See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: tiiuae/falcon-7b
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 3fb514dd17b7df17_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/3fb514dd17b7df17_train_data.json
  type:
    field_instruction: source_text
    field_output: target_text
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/59be0874-2a70-49c3-b1eb-47b501f3a513
hub_repo: null
hub_strategy: null
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 1092
micro_batch_size: 4
mlflow_experiment_name: /tmp/3fb514dd17b7df17_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 2048
special_tokens:
  pad_token: <|endoftext|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.022291076881924166
wandb_entity: null
wandb_mode: online
wandb_name: 44ff7564-3793-4b4d-a62d-9695722eced0
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 44ff7564-3793-4b4d-a62d-9695722eced0
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

59be0874-2a70-49c3-b1eb-47b501f3a513

This model is a fine-tuned version of tiiuae/falcon-7b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0102

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• training_steps: 1092

Training results

Training Loss	Epoch	Step	Validation Loss
5.1817	0.0001	1	0.6095
0.2583	0.0146	100	0.0365
0.2472	0.0292	200	0.0267
0.0897	0.0438	300	0.0204
0.0997	0.0584	400	0.0183
0.1329	0.0730	500	0.0154
0.1117	0.0875	600	0.0139
0.1031	0.1021	700	0.0127
0.1088	0.1167	800	0.0112
0.0854	0.1313	900	0.0106
0.0596	0.1459	1000	0.0102

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1