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": 1.7723829743688115,
|
||
|
|
"learning_rate": 3.9842294026289565e-05,
|
||
|
|
"loss": 0.3465,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.13400381803512573,
|
||
|
|
"step": 5,
|
||
|
|
"valid_targets_mean": 7392.5,
|
||
|
|
"valid_targets_min": 1739
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.6,
|
||
|
|
"grad_norm": 0.420993196339436,
|
||
|
|
"learning_rate": 3.457937254842823e-05,
|
||
|
|
"loss": 0.2816,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.09357594698667526,
|
||
|
|
"step": 10,
|
||
|
|
"valid_targets_mean": 7221.1,
|
||
|
|
"valid_targets_min": 1525
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.9,
|
||
|
|
"grad_norm": 0.2674025472390944,
|
||
|
|
"learning_rate": 2.37476262917145e-05,
|
||
|
|
"loss": 0.2531,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.07050734013319016,
|
||
|
|
"step": 15,
|
||
|
|
"valid_targets_mean": 5955.5,
|
||
|
|
"valid_targets_min": 1382
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"grad_norm": 0.31697038924690435,
|
||
|
|
"learning_rate": 1.1484414168698547e-05,
|
||
|
|
"loss": 0.2287,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.23471158742904663,
|
||
|
|
"step": 20,
|
||
|
|
"valid_targets_mean": 5161.2,
|
||
|
|
"valid_targets_min": 2057
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 6.3,
|
||
|
|
"grad_norm": 0.22086069992222349,
|
||
|
|
"learning_rate": 2.473866399122733e-06,
|
||
|
|
"loss": 0.2215,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.0745329037308693,
|
||
|
|
"step": 25,
|
||
|
|
"valid_targets_mean": 5522.3,
|
||
|
|
"valid_targets_min": 1282
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 7.0,
|
||
|
|
"loss_nan_ranks": 0,
|
||
|
|
"loss_rank_avg": 0.20894506573677063,
|
||
|
|
"step": 28,
|
||
|
|
"total_flos": 5.930599290241024e+16,
|
||
|
|
"train_loss": 0.2524434221642358,
|
||
|
|
"train_runtime": 726.3211,
|
||
|
|
"train_samples_per_second": 3.045,
|
||
|
|
"train_steps_per_second": 0.039,
|
||
|
|
"valid_targets_mean": 5717.0,
|
||
|
|
"valid_targets_min": 1502
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"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": 5.930599290241024e+16,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|