| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 29.933333333333334, |
| "global_step": 210, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.4, |
| "learning_rate": 0.0003, |
| "loss": 145.7069, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_cer": 0.03821196827685652, |
| "eval_loss": 58.014591217041016, |
| "eval_runtime": 4.5419, |
| "eval_samples_per_second": 6.605, |
| "eval_steps_per_second": 0.881, |
| "eval_wer": 0.1686746987951807, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.8, |
| "learning_rate": 0.0003, |
| "loss": 44.761, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_cer": 0.0273972602739726, |
| "eval_loss": 39.7237663269043, |
| "eval_runtime": 4.6109, |
| "eval_samples_per_second": 6.506, |
| "eval_steps_per_second": 0.868, |
| "eval_wer": 0.10040160642570281, |
| "step": 20 |
| }, |
| { |
| "epoch": 4.27, |
| "learning_rate": 0.0003, |
| "loss": 20.1119, |
| "step": 30 |
| }, |
| { |
| "epoch": 4.27, |
| "eval_cer": 0.04830569574621485, |
| "eval_loss": 56.178184509277344, |
| "eval_runtime": 4.589, |
| "eval_samples_per_second": 6.537, |
| "eval_steps_per_second": 0.872, |
| "eval_wer": 0.11646586345381527, |
| "step": 30 |
| }, |
| { |
| "epoch": 5.67, |
| "learning_rate": 0.0003, |
| "loss": 21.7069, |
| "step": 40 |
| }, |
| { |
| "epoch": 5.67, |
| "eval_cer": 0.024513338139870222, |
| "eval_loss": 32.25881576538086, |
| "eval_runtime": 4.5716, |
| "eval_samples_per_second": 6.562, |
| "eval_steps_per_second": 0.875, |
| "eval_wer": 0.07228915662650602, |
| "step": 40 |
| }, |
| { |
| "epoch": 7.13, |
| "learning_rate": 0.0003, |
| "loss": 16.8501, |
| "step": 50 |
| }, |
| { |
| "epoch": 7.13, |
| "eval_cer": 0.020187454938716654, |
| "eval_loss": 26.216426849365234, |
| "eval_runtime": 5.4483, |
| "eval_samples_per_second": 5.506, |
| "eval_steps_per_second": 0.734, |
| "eval_wer": 0.0642570281124498, |
| "step": 50 |
| }, |
| { |
| "epoch": 8.53, |
| "learning_rate": 0.0003, |
| "loss": 16.3551, |
| "step": 60 |
| }, |
| { |
| "epoch": 8.53, |
| "eval_cer": 0.021629416005767843, |
| "eval_loss": 26.32578468322754, |
| "eval_runtime": 4.708, |
| "eval_samples_per_second": 6.372, |
| "eval_steps_per_second": 0.85, |
| "eval_wer": 0.0642570281124498, |
| "step": 60 |
| }, |
| { |
| "epoch": 9.93, |
| "learning_rate": 0.0003, |
| "loss": 13.2955, |
| "step": 70 |
| }, |
| { |
| "epoch": 9.93, |
| "eval_cer": 0.02090843547224225, |
| "eval_loss": 29.824462890625, |
| "eval_runtime": 4.6641, |
| "eval_samples_per_second": 6.432, |
| "eval_steps_per_second": 0.858, |
| "eval_wer": 0.07228915662650602, |
| "step": 70 |
| }, |
| { |
| "epoch": 11.4, |
| "learning_rate": 0.0003, |
| "loss": 10.6388, |
| "step": 80 |
| }, |
| { |
| "epoch": 11.4, |
| "eval_cer": 0.023792357606344627, |
| "eval_loss": 35.78471755981445, |
| "eval_runtime": 4.6396, |
| "eval_samples_per_second": 6.466, |
| "eval_steps_per_second": 0.862, |
| "eval_wer": 0.08032128514056225, |
| "step": 80 |
| }, |
| { |
| "epoch": 12.8, |
| "learning_rate": 0.0003, |
| "loss": 15.3866, |
| "step": 90 |
| }, |
| { |
| "epoch": 12.8, |
| "eval_cer": 0.028118240807498196, |
| "eval_loss": 48.54670333862305, |
| "eval_runtime": 4.711, |
| "eval_samples_per_second": 6.368, |
| "eval_steps_per_second": 0.849, |
| "eval_wer": 0.07228915662650602, |
| "step": 90 |
| }, |
| { |
| "epoch": 14.27, |
| "learning_rate": 0.0003, |
| "loss": 13.6768, |
| "step": 100 |
| }, |
| { |
| "epoch": 14.27, |
| "eval_cer": 0.021629416005767843, |
| "eval_loss": 39.71575927734375, |
| "eval_runtime": 4.6373, |
| "eval_samples_per_second": 6.469, |
| "eval_steps_per_second": 0.863, |
| "eval_wer": 0.06827309236947791, |
| "step": 100 |
| }, |
| { |
| "epoch": 15.67, |
| "learning_rate": 0.0003, |
| "loss": 9.2418, |
| "step": 110 |
| }, |
| { |
| "epoch": 15.67, |
| "eval_cer": 0.018745493871665464, |
| "eval_loss": 28.570499420166016, |
| "eval_runtime": 4.6852, |
| "eval_samples_per_second": 6.403, |
| "eval_steps_per_second": 0.854, |
