|
{ |
|
"best_metric": 2.037567615509033, |
|
"best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_default_r32/checkpoint-16", |
|
"epoch": 0.04126536369131573, |
|
"eval_steps": 8, |
|
"global_step": 16, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0025790852307072333, |
|
"grad_norm": 23.78647804260254, |
|
"learning_rate": 1.25e-05, |
|
"loss": 2.2237, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.010316340922828933, |
|
"grad_norm": 10.68380355834961, |
|
"learning_rate": 5e-05, |
|
"loss": 2.0556, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.020632681845657867, |
|
"grad_norm": 13.343311309814453, |
|
"learning_rate": 0.0001, |
|
"loss": 1.9915, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.020632681845657867, |
|
"eval_loss": 2.0385255813598633, |
|
"eval_runtime": 251.2319, |
|
"eval_samples_per_second": 0.979, |
|
"eval_steps_per_second": 0.979, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.0309490227684868, |
|
"grad_norm": 19.941059112548828, |
|
"learning_rate": 9.997251843068762e-05, |
|
"loss": 2.0389, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 0.04126536369131573, |
|
"grad_norm": 13.265397071838379, |
|
"learning_rate": 9.989010393221656e-05, |
|
"loss": 2.054, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 0.04126536369131573, |
|
"eval_loss": 2.037567615509033, |
|
"eval_runtime": 356.5096, |
|
"eval_samples_per_second": 0.69, |
|
"eval_steps_per_second": 0.69, |
|
"step": 16 |
|
} |
|
], |
|
"logging_steps": 4, |
|
"max_steps": 387, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 8, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 1.923891222557491e+16, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|