| { | |
| "best_metric": 0.4305761158466339, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-2680", | |
| "epoch": 1.9734904270986746, | |
| "global_step": 2680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00019926362297496318, | |
| "loss": 2.073, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00019852724594992638, | |
| "loss": 1.5903, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00019779086892488955, | |
| "loss": 1.4651, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019705449189985275, | |
| "loss": 1.2028, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.5852670349907919, | |
| "eval_loss": 1.2089080810546875, | |
| "eval_runtime": 48.3873, | |
| "eval_samples_per_second": 56.11, | |
| "eval_steps_per_second": 7.027, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019631811487481592, | |
| "loss": 1.1103, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0001955817378497791, | |
| "loss": 1.2374, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019484536082474229, | |
| "loss": 1.1704, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019410898379970546, | |
| "loss": 1.0396, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.6504604051565378, | |
| "eval_loss": 1.0024482011795044, | |
| "eval_runtime": 49.385, | |
| "eval_samples_per_second": 54.976, | |
| "eval_steps_per_second": 6.885, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019337260677466863, | |
| "loss": 0.9898, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019263622974963183, | |
| "loss": 0.9049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.000191899852724595, | |
| "loss": 1.0411, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0001911634756995582, | |
| "loss": 1.031, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.6088397790055249, | |
| "eval_loss": 1.0570313930511475, | |
| "eval_runtime": 52.8083, | |
| "eval_samples_per_second": 51.412, | |
| "eval_steps_per_second": 6.438, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019042709867452136, | |
| "loss": 0.9189, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00018969072164948454, | |
| "loss": 1.0785, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00018895434462444773, | |
| "loss": 0.9283, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0001882179675994109, | |
| "loss": 0.94, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.6386740331491713, | |
| "eval_loss": 0.9433544278144836, | |
| "eval_runtime": 51.6666, | |
| "eval_samples_per_second": 52.548, | |
| "eval_steps_per_second": 6.581, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00018748159057437408, | |
| "loss": 1.0078, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00018674521354933727, | |
| "loss": 0.9582, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00018600883652430044, | |
| "loss": 0.9966, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00018527245949926364, | |
| "loss": 0.8435, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.69097605893186, | |
| "eval_loss": 0.8427824378013611, | |
| "eval_runtime": 51.4559, | |
| "eval_samples_per_second": 52.764, | |
| "eval_steps_per_second": 6.608, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001845360824742268, | |
| "loss": 0.7854, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00018379970544918998, | |
| "loss": 0.9786, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00018306332842415318, | |
| "loss": 0.736, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00018232695139911635, | |
| "loss": 0.851, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.6504604051565378, | |
| "eval_loss": 0.9459872841835022, | |
| "eval_runtime": 51.2583, | |
| "eval_samples_per_second": 52.967, | |
| "eval_steps_per_second": 6.633, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00018159057437407955, | |
| "loss": 0.9344, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00018085419734904272, | |
| "loss": 0.8802, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001801178203240059, | |
| "loss": 0.772, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0001793814432989691, | |
| "loss": 0.9324, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_accuracy": 0.6593001841620626, | |
| "eval_loss": 0.9168590903282166, | |
| "eval_runtime": 52.8324, | |
| "eval_samples_per_second": 51.389, | |
| "eval_steps_per_second": 6.435, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00017864506627393226, | |
| "loss": 0.6804, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00017790868924889543, | |
| "loss": 0.7877, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00017717231222385863, | |
| "loss": 0.9609, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0001764359351988218, | |
| "loss": 0.8963, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.7016574585635359, | |
| "eval_loss": 0.7979860901832581, | |
| "eval_runtime": 50.8385, | |
| "eval_samples_per_second": 53.404, | |
| "eval_steps_per_second": 6.688, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.000175699558173785, | |
| "loss": 0.8531, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00017496318114874817, | |
| "loss": 0.899, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00017422680412371134, | |
| "loss": 0.816, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00017349042709867454, | |
| "loss": 0.7532, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.7031307550644568, | |
| "eval_loss": 0.8063194155693054, | |
| "eval_runtime": 50.4103, | |
| "eval_samples_per_second": 53.858, | |
| "eval_steps_per_second": 6.745, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0001727540500736377, | |
| "loss": 0.8984, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0001720176730486009, | |
| "loss": 0.843, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00017128129602356408, | |
| "loss": 0.7647, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00017054491899852725, | |
| "loss": 0.7937, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.7222836095764272, | |
| "eval_loss": 0.7624691724777222, | |
| "eval_runtime": 51.1567, | |
| "eval_samples_per_second": 53.072, | |
| "eval_steps_per_second": 6.646, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00016980854197349044, | |
| "loss": 0.9007, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00016907216494845361, | |
| "loss": 0.7149, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00016833578792341679, | |
| "loss": 0.8868, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00016759941089837998, | |
| "loss": 0.8365, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.7134438305709024, | |
| "eval_loss": 0.7824127078056335, | |
| "eval_runtime": 50.4947, | |
| "eval_samples_per_second": 53.768, | |
| "eval_steps_per_second": 6.733, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00016686303387334315, | |
| "loss": 0.7985, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00016612665684830635, | |
| "loss": 0.775, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00016539027982326952, | |
| "loss": 0.7912, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0001646539027982327, | |
| "loss": 0.8333, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_accuracy": 0.7263351749539595, | |
| "eval_loss": 0.7473925948143005, | |
| "eval_runtime": 51.4866, | |
| "eval_samples_per_second": 52.732, | |
| "eval_steps_per_second": 6.604, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0001639175257731959, | |
| "loss": 0.8501, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00016318114874815906, | |
| "loss": 0.6558, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00016244477172312223, | |
| "loss": 0.5491, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00016170839469808543, | |
| "loss": 0.6784, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_accuracy": 0.7351749539594844, | |
| "eval_loss": 0.7210445404052734, | |
| "eval_runtime": 51.0697, | |
| "eval_samples_per_second": 53.163, | |
| "eval_steps_per_second": 6.658, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0001609720176730486, | |
| "loss": 0.6781, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001602356406480118, | |
| "loss": 0.7243, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00015949926362297497, | |
| "loss": 0.664, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00015876288659793814, | |
| "loss": 0.7103, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.7204419889502762, | |
| "eval_loss": 0.7556896805763245, | |
| "eval_runtime": 51.9028, | |
| "eval_samples_per_second": 52.309, | |
| "eval_steps_per_second": 6.551, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00015802650957290134, | |
| "loss": 0.7232, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001572901325478645, | |
| "loss": 0.7658, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001565537555228277, | |
| "loss": 0.7853, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00015581737849779088, | |
| "loss": 0.6149, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.7484346224677716, | |
| "eval_loss": 0.6908882260322571, | |
| "eval_runtime": 50.7606, | |
| "eval_samples_per_second": 53.486, | |
| "eval_steps_per_second": 6.698, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00015508100147275405, | |
| "loss": 0.7379, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00015434462444771725, | |
| "loss": 0.7758, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00015360824742268042, | |
| "loss": 0.6333, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0001528718703976436, | |
| "loss": 0.5998, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.7322283609576428, | |
| "eval_loss": 0.7079024314880371, | |
| "eval_runtime": 50.4304, | |
| "eval_samples_per_second": 53.837, | |
| "eval_steps_per_second": 6.742, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00015213549337260679, | |
| "loss": 0.6485, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00015139911634756996, | |
| "loss": 0.8297, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00015066273932253315, | |
| "loss": 0.8158, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00014992636229749633, | |
| "loss": 0.8814, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.6744014732965009, | |
| "eval_loss": 0.922738254070282, | |
| "eval_runtime": 50.6437, | |
| "eval_samples_per_second": 53.61, | |
| "eval_steps_per_second": 6.714, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0001491899852724595, | |
| "loss": 0.764, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0001484536082474227, | |
| "loss": 0.6825, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00014771723122238586, | |
| "loss": 0.6629, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00014698085419734906, | |
| "loss": 0.7405, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.7454880294659301, | |
| "eval_loss": 0.6923764944076538, | |
| "eval_runtime": 50.452, | |
| "eval_samples_per_second": 53.814, | |
| "eval_steps_per_second": 6.739, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00014624447717231223, | |
| "loss": 0.8144, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001455081001472754, | |
| "loss": 0.5245, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0001447717231222386, | |
| "loss": 0.7705, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00014403534609720177, | |
| "loss": 0.7463, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.7182320441988951, | |
| "eval_loss": 0.7563807368278503, | |
| "eval_runtime": 50.5699, | |
| "eval_samples_per_second": 53.688, | |
| "eval_steps_per_second": 6.723, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00014329896907216494, | |
| "loss": 0.7648, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00014256259204712814, | |
| "loss": 0.6192, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001418262150220913, | |
| "loss": 0.872, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0001410898379970545, | |
| "loss": 0.684, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.7337016574585635, | |
| "eval_loss": 0.7226435542106628, | |
| "eval_runtime": 50.2491, | |
| "eval_samples_per_second": 54.031, | |
| "eval_steps_per_second": 6.766, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00014035346097201768, | |
| "loss": 0.7713, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00013961708394698085, | |
| "loss": 0.6673, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00013888070692194405, | |
| "loss": 0.6792, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00013814432989690722, | |
| "loss": 0.6252, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.7524861878453039, | |
| "eval_loss": 0.7024006247520447, | |
| "eval_runtime": 50.1454, | |
| "eval_samples_per_second": 54.143, | |
| "eval_steps_per_second": 6.78, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0001374079528718704, | |
| "loss": 0.7425, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0001366715758468336, | |
| "loss": 0.7285, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00013593519882179676, | |
| "loss": 0.7498, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00013519882179675996, | |
| "loss": 0.754, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.7274401473296501, | |
| "eval_loss": 0.6939438581466675, | |
| "eval_runtime": 51.2872, | |
| "eval_samples_per_second": 52.937, | |
| "eval_steps_per_second": 6.629, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00013446244477172313, | |
| "loss": 0.6756, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001337260677466863, | |
| "loss": 0.636, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001329896907216495, | |
| "loss": 0.5823, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013225331369661267, | |
| "loss": 0.6416, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.7565377532228361, | |
| "eval_loss": 0.6688955426216125, | |
| "eval_runtime": 51.1119, | |
| "eval_samples_per_second": 53.119, | |
| "eval_steps_per_second": 6.652, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013151693667157587, | |
| "loss": 0.6125, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00013078055964653904, | |
| "loss": 0.687, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001300441826215022, | |
| "loss": 0.7283, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001293078055964654, | |
| "loss": 0.7123, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.7558011049723757, | |
| "eval_loss": 0.6540634632110596, | |
| "eval_runtime": 50.4305, | |
| "eval_samples_per_second": 53.836, | |
| "eval_steps_per_second": 6.742, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00012857142857142858, | |
| "loss": 0.6579, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00012783505154639175, | |
| "loss": 0.6744, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00012709867452135494, | |
| "loss": 0.619, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00012636229749631812, | |
| "loss": 0.6599, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.7683241252302025, | |
| "eval_loss": 0.6279534101486206, | |
| "eval_runtime": 50.5314, | |
| "eval_samples_per_second": 53.729, | |
| "eval_steps_per_second": 6.728, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0001256259204712813, | |
| "loss": 0.5407, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00012488954344624448, | |
| "loss": 0.7339, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012415316642120765, | |
| "loss": 0.65, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00012341678939617085, | |
| "loss": 0.7664, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.7521178637200736, | |
| "eval_loss": 0.6760364770889282, | |
| "eval_runtime": 51.3273, | |
| "eval_samples_per_second": 52.896, | |
| "eval_steps_per_second": 6.624, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00012268041237113402, | |
| "loss": 0.667, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00012194403534609722, | |
| "loss": 0.6393, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00012120765832106038, | |
| "loss": 0.7258, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00012047128129602356, | |
| "loss": 0.6898, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.7815837937384899, | |
| "eval_loss": 0.6094613075256348, | |
| "eval_runtime": 50.3809, | |
| "eval_samples_per_second": 53.89, | |
| "eval_steps_per_second": 6.749, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00011973490427098675, | |
| "loss": 0.6158, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00011899852724594993, | |
| "loss": 0.5764, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00011826215022091312, | |
| "loss": 0.6926, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0001175257731958763, | |
| "loss": 0.693, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.7543278084714549, | |
| "eval_loss": 0.6533992290496826, | |
| "eval_runtime": 51.5963, | |
| "eval_samples_per_second": 52.62, | |
| "eval_steps_per_second": 6.59, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00011678939617083948, | |
| "loss": 0.7522, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00011605301914580267, | |
| "loss": 0.451, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00011531664212076583, | |
| "loss": 0.564, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00011458026509572901, | |
| "loss": 0.6738, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.7826887661141805, | |
| "eval_loss": 0.6036040186882019, | |
| "eval_runtime": 50.9403, | |
| "eval_samples_per_second": 53.298, | |
| "eval_steps_per_second": 6.674, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0001138438880706922, | |
| "loss": 0.6272, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00011310751104565538, | |
| "loss": 0.5538, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00011237113402061856, | |
| "loss": 0.555, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00011163475699558175, | |
| "loss": 0.6056, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.7874769797421731, | |
| "eval_loss": 0.584153950214386, | |
| "eval_runtime": 51.1472, | |
| "eval_samples_per_second": 53.082, | |
| "eval_steps_per_second": 6.647, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00011089837997054493, | |
| "loss": 0.5609, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00011016200294550812, | |
| "loss": 0.6141, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0001094256259204713, | |
| "loss": 0.6212, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00010868924889543446, | |
| "loss": 0.746, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.7675874769797422, | |
| "eval_loss": 0.6469867825508118, | |
| "eval_runtime": 50.9954, | |
| "eval_samples_per_second": 53.24, | |
| "eval_steps_per_second": 6.667, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00010795287187039764, | |
| "loss": 0.5153, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00010721649484536083, | |
| "loss": 0.6638, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00010648011782032401, | |
| "loss": 0.6719, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0001057437407952872, | |
| "loss": 0.7612, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.7797421731123388, | |
| "eval_loss": 0.613595724105835, | |
| "eval_runtime": 51.0708, | |
| "eval_samples_per_second": 53.161, | |
| "eval_steps_per_second": 6.657, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00010500736377025038, | |
| "loss": 0.6873, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00010427098674521356, | |
| "loss": 0.5534, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00010353460972017675, | |
| "loss": 0.3849, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0001027982326951399, | |
| "loss": 0.5939, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.7952117863720074, | |
| "eval_loss": 0.5709212422370911, | |
| "eval_runtime": 52.4056, | |
| "eval_samples_per_second": 51.807, | |
| "eval_steps_per_second": 6.488, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00010206185567010309, | |
| "loss": 0.5831, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00010132547864506627, | |
| "loss": 0.6712, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00010058910162002946, | |
| "loss": 0.5368, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.985272459499264e-05, | |
| "loss": 0.5586, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7701657458563536, | |
| "eval_loss": 0.6024866104125977, | |
| "eval_runtime": 50.9747, | |
| "eval_samples_per_second": 53.262, | |
| "eval_steps_per_second": 6.67, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.911634756995583e-05, | |
| "loss": 0.4372, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.8379970544919e-05, | |
| "loss": 0.4076, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.764359351988218e-05, | |
| "loss": 0.3933, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.690721649484537e-05, | |
| "loss": 0.459, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_accuracy": 0.7675874769797422, | |
| "eval_loss": 0.6382798552513123, | |
| "eval_runtime": 51.4385, | |
| "eval_samples_per_second": 52.782, | |
| "eval_steps_per_second": 6.61, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.617083946980855e-05, | |
| "loss": 0.3769, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.543446244477173e-05, | |
| "loss": 0.411, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.46980854197349e-05, | |
| "loss": 0.4444, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.396170839469809e-05, | |
| "loss": 0.3771, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_accuracy": 0.7856353591160221, | |
| "eval_loss": 0.5789190530776978, | |
| "eval_runtime": 50.7253, | |
| "eval_samples_per_second": 53.524, | |
| "eval_steps_per_second": 6.703, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.322533136966127e-05, | |
| "loss": 0.3943, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 9.248895434462446e-05, | |
| "loss": 0.5164, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 9.175257731958763e-05, | |
| "loss": 0.3544, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.101620029455081e-05, | |
| "loss": 0.4272, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.8044198895027624, | |
| "eval_loss": 0.5364444851875305, | |
| "eval_runtime": 52.5287, | |
| "eval_samples_per_second": 51.686, | |
| "eval_steps_per_second": 6.473, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 9.0279823269514e-05, | |
| "loss": 0.3962, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.954344624447718e-05, | |
| "loss": 0.434, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.880706921944035e-05, | |
| "loss": 0.4332, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 8.807069219440354e-05, | |
| "loss": 0.3229, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.8044198895027624, | |
| "eval_loss": 0.5452917814254761, | |
| "eval_runtime": 51.1816, | |
| "eval_samples_per_second": 53.046, | |
| "eval_steps_per_second": 6.643, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 8.733431516936672e-05, | |
| "loss": 0.4162, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 8.65979381443299e-05, | |
| "loss": 0.3704, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.586156111929308e-05, | |
| "loss": 0.6147, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 8.512518409425626e-05, | |
| "loss": 0.3736, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.8092081031307551, | |
| "eval_loss": 0.5403669476509094, | |
| "eval_runtime": 50.5773, | |
| "eval_samples_per_second": 53.68, | |
| "eval_steps_per_second": 6.722, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 8.438880706921944e-05, | |
| "loss": 0.4865, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 8.365243004418263e-05, | |
| "loss": 0.3929, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.291605301914581e-05, | |
| "loss": 0.3426, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 8.217967599410898e-05, | |
| "loss": 0.436, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.8081031307550645, | |
| "eval_loss": 0.5359194278717041, | |
| "eval_runtime": 51.3206, | |
| "eval_samples_per_second": 52.903, | |
| "eval_steps_per_second": 6.625, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 8.144329896907217e-05, | |
| "loss": 0.4115, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 8.070692194403535e-05, | |
| "loss": 0.3784, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.997054491899854e-05, | |
| "loss": 0.3383, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 7.923416789396171e-05, | |
| "loss": 0.3721, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_accuracy": 0.7815837937384899, | |
| "eval_loss": 0.6128984689712524, | |
| "eval_runtime": 51.0569, | |
| "eval_samples_per_second": 53.176, | |
| "eval_steps_per_second": 6.659, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.849779086892489e-05, | |
| "loss": 0.4844, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.776141384388808e-05, | |
| "loss": 0.4011, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.702503681885126e-05, | |
| "loss": 0.4711, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.628865979381443e-05, | |
| "loss": 0.3848, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_accuracy": 0.7985267034990792, | |
| "eval_loss": 0.5785150527954102, | |
| "eval_runtime": 51.7772, | |
| "eval_samples_per_second": 52.436, | |
| "eval_steps_per_second": 6.567, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.555228276877762e-05, | |
| "loss": 0.3771, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.48159057437408e-05, | |
| "loss": 0.5486, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 7.407952871870398e-05, | |
| "loss": 0.3795, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 7.334315169366715e-05, | |
| "loss": 0.393, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_accuracy": 0.8047882136279927, | |
| "eval_loss": 0.5534458756446838, | |
| "eval_runtime": 50.9572, | |
| "eval_samples_per_second": 53.28, | |
| "eval_steps_per_second": 6.672, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 7.260677466863034e-05, | |
| "loss": 0.4113, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.187039764359352e-05, | |
| "loss": 0.3919, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.113402061855671e-05, | |
| "loss": 0.5448, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 7.039764359351989e-05, | |
| "loss": 0.3676, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.7952117863720074, | |
| "eval_loss": 0.5770056247711182, | |
| "eval_runtime": 50.6386, | |
| "eval_samples_per_second": 53.615, | |
| "eval_steps_per_second": 6.714, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 6.966126656848306e-05, | |
| "loss": 0.4213, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 6.892488954344625e-05, | |
| "loss": 0.5673, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 6.818851251840943e-05, | |
| "loss": 0.3463, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.745213549337262e-05, | |
| "loss": 0.4531, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_accuracy": 0.7922651933701658, | |
| "eval_loss": 0.5833007097244263, | |
| "eval_runtime": 51.1166, | |
| "eval_samples_per_second": 53.114, | |
| "eval_steps_per_second": 6.651, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.671575846833579e-05, | |
| "loss": 0.3818, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 6.597938144329897e-05, | |
| "loss": 0.4865, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 6.524300441826216e-05, | |
| "loss": 0.4752, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 6.450662739322534e-05, | |
| "loss": 0.4353, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_accuracy": 0.8136279926335175, | |
| "eval_loss": 0.5133872628211975, | |
| "eval_runtime": 52.3565, | |
| "eval_samples_per_second": 51.856, | |
| "eval_steps_per_second": 6.494, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 6.377025036818851e-05, | |
| "loss": 0.3349, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 6.30338733431517e-05, | |
| "loss": 0.2902, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 6.229749631811488e-05, | |
| "loss": 0.5968, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 6.156111929307806e-05, | |
| "loss": 0.422, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_accuracy": 0.8106813996316758, | |
| "eval_loss": 0.521941065788269, | |
| "eval_runtime": 51.8454, | |
| "eval_samples_per_second": 52.367, | |
| "eval_steps_per_second": 6.558, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.0824742268041234e-05, | |
| "loss": 0.3134, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 6.008836524300442e-05, | |
| "loss": 0.5106, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.93519882179676e-05, | |
| "loss": 0.3672, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.861561119293079e-05, | |
| "loss": 0.4875, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.823572744014733, | |
| "eval_loss": 0.4959632158279419, | |
| "eval_runtime": 51.2335, | |
| "eval_samples_per_second": 52.993, | |
| "eval_steps_per_second": 6.636, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 5.787923416789397e-05, | |
| "loss": 0.3232, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 0.3946, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.6406480117820326e-05, | |
| "loss": 0.368, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.567010309278351e-05, | |
| "loss": 0.2695, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_accuracy": 0.8283609576427255, | |
| "eval_loss": 0.4938356876373291, | |
| "eval_runtime": 51.5366, | |
| "eval_samples_per_second": 52.681, | |
| "eval_steps_per_second": 6.597, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5.4933726067746695e-05, | |
| "loss": 0.3932, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 5.4197349042709865e-05, | |
| "loss": 0.3572, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.346097201767305e-05, | |
| "loss": 0.5021, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.2724594992636234e-05, | |
| "loss": 0.3038, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.8276243093922652, | |
| "eval_loss": 0.48587676882743835, | |
| "eval_runtime": 52.2942, | |
| "eval_samples_per_second": 51.918, | |
| "eval_steps_per_second": 6.502, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 5.198821796759942e-05, | |
| "loss": 0.3817, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.125184094256259e-05, | |
| "loss": 0.3844, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.051546391752577e-05, | |
