200 lines
4.4 KiB
JSON
200 lines
4.4 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 10,
|
||
|
|
"global_step": 21,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.049079754601226995,
|
||
|
|
"grad_norm": 122.0,
|
||
|
|
"learning_rate": 0.0,
|
||
|
|
"loss": 9.9141,
|
||
|
|
"step": 1
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09815950920245399,
|
||
|
|
"grad_norm": 116.0,
|
||
|
|
"learning_rate": 2.0000000000000003e-06,
|
||
|
|
"loss": 9.5173,
|
||
|
|
"step": 2
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.147239263803681,
|
||
|
|
"grad_norm": 120.5,
|
||
|
|
"learning_rate": 4.000000000000001e-06,
|
||
|
|
"loss": 9.8562,
|
||
|
|
"step": 3
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19631901840490798,
|
||
|
|
"grad_norm": 111.5,
|
||
|
|
"learning_rate": 6e-06,
|
||
|
|
"loss": 9.6533,
|
||
|
|
"step": 4
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.24539877300613497,
|
||
|
|
"grad_norm": 92.0,
|
||
|
|
"learning_rate": 8.000000000000001e-06,
|
||
|
|
"loss": 8.3424,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.294478527607362,
|
||
|
|
"grad_norm": 70.0,
|
||
|
|
"learning_rate": 1e-05,
|
||
|
|
"loss": 6.9478,
|
||
|
|
"step": 6
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.34355828220858897,
|
||
|
|
"grad_norm": 62.0,
|
||
|
|
"learning_rate": 9.375000000000001e-06,
|
||
|
|
"loss": 5.7066,
|
||
|
|
"step": 7
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.39263803680981596,
|
||
|
|
"grad_norm": 61.25,
|
||
|
|
"learning_rate": 8.750000000000001e-06,
|
||
|
|
"loss": 4.5008,
|
||
|
|
"step": 8
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.44171779141104295,
|
||
|
|
"grad_norm": 60.0,
|
||
|
|
"learning_rate": 8.125000000000001e-06,
|
||
|
|
"loss": 3.7475,
|
||
|
|
"step": 9
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.49079754601226994,
|
||
|
|
"grad_norm": 48.25,
|
||
|
|
"learning_rate": 7.500000000000001e-06,
|
||
|
|
"loss": 3.3201,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.49079754601226994,
|
||
|
|
"eval_loss": 2.8368520736694336,
|
||
|
|
"eval_model_preparation_time": 0.0198,
|
||
|
|
"eval_runtime": 0.6339,
|
||
|
|
"eval_samples_per_second": 56.79,
|
||
|
|
"eval_steps_per_second": 28.395,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5398773006134969,
|
||
|
|
"grad_norm": 28.75,
|
||
|
|
"learning_rate": 6.875e-06,
|
||
|
|
"loss": 2.5935,
|
||
|
|
"step": 11
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.588957055214724,
|
||
|
|
"grad_norm": 23.25,
|
||
|
|
"learning_rate": 6.25e-06,
|
||
|
|
"loss": 2.6667,
|
||
|
|
"step": 12
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6380368098159509,
|
||
|
|
"grad_norm": 20.375,
|
||
|
|
"learning_rate": 5.625e-06,
|
||
|
|
"loss": 2.5437,
|
||
|
|
"step": 13
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6871165644171779,
|
||
|
|
"grad_norm": 18.25,
|
||
|
|
"learning_rate": 5e-06,
|
||
|
|
"loss": 2.4875,
|
||
|
|
"step": 14
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7361963190184049,
|
||
|
|
"grad_norm": 15.75,
|
||
|
|
"learning_rate": 4.3750000000000005e-06,
|
||
|
|
"loss": 1.9838,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7852760736196319,
|
||
|
|
"grad_norm": 13.125,
|
||
|
|
"learning_rate": 3.7500000000000005e-06,
|
||
|
|
"loss": 2.0393,
|
||
|
|
"step": 16
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8343558282208589,
|
||
|
|
"grad_norm": 13.25,
|
||
|
|
"learning_rate": 3.125e-06,
|
||
|
|
"loss": 2.1152,
|
||
|
|
"step": 17
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8834355828220859,
|
||
|
|
"grad_norm": 12.4375,
|
||
|
|
"learning_rate": 2.5e-06,
|
||
|
|
"loss": 1.8531,
|
||
|
|
"step": 18
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9325153374233128,
|
||
|
|
"grad_norm": 12.875,
|
||
|
|
"learning_rate": 1.8750000000000003e-06,
|
||
|
|
"loss": 1.9776,
|
||
|
|
"step": 19
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9815950920245399,
|
||
|
|
"grad_norm": 13.6875,
|
||
|
|
"learning_rate": 1.25e-06,
|
||
|
|
"loss": 2.035,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9815950920245399,
|
||
|
|
"eval_loss": 2.012303352355957,
|
||
|
|
"eval_model_preparation_time": 0.0198,
|
||
|
|
"eval_runtime": 0.6308,
|
||
|
|
"eval_samples_per_second": 57.075,
|
||
|
|
"eval_steps_per_second": 28.537,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 16.875,
|
||
|
|
"learning_rate": 6.25e-07,
|
||
|
|
"loss": 1.6999,
|
||
|
|
"step": 21
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 1,
|
||
|
|
"max_steps": 21,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 5000,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 511063872884736.0,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|