| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 16.995936461026968, | |
| "global_step": 3834, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 69.52679870560905, | |
| "eval_loss": 0.16526192426681519, | |
| "eval_perfect_accuracy": 0.10141313383208644, | |
| "eval_runtime": 3340.3895, | |
| "eval_samples_per_second": 1.08, | |
| "eval_steps_per_second": 0.045, | |
| "eval_subword_accuracy": 0.5594287933042841, | |
| "eval_token_accuracy": 0.6964062297769533, | |
| "eval_word_accuracy": 0.4744973429025366, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 74.31204559162617, | |
| "eval_loss": 0.11631891876459122, | |
| "eval_perfect_accuracy": 0.14990302022720975, | |
| "eval_runtime": 3269.4249, | |
| "eval_samples_per_second": 1.104, | |
| "eval_steps_per_second": 0.046, | |
| "eval_subword_accuracy": 0.6142826689405244, | |
| "eval_token_accuracy": 0.7389895544659987, | |
| "eval_word_accuracy": 0.532967032967033, | |
| "step": 451 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.6045, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 76.85411902074411, | |
| "eval_loss": 0.09909386187791824, | |
| "eval_perfect_accuracy": 0.18952618453865336, | |
| "eval_runtime": 3381.7443, | |
| "eval_samples_per_second": 1.067, | |
| "eval_steps_per_second": 0.045, | |
| "eval_subword_accuracy": 0.6512524901453821, | |
| "eval_token_accuracy": 0.7695926757936676, | |
| "eval_word_accuracy": 0.5709718317623064, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 79.14575222404268, | |
| "eval_loss": 0.09123498946428299, | |
| "eval_perfect_accuracy": 0.21169298974785258, | |
| "eval_runtime": 3480.4117, | |
| "eval_samples_per_second": 1.037, | |
| "eval_steps_per_second": 0.043, | |
| "eval_subword_accuracy": 0.6773669070746654, | |
| "eval_token_accuracy": 0.7950973443948806, | |
| "eval_word_accuracy": 0.5997376762487704, | |
| "step": 902 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 4.111111111111111e-05, | |
| "loss": 0.1137, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 80.22128752790911, | |
| "eval_loss": 0.08361164480447769, | |
| "eval_perfect_accuracy": 0.23330562482682182, | |
| "eval_runtime": 1727.4258, | |
| "eval_samples_per_second": 2.089, | |
| "eval_steps_per_second": 0.087, | |
| "eval_subword_accuracy": 0.693643785404865, | |
| "eval_token_accuracy": 0.8034821726091302, | |
| "eval_word_accuracy": 0.6188809276376962, | |
| "step": 1127 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 80.47331007551738, | |
| "eval_loss": 0.07888099551200867, | |
| "eval_perfect_accuracy": 0.25270157938487114, | |
| "eval_runtime": 1706.4952, | |
| "eval_samples_per_second": 2.115, | |
| "eval_steps_per_second": 0.088, | |
| "eval_subword_accuracy": 0.6985128247973933, | |
| "eval_token_accuracy": 0.8036176790040939, | |
| "eval_word_accuracy": 0.6238234160841085, | |
| "step": 1353 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.0886, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 81.35496460228585, | |
| "eval_loss": 0.07820562273263931, | |
| "eval_perfect_accuracy": 0.257689110556941, | |
| "eval_runtime": 1696.8822, | |
| "eval_samples_per_second": 2.127, | |
| "eval_steps_per_second": 0.089, | |
| "eval_subword_accuracy": 0.7077631303620004, | |
| "eval_token_accuracy": 0.8164888051893702, | |
| "eval_word_accuracy": 0.6347371318469585, | |
| "step": 1579 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 81.4155970043395, | |
| "eval_loss": 0.07889819890260696, | |
| "eval_perfect_accuracy": 0.2454973676918814, | |
| "eval_runtime": 1666.9192, | |
| "eval_samples_per_second": 2.165, | |
| "eval_steps_per_second": 0.091, | |
| "eval_subword_accuracy": 0.6984033613445378, | |
| "eval_token_accuracy": 0.8176890756302521, | |
| "eval_word_accuracy": 0.6222492535662943, | |
| "step": 1804 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3.222222222222223e-05, | |
| "loss": 0.0768, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 81.70774187369526, | |
| "eval_loss": 0.07279801368713379, | |
| "eval_perfect_accuracy": 0.27847049044056527, | |
| "eval_runtime": 1589.4085, | |
| "eval_samples_per_second": 2.271, | |
| "eval_steps_per_second": 0.095, | |
| "eval_subword_accuracy": 0.7123310810810811, | |
| "eval_token_accuracy": 0.8106841216216216, | |
| "eval_word_accuracy": 0.6412536119137586, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 82.60753055516308, | |
| "eval_loss": 0.07139766961336136, | |
| "eval_perfect_accuracy": 0.29897478525907456, | |
| "eval_runtime": 1630.9624, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.093, | |
| "eval_subword_accuracy": 0.7325649377766642, | |
| "eval_token_accuracy": 0.8289902280130294, | |
| "eval_word_accuracy": 0.6648324437352116, | |
| "step": 2255 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 83.62166331521699, | |
| "eval_loss": 0.06930351257324219, | |
| "eval_perfect_accuracy": 0.30950401773344416, | |
| "eval_runtime": 1612.5857, | |
| "eval_samples_per_second": 2.238, | |
| "eval_steps_per_second": 0.094, | |
| "eval_subword_accuracy": 0.739740238184157, | |
| "eval_token_accuracy": 0.8358438109465123, | |
| "eval_word_accuracy": 0.6716549295774648, | |
| "step": 2481 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0681, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 84.02807089662639, | |
| "eval_loss": 0.06798376142978668, | |
| "eval_perfect_accuracy": 0.3081185924078692, | |
| "eval_runtime": 1680.6877, | |
| "eval_samples_per_second": 2.147, | |
| "eval_steps_per_second": 0.09, | |
| "eval_subword_accuracy": 0.745770423991727, | |
| "eval_token_accuracy": 0.8428541882109617, | |
| "eval_word_accuracy": 0.6793930762489044, | |
| "step": 2707 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 84.42354785113844, | |
| "eval_loss": 0.06762586534023285, | |
| "eval_perfect_accuracy": 0.31698531449154893, | |
| "eval_runtime": 1751.9665, | |
| "eval_samples_per_second": 2.06, | |
| "eval_steps_per_second": 0.086, | |
| "eval_subword_accuracy": 0.7447953313190679, | |
| "eval_token_accuracy": 0.8407350689127105, | |
| "eval_word_accuracy": 0.67777227994942, | |
| "step": 2932 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.0619, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 84.59233008431931, | |
| "eval_loss": 0.06648800522089005, | |
| "eval_perfect_accuracy": 0.32696037683568857, | |
| "eval_runtime": 1738.746, | |
| "eval_samples_per_second": 2.076, | |
| "eval_steps_per_second": 0.087, | |
| "eval_subword_accuracy": 0.7536297454036935, | |
| "eval_token_accuracy": 0.8476946489532349, | |
| "eval_word_accuracy": 0.6883173734610123, | |
| "step": 3158 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 84.49240456076029, | |
| "eval_loss": 0.06621185690164566, | |
| "eval_perfect_accuracy": 0.32917705735660846, | |
| "eval_runtime": 1720.559, | |
| "eval_samples_per_second": 2.098, | |
| "eval_steps_per_second": 0.088, | |
| "eval_subword_accuracy": 0.7496996063807748, | |
| "eval_token_accuracy": 0.8439196188108556, | |
| "eval_word_accuracy": 0.6839415256100242, | |
| "step": 3383 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 1.888888888888889e-05, | |
| "loss": 0.0574, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_bleu": 85.02924908908608, | |
| "eval_loss": 0.06474177539348602, | |
| "eval_perfect_accuracy": 0.343585480742588, | |
| "eval_runtime": 1678.6132, | |
| "eval_samples_per_second": 2.15, | |
| "eval_steps_per_second": 0.09, | |
| "eval_subword_accuracy": 0.7591370348981107, | |
| "eval_token_accuracy": 0.8501773781041169, | |
| "eval_word_accuracy": 0.6949701058636388, | |
| "step": 3609 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_bleu": 84.79034637013908, | |
| "eval_loss": 0.06463250517845154, | |
| "eval_perfect_accuracy": 0.34968135217511775, | |
| "eval_runtime": 1711.9697, | |
| "eval_samples_per_second": 2.108, | |
| "eval_steps_per_second": 0.088, | |
| "eval_subword_accuracy": 0.762099607356892, | |
| "eval_token_accuracy": 0.8493490390576566, | |
| "eval_word_accuracy": 0.7003675462175655, | |
| "step": 3834 | |
| } | |
| ], | |
| "max_steps": 5625, | |
| "num_train_epochs": 25, | |
| "total_flos": 6098403774431232.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |