| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.6273393002441008, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.4999999999999999e-05, | |
| "loss": 11.8163, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 11.8929, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.3499999999999993e-05, | |
| "loss": 12.0064, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 10.5132, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.199999999999999e-05, | |
| "loss": 7.9699, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 8.699999999999999e-05, | |
| "loss": 4.7703, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.000102, | |
| "loss": 3.7982, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000117, | |
| "loss": 3.436, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00013199999999999998, | |
| "loss": 3.2573, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.000147, | |
| "loss": 3.1412, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 3.0855324268341064, | |
| "eval_runtime": 1038.5545, | |
| "eval_samples_per_second": 11.178, | |
| "eval_steps_per_second": 0.699, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.000162, | |
| "loss": 3.0577, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00017699999999999997, | |
| "loss": 3.0443, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00019199999999999998, | |
| "loss": 3.0597, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00020699999999999996, | |
| "loss": 3.061, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00022199999999999998, | |
| "loss": 3.0717, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.000237, | |
| "loss": 3.0299, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00025199999999999995, | |
| "loss": 3.0295, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000267, | |
| "loss": 3.0423, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00028199999999999997, | |
| "loss": 3.0437, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 3.0569, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 3.036912679672241, | |
| "eval_runtime": 1030.0275, | |
| "eval_samples_per_second": 11.271, | |
| "eval_steps_per_second": 0.705, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00029973362930077685, | |
| "loss": 3.0243, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.000299400665926748, | |
| "loss": 3.0236, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0002990677025527192, | |
| "loss": 3.0316, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0002987347391786903, | |
| "loss": 3.0363, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029840177580466146, | |
| "loss": 3.0582, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002980688124306326, | |
| "loss": 3.0144, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00029773584905660376, | |
| "loss": 3.0126, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002974028856825749, | |
| "loss": 3.0035, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.000297069922308546, | |
| "loss": 2.9913, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00029673695893451716, | |
| "loss": 2.9625, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 2.977771520614624, | |
| "eval_runtime": 1042.3792, | |
| "eval_samples_per_second": 11.137, | |
| "eval_steps_per_second": 0.696, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002964039955604883, | |
| "loss": 2.9686, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00029607103218645946, | |
| "loss": 2.8773, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002957380688124306, | |
| "loss": 2.726, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00029540510543840177, | |
| "loss": 2.3671, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002950721420643729, | |
| "loss": 1.9479, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.000294739178690344, | |
| "loss": 1.5549, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00029440621531631517, | |
| "loss": 1.1765, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002940732519422863, | |
| "loss": 0.9155, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00029374028856825747, | |
| "loss": 0.7823, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002934073251942286, | |
| "loss": 0.7715, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.5112669467926025, | |
| "eval_runtime": 1048.8544, | |
| "eval_samples_per_second": 11.068, | |
| "eval_steps_per_second": 0.692, | |
| "eval_wer": 0.718480591159969, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00029307436182019977, | |
| "loss": 0.6635, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002927413984461709, | |
| "loss": 0.6032, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.000292408435072142, | |
| "loss": 0.5401, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00029207547169811317, | |
| "loss": 0.5276, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002917425083240843, | |
| "loss": 0.5713, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0002914095449500555, | |
| "loss": 0.46, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0002910765815760266, | |
| "loss": 0.4508, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002907436182019977, | |
| "loss": 0.447, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00029041065482796893, | |
| "loss": 0.4454, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0002900776914539401, | |
| "loss": 0.4725, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.3072386682033539, | |
| "eval_runtime": 1051.5656, | |
| "eval_samples_per_second": 11.04, | |
| "eval_steps_per_second": 0.69, | |
| "eval_wer": 0.5137640833827487, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0002897447280799112, | |
| "loss": 0.3822, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00028941176470588233, | |
