See axolotl config

axolotl version: 0.4.1

yaml

1adapter: qlora 2auto_resume_from_checkpoints: true 3base_model: fxmarty/really-tiny-falcon-testing 4bf16: auto 5chat_template: llama3 6dataset_prepared_path: null 7dataset_processes: 6 8datasets: 9- data_files: 10 - e87224c8eb065bc2_train_data.json 11 ds_type: json 12 format: custom 13 path: /workspace/input_data/e87224c8eb065bc2_train_data.json 14 type: 15 field_input: intent 16 field_instruction: instruction 17 field_output: response 18 format: '{instruction} {input}' 19 no_input_format: '{instruction}' 20 system_format: '{system}' 21 system_prompt: '' 22debug: null 23deepspeed: null 24early_stopping_patience: 3 25eval_max_new_tokens: 128 26eval_steps: 2000 27eval_table_size: null 28evals_per_epoch: null 29flash_attention: false 30fp16: false 31fsdp: null 32fsdp_config: null 33gradient_accumulation_steps: 1 34gradient_checkpointing: false 35group_by_length: false 36hub_model_id: error577/f4e4984a-1046-4a12-80b7-8c5f44f7a124 37hub_repo: null 38hub_strategy: checkpoint 39hub_token: null 40learning_rate: 0.0002 41load_in_4bit: true 42load_in_8bit: false 43local_rank: null 44logging_steps: 1 45lora_alpha: 16 46lora_dropout: 0.1 47lora_fan_in_fan_out: null 48lora_model_dir: null 49lora_r: 8 50lora_target_linear: true 51lr_scheduler: cosine 52max_grad_norm: 1.0 53max_steps: null 54micro_batch_size: 8 55mlflow_experiment_name: /tmp/e87224c8eb065bc2_train_data.json 56model_type: AutoModelForCausalLM 57num_epochs: 3 58optimizer: adamw_bnb_8bit 59output_dir: miner_id_24 60pad_to_sequence_len: true 61resume_from_checkpoint: null 62s2_attention: null 63sample_packing: false 64save_steps: 2000 65sequence_len: 512 66strict: false 67tf32: false 68tokenizer_type: AutoTokenizer 69train_on_inputs: false 70trust_remote_code: true 71val_set_size: 0.005 72wandb_entity: null 73wandb_mode: online 74wandb_name: 1cd1f24c-cd92-4120-99af-dc7e5e1a2626 75wandb_project: Gradients-On-Demand 76wandb_run: your_name 77wandb_runid: 1cd1f24c-cd92-4120-99af-dc7e5e1a2626 78warmup_steps: 30 79weight_decay: 0.0 80xformers_attention: null 81

f4e4984a-1046-4a12-80b7-8c5f44f7a124

• Loss: 10.9646

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• 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: 30
• num_epochs: 3

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