| { | |
| "best_metric": 40.6424, | |
| "best_model_checkpoint": "M2M100_enfr_FT_wang_2022/checkpoint-256000", | |
| "epoch": 3.130158341994253, | |
| "global_step": 256000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.975551751543682e-05, | |
| "loss": 0.2244, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bleu": 35.6567, | |
| "eval_gen_len": 44.8111, | |
| "eval_loss": 0.13247372210025787, | |
| "eval_runtime": 433.7309, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 0.302, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.951108088280247e-05, | |
| "loss": 0.1522, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bleu": 36.512, | |
| "eval_gen_len": 44.4851, | |
| "eval_loss": 0.1262633502483368, | |
| "eval_runtime": 417.8212, | |
| "eval_samples_per_second": 2.496, | |
| "eval_steps_per_second": 0.314, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.9266613682215567e-05, | |
| "loss": 0.1435, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_bleu": 37.957, | |
| "eval_gen_len": 44.5302, | |
| "eval_loss": 0.12126067280769348, | |
| "eval_runtime": 414.4742, | |
| "eval_samples_per_second": 2.516, | |
| "eval_steps_per_second": 0.316, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.9022177049581222e-05, | |
| "loss": 0.1384, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_bleu": 38.0569, | |
| "eval_gen_len": 44.5034, | |
| "eval_loss": 0.1191168949007988, | |
| "eval_runtime": 415.8111, | |
| "eval_samples_per_second": 2.508, | |
| "eval_steps_per_second": 0.315, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.8777709848994316e-05, | |
| "loss": 0.1345, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_bleu": 38.4966, | |
| "eval_gen_len": 44.8821, | |
| "eval_loss": 0.11713190376758575, | |
| "eval_runtime": 423.2079, | |
| "eval_samples_per_second": 2.465, | |
| "eval_steps_per_second": 0.31, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.853325793238369e-05, | |
| "loss": 0.1213, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_bleu": 39.0368, | |
| "eval_gen_len": 44.6012, | |
| "eval_loss": 0.1156671866774559, | |
| "eval_runtime": 417.9213, | |
| "eval_samples_per_second": 2.496, | |
| "eval_steps_per_second": 0.313, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.8288806015773065e-05, | |
| "loss": 0.1199, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_bleu": 39.6906, | |
| "eval_gen_len": 44.8178, | |
| "eval_loss": 0.11390843987464905, | |
| "eval_runtime": 418.2954, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 0.313, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.804436938313872e-05, | |
| "loss": 0.1195, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_bleu": 39.6284, | |
| "eval_gen_len": 44.8552, | |
| "eval_loss": 0.11290750652551651, | |
| "eval_runtime": 422.1564, | |
| "eval_samples_per_second": 2.471, | |
| "eval_steps_per_second": 0.31, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.7799902182551813e-05, | |
| "loss": 0.1185, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_bleu": 39.0079, | |
| "eval_gen_len": 44.5618, | |
| "eval_loss": 0.11247587949037552, | |
| "eval_runtime": 408.9786, | |
| "eval_samples_per_second": 2.55, | |
| "eval_steps_per_second": 0.32, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.7555465549917468e-05, | |
| "loss": 0.1175, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_bleu": 39.2898, | |
| "eval_gen_len": 44.745, | |
| "eval_loss": 0.1103997528553009, | |
| "eval_runtime": 413.1156, | |
| "eval_samples_per_second": 2.525, | |
| "eval_steps_per_second": 0.317, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.7310998349330562e-05, | |
| "loss": 0.1062, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "eval_bleu": 39.8593, | |
| "eval_gen_len": 45.1151, | |
| "eval_loss": 0.1111496165394783, | |
| "eval_runtime": 415.6075, | |
| "eval_samples_per_second": 2.51, | |
| "eval_steps_per_second": 0.315, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.7066546432719936e-05, | |
| "loss": 0.1047, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_bleu": 39.913, | |
| "eval_gen_len": 44.8102, | |
| "eval_loss": 0.11086419969797134, | |
| "eval_runtime": 418.8388, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 0.313, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 1.682210980008559e-05, | |
| "loss": 0.1055, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_bleu": 40.2278, | |
| "eval_gen_len": 45.2848, | |
| "eval_loss": 0.11025020480155945, | |
| "eval_runtime": 421.9598, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 0.31, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 1.6577657883474966e-05, | |
| "loss": 0.1059, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_bleu": 40.2198, | |
| "eval_gen_len": 45.0719, | |
| "eval_loss": 0.10949720442295074, | |
| "eval_runtime": 416.4666, | |
| "eval_samples_per_second": 2.504, | |
| "eval_steps_per_second": 0.315, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 1.633320596686434e-05, | |
| "loss": 0.106, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_bleu": 39.8973, | |
| "eval_gen_len": 44.954, | |
| "eval_loss": 0.10881481319665909, | |
| "eval_runtime": 421.0124, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 0.311, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1.6088769334229995e-05, | |
| "loss": 0.0971, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_bleu": 40.6424, | |
| "eval_gen_len": 44.9732, | |
| "eval_loss": 0.11019956320524216, | |
| "eval_runtime": 418.4973, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 0.313, | |
| "step": 256000 | |
| } | |
| ], | |
| "max_steps": 1308560, | |
| "num_train_epochs": 16, | |
| "total_flos": 8.283927408492872e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |