sedrickkeh's picture
End of training
b5f2874 verified
raw
history blame
3.93 kB
{
"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": 0.9121807009354492,
"learning_rate": 5e-06,
"loss": 1.2823,
"step": 1
},
{
"epoch": 0.3333333333333333,
"grad_norm": 0.9764893050786124,
"learning_rate": 1e-05,
"loss": 1.3191,
"step": 2
},
{
"epoch": 0.5,
"grad_norm": 0.9284384945644188,
"learning_rate": 9.903926402016153e-06,
"loss": 1.287,
"step": 3
},
{
"epoch": 0.6666666666666666,
"grad_norm": 0.8582970971638989,
"learning_rate": 9.619397662556434e-06,
"loss": 1.2686,
"step": 4
},
{
"epoch": 0.8333333333333334,
"grad_norm": 0.8635519627747134,
"learning_rate": 9.157348061512728e-06,
"loss": 1.294,
"step": 5
},
{
"epoch": 1.0,
"grad_norm": 0.7560256587528488,
"learning_rate": 8.535533905932739e-06,
"loss": 1.2551,
"step": 6
},
{
"epoch": 1.1666666666666667,
"grad_norm": 0.6624523616926192,
"learning_rate": 7.777851165098012e-06,
"loss": 1.2267,
"step": 7
},
{
"epoch": 1.3333333333333333,
"grad_norm": 0.5840695781359019,
"learning_rate": 6.913417161825449e-06,
"loss": 1.2174,
"step": 8
},
{
"epoch": 1.5,
"grad_norm": 0.5208348821330916,
"learning_rate": 5.975451610080643e-06,
"loss": 1.2134,
"step": 9
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.4477585736402009,
"learning_rate": 5e-06,
"loss": 1.1716,
"step": 10
},
{
"epoch": 1.8333333333333335,
"grad_norm": 0.4210137832048049,
"learning_rate": 4.02454838991936e-06,
"loss": 1.1606,
"step": 11
},
{
"epoch": 2.0,
"grad_norm": 0.39480845230736955,
"learning_rate": 3.0865828381745515e-06,
"loss": 1.1714,
"step": 12
},
{
"epoch": 2.1666666666666665,
"grad_norm": 0.3736519264684104,
"learning_rate": 2.2221488349019903e-06,
"loss": 1.1518,
"step": 13
},
{
"epoch": 2.3333333333333335,
"grad_norm": 0.36453793415908753,
"learning_rate": 1.4644660940672628e-06,
"loss": 1.1449,
"step": 14
},
{
"epoch": 2.5,
"grad_norm": 0.35639694150571866,
"learning_rate": 8.426519384872733e-07,
"loss": 1.167,
"step": 15
},
{
"epoch": 2.6666666666666665,
"grad_norm": 0.37214055458414597,
"learning_rate": 3.8060233744356634e-07,
"loss": 1.1749,
"step": 16
},
{
"epoch": 2.8333333333333335,
"grad_norm": 0.34523665157571026,
"learning_rate": 9.607359798384785e-08,
"loss": 1.1597,
"step": 17
},
{
"epoch": 3.0,
"grad_norm": 0.34261372184850886,
"learning_rate": 0.0,
"loss": 1.1474,
"step": 18
},
{
"epoch": 3.0,
"step": 18,
"total_flos": 6.208039628360909e+16,
"train_loss": 1.2118252846929762,
"train_runtime": 1204.5753,
"train_samples_per_second": 1.432,
"train_steps_per_second": 0.015
}
],
"logging_steps": 1.0,
"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": 6.208039628360909e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}