suehyunpark's picture
Upload after 3 epochs
c7ee5f6 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 18,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16666666666666666,
"grad_norm": 280.6432626406138,
"learning_rate": 5e-06,
"loss": 2.9518,
"step": 1
},
{
"epoch": 0.3333333333333333,
"grad_norm": 268.34909690297195,
"learning_rate": 1e-05,
"loss": 2.8263,
"step": 2
},
{
"epoch": 0.5,
"grad_norm": 50.75311462024632,
"learning_rate": 9.903926402016153e-06,
"loss": 1.1917,
"step": 3
},
{
"epoch": 0.6666666666666666,
"grad_norm": 22.75327219021691,
"learning_rate": 9.619397662556434e-06,
"loss": 0.8476,
"step": 4
},
{
"epoch": 0.8333333333333334,
"grad_norm": 6.642460322802831,
"learning_rate": 9.157348061512728e-06,
"loss": 0.541,
"step": 5
},
{
"epoch": 1.0,
"grad_norm": 37.864469102940696,
"learning_rate": 8.535533905932739e-06,
"loss": 0.5717,
"step": 6
},
{
"epoch": 1.0,
"eval_loss": 0.3336825966835022,
"eval_runtime": 1.0612,
"eval_samples_per_second": 9.423,
"eval_steps_per_second": 2.827,
"step": 6
},
{
"epoch": 1.1666666666666667,
"grad_norm": 8.475401484959521,
"learning_rate": 7.777851165098012e-06,
"loss": 0.3759,
"step": 7
},
{
"epoch": 1.3333333333333333,
"grad_norm": 2.4494815853787526,
"learning_rate": 6.913417161825449e-06,
"loss": 0.2862,
"step": 8
},
{
"epoch": 1.5,
"grad_norm": 3.7056738008000853,
"learning_rate": 5.975451610080643e-06,
"loss": 0.2859,
"step": 9
},
{
"epoch": 1.6666666666666665,
"grad_norm": 2.528639375054547,
"learning_rate": 5e-06,
"loss": 0.2525,
"step": 10
},
{
"epoch": 1.8333333333333335,
"grad_norm": 1.681034868123111,
"learning_rate": 4.02454838991936e-06,
"loss": 0.2549,
"step": 11
},
{
"epoch": 2.0,
"grad_norm": 1.9554282109403005,
"learning_rate": 3.0865828381745515e-06,
"loss": 0.2654,
"step": 12
},
{
"epoch": 2.0,
"eval_loss": 0.2001432627439499,
"eval_runtime": 1.1953,
"eval_samples_per_second": 8.366,
"eval_steps_per_second": 2.51,
"step": 12
},
{
"epoch": 2.1666666666666665,
"grad_norm": 1.2686802309348975,
"learning_rate": 2.2221488349019903e-06,
"loss": 0.2087,
"step": 13
},
{
"epoch": 2.3333333333333335,
"grad_norm": 1.199423147704983,
"learning_rate": 1.4644660940672628e-06,
"loss": 0.2293,
"step": 14
},
{
"epoch": 2.5,
"grad_norm": 1.0458147461844773,
"learning_rate": 8.426519384872733e-07,
"loss": 0.2092,
"step": 15
},
{
"epoch": 2.6666666666666665,
"grad_norm": 0.9946553061597486,
"learning_rate": 3.8060233744356634e-07,
"loss": 0.1999,
"step": 16
},
{
"epoch": 2.8333333333333335,
"grad_norm": 0.8890781121405242,
"learning_rate": 9.607359798384785e-08,
"loss": 0.2357,
"step": 17
},
{
"epoch": 3.0,
"grad_norm": 0.9030643285074138,
"learning_rate": 0.0,
"loss": 0.1929,
"step": 18
},
{
"epoch": 3.0,
"eval_loss": 0.19034340977668762,
"eval_runtime": 0.9709,
"eval_samples_per_second": 10.3,
"eval_steps_per_second": 3.09,
"step": 18
},
{
"epoch": 3.0,
"step": 18,
"total_flos": 2848171622400.0,
"train_loss": 0.6625954583287239,
"train_runtime": 666.9986,
"train_samples_per_second": 0.774,
"train_steps_per_second": 0.027
}
],
"logging_steps": 1,
"max_steps": 18,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"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": 2848171622400.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}