| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 1682, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.9970273483947685e-05, | |
| "loss": 9.5696, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.913793103448276e-05, | |
| "loss": 2.0231, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.827586206896552e-05, | |
| "loss": 1.2689, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.741379310344828e-05, | |
| "loss": 1.1252, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.655172413793104e-05, | |
| "loss": 1.0745, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.5689655172413794e-05, | |
| "loss": 1.1696, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.482758620689655e-05, | |
| "loss": 1.0006, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.396551724137931e-05, | |
| "loss": 1.0273, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.3103448275862066e-05, | |
| "loss": 1.1105, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.224137931034483e-05, | |
| "loss": 1.2629, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.9306, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.0517241379310344e-05, | |
| "loss": 0.9233, | |
| "step": 319 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.965517241379311e-05, | |
| "loss": 0.9108, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.8793103448275865e-05, | |
| "loss": 0.9462, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.793103448275862e-05, | |
| "loss": 0.9254, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.7068965517241385e-05, | |
| "loss": 0.7106, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3.620689655172414e-05, | |
| "loss": 0.7342, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.53448275862069e-05, | |
| "loss": 0.7573, | |
| "step": 493 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 0.7451, | |
| "step": 522 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3.3620689655172414e-05, | |
| "loss": 0.714, | |
| "step": 551 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.8191, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.1896551724137935e-05, | |
| "loss": 0.6586, | |
| "step": 609 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.103448275862069e-05, | |
| "loss": 0.7097, | |
| "step": 638 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.017241379310345e-05, | |
| "loss": 0.6619, | |
| "step": 667 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.9310344827586206e-05, | |
| "loss": 0.7002, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.844827586206897e-05, | |
| "loss": 0.6465, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.7586206896551727e-05, | |
| "loss": 0.7087, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 2.672413793103448e-05, | |
| "loss": 0.5199, | |
| "step": 783 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.5862068965517244e-05, | |
| "loss": 0.5276, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.486, | |
| "step": 841 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.042814229249011855, | |
| "eval_loss": 0.5168413519859314, | |
| "eval_runtime": 53945.1969, | |
| "eval_samples_per_second": 0.031, | |
| "eval_steps_per_second": 0.004, | |
| "step": 841 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 2.413793103448276e-05, | |
| "loss": 0.4464, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.327586206896552e-05, | |
| "loss": 0.4375, | |
| "step": 899 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.2413793103448276e-05, | |
| "loss": 0.411, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 2.1551724137931033e-05, | |
| "loss": 0.4487, | |
| "step": 957 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 2.0689655172413793e-05, | |
| "loss": 0.388, | |
| "step": 986 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.9827586206896554e-05, | |
| "loss": 0.4448, | |
| "step": 1015 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.896551724137931e-05, | |
| "loss": 0.5106, | |
| "step": 1044 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.810344827586207e-05, | |
| "loss": 0.3352, | |
| "step": 1073 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 0.3674, | |
| "step": 1102 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.6379310344827585e-05, | |
| "loss": 0.3654, | |
| "step": 1131 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 0.3391, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4655172413793103e-05, | |
| "loss": 0.3538, | |
| "step": 1189 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.3793103448275863e-05, | |
| "loss": 0.3327, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.2931034482758622e-05, | |
| "loss": 0.2937, | |
| "step": 1247 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.206896551724138e-05, | |
| "loss": 0.3375, | |
| "step": 1276 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.1206896551724138e-05, | |
| "loss": 0.3083, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.0344827586206897e-05, | |
| "loss": 0.2782, | |
| "step": 1334 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.482758620689655e-06, | |
| "loss": 0.2708, | |
| "step": 1363 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.620689655172414e-06, | |
| "loss": 0.2553, | |
| "step": 1392 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 7.758620689655173e-06, | |
| "loss": 0.2646, | |
| "step": 1421 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 6.896551724137932e-06, | |
| "loss": 0.2812, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 6.03448275862069e-06, | |
| "loss": 0.243, | |
| "step": 1479 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.172413793103448e-06, | |
| "loss": 0.2263, | |
| "step": 1508 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.310344827586207e-06, | |
| "loss": 0.2212, | |
| "step": 1537 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.448275862068966e-06, | |
| "loss": 0.2268, | |
| "step": 1566 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.586206896551724e-06, | |
| "loss": 0.2148, | |
| "step": 1595 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.724137931034483e-06, | |
| "loss": 0.2399, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 8.620689655172415e-07, | |
| "loss": 0.2214, | |
| "step": 1653 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.2187, | |
| "step": 1682 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.0036363636363636364, | |
| "eval_loss": 0.24321582913398743, | |
| "eval_runtime": 55499.6966, | |
| "eval_samples_per_second": 0.03, | |
| "eval_steps_per_second": 0.004, | |
| "step": 1682 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 1682, | |
| "total_flos": 1.9906356553640313e+19, | |
| "train_loss": 0.6093255166633234, | |
| "train_runtime": 257236.9869, | |
| "train_samples_per_second": 0.052, | |
| "train_steps_per_second": 0.007 | |
| } | |
| ], | |
| "max_steps": 1682, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.9906356553640313e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |