|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.716981132075472, |
|
"eval_steps": 500, |
|
"global_step": 1500, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6083870967741936, |
|
"eval_loss": 0.20758771896362305, |
|
"eval_runtime": 2.1187, |
|
"eval_samples_per_second": 1463.163, |
|
"eval_steps_per_second": 30.679, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.5437103509902954, |
|
"learning_rate": 1.606918238993711e-05, |
|
"loss": 0.3332, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8251612903225807, |
|
"eval_loss": 0.10081501305103302, |
|
"eval_runtime": 2.4164, |
|
"eval_samples_per_second": 1282.908, |
|
"eval_steps_per_second": 26.9, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8741935483870967, |
|
"eval_loss": 0.06718587875366211, |
|
"eval_runtime": 2.2023, |
|
"eval_samples_per_second": 1407.623, |
|
"eval_steps_per_second": 29.515, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"grad_norm": 0.4230486750602722, |
|
"learning_rate": 1.2138364779874214e-05, |
|
"loss": 0.1164, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8990322580645161, |
|
"eval_loss": 0.05146277695894241, |
|
"eval_runtime": 2.2444, |
|
"eval_samples_per_second": 1381.214, |
|
"eval_steps_per_second": 28.961, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"grad_norm": 0.3436492085456848, |
|
"learning_rate": 8.207547169811321e-06, |
|
"loss": 0.0743, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2544, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 8, |
|
"save_steps": 500, |
|
"total_flos": 391368939443328.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.5470523291760685, |
|
"num_train_epochs": 8, |
|
"temperature": 8 |
|
} |
|
} |
|
|