| "loss": 0.3217, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.977908689248896e-05, | |
| "loss": 0.4076, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.825414364640884, | |
| "eval_loss": 0.4927995204925537, | |
| "eval_runtime": 50.7363, | |
| "eval_samples_per_second": 53.512, | |
| "eval_steps_per_second": 6.701, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.9042709867452135e-05, | |
| "loss": 0.431, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 4.830633284241532e-05, | |
| "loss": 0.3252, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 4.75699558173785e-05, | |
| "loss": 0.3145, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.683357879234168e-05, | |
| "loss": 0.4613, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.8338858195211787, | |
| "eval_loss": 0.4798433482646942, | |
| "eval_runtime": 51.6374, | |
| "eval_samples_per_second": 52.578, | |
| "eval_steps_per_second": 6.584, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.6097201767304866e-05, | |
| "loss": 0.3371, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.536082474226804e-05, | |
| "loss": 0.5341, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.462444771723123e-05, | |
| "loss": 0.3712, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.3888070692194405e-05, | |
| "loss": 0.3672, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.8331491712707182, | |
| "eval_loss": 0.47213244438171387, | |
| "eval_runtime": 50.5283, | |
| "eval_samples_per_second": 53.732, | |
| "eval_steps_per_second": 6.729, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.315169366715759e-05, | |
| "loss": 0.3377, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.241531664212077e-05, | |
| "loss": 0.377, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.167893961708395e-05, | |
| "loss": 0.3033, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.094256259204713e-05, | |
| "loss": 0.4058, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.8320441988950277, | |
| "eval_loss": 0.46750709414482117, | |
| "eval_runtime": 51.3576, | |
| "eval_samples_per_second": 52.865, | |
| "eval_steps_per_second": 6.62, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.020618556701031e-05, | |
| "loss": 0.2632, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.946980854197349e-05, | |
| "loss": 0.3504, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.8733431516936675e-05, | |
| "loss": 0.2671, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.799705449189985e-05, | |
| "loss": 0.2954, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_accuracy": 0.8368324125230202, | |
| "eval_loss": 0.45633676648139954, | |
| "eval_runtime": 51.1084, | |
| "eval_samples_per_second": 53.122, | |
| "eval_steps_per_second": 6.653, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.726067746686304e-05, | |
| "loss": 0.3205, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.6524300441826214e-05, | |
| "loss": 0.299, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 3.57879234167894e-05, | |
| "loss": 0.3119, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.5051546391752576e-05, | |
| "loss": 0.3334, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.8324125230202578, | |
| "eval_loss": 0.47254547476768494, | |
| "eval_runtime": 50.5237, | |
| "eval_samples_per_second": 53.737, | |
| "eval_steps_per_second": 6.73, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 3.431516936671576e-05, | |
| "loss": 0.3469, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 3.357879234167894e-05, | |
| "loss": 0.2501, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3.284241531664212e-05, | |
| "loss": 0.3941, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.2106038291605307e-05, | |
| "loss": 0.3151, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.8342541436464088, | |
| "eval_loss": 0.49459975957870483, | |
| "eval_runtime": 51.2803, | |
| "eval_samples_per_second": 52.944, | |
| "eval_steps_per_second": 6.63, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.1369661266568484e-05, | |
| "loss": 0.3946, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.063328424153167e-05, | |
| "loss": 0.3603, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.9896907216494846e-05, | |
| "loss": 0.2634, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.916053019145803e-05, | |
| "loss": 0.3322, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.8401473296500921, | |
| "eval_loss": 0.4636726379394531, | |
| "eval_runtime": 51.0603, | |
| "eval_samples_per_second": 53.172, | |
| "eval_steps_per_second": 6.659, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.8424153166421208e-05, | |
| "loss": 0.3206, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.7687776141384392e-05, | |
| "loss": 0.4357, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.695139911634757e-05, | |
| "loss": 0.2878, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.6215022091310754e-05, | |
| "loss": 0.4057, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.8379373848987108, | |
| "eval_loss": 0.461891233921051, | |
| "eval_runtime": 51.5789, | |