| "loss": 0.3807, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002890788013318535, | |
| "loss": 0.3827, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00028874583795782463, | |
| "loss": 0.3945, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00028841287458379573, | |
| "loss": 0.4519, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002880799112097669, | |
| "loss": 0.3534, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00028774694783573803, | |
| "loss": 0.3731, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002874139844617092, | |
| "loss": 0.3433, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00028708102108768033, | |
| "loss": 0.3547, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002867480577136515, | |
| "loss": 0.4103, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 0.24468904733657837, | |
| "eval_runtime": 1043.2136, | |
| "eval_samples_per_second": 11.128, | |
| "eval_steps_per_second": 0.696, | |
| "eval_wer": 0.4336997673676048, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00028641509433962264, | |
| "loss": 0.3484, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00028608213096559373, | |
| "loss": 0.3775, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.0002857491675915649, | |
| "loss": 0.3207, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00028541620421753604, | |
| "loss": 0.2971, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0002850832408435072, | |
| "loss": 0.3129, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00028475027746947834, | |
| "loss": 0.3143, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002844173140954495, | |
| "loss": 0.3179, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00028408435072142064, | |
| "loss": 0.299, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0002837513873473918, | |
| "loss": 0.2734, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002834184239733629, | |
| "loss": 0.2775, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.2055242359638214, | |
| "eval_runtime": 1053.2634, | |
| "eval_samples_per_second": 11.022, | |
| "eval_steps_per_second": 0.689, | |
| "eval_wer": 0.3768644802262464, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00028308546059933404, | |
| "loss": 0.2878, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002827524972253052, | |
| "loss": 0.307, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00028241953385127634, | |
| "loss": 0.2848, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00028208657047724744, | |
| "loss": 0.2744, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00028175360710321865, | |
| "loss": 0.2839, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002814206437291898, | |
| "loss": 0.2888, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0002810876803551609, | |
| "loss": 0.288, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00028075471698113205, | |
| "loss": 0.2559, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0002804217536071032, | |
| "loss": 0.2498, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00028008879023307435, | |
| "loss": 0.2554, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.19498416781425476, | |
| "eval_runtime": 1045.8615, | |
| "eval_samples_per_second": 11.1, | |
| "eval_steps_per_second": 0.694, | |
| "eval_wer": 0.360272316744971, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00027975582685904545, | |
| "loss": 0.3059, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002794228634850166, | |
| "loss": 0.281, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00027908990011098775, | |
| "loss": 0.2642, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0002787569367369589, | |
| "loss": 0.242, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00027842397336293005, | |
| "loss": 0.2541, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0002780910099889012, | |
| "loss": 0.2614, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00027775804661487236, | |
| "loss": 0.3001, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0002774250832408435, | |
| "loss": 0.2365, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0002770921198668146, | |
| "loss": 0.2373, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00027675915649278575, | |
| "loss": 0.263, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 0.1812964379787445, | |
| "eval_runtime": 1053.0871, | |
| "eval_samples_per_second": 11.024, | |
| "eval_steps_per_second": 0.689, | |
| "eval_wer": 0.3371687269078137, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0002764261931187569, | |
| "loss": 0.289, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00027609322974472806, | |
| "loss": 0.2578, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.0002757602663706992, | |
| "loss": 0.2565, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00027542730299667036, | |
| "loss": 0.2543, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0002750943396226415, | |
| "loss": 0.2404, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002747613762486126, | |
| "loss": 0.2291, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00027442841287458376, | |
| "loss": 0.2636, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0002740954495005549, | |
| "loss": 0.2222, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00027376248612652606, | |
| "loss": 0.2322, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0002734295227524972, | |
| "loss": 0.2294, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 0.1664419174194336, | |
| "eval_runtime": 1047.59, | |
| "eval_samples_per_second": 11.082, | |
| "eval_steps_per_second": 0.693, | |
| "eval_wer": 0.3131756602654746, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 9210, | |
| "num_train_epochs": 15, | |
| "save_steps": 200, | |
| "total_flos": 1.0531620074605271e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |