SeaWolf-AI commited on
Commit
aef00eb
·
verified ·
1 Parent(s): 6c9da2d

Darwin-9B-NEG v1.0: First Native Entropy Gating model (+11.3%p GPQA Greedy)

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model:
4
+ - FINAL-Bench/Darwin-9B-Opus
5
+ tags:
6
+ - darwin
7
+ - darwin-v8
8
+ - darwin-neg
9
+ - native-entropy-gating
10
+ - NEG
11
+ - reasoning
12
+ - self-regulated-reasoning
13
+ - thinking
14
+ - qwen3.5
15
+ - gpqa
16
+ - benchmark
17
+ - open-source
18
+ - apache-2.0
19
+ - hybrid-vigor
20
+ - proto-agi
21
+ - vidraft
22
+ - eval-results
23
+ language:
24
+ - en
25
+ - zh
26
+ - ko
27
+ - ja
28
+ - multilingual
29
+ pipeline_tag: text-generation
30
+ library_name: transformers
31
+ ---
32
+
33
+ # Darwin-9B-NEG — First Native Entropy Gating Model
34
+
35
+ <p align="center">
36
+ <a href="https://huggingface.co/FINAL-Bench/Darwin-9B-NEG"><img src="https://img.shields.io/badge/⭐_Darwin_V8-NEG_Native_Entropy_Gating-gold?style=for-the-badge" alt="NEG"></a>
37
+ <a href="https://huggingface.co/FINAL-Bench/Darwin-9B-Opus"><img src="https://img.shields.io/badge/Base-Darwin--9B--Opus-blue?style=for-the-badge" alt="Base"></a>
38
+ </p>
39
+
40
+ > Qwen3.5-9B backbone | 8.95B params | Thinking Mode | BF16 | Apache 2.0
41
+ > **First NEG-enabled model — self-regulating reasoning at 1x inference cost**
42
+
43
+ ---
44
+
45
+ ## 🎯 About NEG (Native Entropy Gating)
46
+
47
+ **Darwin 모델이 GPQA Diamond(박사급 고급 추론) 벤치마크에서 높은 점수를 얻는 비결**
48
+
49
+ 신기술 **'NEG(Native Entropy Gating)'** 는 AI에게 **"자기 확신 감각"을 아키텍처 차원에서 심어주는 Darwin 독자 기술**입니다. 외부 플러그인이나 서빙 옵션이 아닌, **모델 가중치 자체에 임플란트처럼 내재화된 메커니즘**으로, 모델이 생성 루프 안에서 스스로 불확실한 순간을 감지하고 그 자리에서 답을 다듬습니다.
50
+
51
+ 기존에는 추론 정확도를 높이려면 같은 답을 3~8번 반복 생성해야 했지만, **NEG는 전체 생성의 5% 미만에서만 작동**하므로 추가 비용이 거의 없으며, **추론력(벤치마크로 입증)을 약 10% 이상 끌어올립니다.**
52
+
53
+ 별도 라이브러리·추론 엔진·외부 모듈 없이 **모델 파일 하나만 배포**하면 모든 기능이 함께 동작하므로, 고객사의 기존 온프레미스 인프라를 그대로 쓰면서 성능만 향상됩니다. 추가 GPU 구매·추가 라이선스·추가 운영비 — **모두 필요 없습니다.**
54
+
55
+ 모델 크기를 키우지 않고도 차세대 추론 성능을 얻는 **아키텍처 내재화 접근**, **Darwin-NEG 시리즈**의 핵심 기술입니다.
56
+
57
+ ---
58
+
59
+ ## 📊 PoC Evaluation Results (GPQA Diamond, Greedy mode)
60
+
61
+ Evaluation on **same 80 questions**, **same deterministic Greedy decoding**, **same 1x inference cost**:
62
+
63
+ | Question Set | Baseline (Darwin-9B-Opus) | **NEG-enabled (this model)** | **Δ** |
64
+ |:---:|:---:|:---:|:---:|
65
+ | Q20 | 55.0% | **70.0%** | **+15.0%p** 🔥 |
66
+ | Q40 | 52.5% | **60.0%** | **+7.5%p** ✅ |
67
+ | Q60 | 51.7% | **63.3%** | **+11.6%p** 🔥 |
68
+ | Q80 | 51.2% | **62.5%** | **+11.3%p** 🔥 |
69
+
70
+ **Average Δ: +11.35 percentage points** (GO threshold was +2%p → **5.65× exceeded**)
71
+
72
+ **Gate activation rate**: 4.6-5.1% (conditional, not over-active)
73
+
74
+ ---
75
+
76
+ ## 🏗️ Architecture
77
+
78
+ ```
79
+ Input Text
80
+
81
+ [Darwin-9B-Opus Base (FROZEN)]
82
+
83
+ [Transformer Layers × 32]
84
+
85
+ last hidden state
86
+ ├──▶ NEG-Head (4.2M params) ──▶ predicted_entropy
87
+ │ │
88
+ ▼ ▼
89
+ LM Head ─▶ base_logits ──▶ NEG-Gate (top-k masking)
90
+
91
+ guided_logits
92
+
93
+ next_token
94
+ ```
95
+
96
+ ### Key Specifications
97
+
98
+ | Component | Value |
99
+ |---|---|
100
+ | Base model | Darwin-9B-Opus (Qwen3.5 family) |
101
+ | Total parameters | 8.95 B |
102
+ | NEG-Head parameters | 4.2 M (0.05%) |
103
+ | NEG-Gate parameters | 1 (learnable threshold) |
104
+ | NEG activation rate | 4.8% (typical) |
105
+ | NEG-Head Pearson correlation | 0.8744 |
106
+ | NEG-Gate threshold (learned) | 1.175 |
107
+ | NEG-Gate top_k | 20 |
108
+ | Context | 262,144 tokens |
109
+ | Dtype | bfloat16 |
110
+ | License | Apache 2.0 |
111
+
112
+ ---
113
+
114
+ ## 🚀 Usage
115
+
116
+ ### Quick Start
117
+
118
+ ```python
119
+ from modeling_darwin_neg import load_darwin_neg
120
+ import torch
121
+
122
+ model = load_darwin_neg(
123
+ "FINAL-Bench/Darwin-9B-NEG",
124
+ torch_dtype=torch.bfloat16,
125
+ device_map="auto",
126
+ hf_token="hf_xxx", # required for private repo
127
+ )
128
+
129
+ from transformers import AutoTokenizer
130
+ tok = AutoTokenizer.from_pretrained("FINAL-Bench/Darwin-9B-NEG", trust_remote_code=True)
131
+
132
+ messages = [{"role": "user", "content": "Solve: What is the derivative of sin(x²)?"}]
133
+ text = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
134
+ inputs = tok(text, return_tensors="pt").to(model.device)
135
+ outputs = model.generate(**inputs, max_new_tokens=2048, do_sample=False)
136
+ print(tok.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))
137
+ ```
138
+
139
+ ### Manual loading (more control)
140
+
141
+ ```python
142
+ from transformers import AutoModelForCausalLM, AutoTokenizer
143
+ from modeling_darwin_neg import attach_neg
144
+
145
+ model = AutoModelForCausalLM.from_pretrained(
146
+ "FINAL-Bench/Darwin-9B-NEG",
147
+ torch_dtype=torch.bfloat16, device_map="auto",
148
+ trust_remote_code=True, token="hf_xxx",
149
+ )
150
+ model = attach_neg(model, "FINAL-Bench/Darwin-9B-NEG", hf_token="hf_xxx")
151
+ # NEG is now active — use model.generate() normally
152
+ ```
153
+
154
+ ### How NEG works at runtime
155
+
156
+ NEG is applied at every generation step:
157
+ 1. Model computes hidden state for current position
158
+ 2. NEG-Head predicts the entropy from hidden state
159
+ 3. If predicted_entropy > threshold (1.175), NEG-Gate applies top-k masking (k=20) to logits
160
+ 4. Otherwise, logits pass through unchanged
161
+ 5. argmax or sample next token
162
+
163
+ In typical reasoning traces, NEG activates on **4.6-5.1% of tokens** — only at genuinely ambiguous decision points.
164
+
165
+ ---
166
+
167
+ ## 🔬 Training Procedure
168
+
169
+ NEG was trained via 7-phase pipeline:
170
+
171
+ 1. **Phase 0-1**: Load base Darwin-9B-Opus, compute SHA256 hash for later frozen verification
172
+ 2. **Phase 2**: Collect 30,208 teacher entropy samples from GPQA extended (training set, Diamond excluded)
173
+ 3. **Phase 3**: Joint train NEG-Head + NEG-Gate with MSE (entropy) + 0.3·CE (next-token) loss, 3 epochs
174
+ 4. **Phase 4**: Verify base model hash unchanged (confirmed: 100% frozen)
175
+ 5. **Phase 5**: Evaluate baseline (Darwin-9B-Opus alone) on GPQA Diamond Greedy
176
+ 6. **Phase 6**: Evaluate NEG-enabled model on same GPQA Diamond Greedy
177
+ 7. **Phase 7**: Compare — **+11.3%p sustained improvement confirmed**
178
+
179
+ ### NEG Training Hyperparameters
180
+ - Batch size: 32
181
+ - Learning rate: 1e-4 (AdamW, weight_decay=0)
182
+ - Loss: `loss_ent + 0.3 * loss_ce`
183
+ - Epochs: 3 (early-stop at Pearson > 0.8)
184
+ - Gradient clipping: 1.0
185
+
186
+ ---
187
+
188
+ ## 📦 Files
189
+
190
+ | File | Purpose |
191
+ |---|---|
192
+ | `model-*-of-*.safetensors` | Base Darwin-9B-Opus weights (frozen) |
193
+ | `config.json` | Model config + `neg_config` metadata |
194
+ | `neg_modules.safetensors` | NEG-Head + NEG-Gate weights |
195
+ | `modeling_darwin_neg.py` | Custom loader and `attach_neg` utility |
196
+ | `tokenizer.json`, `tokenizer_config.json` | Tokenizer |
197
+ | `chat_template.jinja` | Chat template (Qwen3.5-style) |
198
+ | `README.md` | This file |
199
+
200
+ ---
201
+
202
+ ## ⚠️ Comparison with MTI
203
+
204
+ Darwin V7 uses external Multi-Turn Iteration (MTI) for reasoning enhancement. NEG is **NOT** a replacement or variant — it's a complementary technique operating at a different level:
205
+
206
+ | Property | External MTI | Darwin V8 NEG |
207
+ |:---|:---:|:---:|
208
+ | Unit of operation | Full answer (problem) | Single token |
209
+ | Signal source | Multiple sampled answers | Internal hidden state |
210
+ | Inference cost | 3-8× | **1×** |
211
+ | External pipeline | Required | **Not required** |
212
+ | Deployment | Complex | **Single file** |
213
+ | Combinable | — | Yes, multiplicative |
214
+
215
+ **NEG × MTI synergy** is expected to yield additional gains when stacked.
216
+
217
+ ---
218
+
219
+ ## 🏆 Darwin Model Family
220
+
221
+ | Model | Base | Params | GPQA Diamond (Greedy) |
222
+ |---|---|---|---|
223
+ | Darwin-9B-Opus | Qwen3.5-9B | 9 B | 51.0% |
224
+ | **Darwin-9B-NEG (this)** | Darwin-9B-Opus | **9 B** | **~62%** (+11.3%p, Greedy only) |
225
+ | Darwin-27B-Opus | Qwen3.5-27B | 27 B | 86.9% (with full 5-phase eval) |
226
+ | Darwin-36B-Opus | Qwen3.6-35B-A3B | 36 B | 88.4% (with full 5-phase eval) |
227
+
228
+ Future: **Darwin-27B-NEG**, **Darwin-36B-NEG** (targeting GPQA 90%+ at 1x cost)
229
+
230
+ ---
231
+
232
+ ## 📚 References
233
+
234
+ - [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://huggingface.co/datasets/Idavidrein/gpqa)
235
+ - Darwin V7 base model: [FINAL-Bench/Darwin-9B-Opus](https://huggingface.co/FINAL-Bench/Darwin-9B-Opus)
236
+ - NEG technical report: (see `reports/` in training repo)
237
+
238
+ ---
239
+
240
+ ## 🙏 Acknowledgments
241
+
242
+ - Qwen Team (base model architecture)
243
+ - FINAL-Bench / VIDRAFT_LAB (Darwin V8 NEG engine + training pipeline)
244
+ - Anthropic Claude Opus 4.6 (reasoning teacher for base distillation)
245
+
246
+ ---
247
+
248
+ ## 📜 Citation
249
+
250
+ ```bibtex
251
+ @misc{darwin-9b-neg,
252
+ title = {Darwin-9B-NEG: First Native Entropy Gating Model},
253
+ author = {FINAL-Bench and VIDRAFT_LAB},
254
+ year = {2026},
255
+ url = {https://huggingface.co/FINAL-Bench/Darwin-9B-NEG},
256
+ note = {Darwin V8, NEG = self-regulating reasoning at 1x inference cost}
257
+ }
258
+ ```
chat_template.jinja ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'text' in item %}
31
+ {{- item.text }}
32
+ {%- else %}
33
+ {{- raise_exception('Unexpected item type in content.') }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- elif content is none or content is undefined %}
37
+ {{- '' }}
38
+ {%- else %}
39
+ {{- raise_exception('Unexpected content type.') }}
40
+ {%- endif %}
41
+ {%- endmacro %}
42
+ {%- if not messages %}
43
+ {{- raise_exception('No messages provided.') }}
44
+ {%- endif %}
45
+ {%- if tools and tools is iterable and tools is not mapping %}
46
+ {{- '<|im_start|>system\n' }}
47
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
48
+ {%- for tool in tools %}
49
+ {{- "\n" }}
50
+ {{- tool | tojson }}
51
+ {%- endfor %}
52
+ {{- "\n</tools>" }}
53
+ {{- '\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>' }}
54
+ {%- if messages[0].role == 'system' %}
55
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
56
+ {%- if content %}
57
+ {{- '\n\n' + content }}
58
+ {%- endif %}
59
+ {%- endif %}
60
+ {{- '<|im_end|>\n' }}
61
+ {%- else %}
62
+ {%- if messages[0].role == 'system' %}
63
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
64
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
68
+ {%- for message in messages[::-1] %}
69
+ {%- set index = (messages|length - 1) - loop.index0 %}
70
+ {%- if ns.multi_step_tool and message.role == "user" %}
71
+ {%- set content = render_content(message.content, false)|trim %}
72
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
73
+ {%- set ns.multi_step_tool = false %}
74
+ {%- set ns.last_query_index = index %}
75
+ {%- endif %}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {%- if ns.multi_step_tool %}
79
+ {{- raise_exception('No user query found in messages.') }}
80
+ {%- endif %}
81
+ {%- for message in messages %}
82
+ {%- set content = render_content(message.content, true)|trim %}
83
+ {%- if message.role == "system" %}
84
+ {%- if not loop.first %}
85
+ {{- raise_exception('System message must be at the beginning.') }}
86
+ {%- endif %}
87
+ {%- elif message.role == "user" %}
88
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
89
+ {%- elif message.role == "assistant" %}
90
+ {%- set reasoning_content = '' %}
91
+ {%- if message.reasoning_content is string %}
92
+ {%- set reasoning_content = message.reasoning_content %}
93
+ {%- else %}
94
+ {%- if '</think>' in content %}
95
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
96
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
97
+ {%- endif %}
98
+ {%- endif %}
99
+ {%- set reasoning_content = reasoning_content|trim %}
100
+ {%- if loop.index0 > ns.last_query_index %}
101
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
102
+ {%- else %}
103
+ {{- '<|im_start|>' + message.role + '\n' + content }}
104
+ {%- endif %}
105
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
106
+ {%- for tool_call in message.tool_calls %}
107
+ {%- if tool_call.function is defined %}
108
+ {%- set tool_call = tool_call.function %}
109
+ {%- endif %}
110
+ {%- if loop.first %}
111
+ {%- if content|trim %}
112
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
113
+ {%- else %}
114
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
115
+ {%- endif %}
116
+ {%- else %}
117
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- endif %}
119
+ {%- if tool_call.arguments is defined %}
120
+ {%- for args_name, args_value in tool_call.arguments|items %}
121
+ {{- '<parameter=' + args_name + '>\n' }}
122
+ {%- 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 %}
123
+ {{- args_value }}
124
+ {{- '\n</parameter>\n' }}
125
+ {%- endfor %}
126
+ {%- endif %}
127
+ {{- '</function>\n</tool_call>' }}
128
+ {%- endfor %}
129
+ {%- endif %}
130
+ {{- '<|im_end|>\n' }}
131
+ {%- elif message.role == "tool" %}
132
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
133
+ {{- '<|im_start|>user' }}
134
+ {%- endif %}
135
+ {{- '\n<tool_response>\n' }}
136
+ {{- content }}
137
+ {{- '\n</tool_response>' }}
138
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
139
+ {{- '<|im_end|>\n' }}
140
+ {%- elif loop.last %}
141
+ {{- '<|im_end|>\n' }}
142
+ {%- endif %}
143
+ {%- else %}
144
+ {{- raise_exception('Unexpected message role.') }}
145
+ {%- endif %}
146
+ {%- endfor %}
147
+ {%- if add_generation_prompt %}
148
+ {{- '<|im_start|>assistant\n' }}
149
+ {%- if enable_thinking is defined and enable_thinking is false %}
150
+ {{- '<think>\n\n</think>\n\n' }}
151
+ {%- else %}
152
+ {{- '<think>\n' }}
153
+ {%- endif %}
154
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForConditionalGeneration"
4
+ ],
5
+ "image_token_id": 248056,
6
+ "model_type": "qwen3_5",
7
+ "text_config": {
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attn_output_gate": true,
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 248044,
13
+ "full_attention_interval": 4,
14
+ "head_dim": 256,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 4096,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 12288,
19
+ "layer_types": [
20
+ "linear_attention",
21
+ "linear_attention",
22
+ "linear_attention",
23
+ "full_attention",
24
+ "linear_attention",
25
+ "linear_attention",
26
+ "linear_attention",
27
+ "full_attention",
28
+ "linear_attention",
29
+ "linear_attention",
30
+ "linear_attention",
31
+ "full_attention",
32
+ "linear_attention",
33
+ "linear_attention",
34
+ "linear_attention",
35
+ "full_attention",
36
+ "linear_attention",
37
+ "linear_attention",
38
+ "linear_attention",
39
+ "full_attention",
40
+ "linear_attention",
41
+ "linear_attention",
42
+ "linear_attention",
43
+ "full_attention",
44
+ "linear_attention",
45
+ "linear_attention",
46
+ "linear_attention",
47
+ "full_attention",
48
+ "linear_attention",
49
+ "linear_attention",
50
+ "linear_attention",
51
+ "full_attention"
52
+ ],
53
+ "linear_conv_kernel_dim": 4,
54
+ "linear_key_head_dim": 128,
55
+ "linear_num_key_heads": 16,
56
+ "linear_num_value_heads": 32,
57
+ "linear_value_head_dim": 128,
58
+ "max_position_embeddings": 262144,
59
+ "mlp_only_layers": [],
60
+ "model_type": "qwen3_5_text",
61
+ "mtp_num_hidden_layers": 1,
62
+ "mtp_use_dedicated_embeddings": false,
63
+ "num_attention_heads": 16,
64
+ "num_hidden_layers": 32,
65
+ "num_key_value_heads": 4,
66
+ "rms_norm_eps": 1e-06,
67
+ "use_cache": true,
68
+ "vocab_size": 248320,
69
+ "mamba_ssm_dtype": "float32",
70
+ "rope_parameters": {
71
+ "mrope_interleaved": true,
72
+ "mrope_section": [
73
+ 11,
74
+ 11,
75
+ 10
76
+ ],
77
+ "rope_type": "default",
78
+ "rope_theta": 10000000,
79
+ "partial_rotary_factor": 0.25
80
+ }
81
+ },
82
+ "tie_word_embeddings": false,
83
+ "transformers_version": "4.57.0.dev0",
84
+ "video_token_id": 248057,
85
+ "vision_config": {
86
+ "deepstack_visual_indexes": [],
87
+ "depth": 27,
88
+ "hidden_act": "gelu_pytorch_tanh",
89
+ "hidden_size": 1152,
90
+ "in_channels": 3,
91
+ "initializer_range": 0.02,
92
+ "intermediate_size": 4304,
93
+ "model_type": "qwen3_5",
94
+ "num_heads": 16,
95
+ "num_position_embeddings": 2304,
96
+ "out_hidden_size": 4096,
97
+ "patch_size": 16,
98
+ "spatial_merge_size": 2,
99
+ "temporal_patch_size": 2
100
+ },
101
+ "vision_end_token_id": 248054,
102
+ "vision_start_token_id": 248053,
103
+ "neg_config": {
104
+ "enabled": true,
105
+ "threshold": 1.175187349319458,
106
+ "top_k": 20,
107
+ "version": "1.0",
108
+ "base_eval_pearson": 0.8744,
109
+ "hidden_size_for_head": 4096
110
+ },
111
+ "_darwin_v8": "Darwin-9B-NEG (Native Entropy Gating)"
112
+ }
darwin_mri_report.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "layers_total": 775,
3
+ "transplant_a": 0,
4
+ "transplant_b": 0,
5
+ "blended": 775,
6
+ "details": {}
7
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bbd456f1367d1d9d7273b0a5735a57ce73f6a56f5a09c3b99a8a607a0a5f65a
3
+ size 5276436216
model.safetensors-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:048129af3b6acde304c92fc262b12db11e25bc23a5187c42a60b0a6ee16749fb
3
+ size 5335161512
model.safetensors-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7283cf97c0bc17a351e1b08ba6b6f3d4c2920704a6d1b1cacfb6ae6510c45730
3
+ size 5368717440
model.safetensors-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53161974a653473c3829f77974fda95d7bbaabc62a9ee309925a663625fdf0ee
3
+ size 3325988568
model.safetensors.index.json ADDED
@@ -0,0 +1,782 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 19306216416
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model.safetensors-00001-of-00004.safetensors",
7
+ "model.language_model.embed_tokens.weight": "model.safetensors-00001-of-00004.safetensors",
8
+ "model.language_model.layers.14.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
9
+ "model.language_model.layers.14.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
10
+ "model.language_model.layers.14.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
11
+ "model.language_model.layers.15.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
12
+ "model.language_model.layers.15.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
13
+ "model.language_model.layers.15.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
14
+ "model.language_model.layers.25.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
15
+ "model.language_model.layers.26.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
16
+ "model.language_model.layers.26.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
17
+ "model.language_model.layers.26.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
18
+ "model.language_model.layers.27.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
19
+ "model.language_model.layers.27.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
20
+ "model.language_model.layers.27.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
21
+ "model.language_model.layers.3.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
22
+ "model.language_model.layers.3.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
23
+ "model.language_model.layers.3.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
24
+ "model.language_model.layers.30.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
25
+ "model.language_model.layers.30.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
26
+ "model.language_model.layers.30.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
27
+ "model.language_model.layers.31.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
28
+ "model.language_model.layers.31.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
29
+ "model.language_model.layers.9.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
30
+ "model.language_model.layers.9.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
31
+ "model.language_model.layers.9.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
32
+ "mtp.layers.0.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
33
+ "mtp.layers.0.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
34
+ "mtp.layers.0.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
35
+ "model.language_model.layers.1.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
36
+ "model.language_model.layers.1.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
37
+ "model.language_model.layers.1.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
38
+ "model.language_model.layers.10.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
39
+ "model.language_model.layers.10.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
40
+ "model.language_model.layers.10.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
41
+ "model.language_model.layers.11.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
42
+ "model.language_model.layers.11.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
43
+ "model.language_model.layers.11.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
44
+ "model.language_model.layers.0.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
45
+ "model.language_model.layers.0.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
46
+ "model.language_model.layers.0.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
47
+ "model.language_model.layers.21.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
48
+ "model.language_model.layers.21.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
49
+ "model.language_model.layers.21.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
50
+ "model.language_model.layers.22.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
51
+ "model.language_model.layers.22.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
52
+ "model.language_model.layers.22.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
53
+ "model.language_model.layers.23.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
54
+ "model.language_model.layers.23.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
55
+ "model.language_model.layers.23.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
56
+ "model.language_model.layers.8.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
57
+ "model.language_model.layers.8.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
58
+ "model.language_model.layers.8.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
59
+ "model.language_model.layers.31.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
60
+ "model.language_model.layers.4.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
61
+ "model.language_model.layers.4.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
62
+ "model.language_model.layers.4.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
63
+ "model.language_model.layers.16.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
64
+ "model.language_model.layers.16.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
65
+ "model.language_model.layers.16.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
66
+ "model.language_model.layers.17.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
67
+ "model.language_model.layers.17.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
68
+ "model.language_model.layers.17.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
69
+ "model.language_model.layers.18.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
70
+ "model.language_model.layers.18.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
71
+ "model.language_model.layers.18.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
72
+ "model.language_model.layers.19.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
73
+ "model.language_model.layers.19.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
74
+ "model.language_model.layers.19.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
75
+ "model.language_model.layers.24.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
76
+ "model.language_model.layers.24.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
77
+ "model.language_model.layers.24.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
78
+ "model.language_model.layers.25.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
79
+ "model.language_model.layers.25.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
80
+ "model.language_model.layers.5.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
81
+ "model.language_model.layers.5.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
82
+ "model.language_model.layers.5.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
83
+ "model.language_model.layers.6.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
84
+ "model.language_model.layers.6.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
85
+ "model.language_model.layers.6.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
86
+ "model.language_model.layers.7.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
87
+ "model.language_model.layers.7.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
88
+ "model.language_model.layers.7.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
89
+ "model.language_model.layers.2.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
90
+ "model.language_model.layers.2.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
91
+ "model.language_model.layers.2.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
92
+ "model.language_model.layers.20.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
93
+ "model.language_model.layers.20.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
94
+ "model.language_model.layers.20.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
95
+ "model.language_model.layers.28.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
96
+ "model.language_model.layers.28.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
97
+ "model.language_model.layers.28.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
98
+ "model.language_model.layers.29.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
99
+ "model.language_model.layers.29.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
100
+ "model.language_model.layers.29.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
101
+ "model.language_model.layers.12.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
102
+ "model.language_model.layers.12.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
103
+ "model.language_model.layers.12.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
104
+ "model.language_model.layers.13.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
105
+ "model.language_model.layers.13.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
106
+ "model.language_model.layers.13.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
107
+ "model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
108
+ "model.language_model.layers.15.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
109
+ "model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
110
+ "model.language_model.layers.27.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
111
+ "model.language_model.layers.3.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
112
+ "model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
113
+ "model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
114
+ "mtp.fc.weight": "model.safetensors-00003-of-00004.safetensors",
115
+ "mtp.layers.0.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
116
+ "model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
117
+ "model.language_model.layers.11.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
118
+ "model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
119
+ "model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
120
+ "model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
121
+ "model.language_model.layers.23.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
122
+ "model.language_model.layers.7.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
123
+ "model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
124
+ "model.language_model.layers.31.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
125
+ "model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
126
+ "model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
127
+ "model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
128
+ "model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
129
+ "model.language_model.layers.19.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
130
+ "model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
131
+ "model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
132
+ "model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
133
+ "model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
134
+ "model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
135
+ "model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
136
+ "model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
137
+ "model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
138
+ "model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
139
+ "model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
140
+ "model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
141
+ "model.visual.merger.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
142
+ "model.visual.merger.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
143
+ "model.language_model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
144
+ "model.language_model.layers.14.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
145
+ "model.language_model.layers.15.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
146
+ "model.language_model.layers.26.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
147
+ "model.language_model.layers.26.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
148
+ "model.language_model.layers.27.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
149
+ "model.language_model.layers.3.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
150
+ "model.language_model.layers.30.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
151
+ "model.language_model.layers.30.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
152
+ "model.language_model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
153
+ "model.language_model.layers.9.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
154
+ "mtp.layers.0.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
155
+ "model.language_model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
156
+ "model.language_model.layers.10.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
157
+ "model.language_model.layers.11.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
158
+ "model.language_model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
159
+ "model.language_model.layers.0.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
160
+ "model.language_model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
161
+ "model.language_model.layers.1.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
162
+ "model.language_model.layers.21.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
163
+ "model.language_model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
164
+ "model.language_model.layers.22.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
165
+ "model.language_model.layers.23.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
166
+ "model.language_model.layers.7.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
167
+ "model.language_model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
168
+ "model.language_model.layers.8.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
169
+ "model.language_model.layers.31.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
170
+ "model.language_model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
171
+ "model.language_model.layers.4.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
172
+ "model.language_model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
173
+ "model.language_model.layers.16.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
174
+ "model.language_model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
175
+ "model.language_model.layers.17.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
176
+ "model.language_model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
177
+ "model.language_model.layers.18.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
178
+ "model.language_model.layers.19.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
179
+ "model.language_model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
180
+ "model.language_model.layers.24.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
181
+ "model.language_model.layers.24.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
182
+ "model.language_model.layers.25.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
183
+ "model.language_model.layers.25.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
184
+ "model.language_model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
185
+ "model.language_model.layers.5.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
186
+ "model.language_model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
187
+ "model.language_model.layers.6.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
188
+ "model.language_model.layers.2.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
189
+ "model.language_model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
190
+ "model.language_model.layers.20.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
191
+ "model.language_model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
192
+ "model.language_model.layers.28.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
193
+ "model.language_model.layers.28.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
194
+ "model.language_model.layers.29.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
195
+ "model.language_model.layers.29.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
196
+ "model.language_model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
197
+ "model.language_model.layers.12.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
198
+ "model.language_model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
199
+ "model.language_model.layers.13.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
200
+ "model.visual.blocks.0.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
201
+ "model.visual.blocks.0.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
202
+ "model.visual.blocks.1.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
203
+ "model.visual.blocks.1.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
204
+ "model.visual.blocks.10.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
205
+ "model.visual.blocks.10.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
206
+ "model.visual.blocks.11.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
207
+ "model.visual.blocks.11.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
208
+ "model.visual.blocks.12.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
209
+ "model.visual.blocks.12.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
210
+ "model.visual.blocks.13.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
211
+ "model.visual.blocks.13.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
212
+ "model.visual.blocks.14.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
213
+ "model.visual.blocks.14.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
214
+ "model.visual.blocks.15.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
215
+ "model.visual.blocks.15.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
216
+ "model.visual.blocks.16.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
217
+ "model.visual.blocks.16.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
218
+ "model.visual.blocks.17.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
219
+ "model.visual.blocks.17.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
220
+ "model.visual.blocks.18.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
221
+ "model.visual.blocks.18.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
222
+ "model.visual.blocks.19.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
223
+ "model.visual.blocks.19.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
224
+ "model.visual.blocks.2.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
225
+ "model.visual.blocks.2.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
226
+ "model.visual.blocks.20.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
227
+ "model.visual.blocks.20.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
228
+ "model.visual.blocks.21.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
229
+ "model.visual.blocks.21.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
230
+ "model.visual.blocks.22.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
231
+ "model.visual.blocks.22.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
232
+ "model.visual.blocks.23.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
233
+ "model.visual.blocks.23.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
234
+ "model.visual.blocks.24.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
235
+ "model.visual.blocks.24.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
236
+ "model.visual.blocks.25.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
237
+ "model.visual.blocks.25.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
238
+ "model.visual.blocks.26.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
239
+ "model.visual.blocks.26.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
240
+ "model.visual.blocks.3.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
241
+ "model.visual.blocks.3.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
242
+ "model.visual.blocks.4.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
243
+ "model.visual.blocks.4.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
244
+ "model.visual.blocks.5.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
245
+ "model.visual.blocks.5.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
246
+ "model.visual.blocks.6.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
247
+ "model.visual.blocks.6.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
248
+ "model.visual.blocks.7.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
249
+ "model.visual.blocks.7.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
250
+ "model.visual.blocks.8.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
251
+ "model.visual.blocks.8.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
252
+ "model.visual.blocks.9.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
253
+ "model.visual.blocks.9.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
254
+ "model.language_model.layers.15.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
255
+ "model.language_model.layers.15.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
256
+ "model.language_model.layers.27.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
257
+ "model.language_model.layers.27.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
258
+ "model.language_model.layers.3.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
259
+ "model.language_model.layers.3.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
260
+ "mtp.layers.0.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
261
+ "mtp.layers.0.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
262
+ "model.language_model.layers.11.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
263
+ "model.language_model.layers.11.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
264
+ "model.language_model.layers.23.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
265
+ "model.language_model.layers.23.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
266
+ "model.language_model.layers.7.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
267
+ "model.language_model.layers.7.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
268
+ "model.language_model.layers.31.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
269
+ "model.language_model.layers.31.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
270
+ "model.language_model.layers.19.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
271
+ "model.language_model.layers.19.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
272
+ "model.visual.blocks.0.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
273
+ "model.visual.blocks.1.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
274
+ "model.visual.blocks.10.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
275
+ "model.visual.blocks.11.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
276
+ "model.visual.blocks.12.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
277
+ "model.visual.blocks.13.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
278
+ "model.visual.blocks.14.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
279
+ "model.visual.blocks.15.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
280
+ "model.visual.blocks.16.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
281
+ "model.visual.blocks.17.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
282
+ "model.visual.blocks.18.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
283
+ "model.visual.blocks.19.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
284
+ "model.visual.blocks.2.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
285
+ "model.visual.blocks.20.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
286
+ "model.visual.blocks.21.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
287
+ "model.visual.blocks.22.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
288
+ "model.visual.blocks.23.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
289
+ "model.visual.blocks.24.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
290
+ "model.visual.blocks.25.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
291
+ "model.visual.blocks.26.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
292
+ "model.visual.blocks.3.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
293
+ "model.visual.blocks.4.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
294
+ "model.visual.blocks.5.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
295
+ "model.visual.blocks.6.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
296
+ "model.visual.blocks.7.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
297
+ "model.visual.blocks.8.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
298
+ "model.visual.blocks.9.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
299
+ "model.visual.pos_embed.weight": "model.safetensors-00004-of-00004.safetensors",
300
+ "model.visual.patch_embed.proj.weight": "model.safetensors-00004-of-00004.safetensors",
301
+ "model.visual.blocks.0.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
302
+ "model.visual.blocks.1.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
303
+ "model.visual.blocks.10.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
304
+ "model.visual.blocks.11.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
305
+ "model.visual.blocks.12.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
306
+ "model.visual.blocks.13.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
307
+ "model.visual.blocks.14.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
308
+ "model.visual.blocks.15.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
309
+ "model.visual.blocks.16.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
310
+ "model.visual.blocks.17.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
311
+ "model.visual.blocks.18.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
312
+ "model.visual.blocks.19.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
313
+ "model.visual.blocks.2.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
314
+ "model.visual.blocks.20.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
315
+ "model.visual.blocks.21.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
316
+ "model.visual.blocks.22.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
317
+ "model.visual.blocks.23.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
318
+ "model.visual.blocks.24.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
319
+ "model.visual.blocks.25.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
320
+ "model.visual.blocks.26.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
321
+ "model.visual.blocks.3.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
322
+ "model.visual.blocks.4.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
323
+ "model.visual.blocks.5.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
324
+ "model.visual.blocks.6.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
325
+ "model.visual.blocks.7.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
326
+ "model.visual.blocks.8.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
327
+ "model.visual.blocks.9.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
328
+ "model.language_model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
329
+ "model.language_model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
330
+ "model.language_model.layers.26.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
331
+ "model.language_model.layers.26.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
332
+ "model.language_model.layers.30.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
333
+ "model.language_model.layers.30.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
334
+ "model.language_model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
335
+ "model.language_model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
336
+ "model.language_model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
337
+ "model.language_model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
338
+ "model.language_model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
339
+ "model.language_model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
340
+ "model.language_model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
341
+ "model.language_model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
342
+ "model.language_model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
343
+ "model.language_model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
344
+ "model.language_model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
345
+ "model.language_model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
346
+ "model.language_model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
347
+ "model.language_model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
348
+ "model.language_model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
349
+ "model.language_model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
350
+ "model.language_model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
351
+ "model.language_model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
352
+ "model.language_model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
353
+ "model.language_model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
354
+ "model.language_model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
355
+ "model.language_model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
356
+ "model.language_model.layers.24.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
357
+ "model.language_model.layers.24.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
358
+ "model.language_model.layers.25.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
359
+ "model.language_model.layers.25.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
360
+ "model.language_model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
361
+ "model.language_model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
362
+ "model.language_model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
363
+ "model.language_model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
364
+ "model.language_model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
365
+ "model.language_model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
366
+ "model.language_model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
367
+ "model.language_model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
368
+ "model.language_model.layers.28.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
369
+ "model.language_model.layers.28.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
370
+ "model.language_model.layers.29.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
371
+ "model.language_model.layers.29.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
372
+ "model.language_model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
373
+ "model.language_model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
374
+ "model.language_model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
375
+ "model.language_model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
376
+ "model.language_model.layers.14.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
377
+ "model.language_model.layers.16.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
378
+ "model.language_model.layers.26.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
379
+ "model.language_model.layers.30.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
380
+ "model.language_model.layers.10.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
381
+ "model.language_model.layers.0.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
382
+ "model.language_model.layers.1.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
383
+ "model.language_model.layers.22.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
384
+ "model.language_model.layers.8.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
385
+ "model.language_model.layers.9.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
386
+ "model.language_model.layers.4.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
387
+ "model.language_model.layers.5.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
388
+ "model.language_model.layers.17.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
389
+ "model.language_model.layers.18.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
390
+ "model.language_model.layers.2.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
391
+ "model.language_model.layers.24.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
392
+ "model.language_model.layers.25.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
393
+ "model.language_model.layers.6.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
394
+ "model.language_model.layers.20.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
395
+ "model.language_model.layers.21.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
396
+ "model.language_model.layers.28.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
397
+ "model.language_model.layers.29.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
398
+ "model.language_model.layers.12.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
399
+ "model.language_model.layers.13.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
400
+ "model.visual.merger.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
401
+ "model.visual.blocks.0.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
402
+ "model.visual.blocks.1.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
403
+ "model.visual.blocks.10.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
404
+ "model.visual.blocks.11.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
405
+ "model.visual.blocks.12.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
406
+ "model.visual.blocks.13.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
407
+ "model.visual.blocks.14.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
408
+ "model.visual.blocks.15.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
409
+ "model.visual.blocks.16.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
410
+ "model.visual.blocks.17.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
411
+ "model.visual.blocks.18.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
412
+ "model.visual.blocks.19.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
413
+ "model.visual.blocks.2.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
414
+ "model.visual.blocks.20.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
415
+ "model.visual.blocks.21.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
416
+ "model.visual.blocks.22.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
417
+ "model.visual.blocks.23.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
418
+ "model.visual.blocks.24.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
419
+ "model.visual.blocks.25.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
420
+ "model.visual.blocks.26.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
421
+ "model.visual.blocks.3.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
422
+ "model.visual.blocks.4.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
423
+ "model.visual.blocks.5.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
424
+ "model.visual.blocks.6.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
425
+ "model.visual.blocks.7.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
426
+ "model.visual.blocks.8.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
427
+ "model.visual.blocks.9.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
428
+ "model.language_model.layers.14.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
429
+ "model.language_model.layers.14.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
430
+ "model.language_model.layers.15.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
431
+ "model.language_model.layers.15.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
432
+ "model.language_model.layers.16.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
433
+ "model.language_model.layers.26.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
434
+ "model.language_model.layers.26.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
435
+ "model.language_model.layers.27.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
436
+ "model.language_model.layers.27.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
437
+ "model.language_model.layers.3.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
438
+ "model.language_model.layers.3.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
439
+ "model.language_model.layers.30.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
440
+ "model.language_model.layers.30.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
441
+ "model.language_model.layers.31.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
442
+ "model.language_model.layers.9.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
443
+ "mtp.layers.0.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
444
+ "mtp.layers.0.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
445
+ "mtp.norm.weight": "model.safetensors-00004-of-00004.safetensors",
446
+ "mtp.pre_fc_norm_embedding.weight": "model.safetensors-00004-of-00004.safetensors",
447
+ "mtp.pre_fc_norm_hidden.weight": "model.safetensors-00004-of-00004.safetensors",
448
+ "model.language_model.layers.10.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
449
+ "model.language_model.layers.10.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
450
+ "model.language_model.layers.11.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
451
+ "model.language_model.layers.11.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
452
+ "model.language_model.layers.0.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
453
+ "model.language_model.layers.0.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
454
+ "model.language_model.layers.1.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
455
+ "model.language_model.layers.1.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
456
+ "model.language_model.norm.weight": "model.safetensors-00004-of-00004.safetensors",
457
+ "model.visual.merger.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
458
+ "model.language_model.layers.21.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
459
+ "model.language_model.layers.22.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
460
+ "model.language_model.layers.22.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
461
+ "model.language_model.layers.23.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
462
+ "model.language_model.layers.23.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
463
+ "model.language_model.layers.7.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
464
+ "model.language_model.layers.8.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
465
+ "model.language_model.layers.8.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
466
+ "model.language_model.layers.9.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
467
+ "model.language_model.layers.31.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
468
+ "model.language_model.layers.4.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
469
+ "model.language_model.layers.4.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
470
+ "model.language_model.layers.5.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
471
+ "model.language_model.layers.16.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
472
+ "model.language_model.layers.17.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
473
+ "model.language_model.layers.17.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
474
+ "model.language_model.layers.18.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
475
+ "model.language_model.layers.18.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
476
+ "model.language_model.layers.19.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
477
+ "model.language_model.layers.19.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
478
+ "model.language_model.layers.2.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
479
+ "model.language_model.layers.24.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
480
+ "model.language_model.layers.24.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
481
+ "model.language_model.layers.25.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
482
+ "model.language_model.layers.25.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
483
+ "model.language_model.layers.5.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
484
+ "model.language_model.layers.6.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
485
+ "model.language_model.layers.6.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
486
+ "model.language_model.layers.7.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
487
+ "model.language_model.layers.2.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
488
+ "model.language_model.layers.20.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
489
+ "model.language_model.layers.20.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
490
+ "model.language_model.layers.21.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
491
+ "model.language_model.layers.28.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
492
+ "model.language_model.layers.28.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
493
+ "model.language_model.layers.29.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
494
+ "model.language_model.layers.29.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
495
+ "model.language_model.layers.12.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
496
+ "model.language_model.layers.12.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
497
+ "model.language_model.layers.13.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
498
+ "model.language_model.layers.13.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
499
+ "model.visual.blocks.0.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
500
+ "model.visual.blocks.1.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
501
+ "model.visual.blocks.10.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
502
+ "model.visual.blocks.11.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
503
+ "model.visual.blocks.12.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
504
+ "model.visual.blocks.13.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
505
+ "model.visual.blocks.14.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
506
+ "model.visual.blocks.15.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
507
+ "model.visual.blocks.16.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
508
+ "model.visual.blocks.17.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
509
+ "model.visual.blocks.18.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
510
+ "model.visual.blocks.19.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
511
+ "model.visual.blocks.2.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
512
+ "model.visual.blocks.20.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
513
+ "model.visual.blocks.21.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
514
+ "model.visual.blocks.22.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
515
+ "model.visual.blocks.23.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
516
+ "model.visual.blocks.24.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
517
+ "model.visual.blocks.25.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
518
+ "model.visual.blocks.26.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
519
+ "model.visual.blocks.3.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
520
+ "model.visual.blocks.4.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
521
+ "model.visual.blocks.5.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
522
+ "model.visual.blocks.6.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
523
+ "model.visual.blocks.7.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
524
+ "model.visual.blocks.8.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
525
+ "model.visual.blocks.9.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
526
+ "model.visual.blocks.0.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
527
+ "model.visual.blocks.0.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
528
+ "model.visual.blocks.0.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
529
+ "model.visual.blocks.0.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
530
+ "model.visual.blocks.0.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
531
+ "model.visual.blocks.0.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
532
+ "model.visual.blocks.1.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
533
+ "model.visual.blocks.1.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
534
+ "model.visual.blocks.1.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
535
+ "model.visual.blocks.1.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
536
+ "model.visual.blocks.1.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
537
+ "model.visual.blocks.1.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
538
+ "model.visual.blocks.10.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
539
+ "model.visual.blocks.10.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
540
+ "model.visual.blocks.10.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
541
+ "model.visual.blocks.10.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
542
+ "model.visual.blocks.10.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
543
+ "model.visual.blocks.10.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
544
+ "model.visual.blocks.11.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
545
+ "model.visual.blocks.11.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
546
+ "model.visual.blocks.11.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
547
+ "model.visual.blocks.11.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
548
+ "model.visual.blocks.11.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
549
+ "model.visual.blocks.11.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
550
+ "model.visual.blocks.12.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
551
+ "model.visual.blocks.12.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
552
+ "model.visual.blocks.12.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
553
+ "model.visual.blocks.12.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
554
+ "model.visual.blocks.12.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
555
+ "model.visual.blocks.12.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
556
+ "model.visual.blocks.13.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
557
+ "model.visual.blocks.13.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
558
+ "model.visual.blocks.13.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
559
+ "model.visual.blocks.13.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
560
+ "model.visual.blocks.13.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
561
+ "model.visual.blocks.13.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
562
+ "model.visual.blocks.14.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
563
+ "model.visual.blocks.14.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
564
+ "model.visual.blocks.14.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
565
+ "model.visual.blocks.14.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
566
+ "model.visual.blocks.14.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
567
+ "model.visual.blocks.14.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
568
+ "model.visual.blocks.15.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
569
+ "model.visual.blocks.15.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
570
+ "model.visual.blocks.15.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
571
+ "model.visual.blocks.15.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
572
+ "model.visual.blocks.15.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
573
+ "model.visual.blocks.15.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
574
+ "model.visual.blocks.16.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
575
+ "model.visual.blocks.16.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
576
+ "model.visual.blocks.16.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
577
+ "model.visual.blocks.16.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
578
+ "model.visual.blocks.16.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
579
+ "model.visual.blocks.16.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
580
+ "model.visual.blocks.17.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
581
+ "model.visual.blocks.17.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
582
+ "model.visual.blocks.17.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
583
+ "model.visual.blocks.17.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
584
+ "model.visual.blocks.17.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
585
+ "model.visual.blocks.17.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
586
+ "model.visual.blocks.18.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
587
+ "model.visual.blocks.18.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
588
+ "model.visual.blocks.18.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
589
+ "model.visual.blocks.18.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
590
+ "model.visual.blocks.18.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
591
+ "model.visual.blocks.18.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
592
+ "model.visual.blocks.19.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
593
+ "model.visual.blocks.19.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
594
+ "model.visual.blocks.19.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
595
+ "model.visual.blocks.19.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
596
+ "model.visual.blocks.19.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
597
+ "model.visual.blocks.19.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
598
+ "model.visual.blocks.2.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
599
+ "model.visual.blocks.2.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
600
+ "model.visual.blocks.2.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
601
+ "model.visual.blocks.2.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
602
+ "model.visual.blocks.2.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
603
+ "model.visual.blocks.2.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
604
+ "model.visual.blocks.20.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
605
+ "model.visual.blocks.20.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
606
+ "model.visual.blocks.20.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
607
+ "model.visual.blocks.20.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
608
+ "model.visual.blocks.20.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
609
+ "model.visual.blocks.20.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
610
+ "model.visual.blocks.21.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
611
+ "model.visual.blocks.21.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
612
+ "model.visual.blocks.21.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
613
+ "model.visual.blocks.21.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
614
+ "model.visual.blocks.21.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
615
+ "model.visual.blocks.21.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
616
+ "model.visual.blocks.22.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
617
+ "model.visual.blocks.22.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
618
+ "model.visual.blocks.22.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
619
+ "model.visual.blocks.22.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
620
+ "model.visual.blocks.22.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
621
+ "model.visual.blocks.22.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
622
+ "model.visual.blocks.23.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
623
+ "model.visual.blocks.23.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
624
+ "model.visual.blocks.23.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
625
+ "model.visual.blocks.23.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
626
+ "model.visual.blocks.23.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
627
+ "model.visual.blocks.23.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
628
+ "model.visual.blocks.24.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
629
+ "model.visual.blocks.24.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
630
+ "model.visual.blocks.24.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
631
+ "model.visual.blocks.24.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
632
+ "model.visual.blocks.24.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
633
+ "model.visual.blocks.24.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
634
+ "model.visual.blocks.25.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
635
+ "model.visual.blocks.25.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
636
+ "model.visual.blocks.25.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
637
+ "model.visual.blocks.25.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
638
+ "model.visual.blocks.25.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
639
+ "model.visual.blocks.25.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
640
+ "model.visual.blocks.26.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
641
+ "model.visual.blocks.26.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
642
+ "model.visual.blocks.26.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
643
+ "model.visual.blocks.26.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
644
+ "model.visual.blocks.26.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
645
+ "model.visual.blocks.26.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
646
+ "model.visual.blocks.3.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
647
+ "model.visual.blocks.3.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
648
+ "model.visual.blocks.3.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
649
+ "model.visual.blocks.3.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
650
+ "model.visual.blocks.3.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
651
+ "model.visual.blocks.3.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
652
+ "model.visual.blocks.4.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
653
+ "model.visual.blocks.4.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
654
+ "model.visual.blocks.4.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
655
+ "model.visual.blocks.4.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
656
+ "model.visual.blocks.4.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
657
+ "model.visual.blocks.4.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
658
+ "model.visual.blocks.5.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
659
+ "model.visual.blocks.5.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
660
+ "model.visual.blocks.5.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
661
+ "model.visual.blocks.5.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
662
+ "model.visual.blocks.5.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
663
+ "model.visual.blocks.5.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
664
+ "model.visual.blocks.6.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
665
+ "model.visual.blocks.6.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
666
+ "model.visual.blocks.6.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
667
+ "model.visual.blocks.6.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
668
+ "model.visual.blocks.6.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
669
+ "model.visual.blocks.6.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
670
+ "model.visual.blocks.7.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
671
+ "model.visual.blocks.7.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
672
+ "model.visual.blocks.7.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
673
+ "model.visual.blocks.7.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
674
+ "model.visual.blocks.7.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
675
+ "model.visual.blocks.7.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
676
+ "model.visual.blocks.8.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
677
+ "model.visual.blocks.8.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
678
+ "model.visual.blocks.8.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
679
+ "model.visual.blocks.8.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
680
+ "model.visual.blocks.8.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
681
+ "model.visual.blocks.8.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
682
+ "model.visual.blocks.9.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
683
+ "model.visual.blocks.9.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
684
+ "model.visual.blocks.9.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
685
+ "model.visual.blocks.9.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
686
+ "model.visual.blocks.9.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
687
+ "model.visual.blocks.9.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
688
+ "model.visual.merger.norm.bias": "model.safetensors-00004-of-00004.safetensors",
689
+ "model.visual.merger.norm.weight": "model.safetensors-00004-of-00004.safetensors",
690
+ "model.visual.patch_embed.proj.bias": "model.safetensors-00004-of-00004.safetensors",
691
+ "model.language_model.layers.14.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
692
+ "model.language_model.layers.15.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
693
+ "model.language_model.layers.15.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
694
+ "model.language_model.layers.26.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
695
+ "model.language_model.layers.27.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
696
+ "model.language_model.layers.27.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
697
+ "model.language_model.layers.3.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
698
+ "model.language_model.layers.3.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
699
+ "model.language_model.layers.30.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
700
+ "model.language_model.layers.9.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
701
+ "mtp.layers.0.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
702
+ "mtp.layers.0.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
703
+ "model.language_model.layers.10.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
704
+ "model.language_model.layers.11.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
705
+ "model.language_model.layers.11.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
706
+ "model.language_model.layers.0.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
707
+ "model.language_model.layers.1.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
708
+ "model.language_model.layers.22.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
709
+ "model.language_model.layers.23.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
710
+ "model.language_model.layers.7.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
711
+ "model.language_model.layers.7.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
712
+ "model.language_model.layers.8.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
713
+ "model.language_model.layers.31.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
714
+ "model.language_model.layers.31.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
715
+ "model.language_model.layers.4.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
716
+ "model.language_model.layers.16.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
717
+ "model.language_model.layers.17.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
718
+ "model.language_model.layers.18.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
719
+ "model.language_model.layers.19.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
720
+ "model.language_model.layers.19.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
721
+ "model.language_model.layers.2.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
722
+ "model.language_model.layers.23.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
723
+ "model.language_model.layers.24.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
724
+ "model.language_model.layers.25.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
725
+ "model.language_model.layers.5.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
726
+ "model.language_model.layers.6.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
727
+ "model.language_model.layers.20.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
728
+ "model.language_model.layers.21.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
729
+ "model.language_model.layers.28.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
730
+ "model.language_model.layers.29.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
731
+ "model.language_model.layers.12.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
732
+ "model.language_model.layers.13.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
733
+ "model.language_model.layers.16.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
734
+ "model.language_model.layers.26.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
735
+ "model.language_model.layers.28.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
736
+ "model.language_model.layers.30.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
737
+ "model.language_model.layers.10.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
738
+ "model.language_model.layers.12.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
739
+ "model.language_model.layers.0.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
740
+ "model.language_model.layers.1.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
741
+ "model.language_model.layers.22.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
742
+ "model.language_model.layers.8.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
743
+ "model.language_model.layers.9.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
744
+ "model.language_model.layers.4.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
745
+ "model.language_model.layers.5.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
746
+ "model.language_model.layers.17.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
747
+ "model.language_model.layers.18.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
748
+ "model.language_model.layers.2.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
749
+ "model.language_model.layers.24.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
750
+ "model.language_model.layers.25.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
751
+ "model.language_model.layers.6.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
752
+ "model.language_model.layers.20.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
753
+ "model.language_model.layers.21.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
754
+ "model.language_model.layers.29.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
755
+ "model.language_model.layers.13.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
756
+ "model.language_model.layers.14.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
757
+ "model.language_model.layers.14.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
758
+ "model.language_model.layers.16.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
759
+ "model.language_model.layers.26.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
760
+ "model.language_model.layers.30.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
761
+ "model.language_model.layers.10.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
762
+ "model.language_model.layers.0.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
763
+ "model.language_model.layers.1.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
764
+ "model.language_model.layers.22.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
765
+ "model.language_model.layers.8.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
766
+ "model.language_model.layers.9.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
767
+ "model.language_model.layers.4.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
768
+ "model.language_model.layers.5.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
769
+ "model.language_model.layers.17.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
770
+ "model.language_model.layers.18.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
771
+ "model.language_model.layers.2.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
772
+ "model.language_model.layers.24.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
773
+ "model.language_model.layers.25.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
774
+ "model.language_model.layers.6.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
775
+ "model.language_model.layers.20.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
776
+ "model.language_model.layers.21.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
777
+ "model.language_model.layers.28.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
778
+ "model.language_model.layers.29.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
779
+ "model.language_model.layers.12.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
780
+ "model.language_model.layers.13.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors"
781
+ }
782
+ }
modeling_darwin_neg.py ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ Darwin-9B-NEG — Native Entropy Gating enabled model.
3
+
4
+ Helper module to attach NEG (Native Entropy Gating) to a Darwin base model.
5
+ Provides:
6
+ - NEGHead : predicts per-token entropy from last hidden state
7
+ - NEGGate : non-monotonic top-k logit masking (effective in greedy decoding)
8
+ - attach_neg(model, path_or_repo) : monkey-patches forward to apply NEG
9
+
10
+ See README.md for usage.
11
+ """
12
+ import os
13
+ import torch
14
+ import torch.nn as nn
15
+ import torch.nn.functional as F
16
+ from safetensors.torch import load_file
17
+
18
+
19
+ class NEGHead(nn.Module):
20
+ """NEG-Head: predicts entropy of next-token distribution.
21
+
22
+ Input: hidden_state [B, H]
23
+ Output: predicted_entropy [B] (>= 0 via softplus)
24
+ """
25
+ def __init__(self, hidden: int, dropout: float = 0.1):
26
+ super().__init__()
27
+ self.proj_down = nn.Linear(hidden, hidden // 4)
28
+ self.act = nn.GELU()
29
+ self.dropout = nn.Dropout(dropout)
30
+ self.proj_out = nn.Linear(hidden // 4, 1)
31
+
32
+ def forward(self, h):
33
+ x = self.proj_down(h)
34
+ x = self.act(x)
35
+ x = self.dropout(x)
36
+ return F.softplus(self.proj_out(x).squeeze(-1))
37
+
38
+
39
+ class NEGGate(nn.Module):
40
+ """NEG-Gate: top-k logit masking (non-monotonic).
41
+
42
+ When predicted_entropy > threshold, restrict logits to top-k candidates.
43
+ This changes argmax (non-monotonic), making NEG effective in greedy decoding.
44
+ """
45
+ def __init__(self, init_threshold: float = 1.175, top_k: int = 20):
46
+ super().__init__()
47
+ self.threshold = nn.Parameter(torch.tensor(init_threshold))
48
+ self.top_k = top_k
49
+
50
+ def forward(self, logits, predicted_entropy):
51
+ activate = (predicted_entropy > self.threshold).float().unsqueeze(-1)
52
+ if activate.sum() == 0:
53
+ return logits
54
+ top_k_vals, top_k_idx = logits.topk(self.top_k, dim=-1)
55
+ masked = torch.full_like(logits, float('-inf'))
56
+ masked.scatter_(-1, top_k_idx, top_k_vals)
57
+ return logits * (1 - activate) + masked * activate
58
+
59
+
60
+ def attach_neg(base_model, neg_path_or_repo, hf_token=None):
61
+ """Attach NEG to a loaded base model.
62
+
63
+ Args:
64
+ base_model: a HuggingFace AutoModelForCausalLM instance
65
+ neg_path_or_repo: local path or HF repo containing neg_modules.safetensors
66
+ hf_token: optional HF token (for private repos)
67
+
68
+ Returns:
69
+ The same model with NEG-Head and NEG-Gate attached and forward() wrapped
70
+ to apply NEG at each generation step.
71
+ """
72
+ # Find neg_modules.safetensors
73
+ neg_file = None
74
+ if os.path.isdir(neg_path_or_repo):
75
+ candidate = os.path.join(neg_path_or_repo, "neg_modules.safetensors")
76
+ if os.path.exists(candidate):
77
+ neg_file = candidate
78
+ if neg_file is None:
79
+ try:
80
+ from huggingface_hub import hf_hub_download
81
+ neg_file = hf_hub_download(
82
+ repo_id=neg_path_or_repo,
83
+ filename="neg_modules.safetensors",
84
+ token=hf_token or os.environ.get("HF_TOKEN"),
85
+ )
86
+ except Exception as e:
87
+ raise FileNotFoundError(
88
+ f"Cannot locate neg_modules.safetensors at {neg_path_or_repo}: {e}"
89
+ )
90
+
91
+ # Determine hidden size and device
92
+ hidden_size = getattr(base_model.config, "hidden_size", None)
93
+ if hidden_size is None:
94
+ hidden_size = getattr(getattr(base_model.config, "text_config", None), "hidden_size", None)
95
+ if hidden_size is None:
96
+ raise ValueError("Could not determine hidden_size from model config.")
97
+
98
+ device = next(base_model.parameters()).device
99
+
100
+ # Load state dict
101
+ state = load_file(neg_file)
102
+ head_sd = {k.replace("head.", "", 1): v for k, v in state.items() if k.startswith("head.")}
103
+ gate_sd = {k.replace("gate.", "", 1): v for k, v in state.items() if k.startswith("gate.")}
104
+
105
+ # Build and load NEG modules
106
+ head = NEGHead(hidden_size).to(device=device, dtype=torch.float32)
107
+ if head_sd:
108
+ head.load_state_dict(head_sd)
109
+ head.eval()
110
+
111
+ # Infer gate params from state
112
+ gate_threshold = gate_sd.get("threshold", torch.tensor(1.175)).item()
113
+ # top_k is not a learnable param; read from metadata if present, else default 20
114
+ top_k = state.get("meta.top_k", torch.tensor(20)).item() if "meta.top_k" in state else 20
115
+ gate = NEGGate(init_threshold=gate_threshold, top_k=int(top_k)).to(
116
+ device=device, dtype=torch.float32
117
+ )
118
+ if gate_sd:
119
+ gate.load_state_dict(gate_sd)
120
+ gate.eval()
121
+
122
+ # Attach
123
+ base_model.neg_head = head
124
+ base_model.neg_gate = gate
125
+
126
+ # Wrap forward
127
+ original_forward = base_model.forward
128
+
129
+ def forward_with_neg(*args, **kwargs):
130
+ # Force hidden states capture
131
+ kwargs["output_hidden_states"] = True
132
+ out = original_forward(*args, **kwargs)
133
+ hidden_states = out.hidden_states
134
+ if hidden_states is None:
135
+ return out
136
+ last_hidden = hidden_states[-1][:, -1].float()
137
+ pred_ent = base_model.neg_head(last_hidden)
138
+ logits = out.logits
139
+ last_logits = logits[:, -1].float()
140
+ guided = base_model.neg_gate(last_logits, pred_ent)
141
+ # Clone and replace last position
142
+ new_logits = logits.clone()
143
+ new_logits[:, -1] = guided.to(logits.dtype)
144
+ out.logits = new_logits
145
+ return out
146
+
147
+ base_model.forward = forward_with_neg
148
+ base_model._neg_attached = True
149
+
150
+ print(f"[Darwin-NEG] NEG attached successfully.")
151
+ print(f"[Darwin-NEG] threshold = {gate.threshold.item():.4f}")
152
+ print(f"[Darwin-NEG] top_k = {gate.top_k}")
153
+ print(f"[Darwin-NEG] head params: {sum(p.numel() for p in head.parameters()):,}")
154
+ return base_model
155
+
156
+
157
+ def load_darwin_neg(repo_or_path, torch_dtype=torch.bfloat16, device_map="auto",
158
+ trust_remote_code=True, hf_token=None, **kwargs):
159
+ """Convenience loader: loads base model + attaches NEG in one call.
160
+
161
+ Example:
162
+ from modeling_darwin_neg import load_darwin_neg
163
+ model = load_darwin_neg("FINAL-Bench/Darwin-9B-NEG", hf_token="hf_...")
164
+ """
165
+ from transformers import AutoModelForCausalLM
166
+ token = hf_token or os.environ.get("HF_TOKEN")
167
+ base = AutoModelForCausalLM.from_pretrained(
168
+ repo_or_path,
169
+ torch_dtype=torch_dtype,
170
+ device_map=device_map,
171
+ trust_remote_code=trust_remote_code,
172
+ token=token,
173
+ low_cpu_mem_usage=True,
174
+ **kwargs,
175
+ )
176
+ return attach_neg(base, repo_or_path, hf_token=token)
neg_modules.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fcc1a5a9f7cdeaf2462af9f6de87ecf7626be8a96287e95bb2a20d63cbcb71a
3
+ size 16785908
preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 16777216,
4
+ "shortest_edge": 65536
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "image_processor_type": "Qwen2VLImageProcessorFast"
21
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
3
+ size 12807982
tokenizer_config.json ADDED
@@ -0,0 +1,305 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "248044": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "248045": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "248046": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "248047": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "248048": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "248049": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "248050": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "248051": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "248052": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "248053": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "248054": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "248055": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "248056": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "248057": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "248058": {
117
+ "content": "<tool_call>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "248059": {
125
+ "content": "</tool_call>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "248060": {
133
+ "content": "<|fim_prefix|>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "248061": {
141
+ "content": "<|fim_middle|>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "248062": {
149
+ "content": "<|fim_suffix|>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "248063": {
157
+ "content": "<|fim_pad|>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "248064": {
165
+ "content": "<|repo_name|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "248065": {
173
+ "content": "<|file_sep|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "248066": {
181
+ "content": "<tool_response>",
182
+ "lstrip": false,
183
+ "normalized": false,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "248067": {
189
+ "content": "</tool_response>",
190
+ "lstrip": false,
191
+ "normalized": false,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "248068": {
197
+ "content": "<think>",
198
+ "lstrip": false,
199
+ "normalized": false,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "248069": {
205
+ "content": "</think>",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ },
212
+ "248070": {
213
+ "content": "<|audio_start|>",
214
+ "lstrip": false,
215
+ "normalized": false,
216
+ "rstrip": false,
217
+ "single_word": false,
218
+ "special": true
219
+ },
220
+ "248071": {
221
+ "content": "<|audio_end|>",
222
+ "lstrip": false,
223
+ "normalized": false,
224
+ "rstrip": false,
225
+ "single_word": false,
226
+ "special": true
227
+ },
228
+ "248072": {
229
+ "content": "<tts_pad>",
230
+ "lstrip": false,
231
+ "normalized": false,
232
+ "rstrip": false,
233
+ "single_word": false,
234
+ "special": true
235
+ },
236
+ "248073": {
237
+ "content": "<tts_text_bos>",
238
+ "lstrip": false,
239
+ "normalized": false,
240
+ "rstrip": false,
241
+ "single_word": false,
242
+ "special": true
243
+ },
244
+ "248074": {
245
+ "content": "<tts_text_eod>",
246
+ "lstrip": false,
247
+ "normalized": false,
248
+ "rstrip": false,
249
+ "single_word": false,
250
+ "special": true
251
+ },
252
+ "248075": {
253
+ "content": "<tts_text_bos_single>",
254
+ "lstrip": false,
255
+ "normalized": false,
256
+ "rstrip": false,
257
+ "single_word": false,
258
+ "special": true
259
+ },
260
+ "248076": {
261
+ "content": "<|audio_pad|>",
262
+ "lstrip": false,
263
+ "normalized": false,
264
+ "rstrip": false,
265
+ "single_word": false,
266
+ "special": true
267
+ }
268
+ },
269
+ "additional_special_tokens": [
270
+ "<|im_start|>",
271
+ "<|im_end|>",
272
+ "<|object_ref_start|>",
273
+ "<|object_ref_end|>",
274
+ "<|box_start|>",
275
+ "<|box_end|>",
276
+ "<|quad_start|>",
277
+ "<|quad_end|>",
278
+ "<|vision_start|>",
279
+ "<|vision_end|>",
280
+ "<|vision_pad|>",
281
+ "<|image_pad|>",
282
+ "<|video_pad|>"
283
+ ],
284
+ "bos_token": null,
285
+ "chat_template": "{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- macro render_content(content, do_vision_count, is_system_content=false) %}\n {%- if content is string %}\n {{- content }}\n {%- elif content is iterable and content is not mapping %}\n {%- for item in content %}\n {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain images.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Picture ' ~ image_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|image_pad|><|vision_end|>' }}\n {%- elif 'video' in item or item.type == 'video' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain videos.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Video ' ~ video_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|video_pad|><|vision_end|>' }}\n {%- elif 'text' in item %}\n {{- item.text }}\n {%- else %}\n {{- raise_exception('Unexpected item type in content.') }}\n {%- endif %}\n {%- endfor %}\n {%- elif content is none or content is undefined %}\n {{- '' }}\n {%- else %}\n {{- raise_exception('Unexpected content type.') }}\n {%- endif %}\n{%- endmacro %}\n{%- if not messages %}\n {{- raise_exception('No messages provided.') }}\n{%- endif %}\n{%- if tools and tools is iterable and tools is not mapping %}\n {{- '<|im_start|>system\\n' }}\n {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\" }}\n {{- '\\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>' }}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {%- if content %}\n {{- '\\n\\n' + content }}\n {%- endif %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {{- '<|im_start|>system\\n' + content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" %}\n {%- set content = render_content(message.content, false)|trim %}\n {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if ns.multi_step_tool %}\n {{- raise_exception('No user query found in messages.') }}\n{%- endif %}\n{%- for message in messages %}\n {%- set content = render_content(message.content, true)|trim %}\n {%- if message.role == \"system\" %}\n {%- if not loop.first %}\n {{- raise_exception('System message must be at the beginning.') }}\n {%- endif %}\n {%- elif message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- set reasoning_content = reasoning_content|trim %}\n {%- if loop.index0 > ns.last_query_index %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content + '\\n</think>\\n\\n' + content }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.first %}\n {%- if content|trim %}\n {{- '\\n\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- else %}\n {{- '<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- else %}\n {{- '\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- if tool_call.arguments is defined %}\n {%- for args_name, args_value in tool_call.arguments|items %}\n {{- '<parameter=' + args_name + '>\\n' }}\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 %}\n {{- args_value }}\n {{- '\\n</parameter>\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n {{- '<|im_end|>\\n' }}\n {%- elif loop.last %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- else %}\n {{- raise_exception('Unexpected message role.') }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- else %}\n {{- '<think>\\n' }}\n {%- endif %}\n{%- endif %}",
286
+ "clean_up_tokenization_spaces": false,
287
+ "eos_token": "<|im_end|>",
288
+ "errors": "replace",
289
+ "model_max_length": 262144,
290
+ "pad_token": "<|endoftext|>",
291
+ "split_special_tokens": false,
292
+ "tokenizer_class": "Qwen2Tokenizer",
293
+ "unk_token": null,
294
+ "add_bos_token": false,
295
+ "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+",
296
+ "extra_special_tokens": {
297
+ "audio_bos_token": "<|audio_start|>",
298
+ "audio_eos_token": "<|audio_end|>",
299
+ "audio_token": "<|audio_pad|>",
300
+ "image_token": "<|image_pad|>",
301
+ "video_token": "<|video_pad|>",
302
+ "vision_bos_token": "<|vision_start|>",
303
+ "vision_eos_token": "<|vision_end|>"
304
+ }
305
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff