109 lines
2.5 KiB
JSON
109 lines
2.5 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0,
|
|
"eval_steps": 25,
|
|
"global_step": 39,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.4,
|
|
"grad_norm": 14.682328659827894,
|
|
"learning_rate": 1.9856159103477085e-05,
|
|
"loss": 0.8891424179077149,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 11.714302479472073,
|
|
"learning_rate": 1.8285096492438424e-05,
|
|
"loss": 0.5139811515808106,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 1.16,
|
|
"grad_norm": 10.208381605756056,
|
|
"learning_rate": 1.5243072835572319e-05,
|
|
"loss": 0.3687877893447876,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 1.56,
|
|
"grad_norm": 10.262843920575644,
|
|
"learning_rate": 1.1270178197468788e-05,
|
|
"loss": 0.21150989532470704,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 1.96,
|
|
"grad_norm": 6.68105063937876,
|
|
"learning_rate": 7.071772287234497e-06,
|
|
"loss": 0.17464923858642578,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 1.96,
|
|
"eval_loss": 0.25689050555229187,
|
|
"eval_runtime": 0.9005,
|
|
"eval_samples_per_second": 46.642,
|
|
"eval_steps_per_second": 6.663,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 2.32,
|
|
"grad_norm": 5.69195152871731,
|
|
"learning_rate": 3.3932527660991877e-06,
|
|
"loss": 0.09203737378120422,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 2.7199999999999998,
|
|
"grad_norm": 4.8584537804368715,
|
|
"learning_rate": 8.87715096118642e-07,
|
|
"loss": 0.0855832576751709,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 3.0,
|
|
"eval_loss": 0.22509032487869263,
|
|
"eval_runtime": 0.8925,
|
|
"eval_samples_per_second": 47.061,
|
|
"eval_steps_per_second": 6.723,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 3.0,
|
|
"step": 39,
|
|
"total_flos": 1274990395392.0,
|
|
"train_loss": 0.3065599493491344,
|
|
"train_runtime": 587.7054,
|
|
"train_samples_per_second": 4.012,
|
|
"train_steps_per_second": 0.066
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 39,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 3,
|
|
"save_steps": 25,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1274990395392.0,
|
|
"train_batch_size": 4,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|