| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 1000, | |
| "global_step": 2250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 42.9058, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 21.7572, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.000215, | |
| "loss": 6.6528, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00029, | |
| "loss": 4.7079, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.000365, | |
| "loss": 4.6824, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00044, | |
| "loss": 4.6861, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0004976923076923077, | |
| "loss": 4.6781, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0004861538461538462, | |
| "loss": 4.657, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.00047461538461538463, | |
| "loss": 4.6171, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00046307692307692304, | |
| "loss": 4.5707, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 0.00045153846153846156, | |
| "loss": 4.4978, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.00044, | |
| "loss": 4.4697, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 0.0004284615384615385, | |
| "loss": 4.4129, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 0.0004169230769230769, | |
| "loss": 4.3792, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0004053846153846154, | |
| "loss": 4.3352, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.00039384615384615386, | |
| "loss": 4.3007, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 0.0003823076923076923, | |
| "loss": 4.2279, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 0.0003707692307692308, | |
| "loss": 4.1306, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 0.00035923076923076924, | |
| "loss": 3.9656, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0003476923076923077, | |
| "loss": 3.62, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 0.00033615384615384616, | |
| "loss": 3.0615, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 0.0003246153846153846, | |
| "loss": 2.5849, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "eval_cer": 0.5984864410342653, | |
| "eval_loss": 2.5845649242401123, | |
| "eval_runtime": 2.3789, | |
| "eval_samples_per_second": 74.826, | |
| "eval_steps_per_second": 5.044, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 0.0003130769230769231, | |
| "loss": 2.1788, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 0.00030153846153846154, | |
| "loss": 1.9325, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 0.00029, | |
| "loss": 1.7492, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 0.00027846153846153846, | |
| "loss": 1.6023, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 0.0002669230769230769, | |
| "loss": 1.4718, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 0.0002553846153846154, | |
| "loss": 1.3751, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 0.00024384615384615384, | |
| "loss": 1.2758, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.0002323076923076923, | |
| "loss": 1.2088, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 0.00022076923076923076, | |
| "loss": 1.1447, | |
| "step": 1395 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 0.00020923076923076922, | |
| "loss": 1.07, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 0.0001976923076923077, | |
| "loss": 1.0118, | |
| "step": 1485 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 0.00018615384615384615, | |
| "loss": 0.9947, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 0.00017461538461538463, | |
| "loss": 0.9557, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 0.00016307692307692307, | |
| "loss": 0.8984, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 0.00015153846153846155, | |
| "loss": 0.8694, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 0.8245, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 0.00012846153846153848, | |
| "loss": 0.7914, | |
| "step": 1755 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 0.00011692307692307692, | |
| "loss": 0.7883, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 0.00010538461538461538, | |
| "loss": 0.7628, | |
| "step": 1845 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 9.384615384615384e-05, | |
| "loss": 0.7449, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 8.23076923076923e-05, | |
| "loss": 0.7114, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 7.076923076923076e-05, | |
| "loss": 0.7224, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "eval_cer": 0.3664073996216103, | |
| "eval_loss": 1.5880260467529297, | |
| "eval_runtime": 2.3696, | |
| "eval_samples_per_second": 75.118, | |
| "eval_steps_per_second": 5.064, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 5.923076923076923e-05, | |
| "loss": 0.703, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 4.769230769230769e-05, | |
| "loss": 0.6685, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 3.615384615384615e-05, | |
| "loss": 0.6755, | |
| "step": 2115 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 2.4615384615384616e-05, | |
| "loss": 0.6411, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 1.3076923076923077e-05, | |
| "loss": 0.6266, | |
| "step": 2205 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.5384615384615385e-06, | |
| "loss": 0.6484, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 2250, | |
| "total_flos": 9.519369792641827e+18, | |
| "train_loss": 3.618984514872233, | |
| "train_runtime": 1635.0894, | |
| "train_samples_per_second": 43.637, | |
| "train_steps_per_second": 1.376 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2250, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 9.519369792641827e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |