See axolotl config

axolotl version: 0.4.1

yaml

1adapter: lora 2auto_resume_from_checkpoints: false 3base_model: EleutherAI/pythia-160m 4bf16: false 5chat_template: llama3 6dataset_prepared_path: null 7datasets: 8- data_files: 9 - 5cf4bc9f2d02ce29_train_data.json 10 ds_type: json 11 format: custom 12 path: /workspace/input_data/5cf4bc9f2d02ce29_train_data.json 13 type: 14 field_instruction: Description 15 field_output: Product Name 16 format: '{instruction}' 17 no_input_format: '{instruction}' 18 system_format: '{system}' 19 system_prompt: '' 20debug: null 21deepspeed: null 22early_stopping_patience: 4 23eval_max_new_tokens: 128 24eval_steps: 100 25eval_table_size: null 26flash_attention: false 27fp16: true 28fsdp: null 29fsdp_config: null 30gradient_accumulation_steps: 4 31gradient_checkpointing: true 32group_by_length: false 33hub_model_id: error577/d71d5b54-5e85-4947-9c78-097fdffe44b5 34hub_repo: null 35hub_strategy: checkpoint 36hub_token: null 37learning_rate: 0.0002 38load_in_4bit: false 39load_in_8bit: false 40local_rank: null 41logging_steps: 1 42lora_alpha: 128 43lora_dropout: 0.05 44lora_fan_in_fan_out: null 45lora_model_dir: null 46lora_r: 64 47lora_target_linear: true 48lr_scheduler: cosine 49max_grad_norm: 1.0 50max_steps: null 51micro_batch_size: 4 52mlflow_experiment_name: /tmp/5cf4bc9f2d02ce29_train_data.json 53model_type: AutoModelForCausalLM 54num_epochs: 10 55optimizer: adamw_torch 56output_dir: miner_id_24 57pad_to_sequence_len: true 58resume_from_checkpoint: null 59s2_attention: null 60sample_packing: false 61save_steps: 100 62sequence_len: 512 63special_tokens: 64 pad_token: <|endoftext|> 65strict: false 66tf32: false 67tokenizer_type: AutoTokenizer 68train_on_inputs: false 69trust_remote_code: true 70val_set_size: 0.005 71wandb_entity: null 72wandb_mode: online 73wandb_name: 451a522d-4518-49f1-886b-2a8292a7075c 74wandb_project: Gradients-On-Demand 75wandb_run: your_name 76wandb_runid: 451a522d-4518-49f1-886b-2a8292a7075c 77warmup_steps: 30 78weight_decay: 0.0 79xformers_attention: null 80

d71d5b54-5e85-4947-9c78-097fdffe44b5

• Loss: 2.2954

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: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 30
• num_epochs: 10
• mixed_precision_training: Native AMP

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