|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 2000.0, |
|
"global_step": 11877, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 2.5e-05, |
|
"loss": 1.5023, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 5e-05, |
|
"loss": 1.124, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 7.500000000000001e-05, |
|
"loss": 1.0616, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.0001, |
|
"loss": 1.0334, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 9.49377341297965e-05, |
|
"loss": 1.0149, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 8.9875468259593e-05, |
|
"loss": 0.9938, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 8.481320238938948e-05, |
|
"loss": 0.987, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 7.975093651918599e-05, |
|
"loss": 0.97, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 7.468867064898249e-05, |
|
"loss": 0.9497, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 6.962640477877899e-05, |
|
"loss": 0.9438, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 6.456413890857548e-05, |
|
"loss": 0.938, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 5.950187303837198e-05, |
|
"loss": 0.9261, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 5.443960716816847e-05, |
|
"loss": 0.924, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 4.937734129796497e-05, |
|
"loss": 0.9165, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 4.431507542776147e-05, |
|
"loss": 0.9034, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 3.925280955755796e-05, |
|
"loss": 0.8966, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"learning_rate": 3.419054368735446e-05, |
|
"loss": 0.8829, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 2.9128277817150957e-05, |
|
"loss": 0.8843, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"learning_rate": 2.4066011946947456e-05, |
|
"loss": 0.8725, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"learning_rate": 1.900374607674395e-05, |
|
"loss": 0.8712, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"learning_rate": 1.3941480206540447e-05, |
|
"loss": 0.8673, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 8.879214336336945e-06, |
|
"loss": 0.8669, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"learning_rate": 3.816948466133442e-06, |
|
"loss": 0.8679, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 11877, |
|
"total_flos": 5.792740247863296e+16, |
|
"train_loss": 0.9618661650365136, |
|
"train_runtime": 9694.4976, |
|
"train_samples_per_second": 19.6, |
|
"train_steps_per_second": 1.225 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 11877, |
|
"num_train_epochs": 3, |
|
"save_steps": 2000, |
|
"total_flos": 5.792740247863296e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|