|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 5.058412621943876, |
|
"eval_steps": 500, |
|
"global_step": 21000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.24087679152113695, |
|
"grad_norm": 0.5082331299781799, |
|
"learning_rate": 0.000991969806472336, |
|
"loss": 1.4852, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.4817535830422739, |
|
"grad_norm": 0.4949457347393036, |
|
"learning_rate": 0.0009839396129446719, |
|
"loss": 1.369, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.7226303745634108, |
|
"grad_norm": 0.5630651116371155, |
|
"learning_rate": 0.0009759094194170079, |
|
"loss": 1.3189, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.9635071660845478, |
|
"grad_norm": 0.5108479261398315, |
|
"learning_rate": 0.000967879225889344, |
|
"loss": 1.2817, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.2043839576056847, |
|
"grad_norm": 0.47163671255111694, |
|
"learning_rate": 0.00095984903236168, |
|
"loss": 1.2511, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.4452607491268217, |
|
"grad_norm": 0.531821072101593, |
|
"learning_rate": 0.0009518188388340159, |
|
"loss": 1.2309, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.6861375406479586, |
|
"grad_norm": 0.4891831874847412, |
|
"learning_rate": 0.0009437886453063518, |
|
"loss": 1.2088, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.9270143321690956, |
|
"grad_norm": 0.4778994619846344, |
|
"learning_rate": 0.0009357584517786879, |
|
"loss": 1.1936, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.1678911236902323, |
|
"grad_norm": 0.5436965823173523, |
|
"learning_rate": 0.0009277282582510239, |
|
"loss": 1.1792, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.4087679152113695, |
|
"grad_norm": 0.45789214968681335, |
|
"learning_rate": 0.0009196980647233599, |
|
"loss": 1.1609, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 2.649644706732506, |
|
"grad_norm": 0.49292871356010437, |
|
"learning_rate": 0.0009116678711956958, |
|
"loss": 1.1465, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 2.8905214982536434, |
|
"grad_norm": 0.4877796769142151, |
|
"learning_rate": 0.0009036376776680318, |
|
"loss": 1.1339, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 3.13139828977478, |
|
"grad_norm": 0.45624956488609314, |
|
"learning_rate": 0.0008956074841403679, |
|
"loss": 1.1235, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 3.3722750812959172, |
|
"grad_norm": 0.45705732703208923, |
|
"learning_rate": 0.0008875772906127038, |
|
"loss": 1.114, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 3.613151872817054, |
|
"grad_norm": 0.4534723460674286, |
|
"learning_rate": 0.0008795470970850398, |
|
"loss": 1.1022, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 3.854028664338191, |
|
"grad_norm": 0.5117238163948059, |
|
"learning_rate": 0.0008715169035573758, |
|
"loss": 1.0936, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 4.094905455859328, |
|
"grad_norm": 0.4832773804664612, |
|
"learning_rate": 0.0008634867100297117, |
|
"loss": 1.0895, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 4.335782247380465, |
|
"grad_norm": 0.5012445449829102, |
|
"learning_rate": 0.0008554565165020477, |
|
"loss": 1.0793, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 4.576659038901602, |
|
"grad_norm": 0.5131984353065491, |
|
"learning_rate": 0.0008474263229743838, |
|
"loss": 1.0737, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 4.817535830422739, |
|
"grad_norm": 0.5085521340370178, |
|
"learning_rate": 0.0008393961294467197, |
|
"loss": 1.068, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 5.058412621943876, |
|
"grad_norm": 0.4861578345298767, |
|
"learning_rate": 0.0008313659359190556, |
|
"loss": 1.0596, |
|
"step": 21000 |
|
} |
|
], |
|
"logging_steps": 1000, |
|
"max_steps": 124530, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 30, |
|
"save_steps": 1000, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 2.1007632729269862e+17, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|