lesso04's picture
Training in progress, step 20, checkpoint
8daf1ff verified
{
"best_metric": NaN,
"best_model_checkpoint": "miner_id_24/checkpoint-10",
"epoch": 0.005279482610704151,
"eval_steps": 5,
"global_step": 20,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0002639741305352075,
"grad_norm": NaN,
"learning_rate": 2e-05,
"loss": 0.0,
"step": 1
},
{
"epoch": 0.0002639741305352075,
"eval_loss": NaN,
"eval_runtime": 219.2258,
"eval_samples_per_second": 7.276,
"eval_steps_per_second": 3.64,
"step": 1
},
{
"epoch": 0.000527948261070415,
"grad_norm": NaN,
"learning_rate": 4e-05,
"loss": 0.0,
"step": 2
},
{
"epoch": 0.0007919223916056226,
"grad_norm": NaN,
"learning_rate": 6e-05,
"loss": 0.0,
"step": 3
},
{
"epoch": 0.00105589652214083,
"grad_norm": NaN,
"learning_rate": 8e-05,
"loss": 0.0,
"step": 4
},
{
"epoch": 0.0013198706526760377,
"grad_norm": NaN,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5
},
{
"epoch": 0.0013198706526760377,
"eval_loss": NaN,
"eval_runtime": 216.4913,
"eval_samples_per_second": 7.368,
"eval_steps_per_second": 3.686,
"step": 5
},
{
"epoch": 0.0015838447832112453,
"grad_norm": NaN,
"learning_rate": 0.00012,
"loss": 0.0,
"step": 6
},
{
"epoch": 0.0018478189137464528,
"grad_norm": NaN,
"learning_rate": 0.00014,
"loss": 0.0,
"step": 7
},
{
"epoch": 0.00211179304428166,
"grad_norm": NaN,
"learning_rate": 0.00016,
"loss": 0.0,
"step": 8
},
{
"epoch": 0.002375767174816868,
"grad_norm": NaN,
"learning_rate": 0.00018,
"loss": 0.0,
"step": 9
},
{
"epoch": 0.0026397413053520754,
"grad_norm": NaN,
"learning_rate": 0.0002,
"loss": 0.0,
"step": 10
},
{
"epoch": 0.0026397413053520754,
"eval_loss": NaN,
"eval_runtime": 218.0065,
"eval_samples_per_second": 7.316,
"eval_steps_per_second": 3.66,
"step": 10
},
{
"epoch": 0.002903715435887283,
"grad_norm": NaN,
"learning_rate": 0.00019781476007338058,
"loss": 0.0,
"step": 11
},
{
"epoch": 0.0031676895664224905,
"grad_norm": NaN,
"learning_rate": 0.0001913545457642601,
"loss": 0.0,
"step": 12
},
{
"epoch": 0.0034316636969576983,
"grad_norm": NaN,
"learning_rate": 0.00018090169943749476,
"loss": 0.0,
"step": 13
},
{
"epoch": 0.0036956378274929057,
"grad_norm": NaN,
"learning_rate": 0.00016691306063588583,
"loss": 0.0,
"step": 14
},
{
"epoch": 0.0039596119580281135,
"grad_norm": NaN,
"learning_rate": 0.00015000000000000001,
"loss": 0.0,
"step": 15
},
{
"epoch": 0.0039596119580281135,
"eval_loss": NaN,
"eval_runtime": 219.8092,
"eval_samples_per_second": 7.256,
"eval_steps_per_second": 3.63,
"step": 15
},
{
"epoch": 0.00422358608856332,
"grad_norm": NaN,
"learning_rate": 0.00013090169943749476,
"loss": 0.0,
"step": 16
},
{
"epoch": 0.004487560219098528,
"grad_norm": NaN,
"learning_rate": 0.00011045284632676536,
"loss": 0.0,
"step": 17
},
{
"epoch": 0.004751534349633736,
"grad_norm": NaN,
"learning_rate": 8.954715367323468e-05,
"loss": 0.0,
"step": 18
},
{
"epoch": 0.005015508480168944,
"grad_norm": NaN,
"learning_rate": 6.909830056250527e-05,
"loss": 0.0,
"step": 19
},
{
"epoch": 0.005279482610704151,
"grad_norm": NaN,
"learning_rate": 5.000000000000002e-05,
"loss": 0.0,
"step": 20
},
{
"epoch": 0.005279482610704151,
"eval_loss": NaN,
"eval_runtime": 217.4099,
"eval_samples_per_second": 7.336,
"eval_steps_per_second": 3.67,
"step": 20
}
],
"logging_steps": 1,
"max_steps": 25,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 10,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 2,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 2
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 3505339571896320.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}