| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 19510, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.8718605843157357e-05, | |
| "loss": 0.5024, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.743721168631472e-05, | |
| "loss": 0.3004, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.6155817529472065e-05, | |
| "loss": 0.2771, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.487442337262942e-05, | |
| "loss": 0.2631, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.359302921578678e-05, | |
| "loss": 0.2525, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.2311635058944134e-05, | |
| "loss": 0.246, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.103024090210149e-05, | |
| "loss": 0.2379, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.974884674525884e-05, | |
| "loss": 0.2329, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.8467452588416197e-05, | |
| "loss": 0.2266, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.718605843157356e-05, | |
| "loss": 0.224, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.590466427473091e-05, | |
| "loss": 0.219, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.462327011788826e-05, | |
| "loss": 0.2148, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.334187596104562e-05, | |
| "loss": 0.2128, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.2060481804202974e-05, | |
| "loss": 0.2118, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.077908764736033e-05, | |
| "loss": 0.2088, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.9497693490517686e-05, | |
| "loss": 0.2042, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.821629933367504e-05, | |
| "loss": 0.2031, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.693490517683239e-05, | |
| "loss": 0.2031, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.565351101998975e-05, | |
| "loss": 0.1993, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 0.17119529843330383, | |
| "eval_runtime": 1684.1894, | |
| "eval_samples_per_second": 46.336, | |
| "eval_steps_per_second": 0.724, | |
| "step": 9755 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 2.4372116863147106e-05, | |
| "loss": 0.1977, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.309072270630446e-05, | |
| "loss": 0.1966, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.1809328549461817e-05, | |
| "loss": 0.1957, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.052793439261917e-05, | |
| "loss": 0.1938, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.9246540235776526e-05, | |
| "loss": 0.1927, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.796514607893388e-05, | |
| "loss": 0.1926, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.6683751922091237e-05, | |
| "loss": 0.1908, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.540235776524859e-05, | |
| "loss": 0.1901, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.4120963608405946e-05, | |
| "loss": 0.1892, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.2839569451563302e-05, | |
| "loss": 0.1886, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.1558175294720656e-05, | |
| "loss": 0.1893, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.0276781137878012e-05, | |
| "loss": 0.1872, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.995386981035367e-06, | |
| "loss": 0.1879, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 7.713992824192722e-06, | |
| "loss": 0.1862, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 6.432598667350077e-06, | |
| "loss": 0.1862, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.151204510507432e-06, | |
| "loss": 0.1861, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.869810353664787e-06, | |
| "loss": 0.185, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.588416196822143e-06, | |
| "loss": 0.1848, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.3070220399794978e-06, | |
| "loss": 0.186, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.5627883136852895e-08, | |
| "loss": 0.1863, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 0.163666233420372, | |
| "eval_runtime": 1684.6721, | |
| "eval_samples_per_second": 46.323, | |
| "eval_steps_per_second": 0.724, | |
| "step": 19510 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 19510, | |
| "total_flos": 3.801797311667896e+17, | |
| "train_loss": 0.21620578862409723, | |
| "train_runtime": 45769.0844, | |
| "train_samples_per_second": 27.281, | |
| "train_steps_per_second": 0.426 | |
| } | |
| ], | |
| "max_steps": 19510, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.801797311667896e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |