|
{ |
|
"best_metric": 26.9203, |
|
"best_model_checkpoint": "./ko-en_mbartLarge_exp20p_linear_decay/checkpoint-11000", |
|
"epoch": 2.552500290056851, |
|
"eval_steps": 1000, |
|
"global_step": 11000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 5e-05, |
|
"loss": 1.557, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 4.941300774829773e-05, |
|
"loss": 1.4396, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_bleu": 21.7052, |
|
"eval_gen_len": 18.6047, |
|
"eval_loss": 1.381515622138977, |
|
"eval_runtime": 1113.9878, |
|
"eval_samples_per_second": 15.473, |
|
"eval_steps_per_second": 0.968, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 4.8826015496595445e-05, |
|
"loss": 1.3838, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 4.823902324489317e-05, |
|
"loss": 1.338, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"eval_bleu": 23.7087, |
|
"eval_gen_len": 18.9939, |
|
"eval_loss": 1.3043569326400757, |
|
"eval_runtime": 1109.0361, |
|
"eval_samples_per_second": 15.542, |
|
"eval_steps_per_second": 0.972, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 4.7652030993190894e-05, |
|
"loss": 1.3081, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 4.706503874148862e-05, |
|
"loss": 1.2938, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"eval_bleu": 24.6339, |
|
"eval_gen_len": 18.8866, |
|
"eval_loss": 1.2555760145187378, |
|
"eval_runtime": 1131.9789, |
|
"eval_samples_per_second": 15.227, |
|
"eval_steps_per_second": 0.952, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 4.6478046489786336e-05, |
|
"loss": 1.2587, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 4.589105423808406e-05, |
|
"loss": 1.251, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"eval_bleu": 25.2975, |
|
"eval_gen_len": 19.0918, |
|
"eval_loss": 1.2229138612747192, |
|
"eval_runtime": 1120.4864, |
|
"eval_samples_per_second": 15.383, |
|
"eval_steps_per_second": 0.962, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 4.530406198638178e-05, |
|
"loss": 1.135, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"learning_rate": 4.47170697346795e-05, |
|
"loss": 0.9843, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"eval_bleu": 25.609, |
|
"eval_gen_len": 18.7589, |
|
"eval_loss": 1.2308591604232788, |
|
"eval_runtime": 1118.1785, |
|
"eval_samples_per_second": 15.415, |
|
"eval_steps_per_second": 0.964, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 4.413007748297723e-05, |
|
"loss": 0.9916, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 4.354308523127495e-05, |
|
"loss": 0.9874, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"eval_bleu": 26.1792, |
|
"eval_gen_len": 18.8287, |
|
"eval_loss": 1.2100664377212524, |
|
"eval_runtime": 1093.6874, |
|
"eval_samples_per_second": 15.76, |
|
"eval_steps_per_second": 0.986, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 4.295609297957267e-05, |
|
"loss": 0.9931, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"learning_rate": 4.2369100727870395e-05, |
|
"loss": 0.9838, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"eval_bleu": 26.024, |
|
"eval_gen_len": 18.4025, |
|
"eval_loss": 1.2053465843200684, |
|
"eval_runtime": 1074.6707, |
|
"eval_samples_per_second": 16.039, |
|
"eval_steps_per_second": 1.003, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 4.178210847616812e-05, |
|
"loss": 0.9931, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 4.1195116224465844e-05, |
|
"loss": 0.9927, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"eval_bleu": 26.3148, |
|
"eval_gen_len": 19.09, |
|
"eval_loss": 1.1906589269638062, |
|
"eval_runtime": 1112.0878, |
|
"eval_samples_per_second": 15.5, |
|
"eval_steps_per_second": 0.969, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"learning_rate": 4.060812397276356e-05, |
|
"loss": 0.9921, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 2.09, |
|
"learning_rate": 4.0021131721061286e-05, |
|
"loss": 0.7835, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.09, |
|
"eval_bleu": 26.5613, |
|
"eval_gen_len": 18.7196, |
|
"eval_loss": 1.2300479412078857, |
|
"eval_runtime": 1083.4234, |
|
"eval_samples_per_second": 15.91, |
|
"eval_steps_per_second": 0.995, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"learning_rate": 3.9434139469359004e-05, |
|
"loss": 0.7268, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 2.32, |
|
"learning_rate": 3.884714721765673e-05, |
|
"loss": 0.7437, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 2.32, |
|
"eval_bleu": 26.8232, |
|
"eval_gen_len": 18.6513, |
|
"eval_loss": 1.2358065843582153, |
|
"eval_runtime": 1078.4504, |
|
"eval_samples_per_second": 15.983, |
|
"eval_steps_per_second": 1.0, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 2.44, |
|
"learning_rate": 3.826015496595445e-05, |
|
"loss": 0.7532, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 2.55, |
|
"learning_rate": 3.767316271425218e-05, |
|
"loss": 0.7585, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 2.55, |
|
"eval_bleu": 26.9203, |
|
"eval_gen_len": 18.7513, |
|
"eval_loss": 1.2291104793548584, |
|
"eval_runtime": 1083.6296, |
|
"eval_samples_per_second": 15.907, |
|
"eval_steps_per_second": 0.995, |
|
"step": 11000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 43090, |
|
"num_train_epochs": 10, |
|
"save_steps": 1000, |
|
"total_flos": 7.62829678313472e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|