140 lines
3.2 KiB
JSON
140 lines
3.2 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 3.83712,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1500,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.256,
|
||
|
|
"grad_norm": 1.265313982963562,
|
||
|
|
"learning_rate": 2.4750000000000002e-05,
|
||
|
|
"loss": 6.332213745117188,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.512,
|
||
|
|
"grad_norm": 1.302055835723877,
|
||
|
|
"learning_rate": 4.975e-05,
|
||
|
|
"loss": 4.7161932373046875,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.768,
|
||
|
|
"grad_norm": 1.1382993459701538,
|
||
|
|
"learning_rate": 7.475e-05,
|
||
|
|
"loss": 4.228371276855468,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.02304,
|
||
|
|
"grad_norm": 1.2706514596939087,
|
||
|
|
"learning_rate": 9.975000000000001e-05,
|
||
|
|
"loss": 3.9627920532226564,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.27904,
|
||
|
|
"grad_norm": 1.1472086906433105,
|
||
|
|
"learning_rate": 0.00012475,
|
||
|
|
"loss": 3.7471749877929685,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.53504,
|
||
|
|
"grad_norm": 1.0958755016326904,
|
||
|
|
"learning_rate": 0.00014975,
|
||
|
|
"loss": 3.609476318359375,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.79104,
|
||
|
|
"grad_norm": 1.1427408456802368,
|
||
|
|
"learning_rate": 0.00017475,
|
||
|
|
"loss": 3.5249368286132814,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.04608,
|
||
|
|
"grad_norm": 2.976166248321533,
|
||
|
|
"learning_rate": 0.00019975,
|
||
|
|
"loss": 3.419891357421875,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.30208,
|
||
|
|
"grad_norm": 1.2115665674209595,
|
||
|
|
"learning_rate": 0.00022475000000000001,
|
||
|
|
"loss": 3.2445193481445314,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.55808,
|
||
|
|
"grad_norm": 1.1276987791061401,
|
||
|
|
"learning_rate": 0.00024975,
|
||
|
|
"loss": 3.2045053100585936,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.81408,
|
||
|
|
"grad_norm": 1.2419933080673218,
|
||
|
|
"learning_rate": 0.00022658677055413672,
|
||
|
|
"loss": 3.169610595703125,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.06912,
|
||
|
|
"grad_norm": 1.1096692085266113,
|
||
|
|
"learning_rate": 0.0001643733149569131,
|
||
|
|
"loss": 3.0363580322265626,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.32512,
|
||
|
|
"grad_norm": 1.0762932300567627,
|
||
|
|
"learning_rate": 8.712059129590328e-05,
|
||
|
|
"loss": 2.8395703125,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.58112,
|
||
|
|
"grad_norm": 1.061633586883545,
|
||
|
|
"learning_rate": 2.4336514286109764e-05,
|
||
|
|
"loss": 2.7858563232421876,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.83712,
|
||
|
|
"grad_norm": 0.9627223610877991,
|
||
|
|
"learning_rate": 2.4673929828605834e-09,
|
||
|
|
"loss": 2.765904541015625,
|
||
|
|
"step": 1500
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 100,
|
||
|
|
"max_steps": 1500,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"save_steps": 200,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 6403909619810304.0,
|
||
|
|
"train_batch_size": 4,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|