timewanderer's picture
Training in progress, step 500
9e243d2 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"eval_steps": 500,
"global_step": 3180,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.5938709677419355,
"eval_loss": 0.19997140765190125,
"eval_runtime": 5.5473,
"eval_samples_per_second": 558.834,
"eval_steps_per_second": 11.717,
"step": 318
},
{
"epoch": 1.5723270440251573,
"grad_norm": 0.5329213738441467,
"learning_rate": 1.685534591194969e-05,
"loss": 0.3229,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.8238709677419355,
"eval_loss": 0.09559512138366699,
"eval_runtime": 5.7978,
"eval_samples_per_second": 534.683,
"eval_steps_per_second": 11.211,
"step": 636
},
{
"epoch": 3.0,
"eval_accuracy": 0.8812903225806452,
"eval_loss": 0.06311405450105667,
"eval_runtime": 5.4988,
"eval_samples_per_second": 563.759,
"eval_steps_per_second": 11.821,
"step": 954
},
{
"epoch": 3.1446540880503147,
"grad_norm": 0.44504690170288086,
"learning_rate": 1.371069182389937e-05,
"loss": 0.1111,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.9006451612903226,
"eval_loss": 0.048286207020282745,
"eval_runtime": 5.5577,
"eval_samples_per_second": 557.787,
"eval_steps_per_second": 11.696,
"step": 1272
},
{
"epoch": 4.716981132075472,
"grad_norm": 0.3021944761276245,
"learning_rate": 1.0566037735849058e-05,
"loss": 0.0699,
"step": 1500
},
{
"epoch": 5.0,
"eval_accuracy": 0.9103225806451613,
"eval_loss": 0.03919174522161484,
"eval_runtime": 5.6898,
"eval_samples_per_second": 544.834,
"eval_steps_per_second": 11.424,
"step": 1590
},
{
"epoch": 6.0,
"eval_accuracy": 0.9132258064516129,
"eval_loss": 0.03407873958349228,
"eval_runtime": 5.5185,
"eval_samples_per_second": 561.746,
"eval_steps_per_second": 11.779,
"step": 1908
},
{
"epoch": 6.289308176100629,
"grad_norm": 0.2565140724182129,
"learning_rate": 7.421383647798742e-06,
"loss": 0.0542,
"step": 2000
},
{
"epoch": 7.0,
"eval_accuracy": 0.9229032258064516,
"eval_loss": 0.03131110221147537,
"eval_runtime": 5.6225,
"eval_samples_per_second": 551.352,
"eval_steps_per_second": 11.561,
"step": 2226
},
{
"epoch": 7.861635220125786,
"grad_norm": 0.27681249380111694,
"learning_rate": 4.276729559748428e-06,
"loss": 0.0468,
"step": 2500
},
{
"epoch": 8.0,
"eval_accuracy": 0.9270967741935484,
"eval_loss": 0.028976548463106155,
"eval_runtime": 5.7358,
"eval_samples_per_second": 540.464,
"eval_steps_per_second": 11.332,
"step": 2544
},
{
"epoch": 9.0,
"eval_accuracy": 0.9283870967741935,
"eval_loss": 0.027913924306631088,
"eval_runtime": 5.5425,
"eval_samples_per_second": 559.315,
"eval_steps_per_second": 11.728,
"step": 2862
},
{
"epoch": 9.433962264150944,
"grad_norm": 0.2555118501186371,
"learning_rate": 1.1320754716981133e-06,
"loss": 0.0434,
"step": 3000
}
],
"logging_steps": 500,
"max_steps": 3180,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"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": 780738843279612.0,
"train_batch_size": 48,
"trial_name": null,
"trial_params": {
"alpha": 0.2755961419937909,
"num_train_epochs": 10,
"temperature": 10
}
}