|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.7, |
|
"eval_steps": 3, |
|
"global_step": 27, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 1.4759451150894165, |
|
"eval_runtime": 1.7224, |
|
"eval_samples_per_second": 9.87, |
|
"eval_steps_per_second": 1.742, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 1.6763296127319336, |
|
"learning_rate": 3e-05, |
|
"loss": 1.4056, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 1.4381709098815918, |
|
"eval_runtime": 1.2057, |
|
"eval_samples_per_second": 14.1, |
|
"eval_steps_per_second": 2.488, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 1.269885778427124, |
|
"learning_rate": 6e-05, |
|
"loss": 1.3039, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"eval_loss": 1.2148373126983643, |
|
"eval_runtime": 1.2078, |
|
"eval_samples_per_second": 14.075, |
|
"eval_steps_per_second": 2.484, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 0.949137806892395, |
|
"learning_rate": 9e-05, |
|
"loss": 1.1307, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"eval_loss": 1.1424036026000977, |
|
"eval_runtime": 1.2138, |
|
"eval_samples_per_second": 14.006, |
|
"eval_steps_per_second": 2.472, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"grad_norm": 0.9828763008117676, |
|
"learning_rate": 9.755282581475769e-05, |
|
"loss": 1.0364, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"eval_loss": 1.1044687032699585, |
|
"eval_runtime": 1.2122, |
|
"eval_samples_per_second": 14.024, |
|
"eval_steps_per_second": 2.475, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"grad_norm": 0.9127976298332214, |
|
"learning_rate": 8.535533905932738e-05, |
|
"loss": 0.8639, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"eval_loss": 1.0964289903640747, |
|
"eval_runtime": 1.2142, |
|
"eval_samples_per_second": 14.001, |
|
"eval_steps_per_second": 2.471, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"grad_norm": 1.0147470235824585, |
|
"learning_rate": 6.545084971874738e-05, |
|
"loss": 0.8137, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"eval_loss": 1.1079833507537842, |
|
"eval_runtime": 1.2155, |
|
"eval_samples_per_second": 13.986, |
|
"eval_steps_per_second": 2.468, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"grad_norm": 0.8224146962165833, |
|
"learning_rate": 4.2178276747988446e-05, |
|
"loss": 0.7955, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_loss": 1.1106747388839722, |
|
"eval_runtime": 1.2161, |
|
"eval_samples_per_second": 13.979, |
|
"eval_steps_per_second": 2.467, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"grad_norm": 0.9233430624008179, |
|
"learning_rate": 2.061073738537635e-05, |
|
"loss": 0.6105, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"eval_loss": 1.1175633668899536, |
|
"eval_runtime": 1.2153, |
|
"eval_samples_per_second": 13.988, |
|
"eval_steps_per_second": 2.468, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"grad_norm": 0.8103647232055664, |
|
"learning_rate": 5.449673790581611e-06, |
|
"loss": 0.5713, |
|
"step": 27 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"eval_loss": 1.127324104309082, |
|
"eval_runtime": 1.2229, |
|
"eval_samples_per_second": 13.902, |
|
"eval_steps_per_second": 2.453, |
|
"step": 27 |
|
} |
|
], |
|
"logging_steps": 3, |
|
"max_steps": 30, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 3, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3.794665429008384e+16, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|