See axolotl config

axolotl version: 0.4.1

yaml

1adapter: lora 2base_model: Qwen/Qwen2.5-1.5B-Instruct 3bf16: true 4chat_template: llama3 5dataset_prepared_path: null 6datasets: 7- data_files: 8 - 70885662a4657c52_train_data.json 9 ds_type: json 10 format: custom 11 path: /workspace/input_data/70885662a4657c52_train_data.json 12 type: 13 field_input: instruction 14 field_instruction: input 15 field_output: output 16 format: '{instruction} {input}' 17 no_input_format: '{instruction}' 18 system_format: '{system}' 19 system_prompt: '' 20debug: null 21deepspeed: null 22early_stopping_patience: 2 23eval_max_new_tokens: 128 24eval_steps: 100 25eval_table_size: null 26flash_attention: true 27fp16: null 28fsdp: null 29fsdp_config: null 30gradient_accumulation_steps: 8 31gradient_checkpointing: true 32group_by_length: false 33hub_model_id: Alphatao/64bc9474-5926-4130-91f7-48b1e2f417d1 34hub_repo: null 35hub_strategy: checkpoint 36hub_token: null 37learning_rate: 0.0002 38load_best_model_at_end: true 39load_in_4bit: false 40load_in_8bit: false 41local_rank: null 42logging_steps: 1 43lora_alpha: 32 44lora_dropout: 0.05 45lora_fan_in_fan_out: null 46lora_model_dir: null 47lora_r: 16 48lora_target_linear: true 49lora_target_modules: 50- q_proj 51- k_proj 52- v_proj 53lr_scheduler: cosine 54max_grad_norm: 1.0 55max_steps: 3060 56micro_batch_size: 4 57mlflow_experiment_name: /tmp/70885662a4657c52_train_data.json 58model_type: AutoModelForCausalLM 59num_epochs: 2 60optimizer: adamw_bnb_8bit 61output_dir: miner_id_24 62pad_to_sequence_len: true 63resume_from_checkpoint: null 64s2_attention: null 65sample_packing: false 66save_steps: 100 67sequence_len: 1024 68strict: false 69tf32: true 70tokenizer_type: AutoTokenizer 71train_on_inputs: false 72trust_remote_code: true 73val_set_size: 0.05 74wandb_entity: null 75wandb_mode: online 76wandb_name: 9c9c1696-fddf-412a-a23d-9af232ad64b9 77wandb_project: Gradients-On-Demand 78wandb_run: your_name 79wandb_runid: 9c9c1696-fddf-412a-a23d-9af232ad64b9 80warmup_steps: 10 81weight_decay: 0.0 82xformers_attention: null 83

64bc9474-5926-4130-91f7-48b1e2f417d1

• Loss: 1.2696

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: 1548

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