149 lines
3.5 KiB
JSON
149 lines
3.5 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 78,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.0641025641025641,
|
||
|
|
"grad_norm": 4.013121128082275,
|
||
|
|
"learning_rate": 1e-05,
|
||
|
|
"loss": 0.5757,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1282051282051282,
|
||
|
|
"grad_norm": 1.0922566652297974,
|
||
|
|
"learning_rate": 1.9989930665413148e-05,
|
||
|
|
"loss": 0.116,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19230769230769232,
|
||
|
|
"grad_norm": 0.5459157824516296,
|
||
|
|
"learning_rate": 1.9639628606958535e-05,
|
||
|
|
"loss": 0.1237,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2564102564102564,
|
||
|
|
"grad_norm": 0.7706058025360107,
|
||
|
|
"learning_rate": 1.880595531856738e-05,
|
||
|
|
"loss": 0.0934,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32051282051282054,
|
||
|
|
"grad_norm": 1.026077389717102,
|
||
|
|
"learning_rate": 1.7530714660036112e-05,
|
||
|
|
"loss": 0.0922,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.38461538461538464,
|
||
|
|
"grad_norm": 0.44153380393981934,
|
||
|
|
"learning_rate": 1.5877852522924733e-05,
|
||
|
|
"loss": 0.0973,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.44871794871794873,
|
||
|
|
"grad_norm": 0.5316270589828491,
|
||
|
|
"learning_rate": 1.3930250316539237e-05,
|
||
|
|
"loss": 0.0888,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5128205128205128,
|
||
|
|
"grad_norm": 0.43476516008377075,
|
||
|
|
"learning_rate": 1.1785568947986368e-05,
|
||
|
|
"loss": 0.0983,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5769230769230769,
|
||
|
|
"grad_norm": 0.16090889275074005,
|
||
|
|
"learning_rate": 9.551351696494854e-06,
|
||
|
|
"loss": 0.0946,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6410256410256411,
|
||
|
|
"grad_norm": 0.16215692460536957,
|
||
|
|
"learning_rate": 7.33963154433325e-06,
|
||
|
|
"loss": 0.0881,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7051282051282052,
|
||
|
|
"grad_norm": 0.27346980571746826,
|
||
|
|
"learning_rate": 5.2613133752700145e-06,
|
||
|
|
"loss": 0.0876,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7692307692307693,
|
||
|
|
"grad_norm": 0.4459897577762604,
|
||
|
|
"learning_rate": 3.4206127406028744e-06,
|
||
|
|
"loss": 0.0874,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8333333333333334,
|
||
|
|
"grad_norm": 0.28861892223358154,
|
||
|
|
"learning_rate": 1.9098300562505266e-06,
|
||
|
|
"loss": 0.0868,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8974358974358975,
|
||
|
|
"grad_norm": 0.25620919466018677,
|
||
|
|
"learning_rate": 8.047222744854943e-07,
|
||
|
|
"loss": 0.0955,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9615384615384616,
|
||
|
|
"grad_norm": 0.03533744439482689,
|
||
|
|
"learning_rate": 1.6070411401370335e-07,
|
||
|
|
"loss": 0.0871,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"step": 78,
|
||
|
|
"total_flos": 2.743502659167191e+17,
|
||
|
|
"train_loss": 0.12591312176141983,
|
||
|
|
"train_runtime": 1812.1381,
|
||
|
|
"train_samples_per_second": 2.749,
|
||
|
|
"train_steps_per_second": 0.043
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 78,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"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": 2.743502659167191e+17,
|
||
|
|
"train_batch_size": 8,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|