| "eval_samples_per_second": 52.638, | |
| "eval_steps_per_second": 6.592, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.547864506627393e-05, | |
| "loss": 0.3853, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.4742268041237116e-05, | |
| "loss": 0.3127, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 2.4005891016200297e-05, | |
| "loss": 0.3005, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.3269513991163478e-05, | |
| "loss": 0.3365, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.8372007366482505, | |
| "eval_loss": 0.45343735814094543, | |
| "eval_runtime": 52.5929, | |
| "eval_samples_per_second": 51.623, | |
| "eval_steps_per_second": 6.465, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.253313696612666e-05, | |
| "loss": 0.285, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.179675994108984e-05, | |
| "loss": 0.3114, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.106038291605302e-05, | |
| "loss": 0.321, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.03240058910162e-05, | |
| "loss": 0.2832, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.8397790055248618, | |
| "eval_loss": 0.4500523805618286, | |
| "eval_runtime": 51.02, | |
| "eval_samples_per_second": 53.214, | |
| "eval_steps_per_second": 6.664, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.9587628865979382e-05, | |
| "loss": 0.3756, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.8851251840942563e-05, | |
| "loss": 0.2677, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.8114874815905744e-05, | |
| "loss": 0.392, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.7378497790868925e-05, | |
| "loss": 0.2512, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.8427255985267035, | |
| "eval_loss": 0.44581884145736694, | |
| "eval_runtime": 50.8126, | |
| "eval_samples_per_second": 53.432, | |
| "eval_steps_per_second": 6.691, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.6642120765832106e-05, | |
| "loss": 0.3217, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.5905743740795287e-05, | |
| "loss": 0.3821, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.5169366715758468e-05, | |
| "loss": 0.2863, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.4432989690721649e-05, | |
| "loss": 0.2347, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_accuracy": 0.8423572744014733, | |
| "eval_loss": 0.44903260469436646, | |
| "eval_runtime": 51.8254, | |
| "eval_samples_per_second": 52.387, | |
| "eval_steps_per_second": 6.56, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.3696612665684833e-05, | |
| "loss": 0.1994, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.2960235640648014e-05, | |
| "loss": 0.2408, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.2223858615611193e-05, | |
| "loss": 0.4364, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.1487481590574374e-05, | |
| "loss": 0.1912, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_accuracy": 0.8449355432780847, | |
| "eval_loss": 0.44017237424850464, | |
| "eval_runtime": 51.6194, | |
| "eval_samples_per_second": 52.597, | |
| "eval_steps_per_second": 6.587, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.0751104565537557e-05, | |
| "loss": 0.3227, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.0014727540500737e-05, | |
| "loss": 0.2769, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 9.278350515463918e-06, | |
| "loss": 0.2485, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 8.5419734904271e-06, | |
| "loss": 0.3679, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.8464088397790055, | |
| "eval_loss": 0.4389094412326813, | |
| "eval_runtime": 50.627, | |
| "eval_samples_per_second": 53.628, | |
| "eval_steps_per_second": 6.716, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 7.80559646539028e-06, | |
| "loss": 0.3111, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 7.069219440353461e-06, | |
| "loss": 0.33, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 6.332842415316642e-06, | |
| "loss": 0.2817, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 5.596465390279824e-06, | |
| "loss": 0.2953, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.8515653775322284, | |
| "eval_loss": 0.4342978596687317, | |
| "eval_runtime": 51.1052, | |
| "eval_samples_per_second": 53.126, | |
| "eval_steps_per_second": 6.653, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.860088365243005e-06, | |
| "loss": 0.247, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.123711340206186e-06, | |
| "loss": 0.3024, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 3.3873343151693666e-06, | |
| "loss": 0.2232, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.650957290132548e-06, | |
| "loss": 0.2982, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8486187845303867, | |
| "eval_loss": 0.4305761158466339, | |
| "eval_runtime": 51.0559, | |
| "eval_samples_per_second": 53.177, | |
| "eval_steps_per_second": 6.659, | |
| "step": 2680 | |
| } | |
| ], | |
| "max_steps": 2716, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.322629783093584e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |