training_params.json

{"model": "TinyPixel/Llama-2-7B-bf16-sharded", "data_path": "wangzhang/sdb", "project_name": "llama_sdb_model_0", "train_split": "train", "valid_split": null, "text_column": "text", "lr": 0.0002, "epochs": 3, "batch_size": 2, "warmup_ratio": 0.1, "gradient_accumulation": 1, "optimizer": "adamw_torch", "scheduler": "linear", "weight_decay": 0.0, "max_grad_norm": 1.0, "seed": 42, "add_eos_token": false, "block_size": -1, "use_peft": true, "lora_r": 16, "lora_alpha": 32, "lora_dropout": 0.05, "logging_steps": -1, "evaluation_strategy": "epoch", "save_total_limit": 1, "save_strategy": "epoch", "auto_find_batch_size": false, "fp16": false, "push_to_hub": true, "use_int8": false, "model_max_length": 2048, "repo_id": "wangzhang/llama2-sdb", "use_int4": true, "trainer": "sft", "target_modules": null, "merge_adapter": false, "username": null, "use_flash_attention_2": false}