|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 9.433962264150944, |
|
"eval_steps": 500, |
|
"global_step": 3000, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5796774193548387, |
|
"eval_loss": 0.1907225400209427, |
|
"eval_runtime": 5.6367, |
|
"eval_samples_per_second": 549.965, |
|
"eval_steps_per_second": 11.532, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.5723270440251573, |
|
"grad_norm": 0.5174282193183899, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 0.3083, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.817741935483871, |
|
"eval_loss": 0.09267137199640274, |
|
"eval_runtime": 5.8965, |
|
"eval_samples_per_second": 525.732, |
|
"eval_steps_per_second": 11.023, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8780645161290322, |
|
"eval_loss": 0.06213448941707611, |
|
"eval_runtime": 6.1922, |
|
"eval_samples_per_second": 500.633, |
|
"eval_steps_per_second": 10.497, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.1446540880503147, |
|
"grad_norm": 0.42976459860801697, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.1074, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9006451612903226, |
|
"eval_loss": 0.047995250672101974, |
|
"eval_runtime": 5.9722, |
|
"eval_samples_per_second": 519.074, |
|
"eval_steps_per_second": 10.884, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.716981132075472, |
|
"grad_norm": 0.29368260502815247, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.0685, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9093548387096774, |
|
"eval_loss": 0.03920818120241165, |
|
"eval_runtime": 5.8584, |
|
"eval_samples_per_second": 529.158, |
|
"eval_steps_per_second": 11.095, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9125806451612903, |
|
"eval_loss": 0.034214429557323456, |
|
"eval_runtime": 5.6445, |
|
"eval_samples_per_second": 549.203, |
|
"eval_steps_per_second": 11.516, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 6.289308176100629, |
|
"grad_norm": 0.25137194991111755, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.0535, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9206451612903226, |
|
"eval_loss": 0.031481869518756866, |
|
"eval_runtime": 5.8755, |
|
"eval_samples_per_second": 527.611, |
|
"eval_steps_per_second": 11.063, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.861635220125786, |
|
"grad_norm": 0.266741544008255, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0463, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9264516129032258, |
|
"eval_loss": 0.029149888083338737, |
|
"eval_runtime": 5.728, |
|
"eval_samples_per_second": 541.205, |
|
"eval_steps_per_second": 11.348, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9283870967741935, |
|
"eval_loss": 0.02811308391392231, |
|
"eval_runtime": 5.6298, |
|
"eval_samples_per_second": 550.644, |
|
"eval_steps_per_second": 11.546, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 9.433962264150944, |
|
"grad_norm": 0.2486511617898941, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0429, |
|
"step": 3000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 780738843279612.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.4529889909540463, |
|
"num_train_epochs": 10, |
|
"temperature": 17 |
|
} |
|
} |
|
|