92 lines
2.3 KiB
JSON
92 lines
2.3 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 7.0,
|
|
"eval_steps": 500,
|
|
"global_step": 21,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 1.8571428571428572,
|
|
"grad_norm": 1.5442416132189545,
|
|
"learning_rate": 3.9696155060244166e-05,
|
|
"loss": 0.6883,
|
|
"loss_nan_ranks": 0,
|
|
"loss_rank_avg": 0.1956656277179718,
|
|
"step": 5,
|
|
"valid_targets_mean": 14087.2,
|
|
"valid_targets_min": 6992
|
|
},
|
|
{
|
|
"epoch": 3.4285714285714284,
|
|
"grad_norm": 0.4775931165678489,
|
|
"learning_rate": 3.0000000000000004e-05,
|
|
"loss": 0.5087,
|
|
"loss_nan_ranks": 0,
|
|
"loss_rank_avg": 0.15360376238822937,
|
|
"step": 10,
|
|
"valid_targets_mean": 13236.8,
|
|
"valid_targets_min": 6992
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"grad_norm": 0.30864673098704515,
|
|
"learning_rate": 1.3159597133486628e-05,
|
|
"loss": 0.4358,
|
|
"loss_nan_ranks": 0,
|
|
"loss_rank_avg": 0.4128803014755249,
|
|
"step": 15,
|
|
"valid_targets_mean": 14522.8,
|
|
"valid_targets_min": 8389
|
|
},
|
|
{
|
|
"epoch": 6.857142857142857,
|
|
"grad_norm": 0.17849191163065167,
|
|
"learning_rate": 1.2061475842818337e-06,
|
|
"loss": 0.4077,
|
|
"loss_nan_ranks": 0,
|
|
"loss_rank_avg": 0.12648391723632812,
|
|
"step": 20,
|
|
"valid_targets_mean": 13797.7,
|
|
"valid_targets_min": 8482
|
|
},
|
|
{
|
|
"epoch": 7.0,
|
|
"loss_nan_ranks": 0,
|
|
"loss_rank_avg": 0.40468284487724304,
|
|
"step": 21,
|
|
"total_flos": 9.83583584503726e+16,
|
|
"train_loss": 0.5073866077831813,
|
|
"train_runtime": 521.1292,
|
|
"train_samples_per_second": 2.875,
|
|
"train_steps_per_second": 0.04,
|
|
"valid_targets_mean": 13947.0,
|
|
"valid_targets_min": 6404
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 21,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 7,
|
|
"save_steps": 50,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 9.83583584503726e+16,
|
|
"train_batch_size": 1,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|