pilotj's picture
Training in progress, step 10000, checkpoint
6651aef verified
raw
history blame
6.46 kB
{
"best_metric": 0.8291246891021729,
"best_model_checkpoint": "/kaggle/working/results/checkpoint-10000",
"epoch": 0.7792410192472532,
"eval_steps": 1000,
"global_step": 10000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03896205096236266,
"grad_norm": 3.4821102619171143,
"learning_rate": 9.610379490376374e-05,
"loss": 0.5436,
"step": 500
},
{
"epoch": 0.07792410192472532,
"grad_norm": 11.18921947479248,
"learning_rate": 9.220758980752748e-05,
"loss": 0.5704,
"step": 1000
},
{
"epoch": 0.07792410192472532,
"eval_loss": 1.0604108572006226,
"eval_runtime": 50.1596,
"eval_samples_per_second": 103.669,
"eval_steps_per_second": 0.817,
"step": 1000
},
{
"epoch": 0.11688615288708798,
"grad_norm": 4.489162445068359,
"learning_rate": 8.831138471129121e-05,
"loss": 0.568,
"step": 1500
},
{
"epoch": 0.15584820384945064,
"grad_norm": 3.448695659637451,
"learning_rate": 8.441517961505494e-05,
"loss": 0.5614,
"step": 2000
},
{
"epoch": 0.15584820384945064,
"eval_loss": 1.175930142402649,
"eval_runtime": 49.9866,
"eval_samples_per_second": 104.028,
"eval_steps_per_second": 0.82,
"step": 2000
},
{
"epoch": 0.1948102548118133,
"grad_norm": 3.9182376861572266,
"learning_rate": 8.051897451881867e-05,
"loss": 0.5523,
"step": 2500
},
{
"epoch": 0.23377230577417596,
"grad_norm": 7.5695905685424805,
"learning_rate": 7.662276942258241e-05,
"loss": 0.5481,
"step": 3000
},
{
"epoch": 0.23377230577417596,
"eval_loss": 1.063319444656372,
"eval_runtime": 50.1363,
"eval_samples_per_second": 103.717,
"eval_steps_per_second": 0.818,
"step": 3000
},
{
"epoch": 0.2727343567365386,
"grad_norm": 7.505063533782959,
"learning_rate": 7.272656432634615e-05,
"loss": 0.555,
"step": 3500
},
{
"epoch": 0.3116964076989013,
"grad_norm": 5.6970295906066895,
"learning_rate": 6.883035923010988e-05,
"loss": 0.5417,
"step": 4000
},
{
"epoch": 0.3116964076989013,
"eval_loss": 1.0108377933502197,
"eval_runtime": 50.1231,
"eval_samples_per_second": 103.744,
"eval_steps_per_second": 0.818,
"step": 4000
},
{
"epoch": 0.35065845866126394,
"grad_norm": 3.7277610301971436,
"learning_rate": 6.493415413387362e-05,
"loss": 0.5422,
"step": 4500
},
{
"epoch": 0.3896205096236266,
"grad_norm": 4.281050205230713,
"learning_rate": 6.103794903763734e-05,
"loss": 0.5257,
"step": 5000
},
{
"epoch": 0.3896205096236266,
"eval_loss": 0.9513921737670898,
"eval_runtime": 50.1094,
"eval_samples_per_second": 103.773,
"eval_steps_per_second": 0.818,
"step": 5000
},
{
"epoch": 0.42858256058598926,
"grad_norm": 3.6584441661834717,
"learning_rate": 5.7141743941401084e-05,
"loss": 0.5195,
"step": 5500
},
{
"epoch": 0.4675446115483519,
"grad_norm": 5.238151550292969,
"learning_rate": 5.3245538845164814e-05,
"loss": 0.5106,
"step": 6000
},
{
"epoch": 0.4675446115483519,
"eval_loss": 0.9227759838104248,
"eval_runtime": 49.9608,
"eval_samples_per_second": 104.082,
"eval_steps_per_second": 0.821,
"step": 6000
},
{
"epoch": 0.5065066625107145,
"grad_norm": 4.445685863494873,
"learning_rate": 4.934933374892854e-05,
"loss": 0.5022,
"step": 6500
},
{
"epoch": 0.5454687134730772,
"grad_norm": 4.609871864318848,
"learning_rate": 4.545312865269228e-05,
"loss": 0.492,
"step": 7000
},
{
"epoch": 0.5454687134730772,
"eval_loss": 0.9493070244789124,
"eval_runtime": 50.0126,
"eval_samples_per_second": 103.974,
"eval_steps_per_second": 0.82,
"step": 7000
},
{
"epoch": 0.5844307644354398,
"grad_norm": 3.851177930831909,
"learning_rate": 4.155692355645601e-05,
"loss": 0.5068,
"step": 7500
},
{
"epoch": 0.6233928153978026,
"grad_norm": 4.094593524932861,
"learning_rate": 3.7660718460219745e-05,
"loss": 0.5033,
"step": 8000
},
{
"epoch": 0.6233928153978026,
"eval_loss": 0.9059617519378662,
"eval_runtime": 49.9788,
"eval_samples_per_second": 104.044,
"eval_steps_per_second": 0.82,
"step": 8000
},
{
"epoch": 0.6623548663601652,
"grad_norm": 3.0767014026641846,
"learning_rate": 3.376451336398348e-05,
"loss": 0.497,
"step": 8500
},
{
"epoch": 0.7013169173225279,
"grad_norm": 3.861445903778076,
"learning_rate": 2.9868308267747215e-05,
"loss": 0.4688,
"step": 9000
},
{
"epoch": 0.7013169173225279,
"eval_loss": 0.8691270351409912,
"eval_runtime": 50.1753,
"eval_samples_per_second": 103.637,
"eval_steps_per_second": 0.817,
"step": 9000
},
{
"epoch": 0.7402789682848905,
"grad_norm": 3.7857894897460938,
"learning_rate": 2.5972103171510948e-05,
"loss": 0.4853,
"step": 9500
},
{
"epoch": 0.7792410192472532,
"grad_norm": 6.56783390045166,
"learning_rate": 2.2075898075274684e-05,
"loss": 0.483,
"step": 10000
},
{
"epoch": 0.7792410192472532,
"eval_loss": 0.8291246891021729,
"eval_runtime": 49.9917,
"eval_samples_per_second": 104.017,
"eval_steps_per_second": 0.82,
"step": 10000
}
],
"logging_steps": 500,
"max_steps": 12833,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 2000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 4.240771055616e+16,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}