      "epoch": 2.7870036101083033,
+      "grad_norm": 0.4738944172859192,
+      "learning_rate": 0.00020123198932762778,
+      "loss": 0.23498407006263733,
+      "step": 387
+    },
+    {
+      "epoch": 2.794223826714801,
+      "grad_norm": 0.48899075388908386,
+      "learning_rate": 0.0002004106654177798,
+      "loss": 0.23857903480529785,
+      "step": 388
+    },
+    {
+      "epoch": 2.8014440433212995,
+      "grad_norm": 0.45867517590522766,
+      "learning_rate": 0.00019958933458222026,
+      "loss": 0.3246588706970215,
+      "step": 389
+    },
+    {
+      "epoch": 2.808664259927798,
+      "grad_norm": 0.5840234160423279,
+      "learning_rate": 0.0001987680106723723,
+      "loss": 0.2935522198677063,
+      "step": 390
+    },
+    {
+      "epoch": 2.815884476534296,
+      "grad_norm": 0.3764658272266388,
+      "learning_rate": 0.00019794670753954204,
+      "loss": 0.2096547782421112,
+      "step": 391
+    },
+    {
+      "epoch": 2.8231046931407944,
+      "grad_norm": 0.44675493240356445,
+      "learning_rate": 0.00019712543903468535,
+      "loss": 0.2988222539424896,
+      "step": 392
+    },
+    {
+      "epoch": 2.8303249097472927,
+      "grad_norm": 0.43063509464263916,
+      "learning_rate": 0.00019630421900817407,
+      "loss": 0.2222481369972229,
+      "step": 393
+    },
+    {
+      "epoch": 2.8375451263537905,
+      "grad_norm": 0.47969475388526917,
+      "learning_rate": 0.00019548306130956242,
+      "loss": 0.26167353987693787,
+      "step": 394
+    },
+    {
+      "epoch": 2.844765342960289,
+      "grad_norm": 0.47235366702079773,
+      "learning_rate": 0.00019466197978735354,
+      "loss": 0.30541300773620605,
+      "step": 395
+    },
+    {
+      "epoch": 2.851985559566787,
+      "grad_norm": 0.47744160890579224,
+      "learning_rate": 0.00019384098828876598,
+      "loss": 0.23645427823066711,
+      "step": 396
+    },
+    {
+      "epoch": 2.859205776173285,
+      "grad_norm": 0.5037774443626404,
+      "learning_rate": 0.00019302010065949988,
+      "loss": 0.26774662733078003,
+      "step": 397
+    },
+    {
+      "epoch": 2.8664259927797833,
+      "grad_norm": 0.4689895510673523,
+      "learning_rate": 0.0001921993307435037,
+      "loss": 0.3405228853225708,
+      "step": 398
+    },
+    {
+      "epoch": 2.8736462093862816,
+      "grad_norm": 0.44282686710357666,
+      "learning_rate": 0.00019137869238274093,
+      "loss": 0.26063668727874756,
+      "step": 399
+    },
+    {
+      "epoch": 2.88086642599278,
+      "grad_norm": 0.4379267990589142,
+      "learning_rate": 0.0001905581994169562,
+      "loss": 0.26478758454322815,
+      "step": 400
+    },
+    {
+      "epoch": 2.888086642599278,
+      "grad_norm": 0.519749641418457,
+      "learning_rate": 0.00018973786568344205,
+      "loss": 0.31215131282806396,
+      "step": 401
+    },
+    {
+      "epoch": 2.895306859205776,
+      "grad_norm": 0.4515770375728607,
+      "learning_rate": 0.000188917705016806,
+      "loss": 0.2881377339363098,
+      "step": 402
+    },
+    {
+      "epoch": 2.9025270758122743,
+      "grad_norm": 0.4650344252586365,
+      "learning_rate": 0.0001880977312487365,
+      "loss": 0.2557958960533142,
+      "step": 403
+    },
+    {
+      "epoch": 2.9097472924187726,
+      "grad_norm": 0.4621173143386841,
+      "learning_rate": 0.0001872779582077702,
+      "loss": 0.23401126265525818,
+      "step": 404
+    },
+    {
+      "epoch": 2.916967509025271,
+      "grad_norm": 0.4094926118850708,
+      "learning_rate": 0.00018645839971905852,
+      "loss": 0.24556361138820648,
+      "step": 405
+    },
+    {
+      "epoch": 2.9241877256317688,
+      "grad_norm": 0.48592710494995117,
+      "learning_rate": 0.0001856390696041345,
+      "loss": 0.27918654680252075,
+      "step": 406
+    },
+    {
+      "epoch": 2.931407942238267,
+      "grad_norm": 0.4039171040058136,
+      "learning_rate": 0.00018481998168067983,
+      "loss": 0.22918914258480072,
+      "step": 407
+    },
+    {
+      "epoch": 2.9386281588447654,
+      "grad_norm": 0.5521522760391235,
+      "learning_rate": 0.0001840011497622917,
+      "loss": 0.35596632957458496,
+      "step": 408
+    },
+    {
+      "epoch": 2.9458483754512637,
+      "grad_norm": 0.5527123212814331,
+      "learning_rate": 0.0001831825876582498,
+      "loss": 0.25526192784309387,
+      "step": 409
+    },
+    {
+      "epoch": 2.953068592057762,
+      "grad_norm": 0.4386238157749176,
+      "learning_rate": 0.0001823643091732837,
+      "loss": 0.2627878189086914,
+      "step": 410
+    },
+    {
+      "epoch": 2.96028880866426,
+      "grad_norm": 0.48318973183631897,
+      "learning_rate": 0.00018154632810733962,
+      "loss": 0.29408901929855347,
+      "step": 411
+    },
+    {
+      "epoch": 2.967509025270758,
+      "grad_norm": 0.5193997025489807,
+      "learning_rate": 0.00018072865825534805,
+      "loss": 0.2762555480003357,
+      "step": 412
+    },
+    {
+      "epoch": 2.9747292418772564,
+      "grad_norm": 0.488490492105484,
+      "learning_rate": 0.000179911313406991,
+      "loss": 0.3119064271450043,
+      "step": 413
+    },
+    {
+      "epoch": 2.9819494584837543,
+      "grad_norm": 0.4461340010166168,
+      "learning_rate": 0.00017909430734646935,
+      "loss": 0.2738839089870453,
+      "step": 414
+    },
+    {
+      "epoch": 2.9891696750902526,
+      "grad_norm": 0.4488433003425598,
+      "learning_rate": 0.0001782776538522704,
+      "loss": 0.25788843631744385,
+      "step": 415
+    },
+    {
+      "epoch": 2.996389891696751,
+      "grad_norm": 0.5179741382598877,
+      "learning_rate": 0.0001774613666969358,
+      "loss": 0.2556946575641632,
+      "step": 416
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.6138741374015808,
+      "learning_rate": 0.00017664545964682878,
+      "loss": 0.23274531960487366,
+      "step": 417
+    },
+    {
+      "epoch": 3.0072202166064983,
+      "grad_norm": 0.27745357155799866,
+      "learning_rate": 0.00017582994646190233,
+      "loss": 0.14754480123519897,
+      "step": 418
+    },
+    {
+      "epoch": 3.0144404332129966,
+      "grad_norm": 0.40759217739105225,
+      "learning_rate": 0.00017501484089546708,
+      "loss": 0.21522817015647888,
+      "step": 419
+    },
+    {
+      "epoch": 3.0216606498194944,
+      "grad_norm": 0.3123788833618164,
+      "learning_rate": 0.00017420015669395938,
+      "loss": 0.15608534216880798,
+      "step": 420
+    },
+    {
+      "epoch": 3.0288808664259927,
+      "grad_norm": 0.4599035978317261,
+      "learning_rate": 0.00017338590759670934,
+      "loss": 0.1309528797864914,
+      "step": 421
+    },
+    {
+      "epoch": 3.036101083032491,
+      "grad_norm": 0.38352325558662415,
+      "learning_rate": 0.00017257210733570916,
+      "loss": 0.1724625676870346,
+      "step": 422
+    },
+    {
+      "epoch": 3.0433212996389893,
+      "grad_norm": 0.44771838188171387,
+      "learning_rate": 0.00017175876963538178,
+      "loss": 0.1661359965801239,
+      "step": 423
+    },
+    {
+      "epoch": 3.050541516245487,
+      "grad_norm": 0.5998123288154602,
+      "learning_rate": 0.00017094590821234906,
+      "loss": 0.1318468153476715,
+      "step": 424
+    },
+    {
+      "epoch": 3.0577617328519855,
+      "grad_norm": 0.47632187604904175,
+      "learning_rate": 0.00017013353677520064,
+      "loss": 0.10464094579219818,
+      "step": 425
+    },
+    {
+      "epoch": 3.064981949458484,
+      "grad_norm": 0.3745962083339691,
+      "learning_rate": 0.00016932166902426288,
+      "loss": 0.10677687078714371,
+      "step": 426
+    },
+    {
+      "epoch": 3.072202166064982,
+      "grad_norm": 0.6340925693511963,
+      "learning_rate": 0.00016851031865136763,
+      "loss": 0.1600916087627411,
+      "step": 427
+    },
+    {
+      "epoch": 3.07942238267148,
+      "grad_norm": 0.8183390498161316,
+      "learning_rate": 0.00016769949933962127,
+      "loss": 0.13513822853565216,
+      "step": 428
+    },
+    {
+      "epoch": 3.0866425992779782,
+      "grad_norm": 0.8068543076515198,
+      "learning_rate": 0.00016688922476317398,
+      "loss": 0.14319300651550293,
+      "step": 429
+    },
+    {
+      "epoch": 3.0938628158844765,
+      "grad_norm": 0.6381204128265381,
+      "learning_rate": 0.00016607950858698946,
+      "loss": 0.1376371830701828,
+      "step": 430
+    },
+    {
+      "epoch": 3.101083032490975,
+      "grad_norm": 0.5576755404472351,
+      "learning_rate": 0.00016527036446661395,
+      "loss": 0.14407199621200562,
+      "step": 431
+    },
+    {
+      "epoch": 3.108303249097473,
+      "grad_norm": 0.6089445948600769,
+      "learning_rate": 0.00016446180604794627,
+      "loss": 0.1575869917869568,
+      "step": 432
+    },
+    {
+      "epoch": 3.115523465703971,
+      "grad_norm": 0.6250684857368469,
+      "learning_rate": 0.0001636538469670077,
+      "loss": 0.17905279994010925,
+      "step": 433
+    },
+    {
+      "epoch": 3.1227436823104693,
+      "grad_norm": 0.4243159592151642,
+      "learning_rate": 0.00016284650084971185,
+      "loss": 0.12401723116636276,
+      "step": 434
+    },
+    {
+      "epoch": 3.1299638989169676,
+      "grad_norm": 0.4369412362575531,
+      "learning_rate": 0.00016203978131163485,
+      "loss": 0.11108125746250153,
+      "step": 435
+    },
+    {
+      "epoch": 3.137184115523466,
+      "grad_norm": 0.45113661885261536,
+      "learning_rate": 0.000161233701957786,
+      "loss": 0.1364113986492157,
+      "step": 436
+    },
+    {
+      "epoch": 3.1444043321299637,
+      "grad_norm": 0.33899277448654175,
+      "learning_rate": 0.00016042827638237806,
+      "loss": 0.1253184676170349,
+      "step": 437
+    },
+    {
+      "epoch": 3.151624548736462,
+      "grad_norm": 0.48023247718811035,
+      "learning_rate": 0.000159623518168598,
+      "loss": 0.1729699671268463,
+      "step": 438
+    },
+    {
+      "epoch": 3.1588447653429603,
+      "grad_norm": 0.4260300397872925,
+      "learning_rate": 0.00015881944088837804,
+      "loss": 0.1630483865737915,
+      "step": 439
+    },
+    {
+      "epoch": 3.1660649819494586,
+      "grad_norm": 0.34896689653396606,
+      "learning_rate": 0.00015801605810216685,
+      "loss": 0.14477582275867462,
+      "step": 440
+    },
+    {
+      "epoch": 3.1732851985559565,
+      "grad_norm": 0.46189841628074646,
+      "learning_rate": 0.00015721338335870057,
+      "loss": 0.16415920853614807,
+      "step": 441
+    },
+    {
+      "epoch": 3.1805054151624548,
+      "grad_norm": 0.40856343507766724,
+      "learning_rate": 0.00015641143019477446,
+      "loss": 0.12028312683105469,
+      "step": 442
+    },
+    {
+      "epoch": 3.187725631768953,
+      "grad_norm": 0.36561018228530884,
+      "learning_rate": 0.00015561021213501483,
+      "loss": 0.12599362432956696,
+      "step": 443
+    },
+    {
+      "epoch": 3.1949458483754514,
+      "grad_norm": 0.4341471493244171,
+      "learning_rate": 0.0001548097426916505,
+      "loss": 0.14446969330310822,
+      "step": 444
+    },
+    {
+      "epoch": 3.2021660649819497,
+      "grad_norm": 0.3864421844482422,
+      "learning_rate": 0.00015401003536428534,
+      "loss": 0.11642799526453018,
+      "step": 445
+    },
+    {
+      "epoch": 3.2093862815884475,
+      "grad_norm": 0.5154263973236084,
+      "learning_rate": 0.00015321110363967035,
+      "loss": 0.17028403282165527,
+      "step": 446
+    },
+    {
+      "epoch": 3.216606498194946,
+      "grad_norm": 0.4077145755290985,
+      "learning_rate": 0.0001524129609914763,
+      "loss": 0.11544163525104523,
+      "step": 447
+    },
+    {
+      "epoch": 3.223826714801444,
+      "grad_norm": 0.41609570384025574,
+      "learning_rate": 0.00015161562088006648,
+      "loss": 0.1203891783952713,
+      "step": 448
+    },
+    {
+      "epoch": 3.2310469314079424,
+      "grad_norm": 0.430624783039093,
+      "learning_rate": 0.00015081909675226975,
+      "loss": 0.09192588925361633,
+      "step": 449
+    },
+    {
+      "epoch": 3.2382671480144403,
+      "grad_norm": 0.5538221001625061,
+      "learning_rate": 0.0001500234020411538,
+      "loss": 0.16532525420188904,
+      "step": 450
+    },
+    {
+      "epoch": 3.2454873646209386,
+      "grad_norm": 0.6360041499137878,
+      "learning_rate": 0.00014922855016579845,
+      "loss": 0.13534685969352722,
+      "step": 451
+    },
+    {
+      "epoch": 3.252707581227437,
+      "grad_norm": 0.556538462638855,
+      "learning_rate": 0.00014843455453106925,
+      "loss": 0.16164641082286835,
+      "step": 452
+    },
+    {
+      "epoch": 3.259927797833935,
+      "grad_norm": 0.8166404366493225,
+      "learning_rate": 0.00014764142852739193,
+      "loss": 0.17865729331970215,
+      "step": 453
+    },
+    {
+      "epoch": 3.2671480144404335,
+      "grad_norm": 0.6231580972671509,
+      "learning_rate": 0.00014684918553052603,
+      "loss": 0.11281149089336395,
+      "step": 454
+    },
+    {
+      "epoch": 3.2743682310469313,
+      "grad_norm": 0.5884938836097717,
+      "learning_rate": 0.00014605783890133948,
+      "loss": 0.15221013128757477,
+      "step": 455
+    },
+    {
+      "epoch": 3.2815884476534296,
+      "grad_norm": 0.40336933732032776,
+      "learning_rate": 0.00014526740198558344,
+      "loss": 0.08725619316101074,
+      "step": 456
+    },
+    {
+      "epoch": 3.288808664259928,
+      "grad_norm": 0.7082973718643188,
+      "learning_rate": 0.0001444778881136671,
+      "loss": 0.1795981079339981,
+      "step": 457
+    },
+    {
+      "epoch": 3.2960288808664258,
+      "grad_norm": 0.44312670826911926,
+      "learning_rate": 0.0001436893106004328,
+      "loss": 0.11532506346702576,
+      "step": 458
+    },
+    {
+      "epoch": 3.303249097472924,
+      "grad_norm": 0.4892312288284302,
+      "learning_rate": 0.0001429016827449316,
+      "loss": 0.15218153595924377,
+      "step": 459
+    },
+    {
+      "epoch": 3.3104693140794224,
+      "grad_norm": 0.36000511050224304,
+      "learning_rate": 0.000142115017830199,
+      "loss": 0.1174839586019516,
+      "step": 460
+    },
+    {
+      "epoch": 3.3176895306859207,
+      "grad_norm": 0.42502933740615845,
+      "learning_rate": 0.00014132932912303082,
+      "loss": 0.1354883909225464,
+      "step": 461
+    },
+    {
+      "epoch": 3.324909747292419,
+      "grad_norm": 0.5004318356513977,
+      "learning_rate": 0.00014054462987375947,
+      "loss": 0.14879979193210602,
+      "step": 462
+    },
+    {
+      "epoch": 3.332129963898917,
+      "grad_norm": 0.5428838729858398,
+      "learning_rate": 0.00013976093331603072,
+      "loss": 0.1601594090461731,
+      "step": 463
+    },
+    {
+      "epoch": 3.339350180505415,
+      "grad_norm": 0.5523360967636108,
+      "learning_rate": 0.00013897825266658015,
+      "loss": 0.1349392980337143,
+      "step": 464
+    },
+    {
+      "epoch": 3.3465703971119134,
+      "grad_norm": 0.4672771394252777,
+      "learning_rate": 0.00013819660112501054,
+      "loss": 0.1563996523618698,
+      "step": 465
+    },
+    {
+      "epoch": 3.3537906137184117,
+      "grad_norm": 0.4747507870197296,
+      "learning_rate": 0.00013741599187356917,
+      "loss": 0.12792187929153442,
+      "step": 466
+    },
+    {
+      "epoch": 3.3610108303249095,
+      "grad_norm": 0.4293171167373657,
+      "learning_rate": 0.00013663643807692547,
+      "loss": 0.13704359531402588,
+      "step": 467
+    },
+    {
+      "epoch": 3.368231046931408,
+      "grad_norm": 0.406646728515625,
+      "learning_rate": 0.00013585795288194902,
+      "loss": 0.14688344299793243,
+      "step": 468
+    },
+    {
+      "epoch": 3.375451263537906,
+      "grad_norm": 0.44870486855506897,
+      "learning_rate": 0.00013508054941748792,
+      "loss": 0.11994664371013641,
+      "step": 469
+    },
+    {
+      "epoch": 3.3826714801444044,
+      "grad_norm": 0.46383994817733765,
+      "learning_rate": 0.00013430424079414728,
+      "loss": 0.13228918612003326,
+      "step": 470
+    },
+    {
+      "epoch": 3.3898916967509027,
+      "grad_norm": 0.873016893863678,
+      "learning_rate": 0.0001335290401040681,
+      "loss": 0.14495904743671417,
+      "step": 471
+    },
+    {
+      "epoch": 3.3971119133574006,
+      "grad_norm": 0.5575613975524902,
+      "learning_rate": 0.0001327549604207066,
+      "loss": 0.17171433568000793,
+      "step": 472
+    },
+    {
+      "epoch": 3.404332129963899,
+      "grad_norm": 0.541860818862915,
+      "learning_rate": 0.00013198201479861353,
+      "loss": 0.14465925097465515,
+      "step": 473
+    },
+    {
+      "epoch": 3.411552346570397,
+      "grad_norm": 0.5210021734237671,
+      "learning_rate": 0.00013121021627321438,
+      "loss": 0.15102122724056244,
+      "step": 474
+    },
+    {
+      "epoch": 3.4187725631768955,
+      "grad_norm": 0.46373456716537476,
+      "learning_rate": 0.00013043957786058904,
+      "loss": 0.15518702566623688,
+      "step": 475
+    },
+    {
+      "epoch": 3.4259927797833933,
+      "grad_norm": 0.48513785004615784,
+      "learning_rate": 0.00012967011255725263,
+      "loss": 0.14110815525054932,
+      "step": 476
+    },
+    {
+      "epoch": 3.4332129963898916,
+      "grad_norm": 0.5956123471260071,
+      "learning_rate": 0.0001289018333399364,
+      "loss": 0.14913466572761536,
+      "step": 477
+    },
+    {
+      "epoch": 3.44043321299639,
+      "grad_norm": 0.5387746095657349,
+      "learning_rate": 0.0001281347531653685,
+      "loss": 0.14882197976112366,
+      "step": 478
+    },
+    {
+      "epoch": 3.4476534296028882,
+      "grad_norm": 0.43160125613212585,
+      "learning_rate": 0.00012736888497005578,
+      "loss": 0.13204768300056458,
+      "step": 479
+    },
+    {
+      "epoch": 3.4548736462093865,
+      "grad_norm": 0.41130170226097107,
+      "learning_rate": 0.0001266042416700656,
+      "loss": 0.13427188992500305,
+      "step": 480
+    },
+    {
+      "epoch": 3.4620938628158844,
+      "grad_norm": 0.5126029253005981,
+      "learning_rate": 0.00012584083616080786,
+      "loss": 0.12719954550266266,
+      "step": 481
+    },
+    {
+      "epoch": 3.4693140794223827,
+      "grad_norm": 0.4994872212409973,
+      "learning_rate": 0.0001250786813168176,
+      "loss": 0.15062202513217926,
+      "step": 482
+    },
+    {
+      "epoch": 3.476534296028881,
+      "grad_norm": 0.4697980284690857,
+      "learning_rate": 0.00012431778999153796,
+      "loss": 0.14715459942817688,
+      "step": 483
+    },
+    {
+      "epoch": 3.483754512635379,
+      "grad_norm": 0.5356115102767944,
+      "learning_rate": 0.00012355817501710328,
+      "loss": 0.12667298316955566,
+      "step": 484
+    },
+    {
+      "epoch": 3.490974729241877,
+      "grad_norm": 0.48629310727119446,
+      "learning_rate": 0.00012279984920412264,
+      "loss": 0.12396648526191711,
+      "step": 485
+    },
+    {
+      "epoch": 3.4981949458483754,
+      "grad_norm": 0.6295841336250305,
+      "learning_rate": 0.00012204282534146414,
+      "loss": 0.1579296588897705,
+      "step": 486
+    },
+    {
+      "epoch": 3.5054151624548737,
+      "grad_norm": 0.524736225605011,
+      "learning_rate": 0.00012128711619603885,
+      "loss": 0.14176616072654724,
+      "step": 487
+    },
+    {
+      "epoch": 3.512635379061372,
+      "grad_norm": 0.4429115355014801,
+      "learning_rate": 0.0001205327345125856,
+      "loss": 0.16075468063354492,
+      "step": 488
+    },
+    {
+      "epoch": 3.51985559566787,
+      "grad_norm": 0.5202926993370056,
+      "learning_rate": 0.00011977969301345627,
+      "loss": 0.155876025557518,
+      "step": 489
+    },
+    {
+      "epoch": 3.527075812274368,
+      "grad_norm": 0.5090999603271484,
+      "learning_rate": 0.00011902800439840091,
+      "loss": 0.1420796513557434,
+      "step": 490
+    },
+    {
+      "epoch": 3.5342960288808665,
+      "grad_norm": 0.5266745686531067,
+      "learning_rate": 0.0001182776813443537,
+      "loss": 0.1911281943321228,
+      "step": 491
+    },
+    {
+      "epoch": 3.5415162454873648,
+      "grad_norm": 0.4100717306137085,
+      "learning_rate": 0.00011752873650521934,
+      "loss": 0.12646690011024475,
+      "step": 492
+    },
+    {
+      "epoch": 3.5487364620938626,
+      "grad_norm": 0.4897909164428711,
+      "learning_rate": 0.00011678118251165929,
+      "loss": 0.14893396198749542,
+      "step": 493
+    },
+    {
+      "epoch": 3.555956678700361,
+      "grad_norm": 0.48633724451065063,
+      "learning_rate": 0.00011603503197087893,
+      "loss": 0.11430861800909042,
+      "step": 494
+    },
+    {
+      "epoch": 3.563176895306859,
+      "grad_norm": 0.3797968327999115,
+      "learning_rate": 0.00011529029746641522,
+      "loss": 0.10336866229772568,
+      "step": 495
+    },
+    {
+      "epoch": 3.5703971119133575,
+      "grad_norm": 0.4714502692222595,
+      "learning_rate": 0.00011454699155792387,
+      "loss": 0.12609894573688507,
+      "step": 496
+    },
+    {
+      "epoch": 3.577617328519856,
+      "grad_norm": 0.5311459898948669,
+      "learning_rate": 0.00011380512678096805,
+      "loss": 0.14762037992477417,
+      "step": 497
+    },
+    {
+      "epoch": 3.5848375451263537,
+      "grad_norm": 0.44842103123664856,
+      "learning_rate": 0.00011306471564680703,
+      "loss": 0.1137317568063736,
+      "step": 498
+    },
+    {
+      "epoch": 3.592057761732852,
+      "grad_norm": 0.6083047986030579,
+      "learning_rate": 0.0001123257706421845,
+      "loss": 0.1538015455007553,
+      "step": 499
+    },
+    {
+      "epoch": 3.5992779783393503,
+      "grad_norm": 0.454277902841568,
+      "learning_rate": 0.00011158830422911882,
+      "loss": 0.13900214433670044,
+      "step": 500
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 770,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.525794067913907e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3feb7d84776eed4718f792285c52122bc3b2372e4e090380a285a8afc9196989
+size 5713

checkpoint-770/README.md

@@ -0,0 +1,210 @@
+---
+base_model: Qwen/Qwen3.5-9B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3.5-9B
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

checkpoint-770/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForConditionalGeneration",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.5-9B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-770/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dc94603f0c6e625c9716499f91c5acf38deff4596e0ed2ca5fc3a4f5126806c
+size 173188512

checkpoint-770/chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

checkpoint-770/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b589081ecc7eb3fc8ca1cffe8dc0cf8b74293cd27dbc861e9aa0caa6e15bfac3
+size 88617029

checkpoint-770/processor_config.json

@@ -0,0 +1,63 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

checkpoint-770/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2922886dc37f9211a4affea61093cbe4db90c0a37e91e7edac63ec0cf7b13d45
+size 14645

checkpoint-770/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6842476510ef1a339b059738a5616f5fe7454531bf7f4f3bc53954af85a4a7f1
+size 1465

checkpoint-770/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:263dcc3feef53d558a357d52941ff3649fa9cd9a3199934376f2bbf615d49ffc
+size 19989610

checkpoint-770/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": false,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

checkpoint-770/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3feb7d84776eed4718f792285c52122bc3b2372e4e090380a285a8afc9196989
+size 5713

eval.json

@@ -0,0 +1,601 @@
+{
+  "checkpoint": "outputs/checkpoint-770",
+  "num_samples": 98,
+  "retrieval_similarity": 0.7239,
+  "tIOU": 0.67,
+  "mAP_0.5": 0.5092,
+  "sim_pass": true,
+  "tiou_pass": true,
+  "map_pass": true,
+  "overall_pass": true,
+  "per_sample": [
+    {
+      "idx": 0,
+      "retrieval_sim": 0.6549,
+      "tIOU": 0.5277,
+      "mAP": 0.45
+    },
+    {
+      "idx": 1,
+      "retrieval_sim": 0.7739,
+      "tIOU": 0.808,
+      "mAP": 1.0
+    },
+    {
+      "idx": 2,
+      "retrieval_sim": 0.6433,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 3,
+      "retrieval_sim": 0.7473,
+      "tIOU": 0.6228,
+      "mAP": 0.45
+    },
+    {
+      "idx": 4,
+      "retrieval_sim": 0.7774,
+      "tIOU": 0.5406,
+      "mAP": 0.25
+    },
+    {
+      "idx": 5,
+      "retrieval_sim": 0.7924,
+      "tIOU": 0.3868,
+      "mAP": 0.05
+    },
+    {
+      "idx": 6,
+      "retrieval_sim": 0.7626,
+      "tIOU": 0.7779,
+      "mAP": 0.75
+    },
+    {
+      "idx": 7,
+      "retrieval_sim": 0.6476,
+      "tIOU": 0.806,
+      "mAP": 0.75
+    },
+    {
+      "idx": 8,
+      "retrieval_sim": 0.7355,
+      "tIOU": 0.752,
+      "mAP": 0.64
+    },
+    {
+      "idx": 9,
+      "retrieval_sim": 0.8859,
+      "tIOU": 0.6579,
+      "mAP": 0.5625
+    },
+    {
+      "idx": 10,
+      "retrieval_sim": 0.7486,
+      "tIOU": 0.7134,
+      "mAP": 0.8
+    },
+    {
+      "idx": 11,
+      "retrieval_sim": 0.6805,
+      "tIOU": 0.5,
+      "mAP": 0.2
+    },
+    {
+      "idx": 12,
+      "retrieval_sim": 0.7698,
+      "tIOU": 0.7164,
+      "mAP": 0.64
+    },
+    {
+      "idx": 13,
+      "retrieval_sim": 0.795,
+      "tIOU": 0.5923,
+      "mAP": 0.25
+    },
+    {
+      "idx": 14,
+      "retrieval_sim": 0.586,
+      "tIOU": 0.397,
+      "mAP": 0.2
+    },
+    {
+      "idx": 15,
+      "retrieval_sim": 0.5982,
+      "tIOU": 0.7678,
+      "mAP": 0.45
+    },
+    {
+      "idx": 16,
+      "retrieval_sim": 0.7116,
+      "tIOU": 0.5205,
+      "mAP": 0.4444
+    },
+    {
+      "idx": 17,
+      "retrieval_sim": 0.6879,
+      "tIOU": 0.3502,
+      "mAP": 0.0833
+    },
+    {
+      "idx": 18,
+      "retrieval_sim": 0.7088,
+      "tIOU": 0.75,
+      "mAP": 0.75
+    },
+    {
+      "idx": 19,
+      "retrieval_sim": 0.5611,
+      "tIOU": 0.7193,
+      "mAP": 0.75
+    },
+    {
+      "idx": 20,
+      "retrieval_sim": 0.5881,
+      "tIOU": 0.6111,
+      "mAP": 0.3333
+    },
+    {
+      "idx": 21,
+      "retrieval_sim": 0.7646,
+      "tIOU": 0.8172,
+      "mAP": 1.0
+    },
+    {
+      "idx": 22,
+      "retrieval_sim": 0.6744,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 23,
+      "retrieval_sim": 0.6437,
+      "tIOU": 0.7193,
+      "mAP": 0.6
+    },
+    {
+      "idx": 24,
+      "retrieval_sim": 0.6583,
+      "tIOU": 0.7193,
+      "mAP": 0.6
+    },
+    {
+      "idx": 25,
+      "retrieval_sim": 0.5824,
+      "tIOU": 0.0227,
+      "mAP": 0.0
+    },
+    {
+      "idx": 26,
+      "retrieval_sim": 0.8244,
+      "tIOU": 0.755,
+      "mAP": 0.8
+    },
+    {
+      "idx": 27,
+      "retrieval_sim": 0.7577,
+      "tIOU": 0.5731,
+      "mAP": 0.2667
+    },
+    {
+      "idx": 28,
+      "retrieval_sim": 0.6997,
+      "tIOU": 0.656,
+      "mAP": 0.8
+    },
+    {
+      "idx": 29,
+      "retrieval_sim": 0.7294,
+      "tIOU": 0.8276,
+      "mAP": 0.64
+    },
+    {
+      "idx": 30,
+      "retrieval_sim": 0.6245,
+      "tIOU": 0.5177,
+      "mAP": 0.3333
+    },
+    {
+      "idx": 31,
+      "retrieval_sim": 0.7857,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 32,
+      "retrieval_sim": 0.9162,
+      "tIOU": 0.3067,
+      "mAP": 0.1111
+    },
+    {
+      "idx": 33,
+      "retrieval_sim": 0.7914,
+      "tIOU": 0.4313,
+      "mAP": 0.0833
+    },
+    {
+      "idx": 34,
+      "retrieval_sim": 0.808,
+      "tIOU": 0.5406,
+      "mAP": 0.25
+    },
+    {
+      "idx": 35,
+      "retrieval_sim": 0.6768,
+      "tIOU": 0.7678,
+      "mAP": 0.45
+    },
+    {
+      "idx": 36,
+      "retrieval_sim": 0.7682,
+      "tIOU": 0.604,
+      "mAP": 0.64
+    },
+    {
+      "idx": 37,
+      "retrieval_sim": 0.8046,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 38,
+      "retrieval_sim": 0.8366,
+      "tIOU": 0.3989,
+      "mAP": 0.2
+    },
+    {
+      "idx": 39,
+      "retrieval_sim": 0.7443,
+      "tIOU": 0.5977,
+      "mAP": 0.3333
+    },
+    {
+      "idx": 40,
+      "retrieval_sim": 0.792,
+      "tIOU": 0.4645,
+      "mAP": 0.2667
+    },
+    {
+      "idx": 41,
+      "retrieval_sim": 0.717,
+      "tIOU": 0.8,
+      "mAP": 0.8
+    },
+    {
+      "idx": 42,
+      "retrieval_sim": 0.7654,
+      "tIOU": 0.5695,
+      "mAP": 0.3333
+    },
+    {
+      "idx": 43,
+      "retrieval_sim": 0.8456,
+      "tIOU": 0.5806,
+      "mAP": 0.2
+    },
+    {
+      "idx": 44,
+      "retrieval_sim": 0.621,
+      "tIOU": 0.6756,
+      "mAP": 0.75
+    },
+    {
+      "idx": 45,
+      "retrieval_sim": 0.5566,
+      "tIOU": 0.5743,
+      "mAP": 0.36
+    },
+    {
+      "idx": 46,
+      "retrieval_sim": 0.6792,
+      "tIOU": 0.5177,
+      "mAP": 0.2667
+    },
+    {
+      "idx": 47,
+      "retrieval_sim": 0.7737,
+      "tIOU": 0.604,
+      "mAP": 0.64
+    },
+    {
+      "idx": 48,
+      "retrieval_sim": 0.7256,
+      "tIOU": 0.6667,
+      "mAP": 0.6667
+    },
+    {
+      "idx": 49,
+      "retrieval_sim": 0.7957,
+      "tIOU": 0.7011,
+      "mAP": 1.0
+    },
+    {
+      "idx": 50,
+      "retrieval_sim": 0.7171,
+      "tIOU": 0.6695,
+      "mAP": 1.0
+    },
+    {
+      "idx": 51,
+      "retrieval_sim": 0.7459,
+      "tIOU": 0.6545,
+      "mAP": 0.2
+    },
+    {
+      "idx": 52,
+      "retrieval_sim": 0.8102,
+      "tIOU": 0.858,
+      "mAP": 0.6667
+    },
+    {
+      "idx": 53,
+      "retrieval_sim": 0.8126,
+      "tIOU": 0.7,
+      "mAP": 0.64
+    },
+    {
+      "idx": 54,
+      "retrieval_sim": 0.8128,
+      "tIOU": 0.6783,
+      "mAP": 0.64
+    },
+    {
+      "idx": 55,
+      "retrieval_sim": 0.6214,
+      "tIOU": 0.6223,
+      "mAP": 0.25
+    },
+    {
+      "idx": 56,
+      "retrieval_sim": 0.729,
+      "tIOU": 0.8333,
+      "mAP": 0.6
+    },
+    {
+      "idx": 57,
+      "retrieval_sim": 0.4982,
+      "tIOU": 0.4498,
+      "mAP": 0.2
+    },
+    {
+      "idx": 58,
+      "retrieval_sim": 0.6194,
+      "tIOU": 0.6983,
+      "mAP": 0.2
+    },
+    {
+      "idx": 59,
+      "retrieval_sim": 0.8932,
+      "tIOU": 0.4533,
+      "mAP": 0.16
+    },
+    {
+      "idx": 60,
+      "retrieval_sim": 0.8175,
+      "tIOU": 0.8333,
+      "mAP": 0.5
+    },
+    {
+      "idx": 61,
+      "retrieval_sim": 0.7364,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 62,
+      "retrieval_sim": 0.5125,
+      "tIOU": 0.6703,
+      "mAP": 0.45
+    },
+    {
+      "idx": 63,
+      "retrieval_sim": 0.7301,
+      "tIOU": 0.4913,
+      "mAP": 0.2667
+    },
+    {
+      "idx": 64,
+      "retrieval_sim": 0.8172,
+      "tIOU": 0.4682,
+      "mAP": 0.0667
+    },
+    {
+      "idx": 65,
+      "retrieval_sim": 0.6517,
+      "tIOU": 0.46,
+      "mAP": 0.2
+    },
+    {
+      "idx": 66,
+      "retrieval_sim": 0.7376,
+      "tIOU": 0.8333,
+      "mAP": 0.75
+    },
+    {
+      "idx": 67,
+      "retrieval_sim": 0.5759,
+      "tIOU": 0.7167,
+      "mAP": 0.1
+    },
+    {
+      "idx": 68,
+      "retrieval_sim": 0.609,
+      "tIOU": 0.4959,
+      "mAP": 0.0833
+    },
+    {
+      "idx": 69,
+      "retrieval_sim": 0.8242,
+      "tIOU": 0.6645,
+      "mAP": 0.5625
+    },
+    {
+      "idx": 70,
+      "retrieval_sim": 0.6911,
+      "tIOU": 0.5938,
+      "mAP": 0.2
+    },
+    {
+      "idx": 71,
+      "retrieval_sim": 0.8479,
+      "tIOU": 0.9167,
+      "mAP": 0.6667
+    },
+    {
+      "idx": 72,
+      "retrieval_sim": 0.7774,
+      "tIOU": 0.6362,
+      "mAP": 0.0833
+    },
+    {
+      "idx": 73,
+      "retrieval_sim": 0.5972,
+      "tIOU": 0.7009,
+      "mAP": 0.36
+    },
+    {
+      "idx": 74,
+      "retrieval_sim": 0.7639,
+      "tIOU": 0.6788,
+      "mAP": 0.2667
+    },
+    {
+      "idx": 75,
+      "retrieval_sim": 0.5832,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 76,
+      "retrieval_sim": 0.6452,
+      "tIOU": 0.6162,
+      "mAP": 0.3333
+    },
+    {
+      "idx": 77,
+      "retrieval_sim": 0.7192,
+      "tIOU": 0.6276,
+      "mAP": 0.45
+    },
+    {
+      "idx": 78,
+      "retrieval_sim": 0.8024,
+      "tIOU": 0.9145,
+      "mAP": 0.8
+    },
+    {
+      "idx": 79,
+      "retrieval_sim": 0.7551,
+      "tIOU": 0.88,
+      "mAP": 0.8
+    },
+    {
+      "idx": 80,
+      "retrieval_sim": 0.6141,
+      "tIOU": 0.7993,
+      "mAP": 1.0
+    },
+    {
+      "idx": 81,
+      "retrieval_sim": 0.698,
+      "tIOU": 0.7751,
+      "mAP": 0.36
+    },
+    {
+      "idx": 82,
+      "retrieval_sim": 0.7071,
+      "tIOU": 0.8,
+      "mAP": 0.8
+    },
+    {
+      "idx": 83,
+      "retrieval_sim": 0.8117,
+      "tIOU": 0.6667,
+      "mAP": 0.1
+    },
+    {
+      "idx": 84,
+      "retrieval_sim": 0.8008,
+      "tIOU": 0.856,
+      "mAP": 1.0
+    },
+    {
+      "idx": 85,
+      "retrieval_sim": 0.7408,
+      "tIOU": 0.6462,
+      "mAP": 0.45
+    },
+    {
+      "idx": 86,
+      "retrieval_sim": 0.7308,
+      "tIOU": 0.6831,
+      "mAP": 0.64
+    },
+    {
+      "idx": 87,
+      "retrieval_sim": 0.6958,
+      "tIOU": 0.7533,
+      "mAP": 0.75
+    },
+    {
+      "idx": 88,
+      "retrieval_sim": 0.8543,
+      "tIOU": 0.8289,
+      "mAP": 1.0
+    },
+    {
+      "idx": 89,
+      "retrieval_sim": 0.7159,
+      "tIOU": 0.5667,
+      "mAP": 0.2
+    },
+    {
+      "idx": 90,
+      "retrieval_sim": 0.837,
+      "tIOU": 0.8276,
+      "mAP": 0.64
+    },
+    {
+      "idx": 91,
+      "retrieval_sim": 0.8093,
+      "tIOU": 0.6382,
+      "mAP": 0.16
+    },
+    {
+      "idx": 92,
+      "retrieval_sim": 0.7542,
+      "tIOU": 1.0,
+      "mAP": 1.0
+    },
+    {
+      "idx": 93,
+      "retrieval_sim": 0.6445,
+      "tIOU": 0.82,
+      "mAP": 1.0
+    },
+    {
+      "idx": 94,
+      "retrieval_sim": 0.712,
+      "tIOU": 0.7667,
+      "mAP": 0.64
+    },
+    {
+      "idx": 95,
+      "retrieval_sim": 0.6122,
+      "tIOU": 0.4707,
+      "mAP": 0.16
+    },
+    {
+      "idx": 96,
+      "retrieval_sim": 0.7239,
+      "tIOU": 0.5453,
+      "mAP": 0.2
+    },
+    {
+      "idx": 97,
+      "retrieval_sim": 0.8103,
+      "tIOU": 0.4778,
+      "mAP": 0.1111
+    }
+  ]
+}

eval.md

@@ -0,0 +1,63 @@
+# v1.0 Verification Results
+
+## Summary
+
+| Metric | Value | Threshold | Result |
+|--------|-------|-----------|--------|
+| retrieval_similarity | 0.7239 | >= 0.60 | PASS |
+| tIOU | 0.6700 | >= 0.40 | PASS |
+| mAP@0.5 | 0.5092 | >= 0.50 | PASS |
+
+**Overall: PASS — v1.0 ACCEPTED**
+
+## Details
+
+- Checkpoint: `outputs/checkpoint-770` (step 770, eval_loss 0.5272)
+- Predictions: `checkpoints/v21_semantic_boundaries/predictions.jsonl` (98 samples)
+- Embedder: OpenAI text-embedding-3-small (production-aligned)
+- Date: 2026-03-19
+
+### Retrieval Similarity
+
+- Mean: 0.7239
+- Median: 0.7355
+- Min: 0.4982, Max: 0.9162
+- Excellent (>= 0.80): 22/98
+- Good (>= 0.60): 87/98
+- Acceptable (>= 0.40): 98/98
+
+### Temporal IoU (tIOU)
+
+- Mean: 0.6700
+- Median: 0.6703
+- Predictions with timestamps: 98/98
+- References with timestamps: 98/98
+
+### mAP@0.5
+
+- Mean: 0.5092
+
+## Production Embeddings (prior run)
+
+For reference, the prior eval with OpenAI text-embedding-3-small on 20 pairs showed:
+- mean_similarity: 0.6781
+- production_ready: true
+
+## Failure Analysis
+
+All metrics pass. v1.0 is accepted for production.
+
+## Per-Sample Scores (first 10)
+
+| # | retrieval_sim | tIOU | mAP@0.5 | pred_events | ref_events |
+|---|--------------|------|---------|-------------|------------|
+| 0 | 0.6549 | 0.5277 | 0.4500 | 5 | 4 |
+| 1 | 0.7739 | 0.8080 | 1.0000 | 5 | 5 |
+| 2 | 0.6433 | 1.0000 | 1.0000 | 4 | 4 |
+| 3 | 0.7473 | 0.6228 | 0.4500 | 5 | 4 |
+| 4 | 0.7774 | 0.5406 | 0.2500 | 4 | 4 |
+| 5 | 0.7924 | 0.3868 | 0.0500 | 4 | 5 |
+| 6 | 0.7626 | 0.7779 | 0.7500 | 4 | 3 |
+| 7 | 0.6476 | 0.8060 | 0.7500 | 4 | 3 |
+| 8 | 0.7355 | 0.7520 | 0.6400 | 5 | 5 |
+| 9 | 0.8859 | 0.6579 | 0.5625 | 4 | 4 |

predictions.jsonl

@@ -0,0 +1,98 @@
+{"example_idx": 0, "prediction": "**2020.6s:** 12 people | A man with a backpack and a woman in a striped shirt stand in the foreground of a lush indoor...\n**2021.9s:** 15 people | The camera pans right to reveal a large decorative garden statue of a woman in a pink dress\n**2023.1s:** 14 people | A man in a white shirt bends over a map in the foreground while a woman and child stand nearby\n**2024.4s:** 12 people | A woman in a white shirt and a young girl in a blue dress enter the immediate foreground\n**2026.9s:** 10 people | A woman in a green dress walks toward the left as the camera continues to pan right", "reference": "**2020.6s:** 12 people | a couple on the left is seen attending to an open suitcase on the walkway\n**2021.9s:** 8 people | Camera pans left to focus on the couple with the suitcase as a woman in a floral shirt walks toward...\n**2024.4s:** 15 people | scene static, no change\n**2029.4s:** 18 people | Final pan past a large white tree trunk reveals a crowded upper-level walkway overlooking an escalator and the lower atrium"}
+{"example_idx": 1, "prediction": "**890.6s:** 20+ people | Wide view of an airport terminal with a massive flight information board and a \"Thai Lion Air\" booth; travelers are...\n**891.9s:** 12 people | The camera pans left, focusing on a couple in the foreground walking past a \"BorPloen\" storefront\n**894.4s:** 15 people | A man in a black t-shirt and a woman in a grey hoodie walk across the frame in front of...\n**896.9s:** 18 people | scene static, no change\n**899.4s:** 10 people | The scene shifts to a \"Boarding Information\" area where passengers wait near stacks of grey cart covers", "reference": "**890.6s:** 25 people | Passengers gathered near a large international arrival flight board and information desks in the terminal hall\n**893.1s:** 12 people | scene static, no change\n**894.4s:** 16 people | Scene shows a \"SEA GAMES\" promotional wall and information booth with passengers walking by\n**896.9s:** 11 people | An elderly man with a cane moves into the foreground as the camera pans toward a \"Grab\" advertisement\n**899.4s:** 15 people | Camera reveals a check-in area with a \"Restricted Area\" sign and passengers standing in a queue"}
+{"example_idx": 2, "prediction": "**2970.6s:** 12 people | a woman in black and a child in a blue shirt walk toward the center\n**2973.1s:** 10 people | A woman in a floral shirt and a man in a blue shirt move into the close foreground on the...\n**2975.6s:** 8 people | A woman in a grey shirt pulls a silver suitcase toward the camera while a young girl in a purple...\n**2978.1s:** 7 people | A woman in a black shirt and a young child in a blue shirt walk through the center of the...", "reference": "**2970.6s:** 12 people | Initial view of the airport walkway featuring a Disney Adventure display and travelers moving toward a decorated blue arch\n**2973.1s:** 10 people | The child in red has left the foreground, and a man in a grey polo shirt is now prominently positioned...\n**2975.6s:** 8 people | scene static, no change\n**2978.1s:** 9 people | A boy in a blue shirt and a woman in a white-and-pink shirt enter the immediate foreground from the left..."}
+{"example_idx": 3, "prediction": "**1080.6s:** 9 people | Customers are seated at tables in a cafe with large windows overlooking an airport runway\n**1083.1s:** 11 people | A staff member in a blue uniform enters the frame from the left as more people gather in the terminal\n**1084.4s:** 15 people | The scene shifts to a busy terminal walkway with a McDonald's restaurant visible in the background\n**1086.9s:** 18 people | A woman with tattoos on her arms and a man with a backpack enter the foreground from the left\n**1089.4s:** 20 people | The crowd moves forward through the terminal past the McDonald's and a Burger King sign", "reference": "**1080.6s:** 10 people | patrons are seated at tables, including a woman with a black suitcase in the center and a couple in the...\n**1084.4s:** 15 people | travelers walk with luggage past a McDonald's, and cleaning staff is visible on the right; Significant scene change to a...\n**1085.6s:** 12 people | people are standing and walking throughout the terminal area\n**1086.9s:** 14 people | A new group of travelers enters the close foreground on the left, including a woman in a \"WHITE FOX\" tank..."}
+{"example_idx": 4, "prediction": "**1460.6s:** 6 people | A man in a white shirt walks across the foreground past a \"KANTIN\" food service area\n**1463.1s:** 7 people | A man with a large green backpack enters the frame, walking toward the left\n**1465.6s:** 8 people | scene static, no change\n**1468.1s:** 12 people | The camera pans left to reveal a \"Home-Made Sweets\" shop and a group of people walking down a corridor in...", "reference": "**1460.6s:** 7 people | A man in a white shirt walks past the \"Rook & Tarry\" bar in a terminal\n**1464.4s:** 5 people | The view pans right to reveal an elevator entrance and several \"Soup Curry\" restaurant posters\n**1468.1s:** 8 people | The camera moves into an open corridor featuring flight information displays and an escalator\n**1469.4s:** 10 people | Multiple travelers walk down a corridor lined with food advertisements and blue storage crates"}
+{"example_idx": 5, "prediction": "**1200.6s:** 15+ people | Tourists gather around a large Christmas tree and a indoor waterfall\n**1203.1s:** 15+ people | A man in a grey shirt and white shorts enters the foreground, moving toward the right\n**1205.6s:** 15+ people | A woman in a bright orange shirt enters the scene from the left, walking toward the Christmas tree\n**1208.1s:** 15+ people | The camera pans left to reveal a \"Shake Shack\" storefront with a queue of people waiting to enter", "reference": "**1200.6s:** 15 people | Crowded indoor garden area near a large decorated Christmas tree\n**1201.9s:** 15 people | A man in a maroon shirt and backpack walks across the foreground, momentarily obscuring the center of the frame\n**1204.4s:** 12 people | focus shifts to a woman in an orange shirt and a girl in a blue shirt standing nearby\n**1206.9s:** 14 people | The girl in the blue shirt walks across the foreground with a luggage cart while other tourists pose for photos...\n**1209.4s:** 10 people | scene static, no change"}
+{"example_idx": 6, "prediction": "**560.6s:** 18 people | Travelers rest on benches and walk through a lush indoor garden featuring a large decorated Christmas tree and a \"KANTIN\"...\n**563.1s:** 22 people | scene static, no change\n**565.6s:** 25 people | scene static, no change\n**568.1s:** 28 people | The perspective shifts further right to reveal the \"Garden and Bar\" area, with people gathering near a lounge and the...", "reference": "**560.6s:** 18 people | Travellers rest with luggage on the left while others walk near a large Christmas tree and a festive wooden cabin...\n**564.4s:** 22 people | scene static, no change\n**568.1s:** 28 people | The scene shifts significantly to focus on an escalator area and a crowded walkway overlooking the multi-level vertical forest"}
+{"example_idx": 7, "prediction": "**1620.6s:** 40+ people | High-angle view of a multi-level shopping mall with people on escalators and various retail floors\n**1623.1s:** 45+ people | scene static, no change\n**1625.6s:** 50+ people | scene static, no change\n**1629.4s:** 55+ people | scene static, no change", "reference": "**1620.6s:** 45 people | A wide-angle view of a multi-story shopping mall with people walking on various levels and using an escalator in the...\n**1624.4s:** 40 people | The camera tilts down to focus on the lower level, showing a LEGO display stand and a large group of...\n**1629.4s:** 55 people | The camera tilts back up to a wider shot, revealing the upper levels again and several new groups of people..."}
+{"example_idx": 8, "prediction": "**540.6s:** 15+ people | A woman in a purple shirt walks away through a busy food court while a staff member in a blue...\n**541.9s:** 15+ people | scene static, no change\n**543.1s:** 15+ people | An elderly woman in a black floral shirt and a young girl enter the foreground from the right as the...\n**546.9s:** 15+ people | scene static, no change\n**549.4s:** 15+ people | scene static, no change", "reference": "**540.6s:** 18 people | A woman in a purple shirt walks through a crowded airport dining area past a man in a wheelchair with...\n**541.9s:** 22 people | scene static, no change\n**546.9s:** 10 people | scene static, no change\n**548.1s:** 8 people | A \"Potato Corner\" food stall comes into view as the woman in purple continues walking toward it\n**549.4s:** 11 people | scene static, no change"}
+{"example_idx": 9, "prediction": "**1590.6s:** 15 people | Wide view of an airport terminal with passengers sitting in a waiting area and others walking in the distance\n**1594.4s:** 12 people | scene static, no change\n**1596.9s:** 10 people | The woman in white walks past a shop on the left while a man with a suitcase stands in the...\n**1599.4s:** 12 people | A robotic floor cleaner is now visible in the foreground as the woman in white exits the frame", "reference": "**1590.6s:** 9 people | Passengers walk through the terminal and sit in the waiting area while a robotic cleaner operates in the background\n**1593.1s:** 11 people | Two passengers, one in a white outfit and another in a tan shirt with a suitcase, walk toward the foreground\n**1595.6s:** 12 people | scene static, no change\n**1599.4s:** 10 people | scene static, no change"}
+{"example_idx": 10, "prediction": "**530.6s:** 4 people | A robotic cleaner with a cartoon bear head stands in a wide airport corridor while a man in black walks...\n**531.9s:** 6 people | scene static, no change\n**534.4s:** 5 people | scene static, no change\n**536.9s:** 12 people | The camera reveals a busy food court seating area with travelers, including a woman with a stroller, eating at tables...\n**539.4s:** 15 people | A staff member in a purple uniform walks into the foreground of the bustling food court", "reference": "**530.6s:** 6 people | A wide hallway view with a cleaning robot in the foreground and several people walking in the distance toward a...\n**534.4s:** 7 people | As the camera approaches the food court, a staff member in a purple shirt walks across the frame while a...\n**538.1s:** 10 people | A woman in a white jacket enters the foreground from the right, passing by the Burger King counter where a...\n**539.4s:** 12 people | The staff member in purple returns to the foreground, and the background reveals a crowded dining area with many people..."}
+{"example_idx": 11, "prediction": "**670.6s:** 12 people | a woman and child walk toward the left while others sit or stand near the food court entrance\n**674.4s:** 10 people | The camera pans right, focusing on a man with a large backpack standing near the railing as the woman and...\n**676.9s:** 5 people | scene static, no change\n**679.4s:** 15 people | The camera completes its pan to show the man with the backpack in the foreground while multiple people are seen...", "reference": "**670.6s:** 11 people | A woman in white pushes a stroller through a transit hall; Travelers on the right walk up a ramp past...\n**671.9s:** 10 people | The woman with the stroller moves significantly closer to the foreground\n**673.1s:** 9 people | The woman and stroller exit the frame as the camera begins to pan right\n**674.4s:** 8 people | scene static, no change\n**679.4s:** 11 people | A man with a large backpack and a travel bag enters the foreground from the right as the camera continues..."}
+{"example_idx": 12, "prediction": "**1110.6s:** 6 people | View of a terminal corridor featuring \"Pang Chat\" and \"Dean & DeLuca\" shops with people walking and standing near the...\n**1111.9s:** 7 people | scene static, no change\n**1113.1s:** 15 people | A woman in a pink shirt appears in the foreground, pushing a luggage cart through the corridor\n**1114.4s:** 12 people | The camera pans right, showing the woman with the cart moving toward the right side of the frame as more...\n**1116.9s:** 14 people | scene static, no change", "reference": "**1110.6s:** 8 people | a woman browses bags while a man in a black jacket walks toward the camera\n**1111.9s:** 12 people | a man in a brown jacket pulls a suitcase toward the camera\n**1113.1s:** 17 people | A man in a pink shirt enters the foreground center, pushing a luggage cart toward the camera\n**1116.9s:** 21 people | focus shifts to a man in a white shirt at the cafe counter and a growing crowd in the background\n**1119.4s:** 21 people | A man in a white shirt and tan pants enters the scene from the right, walking toward the distant gates"}
+{"example_idx": 13, "prediction": "**610.6s:** 15 people | People walk through a lush indoor garden past the \"Polo Ralph Lauren\" store\n**613.1s:** 22 people | A young boy in a black shirt runs across the foreground\n**615.6s:** 35 people | The camera pans to reveal a large crowd gathered at a viewing gallery near an escalator\n**618.1s:** 45 people | A massive crowd is visible on multiple levels as the camera reveals the full scale of the indoor waterfall", "reference": "**610.6s:** 15 people | A crowd gathers in a lush indoor garden near a rustic \"Polo Ralph Lauren\" cabin and a large decorated Christmas...\n**613.1s:** 25 people | scene static, no change\n**616.9s:** 50 people | A significant shift reveals a massive indoor waterfall in the background with a large crowd gathered along a glass railing...\n**619.4s:** 55 people | The pan concludes with a full view of the waterfall cascading into the center, with visitors appearing on escalators and..."}
+{"example_idx": 14, "prediction": "**1600.6s:** 18 people | A man in a white t-shirt and black shorts sits on benches in an airport terminal, looking at his phone\n**1601.9s:** 20 people | scene static, no change\n**1603.1s:** 22 people | scene static, no change\n**1608.1s:** 25 people | A woman in a white jacket and a man in a blue shirt walk toward the camera through the terminal...", "reference": "**1600.6s:** 12 people | Initial view of an airport terminal waiting area featuring an \"Au We White Coffee\" stall, with a floor-cleaning robot in...\n**1603.1s:** 14 people | scene static, no change\n**1604.4s:** 15 people | scene static, no change\n**1606.9s:** 12 people | scene static, no change\n**1609.4s:** 13 people | The \"Starbucks\" stall is now fully visible on the right with several people nearby, as other travelers traverse the open..."}
+{"example_idx": 15, "prediction": "**2910.6s:** 12 people | Visitors walk through a lush indoor garden featuring a large glass-and-steel structure and tropical foliage\n**2913.1s:** 14 people | A woman in a yellow shirt moves into the close foreground while a man in a white shirt and baseball...\n**2914.4s:** 15 people | A young girl in a blue shirt runs across the path as a woman in a white dress enters the...\n**2916.9s:** 12 people | The woman in the white dress moves to the center of the frame as the camera pans, following her movement...\n**2919.4s:** 14 people | scene static, no change", "reference": "**2910.6s:** 12 people | A group of people waits in line at the entrance of a botanical attraction, with a stroller parked in the...\n**2913.1s:** 15 people | scene static, no change\n**2914.4s:** 14 people | scene static, no change\n**2918.1s:** 12 people | scene static, no change"}
+{"example_idx": 16, "prediction": "**710.6s:** 1 people | A man in a blue shirt walks across the terminal with a black suitcase\n**713.1s:** 1 people | scene static, no change\n**716.9s:** 40+ people | The camera pans right, revealing a large crowd of passengers and the \"Arrivals\" section on the lower level of the...", "reference": "**710.6s:** 0 people | Empty airport hallway with a marble pillar and a luggage cart; busy terminal levels are visible in the background\n**711.9s:** 1 people | A man in a black shirt enters the hallway from the left, pulling a black suitcase\n**718.1s:** 0 people | scene static, no change"}
+{"example_idx": 17, "prediction": "**570.6s:** 6 people | Travelers walk through a corridor past a large pillar and a \"Visit Iceland\" poster\n**571.9s:** 15 people | A group of people gathers near a \"Castel\" booth in a busy terminal\n**573.1s:** 60+ people | High-angle view of a massive crowd in a large airport concourse with various booths and seating areas", "reference": "**570.6s:** 40+ people | Camera pans right from behind a large pillar, revealing a crowded station hall with travelers sitting on central benches\n**573.1s:** 70+ people | A traveler in a white jacket pulls a green suitcase across the tiled floor as the camera reveals the full...\n**574.4s:** 75+ people | A man in a grey hoodie and black backpack enters the lower-left foreground, walking toward the center of the hall\n**576.9s:** 80+ people | The camera pans further right to reveal a large glass elevator structure and several new pedestrians walking in the distance"}
+{"example_idx": 18, "prediction": "**1690.6s:** 0 people | View of an airport runway through large glass windows\n**1694.4s:** 0 people | scene static, no change\n**1696.9s:** 0 people | A Japan Airlines airplane is taxiing across the foreground of the airfield", "reference": "**1690.6s:** 0 people | View from inside an airport terminal through glass windows overlooking the tarmac at dusk\n**1694.4s:** 0 people | Scene transition to a direct view of the runway where a white airplane is visible in the distance\n**1695.6s:** 1 people | A ground crew member appears near the bottom right as the wing of a second airplane enters the frame\n**1696.9s:** 2 people | A large airplane becomes fully visible on the right tarmac with two ground crew members standing nearby"}
+{"example_idx": 19, "prediction": "**610.6s:** 6 people | A man in a grey shirt and a woman sit at a table in the foreground, while others queue at...\n**613.1s:** 5 people | scene static, no change\n**616.9s:** 3 people | scene static, no change\n**619.4s:** 4 people | A man in a black shirt stands at the Subway counter as another man walks down the hallway on the...", "reference": "**610.6s:** 8 people | wide view of a food court featuring Pepper Lunch and Subway with several patrons at counters and tables\n**614.4s:** 3 people | zoomed-in view of the Pepper Lunch counter as a staff member prepares an order\n**619.4s:** 2 people | focused view of the Subway counter as a man with a silver suitcase approaches to order"}
+{"example_idx": 20, "prediction": "**2210.6s:** 12 people | A woman in a white hoodie walks through a lush indoor garden restaurant with other patrons seated at tables\n**2213.1s:** 14 people | The woman in the white hoodie moves further into the scene as more people appear in the background seating area\n**2215.6s:** 15 people | A man in a purple shirt and a woman in a black top enter the mid-ground, walking across the patio\n**2219.4s:** 12 people | scene static, no change", "reference": "**2210.6s:** 11 people | scene static, no change\n**2211.9s:** 10 people | scene static, no change\n**2213.1s:** 9 people | scene static, no change"}
+{"example_idx": 21, "prediction": "**1130.6s:** 15+ people | Passengers wait in a terminal hallway; a couple walks away from the camera toward gates 1-6\n**1135.6s:** 15+ people | scene static, no change\n**1139.4s:** 15+ people | scene static, no change", "reference": "**1130.6s:** 12 people | Travelers sit on hallway benches while others walk toward distant gates past a Christmas tree\n**1136.9s:** 11 people | scene static, no change\n**1139.4s:** 10 people | Subtitles appear at the bottom as the camera nears gate signs and a \"Free Charging\" station"}
+{"example_idx": 22, "prediction": "**390.6s:** 50+ people | A man in a white hat and a woman with a black backpack stand in the foreground as a large...\n**393.1s:** 50+ people | The foreground women exit the frame, revealing a boy in a yellow jersey and a girl in a red shirt...\n**395.6s:** 50+ people | The camera tilts upward, revealing the massive glass dome ceiling and the upper section of the waterfall\n**398.1s:** 50+ people | The view pans up to show the top of the waterfall cascading into the large circular pool below", "reference": "**390.6s:** 25 people | Visitors gather near the base of the indoor waterfall\n**393.1s:** 40 people | The camera tilts up, revealing tourists with large red and blue suitcases in the foreground and a growing crowd around...\n**395.6s:** 65 people | As the view expands, the pedestrian bridge and upper levels of the indoor forest become visible, showing a high density...\n**398.1s:** 85 people | scene static, no change"}
+{"example_idx": 23, "prediction": "**1240.6s:** 25 people | scene static, no change\n**1243.1s:** 20 people | The camera pans right, bringing a man in a black shirt and backpack into the close foreground\n**1245.6s:** 18 people | A woman in a blue shirt enters the foreground from the right as the camera continues to track right\n**1248.1s:** 22 people | A man in a patterned black shirt and a woman in a blue hijab enter the frame from the right...\n**1249.4s:** 30 people | The camera reveals a large digital directory sign and a second Starbucks location as the crowd moves through the concourse", "reference": "**1240.6s:** 40+ people | Busy mall corridor with shoppers walking past a \"HANDS\" store and a Starbucks in the background\n**1244.4s:** 35+ people | Approaching a central escalator where people are descending; the perspective on the surrounding shops shifts closer\n**1249.4s:** 25+ people | View transitions into an open atrium featuring a large vertical advertisement banner"}
+{"example_idx": 24, "prediction": "**40.6s:** 12 people | Wide view of an airport terminal with a pigeon on the floor and travelers with luggage walking in the distance\n**43.1s:** 18 people | The camera pans left to reveal a crowded walkway and a \"CardX\" credit card booth on the left side\n**45.6s:** 14 people | Two women in red dresses enter the foreground from the left as the camera approaches the Starbucks counter\n**46.9s:** 16 people | One of the women in red walks out of frame while a woman in a black dress enters from the...\n**49.4s:** 15 people | A man in a black cap and a woman in a black dress walk across the foreground in front of...", "reference": "**40.6s:** 14 people | A man in a dark suit pulls a suitcase toward the right while a monk in maroon robes walks in...\n**44.4s:** 10 people | scene static, no change\n**49.4s:** 20 people | A woman with a backpack enters the right foreground as the camera focuses on a crowded Starbucks seating area and..."}
+{"example_idx": 25, "prediction": "**3000.6s:** 12 people | scene static, no change", "reference": "**2990.6s:** 15 people | A man in a grey polo shirt and a woman in a blue floral dress stand in the foreground of...\n**2993.1s:** 12 people | The foreground subjects change as a woman in a black dress and a young girl in a light blue dress...\n**2995.6s:** 10 people | scene static, no change\n**2998.1s:** 8 people | scene static, no change"}
+{"example_idx": 26, "prediction": "**1970.6s:** 12 people | A group of people walk through a mall corridor past \"The Hub\" and \"Jack's Place\" cafes\n**1973.1s:** 15 people | The camera pans left, revealing a \"Red Fire Brigade\" themed train display and more shoppers\n**1975.6s:** 8 people | scene static, no change\n**1976.9s:** 6 people | Two women are seated at a table outside \"Jack's Place\" as the camera continues to pan\n**1979.4s:** 10 people | The view shifts to an escalator where several people are ascending to the next level of the mall", "reference": "**1970.6s:** 22 people | Initial view of a busy mall lobby with people queuing at \"LOBBY-E\" and walking past \"Andersen's\" bakery\n**1971.9s:** 28 people | scene static, no change\n**1976.9s:** 12 people | Perspective shifts to the entrance of \"Jack's Place\" and a \"Sanrio\" store featuring a large Hello Kitty digital display\n**1979.4s:** 4 people | Scene changes significantly as the camera moves onto an upward escalator, showing people standing on the steps ahead"}
+{"example_idx": 27, "prediction": "**120.6s:** 1 people | Close-up of a self-service check-in kiosk screen displaying options for AirAsia and other airlines\n**121.9s:** 7 people | The camera pans right, revealing a wider view of the airport terminal with several passengers walking in the distance\n**124.4s:** 15 people | A large informational text overlay appears at the bottom of the screen as the camera continues to pan, showing a...\n**125.6s:** 20 people | The scene reveals a currency exchange area with \"Check-in\" and \"Exchange\" signs, and a significant increase in visible passenger traffic\n**129.4s:** 25 people | The camera pans further right to show a long queue of passengers waiting at the currency exchange counter", "reference": "**120.6s:** 3 people | Close-up of a self-service check-in kiosk at the airport with distant staff at counters\n**124.4s:** 12 people | Camera pans right to reveal a traveler sitting on a luggage cart in the main terminal concourse\n**126.9s:** 15 people | Perspective shifts toward currency exchange booths and a designated counter for people with disabilities"}
+{"example_idx": 28, "prediction": "**730.6s:** 0 people | scene static, no change\n**735.6s:** 1 people | A man wearing a white mask and a black backpack enters the scene, pulling a blue suitcase through the shop...\n**736.9s:** 4 people | The man with the suitcase moves toward the left as other people become visible in the background shopping area\n**739.4s:** 6 people | The man with the suitcase exits to the left, leaving a woman in a black shirt visible in the background", "reference": "**730.6s:** 0 people | Close-up view of the \"SKY MUSEUM\" entrance wall with a monitor displaying aircraft footage\n**733.1s:** 0 people | The camera pans left, revealing a model airplane and a rope barrier in a brightly lit museum hallway\n**734.4s:** 12 people | Wide shot of the museum interior showing shop counters, various exhibits, and several visitors walking through the area\n**736.9s:** 10 people | A man wearing a face mask and a backpack enters the foreground from the left, pulling a grey suitcase past...\n**739.4s:** 8 people | The man with the suitcase moves deeper into the museum as the camera pans further right to reveal additional exhibit..."}
+{"example_idx": 29, "prediction": "**1100.6s:** 15 people | Travelers walk through a busy airport corridor past McDonald's and a PangCha stall\n**1101.9s:** 12 people | scene static, no change\n**1103.1s:** 10 people | A woman in a black top walks away with a suitcase as the camera pans right, bringing a PangCha customer...\n**1106.9s:** 7 people | scene static, no change\n**1109.4s:** 9 people | A woman in a floral shirt appears in the foreground looking at her phone as the camera reveals more of...", "reference": "**1100.6s:** 15 people | Travelers stand and walk in a wide airport corridor near retail shops and seating areas\n**1101.9s:** 12 people | A woman in a pink shirt walks quickly across the foreground from right to left\n**1104.4s:** 10 people | scene static, no change\n**1106.9s:** 8 people | scene static, no change\n**1109.4s:** 9 people | A traveler browses merchandise at the \"Tastes of Thailand\" entrance while staff work behind the tea counter"}
+{"example_idx": 30, "prediction": "**510.6s:** 35 people | High-angle view of a busy airport terminal with passengers walking through the hall and others gathered near check-in counters\n**513.1s:** 35 people | scene static, no change\n**516.9s:** 35 people | scene static, no change\n**519.4s:** 35 people | scene static, no change", "reference": "**510.6s:** 24 people | a woman in white walks towards the foreground while others use check-in kiosks\n**514.4s:** 22 people | A man in a blue shirt moves into the center foreground as the flow of passengers continues through the hall\n**518.1s:** 28 people | A large group of travelers walks down the main corridor, and several new passengers appear at the self-service kiosks on..."}
+{"example_idx": 31, "prediction": "**140.6s:** 15 people | People walk through a bright mall corridor near an escalator and a \"HANDS\" store\n**143.1s:** 14 people | A woman in a grey coat walks toward the camera as others continue through the shopping area\n**145.6s:** 16 people | A man in a white t-shirt enters from the right while a man in a grey hoodie walks away\n**148.1s:** 15 people | A man in a grey t-shirt walks toward the camera as the crowd moves deeper into the mall\n**149.4s:** 14 people | A woman with a green backpack appears on the right as the man in the grey t-shirt moves into the...", "reference": "**140.6s:** 18 people | Pedestrians walk through a mall concourse past an escalator\n**143.1s:** 20 people | scene static, no change\n**145.6s:** 15 people | A woman pushing a blue luggage trolley enters the foreground, followed by a man in a light green shirt\n**148.1s:** 12 people | A woman in a black shirt and cap pushing a trolley enters the close foreground, obscuring much of the background...\n**149.4s:** 16 people | A man pushing a luggage trolley while looking at his phone dominates the foreground"}
+{"example_idx": 32, "prediction": "**300.6s:** 50+ people | Wide shot of the indoor waterfall with tourists gathered at the base and along multiple levels of the surrounding garden\n**303.1s:** 30+ people | The camera tilts upward, shifting focus from the crowd to the top of the waterfall and the glass dome ceiling\n**304.4s:** 0 people | scene static, no change", "reference": "**300.6s:** 20+ people | Wide shot of the indoor waterfall base with visitors gathered around the viewing area and on multi-level balconies\n**301.9s:** 10+ people | Camera tilts upward to focus on the top of the central waterfall and the glass dome roof, showing fewer people...\n**309.4s:** 15+ people | scene static, no change"}
+{"example_idx": 33, "prediction": "**30.6s:** 12 people | A man in a grey shirt walks through a food court with patrons seated at tables\n**35.6s:** 18 people | The scene shifts to a wide terminal view where a pigeon stands on the floor as travelers with luggage move...\n**36.9s:** 22 people | A woman in a patterned dress and a man pulling a blue suitcase approach the camera through the terminal walkway", "reference": "**30.6s:** 8 people | A woman walks through a food court past a noodle stall towards a seating area with multiple diners\n**33.1s:** 1 people | Close-up view of four prepared bowls of noodles displayed on a wooden counter with a staff member in the background\n**36.9s:** 20 people | A wide view of an airport terminal hallway featuring a pigeon in the foreground and travelers walking past currency exchange...\n**38.1s:** 22 people | A man in a blazer walks toward the camera while pulling a blue suitcase and talking on a phone, followed..."}
+{"example_idx": 34, "prediction": "**240.6s:** 12 people | Visitors stand on a wooden walkway and a lower stone platform, some with luggage trolleys, overlooking a lush indoor garden...\n**243.1s:** 15 people | A man carrying a young child in a pink dress enters the scene from the right as the camera pans...\n**245.6s:** 14 people | The camera reveals the massive glass dome ceiling and the top of the waterfall as people continue to gather at...\n**248.1s:** 18 people | A significant shift in perspective shows a dense crowd of new visitors in the foreground, including a woman in a...", "reference": "**240.6s:** 10 people | Visitors walk along a lush indoor balcony overlooking a large central waterfall under a glass dome\n**244.4s:** 12 people | The camera moves closer to the crowd, revealing more individuals gathered near the railing to view the waterfall\n**248.1s:** 15 people | Foreground focus shifts to a man in a white shirt and another with a pink backpack as the camera advances...\n**249.4s:** 16 people | A man in a red shirt enters the foreground from the right, holding up a phone to take a photograph"}
+{"example_idx": 35, "prediction": "**1150.6s:** 12 people | A woman in a grey top walks past a \"LOBBYC\" entrance while a man with a red backpack moves toward...\n**1151.9s:** 28 people | The camera pans to a crowded Apple Store area where a man in a grey shirt walks past the glass...\n**1154.4s:** 32 people | A man in a grey shirt enters the close foreground walking right, while a large crowd gathers near the store...\n**1156.9s:** 35 people | A man in a black \"Berkeley\" t-shirt and a woman in a grey shirt walk toward the camera through the...\n**1159.4s:** 38 people | A young boy in a blue shirt enters the foreground from the right as the crowd density in the background...", "reference": "**1150.6s:** 10+ people | scene static, no change\n**1151.9s:** 12+ people | Camera pans left as a man in a black shirt and red backpack walks across the foreground\n**1154.4s:** 25+ people | Camera reveals a crowded retail space\n**1158.1s:** 35+ people | scene static, no change"}
+{"example_idx": 36, "prediction": "**2710.6s:** 6 people | A family group stands on a wooden observation deck around a large tree trunk, with a woman in a white...\n**2713.1s:** 8 people | A man in a black t-shirt and a woman in a floral dress enter the scene from the left, while...\n**2715.6s:** 4 people | The camera pans left, shifting focus from the visitors to the surrounding greenery and the distant waterfall\n**2716.9s:** 0 people | scene static, no change\n**2719.4s:** 0 people | scene static, no change", "reference": "**2710.6s:** 7 people | a man in a light green shirt sits in the foreground\n**2711.9s:** 6 people | The foreground man has left, and a family group of three is now clearly visible posing behind the woman taking...\n**2713.1s:** 7 people | A woman in a brown floral dress enters the scene on the right while the family continues to pose\n**2714.4s:** 9 people | Two more women join the posing group as the woman in the hijab prepares to take their picture\n**2719.4s:** 20+ people | The camera completes its pan, revealing a massive crowd gathered at the base of the waterfall in the distance"}
+{"example_idx": 37, "prediction": "**2650.6s:** 12 people | A purple floral elephant statue sprays water in a lush indoor garden while visitors walk along a curved path\n**2653.1s:** 15 people | The camera pans right to reveal a Disney-themed photo op area featuring Ariel and other Disney characters\n**2655.6s:** 0 people | scene static, no change", "reference": "**2650.6s:** 8 people | A purple floral elephant statue stands beside a pond where people walk along a stone path\n**2653.1s:** 9 people | The camera pans right to reveal travelers with suitcases standing near a winding walkway surrounded by lush greenery\n**2655.6s:** 2 people | scene static, no change"}
+{"example_idx": 38, "prediction": "**1030.6s:** 0 people | View of the \"Super Lounge\" entrance hallway with doors and a fire extinguisher\n**1031.9s:** 0 people | scene static, no change\n**1034.4s:** 1 people | A person with a bright green bag appears, standing by the window\n**1039.4s:** 1 people | scene static, no change", "reference": "**1030.6s:** 0 people | View of an empty hallway leading to the entrance of the \"Super Lounge.\"\n**1033.1s:** 1 people | A person is visible through the glass doors, standing by a window overlooking the airfield\n**1034.4s:** 2 people | Two individuals are now seen in the observation area beyond the automatic doors\n**1035.6s:** 3 people | scene static, no change\n**1036.9s:** 1 people | A person with a bright green bag stands alone by the window, looking out at the airfield"}
+{"example_idx": 39, "prediction": "**2870.6s:** 3 people | A man and woman walk along a wooden path through a lush indoor garden, passing a large ornamental tree on...\n**2873.1s:** 12 people | scene static, no change\n**2874.4s:** 18 people | As the couple continues down the path, more visitors appear on an escalator to the left and in the background...\n**2878.1s:** 25 people | The camera pans left, revealing a crowded escalator and a larger crowd of visitors in the background shopping area", "reference": "**2870.6s:** 4 people | Three individuals walk away on a wooden path through a lush indoor garden\n**2874.4s:** 8 people | An elevated mesh walking net appears on the upper left with several people walking across it\n**2879.4s:** 12 people | Crowd density increases as more visitors become visible on the overhead net and a secondary walkway on the far left"}
+{"example_idx": 40, "prediction": "**810.6s:** 6 people | A man carries a child past large windows while others sit at a cafe with model airplane displays\n**813.1s:** 12 people | View transitions to a lobby area with a large overhead airplane sculpture and travelers with luggage\n**816.9s:** 15 people | A man in a light blue shirt walks across the foreground of a busy terminal waiting area", "reference": "**810.6s:** 8 people | Visitors in an aviation museum view model aircraft and observe the airfield through large windows\n**811.9s:** 7 people | A man in a grey vest walks across the room while others remain at the observation windows\n**814.4s:** 2 people | A close-up of a white biplane model with a visitor's reflection visible in the background glass\n**816.9s:** 20+ people | Scene changes to a busy airport terminal area where travelers with luggage walk past a crowded seating area\n**818.1s:** 20+ people | a man with a black messenger bag walks toward the right"}
+{"example_idx": 41, "prediction": "**1330.6s:** 1 people | Close-up of an Air Malaysia aircraft's nose and cockpit on the tarmac with a ground crew member visible on the...\n**1335.6s:** 0 people | scene static, no change\n**1336.9s:** 0 people | scene static, no change\n**1339.4s:** 0 people | scene static, no change", "reference": "**1330.6s:** 3 people | The nose of an AirAsia airplane is visible with two pilots in the cockpit and a ground crew member signaling...\n**1333.1s:** 2 people | scene static, no change\n**1335.6s:** 2 people | The airplane continues moving, revealing a large tiger graphic on the fuselage\n**1336.9s:** 0 people | The cockpit is no longer visible\n**1339.4s:** 0 people | The rear section of the plane passes by, showing the registration 9M-RAH and graphics of a sun bear and a..."}
+{"example_idx": 42, "prediction": "**700.6s:** 9 people | People are seated at tables in a spacious lounge area with large windows and a staircase\n**701.9s:** 7 people | scene static, no change\n**703.1s:** 10 people | The camera pans to a hallway where several people are seated on a bench, and a woman walks toward the...\n**708.1s:** 0 people | scene static, no change", "reference": "**700.6s:** 7 people | Travelers are seated at tables in an airport dining area with large windows overlooking the tarmac\n**703.1s:** 5 people | The camera pans to a hallway where tables are stacked against a wall and a group of travelers waits with...\n**706.9s:** 6 people | scene static, no change"}
+{"example_idx": 43, "prediction": "**460.6s:** 4 people | A large Gundam statue is visible in a display case within a busy terminal\n**461.9s:** 12 people | The scene shifts to a crowded terminal concourse with travelers walking past luggage carts and information boards\n**466.9s:** 25 people | A dense crowd of passengers moves through the hall as a subtitle about exploring upper levels appears on screen\n**468.1s:** 30+ people | The camera pans to reveal a large \"Toilets\" sign and a massive crowd of people in the background", "reference": "**460.6s:** 6 people | A large white and red Gundam statue is displayed behind a white chain fence in an airport terminal\n**461.9s:** 5 people | scene static, no change\n**463.1s:** 12 people | a woman in a denim vest pulls a red suitcase towards the right\n**464.4s:** 10 people | scene static, no change\n**468.1s:** 8 people | On-screen text appears as the camera follows the group towards the upper levels of the terminal"}
+{"example_idx": 44, "prediction": "**710.6s:** 0 people | Wide view of the airport tarmac with an ANA plane parked at the gate and a service vehicle near the...\n**713.1s:** 0 people | The camera pans right, following an ANA plane as it taxis across the tarmac\n**716.9s:** 0 people | scene static, no change\n**719.4s:** 0 people | scene static, no change", "reference": "**710.6s:** 0 people | An ANA propeller plane is taxiing on the tarmac, viewed from behind a large foreground pillar\n**718.1s:** 0 people | Close-up of a warning sign in Japanese (\"Please do not touch\") on a glass partition with a JAL aircraft tail...\n**719.4s:** 0 people | Wide view of a Japan Airlines (JAL) aircraft docked at a gate with a jet bridge, and a second plane..."}
+{"example_idx": 45, "prediction": "**330.6s:** 12 people | a woman in a black hijab stands on the left\n**331.9s:** 14 people | A woman in a brown jacket enters the foreground left as the woman in the hijab moves further into the...\n**333.1s:** 15 people | A man in a white t-shirt and a woman in a white dress walk toward the camera; a group of...\n**335.6s:** 12 people | A man with a large brown bag enters from the right as the previous group moves past the camera\n**338.1s:** 10 people | A woman in a red uniform appears on the right as a man in a black shirt walks toward the...", "reference": "**330.6s:** 18 people | Travelers with luggage carts walk through the airport terminal near the international departures sign\n**333.1s:** 17 people | A subtitle overlay appears at the bottom of the frame as travelers move deeper into the hall\n**334.4s:** 15 people | scene static, no change\n**336.9s:** 14 people | scene static, no change\n**338.1s:** 12 people | The camera pans to reveal a \"Care for safety\" booth and airport staff in red uniforms walking in the background"}
+{"example_idx": 46, "prediction": "**2420.6s:** 6 people | A woman in a black dress and a woman in a blue patterned dress walk across a glass-floored bridge in...\n**2423.1s:** 7 people | A woman in a teal dress enters the scene from the right as the other two continue walking toward the...\n**2424.4s:** 8 people | A man in a white shirt and a woman in a light-colored dress appear in the background, walking toward the...\n**2425.6s:** 9 people | A woman with a backpack enters from the left foreground, and a man is seen sitting on the glass floor...\n**2428.1s:** 7 people | scene static, no change", "reference": "**2420.6s:** 7 people | Visitors walk across an elevated glass canopy bridge within a large indoor botanical garden\n**2424.4s:** 8 people | A visitor sits down on the glass floor to pose for a photo as a man in a white shirt...\n**2429.4s:** 7 people | scene static, no change"}
+{"example_idx": 47, "prediction": "**210.6s:** 15+ people | Wide view of a multi-level mall with a \"COCORICE\" cafe and various shoppers in the distance\n**213.1s:** 12 people | Perspective shifts to a hallway where a man in a yellow shirt and a woman walk away from the camera\n**215.6s:** 14 people | A man in a grey shirt carrying a blue bag enters from the right as the digital screen updates to...\n**218.1s:** 16 people | A woman in a grey top enters from the right foreground, and a group of people gathers near an escalator...\n**219.4s:** 12 people | The camera pans to follow the woman in the grey top as she walks past a shop labeled \"CENTRAL WORLD\"...", "reference": "**210.6s:** 12 people | View of a \"COUCOU\" restaurant and shoppers on multiple mall levels from an upper-floor balcony\n**211.9s:** 10 people | scene static, no change\n**213.1s:** 8 people | Perspective shifts to follow a couple walking past a large digital billboard advertising \"Quick.\"\n**214.4s:** 7 people | A man in a blue cap with a backpack enters the foreground as the couple continues walking\n**219.4s:** 15 people | The group reaches an elevator lobby where several people, including one with a stroller, are waiting"}
+{"example_idx": 48, "prediction": "**1370.6s:** 1 people | A Malaysia Airlines aircraft is parked at the gate with its cargo hold open, surrounded by several white baggage carts...\n**1376.9s:** 0 people | The scene changes to a different angle showing an AirAsia aircraft taxiing on the tarmac with its landing and taxi...", "reference": "**1370.6s:** 2 people | ground crew members are loading luggage into the cargo hold of an AirAsia aircraft via a conveyor belt system\n**1374.4s:** 0 people | scene static, no change\n**1376.9s:** 0 people | an AirAsia Airbus A320 is shown stationary on the tarmac with aircraft hangars and trees in the background"}
+{"example_idx": 49, "prediction": "**2160.6s:** 12 people | Visitors stand on an elevated bridge overlooking a large indoor waterfall under a geometric glass dome\n**2165.6s:** 15 people | A Skytrain enters the bottom right foreground as more visitors appear on the bridge\n**2169.4s:** 8 people | scene static, no change", "reference": "**2160.6s:** 11 people | View of the Rain Vortex and bridge with tourists walking and taking photos under the glass dome\n**2163.1s:** 12 people | scene static, no change\n**2169.4s:** 7 people | scene static, no change"}
+{"example_idx": 50, "prediction": "**800.6s:** 2 people | View of an airport museum display case featuring model airplanes\n**803.1s:** 1 people | A person in a white jacket appears on the left, looking out the window at the airfield\n**806.9s:** 6 people | The scene shifts to an airport observation deck where several people are gathered, with one individual taking a photo\n**809.4s:** 10 people | A wider view of the observation deck reveals a crowded lounge area with passengers seated and others walking in the...", "reference": "**800.6s:** 0 people | A display area features model airplanes and a \"ROKON JET\" poster behind blue stanchion ropes\n**801.9s:** 1 people | A woman wearing a black hat and a baby carrier stands by a large window looking out at the airport...\n**805.6s:** 0 people | The camera pans away from the window to show a clear view of parked ANA aircraft and airport service vehicles...\n**809.4s:** 2 people | The woman is joined by a man at the window as they observe the airplanes together"}
+{"example_idx": 51, "prediction": "**380.6s:** 15 people | A man in a grey shirt and cap walks through a lush indoor garden while looking at his phone\n**381.9s:** 45 people | The man continues walking as a large crowd gathers around a central fountain in the background plaza\n**384.4s:** 55 people | scene static, no change\n**385.6s:** 50 people | The camera pans right, showing the crowd density near the fountain and a person in an orange shirt beginning to...\n**388.1s:** 60 people | A woman with a baby carrier enters the frame from the right as the camera focuses on the staircase and...", "reference": "**380.6s:** 30+ people | A man in a striped shirt walks up a staircase lined with lush green plants, while a couple sits on...\n**384.4s:** 25+ people | scene static, no change\n**386.9s:** 50+ people | The scene shifts to a wide view of the central indoor waterfall (Rain Vortex), with a large crowd of tourists...\n**388.1s:** 55+ people | A person with a black backpack enters the foreground on the right as the camera continues to pan across the..."}
+{"example_idx": 52, "prediction": "**1270.6s:** 0 people | A Japan Airlines airplane is in flight, ascending into a clear blue sky\n**1276.9s:** 0 people | The airplane continues to climb, moving toward the upper right of the frame\n**1279.4s:** 0 people | scene static, no change", "reference": "**1270.6s:** 0 people | A Jetstar airplane is seen ascending into the sky immediately after takeoff\n**1279.4s:** 0 people | scene static, no change"}
+{"example_idx": 53, "prediction": "**2000.6s:** 12 people | A man in a blue shirt leans over a table in a cafe area while others walk in the background\n**2001.9s:** 25 people | The camera pans right, revealing a large crowd of people walking through a bright indoor plaza with trees\n**2004.4s:** 30 people | A woman in a black shirt walks toward the camera in the foreground as the crowd moves through the center...\n**2008.1s:** 35 people | A man in a dark shirt and a woman pulling a suitcase walk toward the camera amidst the flowing crowd\n**2009.4s:** 40 people | The camera pans further left to reveal the \"CHECKING-IN\" entrance as a young girl in a white shirt walks into...", "reference": "**2000.6s:** 16 people | Customers dine at an outdoor cafe while travelers walk past a circular, glass-walled restaurant under a large glass dome\n**2001.9s:** 20 people | scene static, no change\n**2003.1s:** 24 people | A man in a blue shirt appears in the left foreground while the central walkway becomes more crowded with travelers\n**2006.9s:** 28 people | A woman in a black t-shirt and a man pulling a silver suitcase walk directly toward the camera from the...\n**2009.4s:** 32 people | The travelers with the suitcase move into the close foreground as a new woman in a white t-shirt enters from..."}
+{"example_idx": 54, "prediction": "**1210.6s:** 15 people | scene static, no change\n**1211.9s:** 22 people | The camera pans left, revealing a crowded mall corridor and a man in a red shirt walking away\n**1214.4s:** 31 people | A wide view of the concourse shows an escalator and a Skechers store, with many travelers and shoppers moving through...\n**1216.9s:** 26 people | A woman in a white shirt pulling a red suitcase enters the foreground while the camera continues to pan\n**1219.4s:** 20 people | scene static, no change", "reference": "**1210.6s:** 18 people | Crowded mall corridor near a Shake Shack\n**1213.1s:** 20 people | A man in a red shirt moves to the left counter while a man in a white shirt and a...\n**1215.6s:** 19 people | The man in the dark grey shirt approaches the camera closely as a woman with a bright red suitcase enters...\n**1216.9s:** 18 people | A young girl in a pink shirt enters the foreground carrying a pink bag\n**1219.4s:** 21 people | A man in a dark polo shirt moves into the foreground on the left while the crowd near the escalator..."}
+{"example_idx": 55, "prediction": "**440.6s:** 12 people | A man in a white hoodie walks through a terminal past luggage lockers and seated travelers\n**443.1s:** 14 people | A woman with a black backpack enters from the left as the man in the white hoodie continues toward the...\n**445.6s:** 18 people | The camera pans right to reveal a dense crowd gathered near a shop in the distance\n**448.1s:** 22 people | A detailed directional sign for \"4F Food Court\" and various food stalls appears in the foreground as the crowd moves...", "reference": "**440.6s:** 30+ people | others sit and stand in the large station hall\n**444.4s:** 35+ people | Two women, one in a maroon shirt and one in a black shirt, enter the left foreground\n**446.9s:** 30+ people | The foreground women have exited the frame, revealing a large crowd gathered in the distance near the JR station entrance\n**448.1s:** 25+ people | A woman in a long skirt and backpack walks from left to right"}
+{"example_idx": 56, "prediction": "**1070.6s:** 15 people | Two attendants in red uniforms walk away through a terminal corridor past a \"Bakery & Coffee\" shop\n**1074.4s:** 12 people | The camera pans right, revealing the \"LukKarn Thai Royal Cooking\" restaurant and potential diners seated at outdoor-style tables\n**1076.9s:** 10 people | A closer view of the restaurant's entrance shows a man in a blue shirt standing near the seating area\n**1078.1s:** 12 people | The view shifts to the restaurant's interior, showing patrons seated at tables with large windows overlooking an airfield\n**1079.4s:** 15 people | A detailed view of the dining area reveals passengers eating at tables equipped with large glass panels for viewing the...", "reference": "**1070.6s:** 18 people | A crowded airport terminal hallway with travelers\n**1074.4s:** 10 people | The camera pans right toward the LukKaiThong restaurant entrance, where staff stand near a cleaning cart and patrons are seated...\n**1079.4s:** 6 people | scene static, no change"}
+{"example_idx": 57, "prediction": "**30.6s:** 18 people | People gather around a large Christmas tree in an indoor garden featuring lush greenery and a waterfall\n**31.9s:** 16 people | A subtitle about airplane noise appears as the crowd density near the tree base decreases\n**35.6s:** 10 people | The subtitle disappears and the man in the light blue shirt raises his phone to capture the scene\n**38.1s:** 6 people | The man in the blue shirt moves out of view, revealing more of the tree and the surrounding forest-like environment", "reference": "**30.6s:** 18 people | scene static, no change\n**33.1s:** 15 people | A woman in a black patterned dress walks toward the right while a man in a white polo shirt moves...\n**34.4s:** 14 people | The man in the white polo shirt reaches the immediate foreground as a woman in a blue shirt poses with...\n**35.6s:** 12 people | The foreground man exits the frame, and the crowd thins out near the central walkway\n**39.4s:** 8 people | scene static, no change"}
+{"example_idx": 58, "prediction": "**180.6s:** 12 people | a woman in a grey hoodie and another in a yellow shirt walk away in the center\n**181.9s:** 13 people | A woman in a red shirt enters from the right, walking toward the center\n**184.4s:** 12 people | The woman in red exits right; a woman with a large patterned bag enters from the left\n**185.6s:** 11 people | The woman with the patterned bag moves into the center foreground, partially obscuring the view of the concourse\n**188.1s:** 10 people | The woman with the patterned bag exits right, revealing a woman in a black shirt walking toward the camera on...", "reference": "**180.6s:** 15+ people | scene static, no change\n**181.9s:** 15+ people | scene static, no change\n**183.1s:** 15+ people | A woman and a young boy walk towards the camera, while a woman in a dark blue hoodie and patterned...\n**185.6s:** 12+ people | A worker pushing a cart with a large red bin enters the scene from the right as the pedestrians continue..."}
+{"example_idx": 59, "prediction": "**990.6s:** 0 people | Close-up of mannequin legs in a museum display case wearing historical flight attendant uniforms\n**993.1s:** 0 people | Wide view of the \"History of Airports\" exhibit with model aircraft and vintage luggage on display\n**995.6s:** 2 people | A large model airplane is shown in a central display case while two visitors are visible in the background looking...\n**996.9s:** 4 people | A vintage white jacket is displayed next to a \"Vintage Jackets\" sign as more visitors appear in the background\n**999.4s:** 6 people | A group of visitors is gathered by a large glass partition in the background observing the exhibit", "reference": "**990.6s:** 0 people | A museum display featuring mannequins in historical airline uniforms behind a glass case\n**991.9s:** 5 people | A blurred view of several people walking through a brightly lit concourse or hallway\n**993.1s:** 0 people | A museum exhibition area with wall-mounted displays and model aircraft in glass cases\n**994.4s:** 2 people | A man and a woman stand by a museum exhibit, looking at displays of historical items and model planes\n**998.1s:** 3 people | A group of individuals stand in the background of a museum space, looking out through large windows"}
+{"example_idx": 60, "prediction": "**400.6s:** 100+ people | Wide shot of the Jewel Changi Rain Vortex waterfall with a large crowd of visitors gathered around the viewing area...\n**403.1s:** 50+ people | The camera tilts upward, focusing on the top of the waterfall as it flows from the glass dome ceiling, shifting...", "reference": "**400.6s:** 100+ people | A large crowd of tourists gathers at the base of the Jewel Changi Rain Vortex waterfall, surrounded by lush indoor..."}
+{"example_idx": 61, "prediction": "**940.6s:** 0 people | Close-up view of a historical timeline board showing text and dates related to New Chitose Airport's opening in 1988\n**943.1s:** 0 people | The camera pans right, revealing more details of the 1988 and 1994 timeline entries\n**945.6s:** 0 people | The pan continues to the left, bringing early timeline entries from 1979, 1981, and 1982 into view\n**948.1s:** 0 people | Significant shift to a different section of the board showcasing prehistoric and early aviation imagery, including a photo of a...\n**949.4s:** 0 people | The final pan reveals older photographs and illustrations, including a cartoon-style drawing of an early aircraft and a photo of...", "reference": "**940.6s:** 0 people | A wall-mounted informational display shows the history of New Chitose Airport, specifically the 1988 section titled \"New Chitose Airport is...\n**943.1s:** 0 people | scene static, no change\n**945.6s:** 0 people | scene static, no change\n**948.1s:** 0 people | Panning further left reveals historical entries from 1951, 1963, and 1978, accompanied by archival photos of the airport's development\n**949.4s:** 0 people | scene static, no change"}
+{"example_idx": 62, "prediction": "**2960.6s:** 7 people | A man in black shorts and a woman in a grey shirt walk through a lush indoor garden with a...\n**2963.1s:** 11 people | scene static, no change\n**2965.6s:** 10 people | scene static, no change\n**2968.1s:** 12 people | scene static, no change", "reference": "**2960.6s:** 6 people | A walkway in an indoor garden with a child in the foreground and several people walking in the distance\n**2963.1s:** 9 people | A man with a backpack walks into the foreground while the camera focuses on a platform with blue gift boxes\n**2964.4s:** 7 people | A man carrying a child on his shoulders walks across the decorated platform alongside other visitors\n**2966.9s:** 7 people | A child in a red outfit sits on the floor near a blue structure while people walk toward a distant...\n**2968.1s:** 10 people | A group of people walks along a tree-lined path toward a large decorative archway"}
+{"example_idx": 63, "prediction": "**1810.6s:** 25 people | A man in a black t-shirt pulls a suitcase through an airport terminal past a food court and seated travelers\n**1813.1s:** 5 people | scene static, no change\n**1819.4s:** 10 people | The camera pans right to reveal a \"Ya Kun Kaya Toast\" stall and a seating area where several passengers are...", "reference": "**1810.6s:** 18 people | A man in a black shirt pulls two suitcases down a wide airport corridor past food stalls and seating areas\n**1811.9s:** 12 people | A traveler with a backpack stops at the \"Thai Street Food\" stall while the man with suitcases continues down the...\n**1815.6s:** 10 people | A woman in a black shirt joins the man at the food stall counter\n**1816.9s:** 9 people | The man with the backpack departs, leaving the woman at the counter as the camera pans toward the next stall\n**1819.4s:** 14 people | A man in a blue polo shirt appears at the \"Au We White Coffee\" counter as the man with suitcases..."}
+{"example_idx": 64, "prediction": "**570.6s:** 2 people | A staff member stands behind a noodle shop counter beneath large backlit menu displays\n**573.1s:** 4 people | scene static, no change\n**575.6s:** 1 people | The camera pans into a hallway, showing a person walking in the distance\n**576.9s:** 5 people | A group of people are seen sitting at tables in a dining area with luggage placed against the wall\n**579.4s:** 8 people | The KFC logo becomes clearly visible as the camera reveals more of the dining area and additional patrons", "reference": "**570.6s:** 2 people | A customer stands at the Imm Rice & Noodle counter while a staff member works behind it\n**574.4s:** 1 people | scene static, no change\n**578.1s:** 2 people | The view transitions to a seating area near a KFC where suitcases are parked by a wall"}
+{"example_idx": 65, "prediction": "**1590.6s:** 2 people | View of the \"Royce Chocolate World\" storefront with people walking in the background\n**1593.1s:** 11 people | scene static, no change\n**1594.4s:** 10 people | scene static, no change\n**1596.9s:** 10 people | scene static, no change", "reference": "**1590.6s:** 3 people | a man with a backpack walks past a circular display\n**1591.9s:** 4 people | Transition to a different area of the shop\n**1593.1s:** 6 people | two women are packing items into a bag; New view of the service counter\n**1595.6s:** 8 people | pedestrians are walking through the corridor\n**1598.1s:** 11 people | a man pushes a cart with multiple suitcases on the left; Crowd density increases in the hallway"}
+{"example_idx": 66, "prediction": "**2760.6s:** 6 people | A woman in a white vest and blue hat pulls a suitcase and carries a bag walks through a lush...\n**2763.1s:** 4 people | scene static, no change\n**2765.6s:** 15 people | A long queue of people is visible on the left side under a large, twisted metal sculpture\n**2768.1s:** 20 people | A woman and a young child enter the scene from the right, walking towards the camera under the sculpture", "reference": "**2760.6s:** 11 people | A man with a suitcase and paper bag walks through an indoor garden past a Marvel exhibition\n**2765.6s:** 8 people | scene static, no change\n**2768.1s:** 10 people | A woman and a child enter the foreground on the right near a large, reflective metal walkway"}
+{"example_idx": 67, "prediction": "**1850.6s:** 25 people | a man in a white t-shirt and black cap is seated on the right looking at his phone\n**1851.9s:** 24 people | The man in the white t-shirt stands up and walks toward the background as a woman in a black shirt...\n**1853.1s:** 23 people | A man in a red shirt is now seen seated in the foreground right, looking at his phone\n**1854.4s:** 22 people | A man in a black shirt with a yellow scarf enters the frame from the right foreground\n**1856.9s:** 21 people | A man in a black shirt and blue jeans walks across the foreground from right to left", "reference": "**1850.6s:** 32 people | A crowded airport waiting area with multiple rows of seating\n**1851.9s:** 31 people | The digital advertisement on the central pillar has changed from a car image to blue text"}
+{"example_idx": 68, "prediction": "**2920.6s:** 11 people | Crowded indoor garden walkway with visitors walking in multiple directions near a large cartoon bird sculpture\n**2923.1s:** 9 people | scene static, no change\n**2924.4s:** 10 people | A woman in a black shirt and grey skirt enters the foreground from the right\n**2926.9s:** 8 people | A person pushing a stroller becomes visible in the center of the path as the foreground woman moves out of...", "reference": "**2920.6s:** 12 people | A crowd of visitors walks towards the camera in a lush indoor garden atrium featuring a glass dome ceiling and...\n**2923.1s:** 8 people | The foreground crowd moves out of view, revealing a \"Jewel-Rassic Quest\" digital sign and a couple walking away down the...\n**2929.4s:** 10 people | A man carrying an infant in a front carrier enters the scene from the right foreground while others continue walking..."}
+{"example_idx": 69, "prediction": "**0.6s:** 0 people | scene static, no change\n**1.9s:** 0 people | An AirAsia aircraft is seen landing on the runway\n**4.4s:** 0 people | A Thai Lion Air plane is taxiing across the foreground\n**8.1s:** 0 people | A second Lion Air aircraft is visible on the tarmac in the background", "reference": "**0.6s:** 0 people | An AirAsia aircraft lands on a runway at an airport during twilight\n**4.4s:** 0 people | Overlay text \"TOUR FROM HOME TV PRESENTS\" appears as an AirAsia plane taxies toward the camera\n**6.9s:** 5 people | Ground crew members in safety vests work near the cargo hold of a Myanmar Airways International aircraft\n**8.1s:** 0 people | A Thai Lion Air plane taxies on the runway with another aircraft visible in the background"}
+{"example_idx": 70, "prediction": "**60.6s:** 0 people | Initial view of a modern airport terminal interior featuring a large glass elevator and complex steel truss architecture\n**61.9s:** 6 people | a \"LOVE\" sculpture is visible on the left with travelers walking in the background\n**64.4s:** 12 people | scene static, no change\n**66.9s:** 15 people | A group of travelers, including a woman in a red dress and others with luggage trolleys, gathers on the right...\n**68.1s:** 12 people | scene static, no change", "reference": "**60.6s:** 6 people | High-angle wide shot of the New Chitose Airport lobby featuring a large glass elevator structure and an intricate lattice ceiling\n**61.9s:** 12 people | Perspective shifts to ground level near a \"HOKKAIDO LOVE!\" sign where travelers are walking through the terminal\n**66.9s:** 14 people | A woman pushing a stroller enters the foreground from the right, moving across the lobby\n**69.4s:** 18 people | The scene shifts to show people using an escalator and walking towards the \"Gourmet World\" area"}
+{"example_idx": 71, "prediction": "**1210.6s:** 12 people | View of an airport terminal featuring a long moving walkway on the left and a seating area on the right...\n**1216.9s:** 18 people | scene static, no change\n**1219.4s:** 22 people | Continued pan shows a dense crowd of travelers seated in a waiting area near large windows", "reference": "**1210.6s:** 14 people | Wide view of an airport corridor with a moving walkway on the left and several travelers walking in the distance\n**1216.9s:** 28 people | scene static, no change"}
+{"example_idx": 72, "prediction": "**2540.6s:** 12 people | A woman in black and a man in a grey shirt pose for a photo on the high glass walkway...\n**2544.4s:** 15 people | scene static, no change\n**2545.6s:** 8 people | scene static, no change\n**2548.1s:** 25 people | The camera pans down to reveal a large crowd of people gathered at the base of the waterfall on the...", "reference": "**2540.6s:** 12 people | A man and woman are positioned on a glass walkway within a lush indoor garden under a large geometric glass...\n**2544.4s:** 32 people | scene static, no change\n**2546.9s:** 65 people | The couple on the walkway exits the frame as the camera focuses on the full waterfall and the dense crowds..."}
+{"example_idx": 73, "prediction": "**1100.6s:** 18 people | A man in a yellow mascot costume stands near an escalator while travelers walk through a bright airport terminal past...\n**1101.9s:** 16 people | A couple carrying a child walks toward the camera as the mascot moves to the far left of the frame\n**1104.4s:** 15 people | The couple with the child passes into the foreground\n**1105.6s:** 14 people | scene static, no change\n**1108.1s:** 12 people | The children move toward the background as a man in a black t-shirt enters the scene from the right", "reference": "**1100.6s:** 14 people | A man walks with a purple suitcase past a Coach store as other travelers move through the mall\n**1101.9s:** 16 people | The man with the suitcase moves into the right foreground while more people walk towards the terminal\n**1105.6s:** 10 people | A woman in a black shirt enters the foreground while other travelers move towards the terminal gates\n**1108.1s:** 13 people | The woman in blue approaches as two men wheeling large suitcases walk towards the foreground\n**1109.4s:** 10 people | The woman in blue exits the frame as the two men with suitcases move into the foreground"}
+{"example_idx": 74, "prediction": "**620.6s:** 4 people | Customers order at a Subway restaurant counter\n**624.4s:** 6 people | scene static, no change\n**626.9s:** 8 people | A man in a grey shirt appears in the foreground as the camera continues to pan right, revealing more of...\n**628.1s:** 10 people | A woman in a black and white patterned shirt enters the frame from the right\n**629.4s:** 12 people | A man in a blue shirt walks into the scene from the far right", "reference": "**620.6s:** 4 people | Front view of a Subway counter with two staff members preparing food and two customers waiting\n**626.9s:** 4 people | scene static, no change\n**628.1s:** 4 people | Close-up of the noodle shop counter as a customer in a plaid shirt interacts with the staff"}
+{"example_idx": 75, "prediction": "**800.6s:** 12 people | Travelers with luggage stand near a storage area and information desk in a brightly lit terminal\n**803.1s:** 15 people | A man in a white shirt and straw hat stands by a blue suitcase as the camera pans toward a...\n**805.6s:** 18 people | A woman in a black shirt and pink mask enters the foreground while the camera continues to pan right, revealing...\n**808.1s:** 20 people | A man in a black t-shirt and a woman in a black shirt with a white shopping bag enter the...", "reference": "**800.6s:** 15 people | a man pushes a luggage cart in the center\n**803.1s:** 22 people | scene static, no change\n**805.6s:** 18 people | The view transitions to a hallway featuring a large vertical green wall on the right as people walk toward an...\n**808.1s:** 14 people | Travelers with luggage head toward the 'Customs Inspection' zone"}
+{"example_idx": 76, "prediction": "**2950.6s:** 12 people | Visitors stand near a large mesh rope bridge and elevated green walkway in a sunlit indoor garden\n**2953.1s:** 18 people | scene static, no change\n**2958.1s:** 14 people | scene static, no change", "reference": "**2950.6s:** 6 people | Visitors pass through an entrance gate into an indoor park featuring a large geometric glass roof and lush greenery\n**2953.1s:** 15 people | The camera pans right to reveal an elevated netting structure where numerous people are walking and climbing high above the...\n**2956.9s:** 18 people | scene static, no change\n**2959.4s:** 12 people | Visitors walk along a curved pathway on the ground level, surrounded by dense tropical plants with the netting structure visible..."}
+{"example_idx": 77, "prediction": "**400.6s:** 12 people | Travelers walk through a terminal hall under a \"Departures\" sign\n**403.1s:** 10 people | A woman in a yellow shirt pulls a suitcase toward the left foreground\n**405.6s:** 8 people | A staff member in a blue vest walks across the center of the hall\n**408.1s:** 15 people | The view transitions to an escalator looking down at a crowded lower level", "reference": "**400.6s:** 10 people | A group of travelers stands with luggage near an autonomous cleaning robot in a spacious airport terminal lobby\n**404.4s:** 12 people | A staff member in a safety vest walks directly across the foreground, momentarily obscuring the view of the terminal\n**405.6s:** 0 people | scene static, no change\n**406.9s:** 15 people | A staff member in a blue uniform pushes a luggage trolley on the right while a crowd of travelers is...\n**408.1s:** 16 people | A man pushing a stroller and another with a luggage cart walk towards the camera through the terminal corridor"}
+{"example_idx": 78, "prediction": "**2250.6s:** 12 people | A woman in red plaid pants walks left carrying a baby in a carrier\n**2253.1s:** 14 people | A man in a grey shirt pushing a stroller enters from the background\n**2255.6s:** 15 people | A woman in a black shirt and a man in a blue shirt walk toward the camera\n**2256.9s:** 14 people | The woman in the black shirt moves into the close foreground on the left\n**2259.4s:** 12 people | scene static, no change", "reference": "**2250.6s:** 12 people | A woman in plaid walks toward the camera through a lush indoor garden while travelers rest on a bench with...\n**2253.1s:** 14 people | A woman pushing a stroller enters the walkway from the right, moving toward the foreground\n**2256.9s:** 15 people | A man pushing a second stroller appears in the background as the first stroller moves out of frame\n**2259.4s:** 13 people | The man with the stroller reaches the foreground while a man in a blue shirt walks across the background near..."}
+{"example_idx": 79, "prediction": "**270.6s:** 8 people | A man in a black shirt walks past a shop towards a large terminal hall\n**271.9s:** 12 people | scene static, no change\n**273.1s:** 18 people | The perspective shifts to a wide view of the terminal, revealing a man in a black jacket walking away and...\n**275.6s:** 22 people | A woman with a white shopping bag enters the foreground from the right as the camera pans\n**278.1s:** 25 people | A man in a blue denim jacket pulling a red suitcase walks towards the camera through the busy terminal", "reference": "**270.6s:** 12 people | View of a souvenir shop with several shoppers\n**273.1s:** 14 people | The camera pans left, revealing a woman in a white top and black cardigan in the foreground right\n**275.6s:** 20 people | travelers with luggage are visible in the background\n**278.1s:** 30 people | In the main concourse, crowd density increases as a man with a red suitcase and other travelers navigate the open..."}
+{"example_idx": 80, "prediction": "**2670.6s:** 12 people | scene static, no change\n**2674.4s:** 10 people | The three women have crossed the foreground, and two new women with suitcases enter from the right\n**2678.1s:** 8 people | A family group with several children and suitcases gathers on the path as the previous pedestrians move further away", "reference": "**2670.6s:** 11 people | Travelers walk past a large purple floral elephant sculpture in a lush indoor garden setting\n**2673.1s:** 13 people | Camera shifts to follow travelers with luggage as they navigate a curved wooden walkway\n**2678.1s:** 8 people | The path opens into a wider thoroughfare where travelers with multiple colorful suitcases are gathered"}
+{"example_idx": 81, "prediction": "**1680.6s:** 10 people | Visitors stand on an observation deck, taking photos near a large indoor waterfall\n**1681.9s:** 8 people | A woman holding a baby appears in the foreground as the camera pans right\n**1683.1s:** 6 people | A man in a green shirt enters the foreground to take a photo\n**1684.4s:** 5 people | The man in green moves left, and the woman with the baby continues taking a picture\n**1685.6s:** 0 people | scene static, no change", "reference": "**1680.6s:** 12 people | Tourists gather on a balcony near a large indoor waterfall\n**1681.9s:** 10 people | A woman in an orange top moves into the foreground as others adjust their positions for photos\n**1683.1s:** 8 people | A man in a green shirt enters from the left while a woman poses with a purple teddy bear\n**1684.4s:** 6 people | scene static, no change\n**1686.9s:** 30+ people | The foreground group has moved out of view, revealing a wide perspective of the waterfall and crowds on lower levels"}
+{"example_idx": 82, "prediction": "**750.6s:** 8 people | A man pushes a luggage cart toward the camera in an airport terminal\n**751.9s:** 12 people | scene static, no change\n**754.4s:** 15 people | scene static, no change\n**759.4s:** 20 people | A wider view of the terminal reveals more passengers walking through the concourse near the escalator", "reference": "**750.6s:** 11 people | View of a terminal hallway with \"Cafe Amazon\" on the left, a cleaning robot, and travelers with luggage\n**751.9s:** 11 people | scene static, no change\n**753.1s:** 7 people | A woman in a white shirt appears on the right as the camera approaches the escalator area\n**754.4s:** 5 people | Focus shifts to the \"Domestic Departures\" sign and a woman in black walking toward the background\n**759.4s:** 0 people | Close-up view of the directory on the Terminal Guide pillar"}
+{"example_idx": 83, "prediction": "**600.6s:** 2 people | View of a KFC service counter with two employees standing behind the counter\n**603.1s:** 2 people | scene static, no change\n**605.6s:** 1 people | scene static, no change\n**606.9s:** 7 people | The scene shifts to a food court corridor, revealing a \"RADIO\" storefront and several patrons walking or seated at tables...\n**608.1s:** 9 people | A closer view of the food court seating area shows a man and two children at a table in the...", "reference": "**600.6s:** 2 people | Two staff members stand behind a KFC service counter under a bright menu display\n**608.1s:** 8 people | scene static, no change"}
+{"example_idx": 84, "prediction": "**950.6s:** 0 people | Close-up of a historical timeline board showing black-and-white photos and text about aviation milestones\n**954.4s:** 0 people | The scene shifts to a museum hallway featuring a glass display case with vintage flight attendant mannequins\n**956.9s:** 0 people | A wide view of the hallway reveals a \"Toilets\" sign and a cartoonish blue bird graphic on the wall\n**958.1s:** 0 people | scene static, no change\n**959.4s:** 0 people | scene static, no change", "reference": "**950.6s:** 0 people | View of a historical aviation exhibit wall with photos and text detailing airport history from 1926 to 1978\n**955.6s:** 0 people | scene static, no change\n**956.9s:** 0 people | A display case containing five mannequins dressed in historical flight attendant uniforms comes into view\n**958.1s:** 1 people | A woman with a backpack enters the scene on the left, walking toward the \"Airport Theater\" while holding up a...\n**959.4s:** 0 people | scene static, no change"}
+{"example_idx": 85, "prediction": "**300.6s:** 15 people | a woman in a grey top and white bag walks left while a man with a backpack stands near a...\n**303.1s:** 12 people | A woman in a red dress enters from the left as others continue walking through the terminal\n**305.6s:** 10 people | A woman pulling a white suitcase walks away from the camera towards the center of the hall\n**308.1s:** 12 people | A woman in a white shirt and black pants enters from the right, walking across the geometric patterned floor", "reference": "**300.6s:** 25 people | Wide view of a busy airport terminal with a souvenir shop and statue\n**303.1s:** 22 people | scene static, no change\n**305.6s:** 18 people | The foreground clears, revealing a woman with a white suitcase walking toward the background shops\n**306.9s:** 20 people | scene static, no change\n**309.4s:** 22 people | A man with a large red suitcase enters from the left while crowd density increases near the background stores"}
+{"example_idx": 86, "prediction": "**320.6s:** 12 people | A man in a green cap and a woman in a black shirt walk away in the foreground of a...\n**321.9s:** 14 people | A woman in a white jacket and a man pulling a suitcase enter the scene from the left\n**323.1s:** 15 people | A man in a plaid shirt walks toward the camera while a woman in a black-and-white patterned top crosses the...\n**324.4s:** 14 people | A man in a light blue shirt enters from the left as others continue walking toward the shop\n**328.1s:** 12 people | A woman with a baby in a carrier and a man in a black shirt walk toward the camera", "reference": "**320.6s:** 15 people | Crowded airport hallway with travelers walking past confectionery shops and food stalls\n**321.9s:** 14 people | A woman with a suitcase and white shopping bag enters the foreground from the right\n**323.1s:** 16 people | A man in a plaid shirt enters the scene from the right foreground, walking towards the center\n**325.6s:** 14 people | A man in a grey t-shirt and shorts enters from the right foreground, walking towards the left\n**329.4s:** 13 people | Two women enter from the right foreground, walking quickly across the hallway"}
+{"example_idx": 87, "prediction": "**830.6s:** 2 people | A man and a girl are seen inside a glass elevator shaft surrounded by lush vertical gardens and trees\n**835.6s:** 1 people | scene static, no change\n**838.1s:** 4 people | The scene shifts to an indoor terminal corridor where a couple walks past a McDonald's on the left and a...\n**839.4s:** 5 people | A young woman in a pink shirt enters the foreground from the right as the couple continues down the hallway", "reference": "**830.6s:** 10 people | View of a multi-level atrium with a glass dome ceiling, greenery, and glass elevators\n**834.4s:** 4 people | The perspective shifts to ground level, showing a child on a stone path amidst indoor bamboo\n**839.4s:** 10 people | The scene transitions to a mall corridor with shops and people walking"}
+{"example_idx": 88, "prediction": "**2550.6s:** 40+ people | A Skytrain passes through the foreground tunnel as people gather around the base of the indoor waterfall\n**2553.1s:** 40+ people | The Skytrain has exited the foreground, providing a clear view of the circular water feature and the surrounding crowd\n**2556.9s:** 45+ people | scene static, no change\n**2559.4s:** 45+ people | scene static, no change", "reference": "**2550.6s:** 100+ people | the Skytrain track in the foreground is empty\n**2553.1s:** 100+ people | scene static, no change\n**2555.6s:** 100+ people | The Skytrain passes directly in front of the waterfall as it continues its crossing\n**2559.4s:** 100+ people | scene static, no change"}
+{"example_idx": 89, "prediction": "**890.6s:** 12 people | A man in a plaid shirt stands by a window looking at airplanes while others wait in a line near...\n**893.1s:** 15 people | A woman in a light blue shirt enters the foreground from the right as the man in the plaid shirt...\n**894.4s:** 14 people | A man pushing a luggage cart and a woman in a black hijab walk toward the background\n**896.9s:** 12 people | A woman in a grey sweatshirt enters from the left foreground\n**899.4s:** 10 people | Several people, including a man in a black shirt and a woman in a light blue top, walk toward the...", "reference": "**890.6s:** 17 people | Initial view of an airport observation lounge with travelers looking out large windows and walking through the hall\n**893.1s:** 13 people | A woman in a white t-shirt walks across the foreground while a man pushes a luggage trolley down the center...\n**895.6s:** 12 people | the man with the luggage trolley continues away from the camera\n**898.1s:** 12 people | scene static, no change"}
+{"example_idx": 90, "prediction": "**2090.6s:** 12 people | scene static, no change\n**2091.9s:** 14 people | The camera pans right, revealing a woman in a black top and a man in a white shirt walking away...\n**2093.1s:** 16 people | A family with a child in a stroller becomes the central subject as the camera follows their movement through the...\n**2096.9s:** 18 people | The family with the stroller moves into the foreground left, while a couple walks towards the camera from the background...\n**2099.4s:** 15 people | scene static, no change", "reference": "**2090.6s:** 8 people | a woman with a stroller is on the right\n**2091.9s:** 9 people | a man in a green patterned shirt appears on the right\n**2094.4s:** 13 people | A family with a stroller appears in the middle ground; crowd density increases in the background\n**2096.9s:** 15 people | two women appear walking toward the camera from the background right\n**2099.4s:** 16 people | The family with the stroller moves toward the left edge"}
+{"example_idx": 91, "prediction": "**1040.6s:** 1 people | A woman with a bright green bag stands by a window, looking out at an airfield\n**1041.9s:** 2 people | A woman in a black dress descends an indoor staircase while carrying a patterned bag\n**1044.4s:** 2 people | Outdoors, a woman in a black dress with a floral bag walks toward the camera on a paved path\n**1045.6s:** 3 people | Two women ascend an outdoor staircase, one pulling a silver suitcase and the other carrying a white shopping bag\n**1048.1s:** 3 people | At the top of the stairs, the women move toward a gate surrounded by a tall metal fence and surveillance...", "reference": "**1040.6s:** 1 people | Standing by a large window overlooking an airfield, carrying a bright green bag\n**1041.9s:** 1 people | the person with the green bag is partially visible on the left; Perspective shifts to a tiled wall with a...\n**1044.4s:** 2 people | On an outdoor staircase, a woman in a dark dress descends with luggage while a blonde person ascends with a...\n**1046.9s:** 2 people | The woman with the luggage descends to the bottom left as the blonde person continues climbing the stairs\n**1049.4s:** 3 people | The blonde person reaches the top of the stairs, following another individual with a white suitcase"}
+{"example_idx": 92, "prediction": "**1420.6s:** 15 people | A man in a white shirt and a woman with a yellow suitcase stand near a \"Tokyu\" shop entrance while...\n**1423.1s:** 12 people | A person with a colorful backpack enters the foreground, looking at a digital display kiosk as the crowd in the...\n**1425.6s:** 10 people | The camera pans right towards a \"Smoking Room\" sign as people walk down a corridor lined with shops\n**1428.1s:** 12 people | A group of people, including a couple in matching grey shirts, stands near a shop entrance as the camera continues...", "reference": "**1420.6s:** 12 people | A family is posing for a photo while travelers with luggage walk past shops\n**1423.1s:** 15 people | A group of people is visible further down the hallway near a restaurant entrance\n**1425.6s:** 18 people | Several people are queued or walking in the distance\n**1428.1s:** 14 people | A group of travelers with suitcases stands near the restaurant entrance"}
+{"example_idx": 93, "prediction": "**650.6s:** 6 people | A man in a grey shirt walks away through a food court while a woman stands at a Thai food...\n**653.1s:** 12 people | The man continues walking past a seating area where several people are eating at tables on an artificial lawn\n**656.9s:** 15 people | scene static, no change\n**659.4s:** 0 people | scene static, no change", "reference": "**650.6s:** 7 people | A woman in a grey vest stands at a food counter while another woman in dark blue walks away down...\n**651.9s:** 8 people | The woman in the grey vest exits the scene as the camera moves forward, revealing a man in a grey...\n**656.9s:** 14 people | A staff member in a white shirt and face mask appears behind the food stall counter\n**659.4s:** 16 people | scene static, no change"}
+{"example_idx": 94, "prediction": "**870.6s:** 0 people | Close-up of a decorative wall featuring a large map and model airplane\n**871.9s:** 5 people | scene static, no change\n**874.4s:** 15 people | scene static, no change\n**876.9s:** 20 people | Further right reveals a man in a black shirt walking toward the camera and a crowded background hallway\n**879.4s:** 25 people | a large crowd of travelers with luggage is visible near distant gates", "reference": "**870.6s:** 0 people | View of an airport terminal hall featuring a large abstract mural, decorative flags, and a small wooden shrine\n**871.9s:** 12 people | scene static, no change\n**874.4s:** 20 people | Panning continues to show a \"TAXI SERVICE\" counter and a wider view of the terminal with increased crowd density\n**875.6s:** 8 people | scene static, no change\n**876.9s:** 11 people | A woman walks across the foreground past the \"True\" booth while more people gather near the currency exchange and background..."}
+{"example_idx": 95, "prediction": "**1180.6s:** 15 people | a man in a red shirt walks toward the camera while a man and two children stand near a luggage...\n**1181.9s:** 12 people | The man in the red shirt moves into the close foreground as the family near the coffee shop begins to...\n**1183.1s:** 10 people | scene static, no change\n**1186.9s:** 8 people | scene static, no change\n**1189.4s:** 7 people | A man in a grey shirt and green cap bends over to organize a luggage cart in the foreground as...", "reference": "**1180.6s:** 10 people | Travelers move through an airport corridor past a \"Mivana Coffee\" shop\n**1183.1s:** 0 people | scene static, no change\n**1184.4s:** 7 people | New viewpoint shows a \"Gates 1-6\" sign and an escalator as travelers walk toward the camera\n**1185.6s:** 1 people | The corridor clears of walking travelers, and a text overlay regarding boarding gates appears\n**1186.9s:** 2 people | A man begins descending the escalator while another traveler stands nearby with a luggage cart"}
+{"example_idx": 96, "prediction": "**250.6s:** 6 people | A man in a black shirt checks his phone near a suitcase while a woman with a backpack walks through...\n**251.9s:** 10 people | The camera pans right to reveal a man in a white t-shirt and black backpack walking toward the foreground\n**254.4s:** 12 people | A woman in a blue denim jacket walks away as the camera tracks a growing crowd of travelers moving through...\n**258.1s:** 8 people | scene static, no change", "reference": "**250.6s:** 3 people | a woman in a green coat walks across the terminal floor near departure signs\n**251.9s:** 8 people | a woman in a white dress pulls a silver suitcase through the hall\n**253.1s:** 12 people | a crowd is visible in the background\n**255.6s:** 10 people | scene static, no change\n**259.4s:** 9 people | scene static, no change"}
+{"example_idx": 97, "prediction": "**1140.6s:** 0 people | A Korean Air passenger jet is taxiing on the airport runway, moving from left to right\n**1145.6s:** 0 people | scene static, no change\n**1146.9s:** 0 people | The Korean Air aircraft continues to taxi, turning slightly as it moves further down the runway", "reference": "**1140.6s:** 0 people | A Japan Airlines airplane is taxiing on the tarmac, viewed from its rear-left side\n**1143.1s:** 0 people | scene static, no change\n**1145.6s:** 0 people | scene static, no change"}

processor_config.json

@@ -0,0 +1,63 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "data_format": "channels_first",
+    "default_to_square": true,
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
+size 19989343

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": false,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}