|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.6187161639597835, |
|
"eval_steps": 200, |
|
"global_step": 200, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 0.5968947410583496, |
|
"learning_rate": 9.896800825593395e-05, |
|
"loss": 2.3935, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 0.6443502306938171, |
|
"learning_rate": 9.793601651186791e-05, |
|
"loss": 2.1763, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 0.4056578576564789, |
|
"learning_rate": 9.690402476780186e-05, |
|
"loss": 1.9601, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 0.24697279930114746, |
|
"learning_rate": 9.587203302373582e-05, |
|
"loss": 1.9009, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.24417832493782043, |
|
"learning_rate": 9.484004127966977e-05, |
|
"loss": 1.8698, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"grad_norm": 0.19971579313278198, |
|
"learning_rate": 9.380804953560372e-05, |
|
"loss": 1.8502, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 0.21794500946998596, |
|
"learning_rate": 9.277605779153768e-05, |
|
"loss": 1.8139, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 0.20081552863121033, |
|
"learning_rate": 9.174406604747162e-05, |
|
"loss": 1.8246, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"grad_norm": 0.2045777291059494, |
|
"learning_rate": 9.071207430340559e-05, |
|
"loss": 1.8009, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 0.21513579785823822, |
|
"learning_rate": 8.968008255933953e-05, |
|
"loss": 1.7745, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 0.2838321924209595, |
|
"learning_rate": 8.864809081527348e-05, |
|
"loss": 1.7831, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 0.19812746345996857, |
|
"learning_rate": 8.761609907120744e-05, |
|
"loss": 1.7554, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.30143260955810547, |
|
"learning_rate": 8.658410732714138e-05, |
|
"loss": 1.7707, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 0.21341949701309204, |
|
"learning_rate": 8.555211558307535e-05, |
|
"loss": 1.7476, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.24006041884422302, |
|
"learning_rate": 8.452012383900929e-05, |
|
"loss": 1.7653, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.25095027685165405, |
|
"learning_rate": 8.348813209494324e-05, |
|
"loss": 1.7438, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 0.2602318525314331, |
|
"learning_rate": 8.24561403508772e-05, |
|
"loss": 1.7728, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"grad_norm": 0.26738253235816956, |
|
"learning_rate": 8.142414860681114e-05, |
|
"loss": 1.7433, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 0.25230053067207336, |
|
"learning_rate": 8.039215686274511e-05, |
|
"loss": 1.7304, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"grad_norm": 0.2360549122095108, |
|
"learning_rate": 7.936016511867905e-05, |
|
"loss": 1.7297, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"eval_loss": 1.744746446609497, |
|
"eval_runtime": 294.0669, |
|
"eval_samples_per_second": 35.172, |
|
"eval_steps_per_second": 4.397, |
|
"step": 200 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 969, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 200, |
|
"total_flos": 8.880783974203392e+16, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|