| "eval_wer": 0.05622489959839357, |
| "step": 110 |
| }, |
| { |
| "epoch": 17.13, |
| "learning_rate": 0.0003, |
| "loss": 12.3437, |
| "step": 120 |
| }, |
| { |
| "epoch": 17.13, |
| "eval_cer": 0.023071377072819033, |
| "eval_loss": 34.8448486328125, |
| "eval_runtime": 4.6583, |
| "eval_samples_per_second": 6.44, |
| "eval_steps_per_second": 0.859, |
| "eval_wer": 0.07228915662650602, |
| "step": 120 |
| }, |
| { |
| "epoch": 18.53, |
| "learning_rate": 0.0003, |
| "loss": 7.2282, |
| "step": 130 |
| }, |
| { |
| "epoch": 18.53, |
| "eval_cer": 0.023071377072819033, |
| "eval_loss": 49.3336181640625, |
| "eval_runtime": 4.7123, |
| "eval_samples_per_second": 6.366, |
| "eval_steps_per_second": 0.849, |
| "eval_wer": 0.08032128514056225, |
| "step": 130 |
| }, |
| { |
| "epoch": 19.93, |
| "learning_rate": 0.0003, |
| "loss": 11.1077, |
| "step": 140 |
| }, |
| { |
| "epoch": 19.93, |
| "eval_cer": 0.022350396539293438, |
| "eval_loss": 46.2687873840332, |
| "eval_runtime": 4.6405, |
| "eval_samples_per_second": 6.465, |
| "eval_steps_per_second": 0.862, |
| "eval_wer": 0.0642570281124498, |
| "step": 140 |
| }, |
| { |
| "epoch": 21.4, |
| "learning_rate": 0.0003, |
| "loss": 11.6916, |
| "step": 150 |
| }, |
| { |
| "epoch": 21.4, |
| "eval_cer": 0.025234318673395817, |
| "eval_loss": 38.663455963134766, |
| "eval_runtime": 4.6616, |
| "eval_samples_per_second": 6.436, |
| "eval_steps_per_second": 0.858, |
| "eval_wer": 0.08835341365461848, |
| "step": 150 |
| }, |
| { |
| "epoch": 22.8, |
| "learning_rate": 0.0003, |
| "loss": 7.708, |
| "step": 160 |
| }, |
| { |
| "epoch": 22.8, |
| "eval_cer": 0.025234318673395817, |
| "eval_loss": 30.76764488220215, |
| "eval_runtime": 4.6877, |
| "eval_samples_per_second": 6.4, |
| "eval_steps_per_second": 0.853, |
| "eval_wer": 0.08032128514056225, |
| "step": 160 |
| }, |
| { |
| "epoch": 24.27, |
| "learning_rate": 0.0003, |
| "loss": 8.0417, |
| "step": 170 |
| }, |
| { |
| "epoch": 24.27, |
| "eval_cer": 0.024513338139870222, |
| "eval_loss": 32.57201385498047, |
| "eval_runtime": 4.6348, |
| "eval_samples_per_second": 6.473, |
| "eval_steps_per_second": 0.863, |
| "eval_wer": 0.060240963855421686, |
| "step": 170 |
| }, |
| { |
| "epoch": 25.67, |
| "learning_rate": 0.0003, |
| "loss": 7.4786, |
| "step": 180 |
| }, |
| { |
| "epoch": 25.67, |
| "eval_cer": 0.03749098774333093, |
| "eval_loss": 60.628662109375, |
| "eval_runtime": 4.6417, |
| "eval_samples_per_second": 6.463, |
| "eval_steps_per_second": 0.862, |
| "eval_wer": 0.10843373493975904, |
| "step": 180 |
| }, |
| { |
| "epoch": 27.13, |
| "learning_rate": 0.0003, |
| "loss": 10.2845, |
| "step": 190 |
| }, |
| { |
| "epoch": 27.13, |
| "eval_cer": 0.03316510454217736, |
| "eval_loss": 46.47388458251953, |
| "eval_runtime": 4.694, |
| "eval_samples_per_second": 6.391, |
| "eval_steps_per_second": 0.852, |
| "eval_wer": 0.12048192771084337, |
| "step": 190 |
| }, |
| { |
| "epoch": 28.53, |
| "learning_rate": 0.0003, |
| "loss": 10.9119, |
| "step": 200 |
| }, |
| { |
| "epoch": 28.53, |
| "eval_cer": 0.0273972602739726, |
| "eval_loss": 36.39248275756836, |
| "eval_runtime": 4.6689, |
| "eval_samples_per_second": 6.425, |
| "eval_steps_per_second": 0.857, |
| "eval_wer": 0.08433734939759036, |
| "step": 200 |
| }, |
| { |
| "epoch": 29.93, |
| "learning_rate": 0.0003, |
| "loss": 8.9316, |
| "step": 210 |
| }, |
| { |
| "epoch": 29.93, |
| "eval_cer": 0.024513338139870222, |
| "eval_loss": 30.038408279418945, |
| "eval_runtime": 4.6695, |
| "eval_samples_per_second": 6.425, |
| "eval_steps_per_second": 0.857, |
| "eval_wer": 0.06827309236947791, |
| "step": 210 |
| }, |
| { |
| "epoch": 29.93, |
| "step": 210, |
| "total_flos": 1.2113636824678656e+18, |
| "train_loss": 20.16470682053339, |
| "train_runtime": 1180.2334, |
| "train_samples_per_second": 3.05, |
| "train_steps_per_second": 0.178 |
| } |
| ], |
| "max_steps": 210, |
| "num_train_epochs": 30, |
| "total_flos": 1.2113636824678656e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
