115 lines
3.1 KiB
JSON
115 lines
3.1 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0,
|
|
"eval_steps": 500,
|
|
"global_step": 81,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 0.7260835453867912,
|
|
"epoch": 0.37735849056603776,
|
|
"grad_norm": 0.9797868132591248,
|
|
"learning_rate": 0.00018,
|
|
"loss": 1.0297086715698243,
|
|
"mean_token_accuracy": 0.8047336474061012,
|
|
"num_tokens": 41605.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 0.6147415287792682,
|
|
"epoch": 0.7547169811320755,
|
|
"grad_norm": 0.8319393992424011,
|
|
"learning_rate": 0.00019217488001088784,
|
|
"loss": 0.610823392868042,
|
|
"mean_token_accuracy": 0.8518229335546493,
|
|
"num_tokens": 86144.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 0.5127407720214442,
|
|
"epoch": 1.1132075471698113,
|
|
"grad_norm": 0.898364782333374,
|
|
"learning_rate": 0.0001666935530836651,
|
|
"loss": 0.47104392051696775,
|
|
"mean_token_accuracy": 0.8734802854688544,
|
|
"num_tokens": 125093.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 0.35390120558440685,
|
|
"epoch": 1.490566037735849,
|
|
"grad_norm": 0.6483786702156067,
|
|
"learning_rate": 0.0001283661778334297,
|
|
"loss": 0.34734306335449217,
|
|
"mean_token_accuracy": 0.9023109719157218,
|
|
"num_tokens": 167344.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 0.2861595153808594,
|
|
"epoch": 1.8679245283018868,
|
|
"grad_norm": 0.6667284965515137,
|
|
"learning_rate": 8.457510670346976e-05,
|
|
"loss": 0.27520730495452883,
|
|
"mean_token_accuracy": 0.9225298032164574,
|
|
"num_tokens": 209654.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 0.21439368748351148,
|
|
"epoch": 2.2264150943396226,
|
|
"grad_norm": 0.9153143763542175,
|
|
"learning_rate": 4.375507123592194e-05,
|
|
"loss": 0.19489681720733643,
|
|
"mean_token_accuracy": 0.9477020047212902,
|
|
"num_tokens": 250416.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 0.1475609978660941,
|
|
"epoch": 2.6037735849056602,
|
|
"grad_norm": 0.5222585201263428,
|
|
"learning_rate": 1.3768542747997215e-05,
|
|
"loss": 0.13845933675765992,
|
|
"mean_token_accuracy": 0.9617140784859657,
|
|
"num_tokens": 293209.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 0.14689538963139057,
|
|
"epoch": 2.981132075471698,
|
|
"grad_norm": 0.6572704315185547,
|
|
"learning_rate": 3.913177925055189e-07,
|
|
"loss": 0.13648871183395386,
|
|
"mean_token_accuracy": 0.9615253210067749,
|
|
"num_tokens": 335896.0,
|
|
"step": 80
|
|
}
|
|
],
|
|
"logging_steps": 10,
|
|
"max_steps": 81,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 3,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.0410264608835584e+16,
|
|
"train_batch_size": 2,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|