timewanderer's picture
Training in progress, step 2000
42ba436 verified
raw
history blame
4.06 kB
{
"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": 0.6129032258064516,
"eval_loss": 0.22857815027236938,
"eval_runtime": 5.1291,
"eval_samples_per_second": 604.398,
"eval_steps_per_second": 12.673,
"step": 318
},
{
"epoch": 1.5723270440251573,
"grad_norm": 0.5715988874435425,
"learning_rate": 1.685534591194969e-05,
"loss": 0.368,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.8406451612903226,
"eval_loss": 0.10308429598808289,
"eval_runtime": 5.3537,
"eval_samples_per_second": 579.034,
"eval_steps_per_second": 12.141,
"step": 636
},
{
"epoch": 3.0,
"eval_accuracy": 0.8909677419354839,
"eval_loss": 0.06482071429491043,
"eval_runtime": 5.1876,
"eval_samples_per_second": 597.576,
"eval_steps_per_second": 12.53,
"step": 954
},
{
"epoch": 3.1446540880503147,
"grad_norm": 0.4861523509025574,
"learning_rate": 1.371069182389937e-05,
"loss": 0.1205,
"step": 1000
},
{
"epoch": 4.0,
"eval_accuracy": 0.9032258064516129,
"eval_loss": 0.04816382750868797,
"eval_runtime": 5.1502,
"eval_samples_per_second": 601.919,
"eval_steps_per_second": 12.621,
"step": 1272
},
{
"epoch": 4.716981132075472,
"grad_norm": 0.3236384689807892,
"learning_rate": 1.0566037735849058e-05,
"loss": 0.0727,
"step": 1500
},
{
"epoch": 5.0,
"eval_accuracy": 0.9145161290322581,
"eval_loss": 0.0385914221405983,
"eval_runtime": 5.275,
"eval_samples_per_second": 587.677,
"eval_steps_per_second": 12.322,
"step": 1590
},
{
"epoch": 6.0,
"eval_accuracy": 0.922258064516129,
"eval_loss": 0.03349597379565239,
"eval_runtime": 5.1757,
"eval_samples_per_second": 598.956,
"eval_steps_per_second": 12.559,
"step": 1908
},
{
"epoch": 6.289308176100629,
"grad_norm": 0.27145814895629883,
"learning_rate": 7.421383647798742e-06,
"loss": 0.0555,
"step": 2000
},
{
"epoch": 7.0,
"eval_accuracy": 0.927741935483871,
"eval_loss": 0.030705930665135384,
"eval_runtime": 5.2177,
"eval_samples_per_second": 594.135,
"eval_steps_per_second": 12.458,
"step": 2226
},
{
"epoch": 7.861635220125786,
"grad_norm": 0.2963285446166992,
"learning_rate": 4.276729559748428e-06,
"loss": 0.0477,
"step": 2500
},
{
"epoch": 8.0,
"eval_accuracy": 0.93,
"eval_loss": 0.028481999412178993,
"eval_runtime": 5.2897,
"eval_samples_per_second": 586.041,
"eval_steps_per_second": 12.288,
"step": 2544
},
{
"epoch": 9.0,
"eval_accuracy": 0.9325806451612904,
"eval_loss": 0.027418918907642365,
"eval_runtime": 5.1874,
"eval_samples_per_second": 597.607,
"eval_steps_per_second": 12.53,
"step": 2862
},
{
"epoch": 9.433962264150944,
"grad_norm": 0.2736051380634308,
"learning_rate": 1.1320754716981133e-06,
"loss": 0.0442,
"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": 0.34170044466363136,
"num_train_epochs": 10,
"temperature": 5
}
}