VERSIL91's picture
Training in progress, step 20, checkpoint
a2e0a2e verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.078412153883852,
"eval_steps": 5,
"global_step": 20,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0039206076941926,
"grad_norm": 1.108206868171692,
"learning_rate": 1e-05,
"loss": 0.9312,
"step": 1
},
{
"epoch": 0.0039206076941926,
"eval_loss": 1.099448561668396,
"eval_runtime": 64.3877,
"eval_samples_per_second": 6.678,
"eval_steps_per_second": 3.339,
"step": 1
},
{
"epoch": 0.0078412153883852,
"grad_norm": 1.2041150331497192,
"learning_rate": 2e-05,
"loss": 1.0575,
"step": 2
},
{
"epoch": 0.0117618230825778,
"grad_norm": 1.2153946161270142,
"learning_rate": 3e-05,
"loss": 1.0726,
"step": 3
},
{
"epoch": 0.0156824307767704,
"grad_norm": 1.1603235006332397,
"learning_rate": 4e-05,
"loss": 1.0766,
"step": 4
},
{
"epoch": 0.019603038470963,
"grad_norm": 1.2363840341567993,
"learning_rate": 5e-05,
"loss": 1.0818,
"step": 5
},
{
"epoch": 0.019603038470963,
"eval_loss": 0.9847230315208435,
"eval_runtime": 6.9547,
"eval_samples_per_second": 61.829,
"eval_steps_per_second": 30.914,
"step": 5
},
{
"epoch": 0.0235236461651556,
"grad_norm": 1.3062018156051636,
"learning_rate": 6e-05,
"loss": 1.1497,
"step": 6
},
{
"epoch": 0.0274442538593482,
"grad_norm": 0.8951370120048523,
"learning_rate": 7e-05,
"loss": 0.9674,
"step": 7
},
{
"epoch": 0.0313648615535408,
"grad_norm": 0.9190327525138855,
"learning_rate": 8e-05,
"loss": 0.8878,
"step": 8
},
{
"epoch": 0.035285469247733396,
"grad_norm": 0.7922409772872925,
"learning_rate": 9e-05,
"loss": 0.7617,
"step": 9
},
{
"epoch": 0.039206076941926,
"grad_norm": 0.9028505086898804,
"learning_rate": 0.0001,
"loss": 0.8743,
"step": 10
},
{
"epoch": 0.039206076941926,
"eval_loss": 0.7476094365119934,
"eval_runtime": 6.9611,
"eval_samples_per_second": 61.772,
"eval_steps_per_second": 30.886,
"step": 10
},
{
"epoch": 0.0431266846361186,
"grad_norm": 0.7901967167854309,
"learning_rate": 9.755282581475769e-05,
"loss": 0.7681,
"step": 11
},
{
"epoch": 0.0470472923303112,
"grad_norm": 0.7195383310317993,
"learning_rate": 9.045084971874738e-05,
"loss": 0.7462,
"step": 12
},
{
"epoch": 0.0509679000245038,
"grad_norm": 0.7095159292221069,
"learning_rate": 7.938926261462366e-05,
"loss": 0.6836,
"step": 13
},
{
"epoch": 0.0548885077186964,
"grad_norm": 0.5935850143432617,
"learning_rate": 6.545084971874738e-05,
"loss": 0.5753,
"step": 14
},
{
"epoch": 0.058809115412888996,
"grad_norm": 0.5951647758483887,
"learning_rate": 5e-05,
"loss": 0.5539,
"step": 15
},
{
"epoch": 0.058809115412888996,
"eval_loss": 0.5578025579452515,
"eval_runtime": 6.9612,
"eval_samples_per_second": 61.771,
"eval_steps_per_second": 30.886,
"step": 15
},
{
"epoch": 0.0627297231070816,
"grad_norm": 0.633608877658844,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.5572,
"step": 16
},
{
"epoch": 0.0666503308012742,
"grad_norm": 0.6298178434371948,
"learning_rate": 2.061073738537635e-05,
"loss": 0.5276,
"step": 17
},
{
"epoch": 0.07057093849546679,
"grad_norm": 0.6569718718528748,
"learning_rate": 9.549150281252633e-06,
"loss": 0.5219,
"step": 18
},
{
"epoch": 0.07449154618965939,
"grad_norm": 0.6659846305847168,
"learning_rate": 2.4471741852423237e-06,
"loss": 0.5346,
"step": 19
},
{
"epoch": 0.078412153883852,
"grad_norm": 0.695641815662384,
"learning_rate": 0.0,
"loss": 0.4932,
"step": 20
},
{
"epoch": 0.078412153883852,
"eval_loss": 0.5049919486045837,
"eval_runtime": 6.9646,
"eval_samples_per_second": 61.741,
"eval_steps_per_second": 30.871,
"step": 20
}
],
"logging_steps": 1,
"max_steps": 20,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 5,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 5501993254649856.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}