103 lines
2.6 KiB
JSON
103 lines
2.6 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 7.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 28,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 1.3,
|
||
|
|
"grad_norm": 5.2073807743444815,
|
||
|
|
"learning_rate": 3.9842294026289565e-05,
|
||
|
|
"loss": 0.4525,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.13143858313560486,
|
||
|
|
"step": 5,
|
||
|
|
"valid_targets_mean": 8372.7,
|
||
|
|
"valid_targets_min": 4715
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.6,
|
||
|
|
"grad_norm": 0.5330107824685453,
|
||
|
|
"learning_rate": 3.457937254842823e-05,
|
||
|
|
"loss": 0.3407,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.10134033858776093,
|
||
|
|
"step": 10,
|
||
|
|
"valid_targets_mean": 8739.9,
|
||
|
|
"valid_targets_min": 1991
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.9,
|
||
|
|
"grad_norm": 0.3470100838967144,
|
||
|
|
"learning_rate": 2.37476262917145e-05,
|
||
|
|
"loss": 0.2899,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.1029006689786911,
|
||
|
|
"step": 15,
|
||
|
|
"valid_targets_mean": 8878.3,
|
||
|
|
"valid_targets_min": 4021
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"grad_norm": 0.35052719257337833,
|
||
|
|
"learning_rate": 1.1484414168698547e-05,
|
||
|
|
"loss": 0.2594,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.2442539930343628,
|
||
|
|
"step": 20,
|
||
|
|
"valid_targets_mean": 8485.8,
|
||
|
|
"valid_targets_min": 2261
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 6.3,
|
||
|
|
"grad_norm": 0.22866100837407657,
|
||
|
|
"learning_rate": 2.473866399122733e-06,
|
||
|
|
"loss": 0.2443,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.0720648244023323,
|
||
|
|
"step": 25,
|
||
|
|
"valid_targets_mean": 7971.5,
|
||
|
|
"valid_targets_min": 1792
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 7.0,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.25454777479171753,
|
||
|
|
"step": 28,
|
||
|
|
"total_flos": 1.5644485162421453e+17,
|
||
|
|
"train_loss": 0.31890720554760527,
|
||
|
|
"train_runtime": 744.463,
|
||
|
|
"train_samples_per_second": 2.971,
|
||
|
|
"train_steps_per_second": 0.038,
|
||
|
|
"valid_targets_mean": 8619.0,
|
||
|
|
"valid_targets_min": 1488
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 28,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 7,
|
||
|
|
"save_steps": 300,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.5644485162421453e+17,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|