| { | |
| "best_metric": 0.8710034489631653, | |
| "best_model_checkpoint": "./checkpoint-400", | |
| "epoch": 12.5, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.3709, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 2.8912, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 2.2169, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 1.6923, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 9.407407407407408e-06, | |
| "loss": 1.3934, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 9.037037037037037e-06, | |
| "loss": 1.2336, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 1.1522, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.296296296296297e-06, | |
| "loss": 1.1339, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.925925925925926e-06, | |
| "loss": 0.9576, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.9153, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 1.0240015983581543, | |
| "eval_runtime": 182.7755, | |
| "eval_samples_per_second": 2.801, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 68.9863608183509, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 7.185185185185186e-06, | |
| "loss": 0.8884, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 6.814814814814815e-06, | |
| "loss": 0.8334, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 6.444444444444445e-06, | |
| "loss": 0.7989, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.0740740740740745e-06, | |
| "loss": 0.7718, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 5.7037037037037045e-06, | |
| "loss": 0.7671, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.8208, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 4.962962962962964e-06, | |
| "loss": 0.7219, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 4.592592592592593e-06, | |
| "loss": 0.7034, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 4.222222222222223e-06, | |
| "loss": 0.6966, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.851851851851852e-06, | |
| "loss": 0.6865, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8967972993850708, | |
| "eval_runtime": 172.9658, | |
| "eval_samples_per_second": 2.96, | |
| "eval_steps_per_second": 0.093, | |
| "eval_wer": 61.7660411622276, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 6.255319148936171e-06, | |
| "loss": 0.6414, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 6.042553191489362e-06, | |
| "loss": 0.6552, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 5.829787234042553e-06, | |
| "loss": 0.6445, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 5.617021276595746e-06, | |
| "loss": 0.622, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 5.404255319148937e-06, | |
| "loss": 0.6004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 5.191489361702128e-06, | |
| "loss": 0.5882, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 4.9787234042553195e-06, | |
| "loss": 0.6016, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.765957446808511e-06, | |
| "loss": 0.5682, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 4.553191489361702e-06, | |
| "loss": 0.5672, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 4.340425531914894e-06, | |
| "loss": 0.5474, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_loss": 0.8744030594825745, | |
| "eval_runtime": 190.8413, | |
| "eval_samples_per_second": 2.683, | |
| "eval_steps_per_second": 0.084, | |
| "eval_wer": 60.55538740920097, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 4.127659574468085e-06, | |
| "loss": 0.5432, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.914893617021277e-06, | |
| "loss": 0.534, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 3.702127659574468e-06, | |
| "loss": 0.5164, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 3.48936170212766e-06, | |
| "loss": 0.5223, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 3.276595744680851e-06, | |
| "loss": 0.5126, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.0638297872340428e-06, | |
| "loss": 0.5059, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 2.8510638297872346e-06, | |
| "loss": 0.4958, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.6382978723404256e-06, | |
| "loss": 0.4928, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 2.425531914893617e-06, | |
| "loss": 0.4856, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.2127659574468085e-06, | |
| "loss": 0.4646, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8710034489631653, | |
| "eval_runtime": 173.1105, | |
| "eval_samples_per_second": 2.958, | |
| "eval_steps_per_second": 0.092, | |
| "eval_wer": 60.05599273607748, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.4711, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 1.7872340425531918e-06, | |
| "loss": 0.472, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 1.5744680851063832e-06, | |
| "loss": 0.4745, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 1.3617021276595746e-06, | |
| "loss": 0.4546, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 1.148936170212766e-06, | |
| "loss": 0.4702, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 9.361702127659575e-07, | |
| "loss": 0.448, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.75, | |
| "learning_rate": 7.234042553191489e-07, | |
| "loss": 0.4584, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 5.106382978723404e-07, | |
| "loss": 0.4719, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.25, | |
| "learning_rate": 2.9787234042553196e-07, | |
| "loss": 0.45, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 8.510638297872341e-08, | |
| "loss": 0.4557, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 0.873198390007019, | |
| "eval_runtime": 187.2044, | |
| "eval_samples_per_second": 2.735, | |
| "eval_steps_per_second": 0.085, | |
| "eval_wer": 59.465799031477, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "step": 500, | |
| "total_flos": 2.02648572002304e+18, | |
| "train_loss": 0.31471561336517334, | |
| "train_runtime": 1275.0957, | |
| "train_samples_per_second": 25.096, | |
| "train_steps_per_second": 0.392 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 13, | |
| "total_flos": 2.02648572002304e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |