See axolotl config

axolotl version: 0.4.1

yaml

1adapter: lora 2base_model: JackFram/llama-68m 3bf16: true 4chat_template: llama3 5dataset_prepared_path: null 6datasets: 7- data_files: 8 - 3b022bbf876bb022_train_data.json 9 ds_type: json 10 format: custom 11 path: /workspace/input_data/3b022bbf876bb022_train_data.json 12 type: 13 field_instruction: instruction 14 field_output: output 15 format: '{instruction}' 16 no_input_format: '{instruction}' 17 system_format: '{system}' 18 system_prompt: '' 19debug: null 20deepspeed: null 21early_stopping_patience: 2 22eval_max_new_tokens: 128 23eval_steps: 100 24eval_table_size: null 25flash_attention: true 26fp16: null 27fsdp: null 28fsdp_config: null 29gradient_accumulation_steps: 8 30gradient_checkpointing: true 31group_by_length: false 32hub_model_id: Alphatao/48b93c3f-b584-4eea-af10-bdcd3b475793 33hub_repo: null 34hub_strategy: checkpoint 35hub_token: null 36learning_rate: 0.0002 37load_best_model_at_end: true 38load_in_4bit: false 39load_in_8bit: false 40local_rank: null 41logging_steps: 1 42lora_alpha: 32 43lora_dropout: 0.05 44lora_fan_in_fan_out: null 45lora_model_dir: null 46lora_r: 16 47lora_target_linear: true 48lora_target_modules: 49- q_proj 50- k_proj 51- v_proj 52lr_scheduler: cosine 53max_grad_norm: 1.0 54max_steps: 2520 55micro_batch_size: 4 56mlflow_experiment_name: /tmp/3b022bbf876bb022_train_data.json 57model_type: AutoModelForCausalLM 58num_epochs: 2 59optimizer: adamw_bnb_8bit 60output_dir: miner_id_24 61pad_to_sequence_len: true 62resume_from_checkpoint: null 63s2_attention: null 64sample_packing: false 65save_steps: 100 66sequence_len: 2048 67special_tokens: 68 pad_token: </s> 69strict: false 70tf32: true 71tokenizer_type: AutoTokenizer 72train_on_inputs: false 73trust_remote_code: true 74val_set_size: 0.04 75wandb_entity: null 76wandb_mode: online 77wandb_name: 9a2de9cd-181f-4b68-b9fa-bbcc319584cc 78wandb_project: Gradients-On-Demand 79wandb_run: your_name 80wandb_runid: 9a2de9cd-181f-4b68-b9fa-bbcc319584cc 81warmup_steps: 10 82weight_decay: 0.0 83xformers_attention: null 84

48b93c3f-b584-4eea-af10-bdcd3b475793

• Loss: 2.4406

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• 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: 2520

Training results

Framework versions

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