| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.049582312093116344, |
| "epoch": 0.016, |
| "grad_norm": 0.40234375, |
| "learning_rate": 7.2e-07, |
| "loss": 0.0306, |
| "mean_token_accuracy": 0.9906190823763609, |
| "num_tokens": 282849.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.05075581485871226, |
| "epoch": 0.032, |
| "grad_norm": 0.3828125, |
| "learning_rate": 1.52e-06, |
| "loss": 0.0311, |
| "mean_token_accuracy": 0.990446624532342, |
| "num_tokens": 561726.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.049523813289124516, |
| "epoch": 0.048, |
| "grad_norm": 0.388671875, |
| "learning_rate": 2.3200000000000002e-06, |
| "loss": 0.0309, |
| "mean_token_accuracy": 0.9905388820916414, |
| "num_tokens": 842092.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.04796491010347381, |
| "epoch": 0.064, |
| "grad_norm": 0.4453125, |
| "learning_rate": 3.12e-06, |
| "loss": 0.0299, |
| "mean_token_accuracy": 0.9908089976757765, |
| "num_tokens": 1133349.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.04753222216386348, |
| "epoch": 0.08, |
| "grad_norm": 0.5859375, |
| "learning_rate": 3.920000000000001e-06, |
| "loss": 0.0301, |
| "mean_token_accuracy": 0.9906909823417663, |
| "num_tokens": 1422340.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.04738440909422934, |
| "epoch": 0.096, |
| "grad_norm": 0.51953125, |
| "learning_rate": 4.7200000000000005e-06, |
| "loss": 0.0303, |
| "mean_token_accuracy": 0.9907543085515499, |
| "num_tokens": 1711821.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.04916054802015424, |
| "epoch": 0.112, |
| "grad_norm": 0.478515625, |
| "learning_rate": 5.5200000000000005e-06, |
| "loss": 0.0314, |
| "mean_token_accuracy": 0.990179405733943, |
| "num_tokens": 1991855.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.04721315624192357, |
| "epoch": 0.128, |
| "grad_norm": 0.439453125, |
| "learning_rate": 6.3200000000000005e-06, |
| "loss": 0.0299, |
| "mean_token_accuracy": 0.990847859159112, |
| "num_tokens": 2288720.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.05030787743162364, |
| "epoch": 0.144, |
| "grad_norm": 0.44140625, |
| "learning_rate": 7.1200000000000004e-06, |
| "loss": 0.0323, |
| "mean_token_accuracy": 0.9902282193303108, |
| "num_tokens": 2564352.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.05056912547443062, |
| "epoch": 0.16, |
| "grad_norm": 0.41796875, |
| "learning_rate": 7.92e-06, |
| "loss": 0.0327, |
| "mean_token_accuracy": 0.9901061341166496, |
| "num_tokens": 2841876.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 0.24093767249584197, |
| "eval_chemistry_loss": 1.4594624042510986, |
| "eval_chemistry_mean_token_accuracy": 0.7775518426895142, |
| "eval_chemistry_num_tokens": 2841876.0, |
| "eval_chemistry_runtime": 34.757, |
| "eval_chemistry_samples_per_second": 14.386, |
| "eval_chemistry_steps_per_second": 3.596, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 0.18010181015729904, |
| "eval_math_loss": 1.3218717575073242, |
| "eval_math_mean_token_accuracy": 0.8153265738487243, |
| "eval_math_num_tokens": 2841876.0, |
| "eval_math_runtime": 35.7695, |
| "eval_math_samples_per_second": 13.978, |
| "eval_math_steps_per_second": 3.495, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.04988773693330586, |
| "epoch": 0.176, |
| "grad_norm": 0.431640625, |
| "learning_rate": 8.720000000000001e-06, |
| "loss": 0.0322, |
| "mean_token_accuracy": 0.9902508158236742, |
| "num_tokens": 3129237.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.05084227197803557, |
| "epoch": 0.192, |
| "grad_norm": 0.45703125, |
| "learning_rate": 9.52e-06, |
| "loss": 0.0329, |
| "mean_token_accuracy": 0.9902640093117953, |
| "num_tokens": 3414875.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.052573437918908894, |
| "epoch": 0.208, |
| "grad_norm": 0.423828125, |
| "learning_rate": 1.0320000000000001e-05, |
| "loss": 0.0339, |
| "mean_token_accuracy": 0.9898341007530689, |
| "num_tokens": 3695623.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.05318723081145436, |
| "epoch": 0.224, |
| "grad_norm": 0.5859375, |
| "learning_rate": 1.1120000000000002e-05, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9897902756929398, |
| "num_tokens": 3977666.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.052078232262283565, |
| "epoch": 0.24, |
| "grad_norm": 0.70703125, |
| "learning_rate": 1.1920000000000001e-05, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.9899859059602022, |
| "num_tokens": 4269287.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.05714732928900048, |
| "epoch": 0.256, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.2720000000000002e-05, |
| "loss": 0.0369, |
| "mean_token_accuracy": 0.9894224800169468, |
| "num_tokens": 4558632.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.0642960250377655, |
| "epoch": 0.272, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.3520000000000003e-05, |
| "loss": 0.0424, |
| "mean_token_accuracy": 0.9880315110087394, |
| "num_tokens": 4839270.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.07618674396071583, |
| "epoch": 0.288, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.432e-05, |
| "loss": 0.0515, |
| "mean_token_accuracy": 0.9856673073023557, |
| "num_tokens": 5125447.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.08972589983604848, |
| "epoch": 0.304, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.5120000000000001e-05, |
| "loss": 0.0616, |
| "mean_token_accuracy": 0.98250596486032, |
| "num_tokens": 5415925.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.09321089587174355, |
| "epoch": 0.32, |
| "grad_norm": 2.40625, |
| "learning_rate": 1.5920000000000003e-05, |
| "loss": 0.0662, |
| "mean_token_accuracy": 0.9811959642916918, |
| "num_tokens": 5710178.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.26300008833408356, |
| "eval_chemistry_loss": 1.3655120134353638, |
| "eval_chemistry_mean_token_accuracy": 0.7764006156921387, |
| "eval_chemistry_num_tokens": 5710178.0, |
| "eval_chemistry_runtime": 34.6925, |
| "eval_chemistry_samples_per_second": 14.412, |
| "eval_chemistry_steps_per_second": 3.603, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 0.21227243894338607, |
| "eval_math_loss": 1.2128866910934448, |
| "eval_math_mean_token_accuracy": 0.8105128321647644, |
| "eval_math_num_tokens": 5710178.0, |
| "eval_math_runtime": 35.7511, |
| "eval_math_samples_per_second": 13.986, |
| "eval_math_steps_per_second": 3.496, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.1054203346837312, |
| "epoch": 0.336, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.672e-05, |
| "loss": 0.0748, |
| "mean_token_accuracy": 0.9779953259974719, |
| "num_tokens": 5993843.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.10738181113265455, |
| "epoch": 0.352, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.752e-05, |
| "loss": 0.077, |
| "mean_token_accuracy": 0.9772336103022099, |
| "num_tokens": 6279137.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.11388949756510555, |
| "epoch": 0.368, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.832e-05, |
| "loss": 0.082, |
| "mean_token_accuracy": 0.9750573400408029, |
| "num_tokens": 6562387.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.11353842569515109, |
| "epoch": 0.384, |
| "grad_norm": 2.5, |
| "learning_rate": 1.912e-05, |
| "loss": 0.0844, |
| "mean_token_accuracy": 0.9745409067720174, |
| "num_tokens": 6851520.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.12373326132073999, |
| "epoch": 0.4, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.9920000000000002e-05, |
| "loss": 0.0929, |
| "mean_token_accuracy": 0.9715707816183567, |
| "num_tokens": 7138932.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.1284570125862956, |
| "epoch": 0.416, |
| "grad_norm": 2.8125, |
| "learning_rate": 1.9920000000000002e-05, |
| "loss": 0.0971, |
| "mean_token_accuracy": 0.9700299393385649, |
| "num_tokens": 7420720.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.1309962844941765, |
| "epoch": 0.432, |
| "grad_norm": 3.0, |
| "learning_rate": 1.9831111111111113e-05, |
| "loss": 0.1024, |
| "mean_token_accuracy": 0.9681163311004639, |
| "num_tokens": 7695518.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.1339765153825283, |
| "epoch": 0.448, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.9742222222222223e-05, |
| "loss": 0.1012, |
| "mean_token_accuracy": 0.9680692821741104, |
| "num_tokens": 7983543.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.13085799617692828, |
| "epoch": 0.464, |
| "grad_norm": 2.625, |
| "learning_rate": 1.9653333333333334e-05, |
| "loss": 0.0988, |
| "mean_token_accuracy": 0.968861211463809, |
| "num_tokens": 8271787.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.13468673978932202, |
| "epoch": 0.48, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.9564444444444444e-05, |
| "loss": 0.1021, |
| "mean_token_accuracy": 0.9679039020091296, |
| "num_tokens": 8556170.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.2733586312532425, |
| "eval_chemistry_loss": 1.3263864517211914, |
| "eval_chemistry_mean_token_accuracy": 0.7758925561904907, |
| "eval_chemistry_num_tokens": 8556170.0, |
| "eval_chemistry_runtime": 34.7028, |
| "eval_chemistry_samples_per_second": 14.408, |
| "eval_chemistry_steps_per_second": 3.602, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.2326348875761032, |
| "eval_math_loss": 1.1490864753723145, |
| "eval_math_mean_token_accuracy": 0.8080902843475342, |
| "eval_math_num_tokens": 8556170.0, |
| "eval_math_runtime": 35.7382, |
| "eval_math_samples_per_second": 13.991, |
| "eval_math_steps_per_second": 3.498, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.13333711712621152, |
| "epoch": 0.496, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.9475555555555558e-05, |
| "loss": 0.1, |
| "mean_token_accuracy": 0.9678985115140677, |
| "num_tokens": 8840590.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.13105960306711495, |
| "epoch": 0.512, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.938666666666667e-05, |
| "loss": 0.0989, |
| "mean_token_accuracy": 0.9688469890505076, |
| "num_tokens": 9131903.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.12777495444752276, |
| "epoch": 0.528, |
| "grad_norm": 2.78125, |
| "learning_rate": 1.929777777777778e-05, |
| "loss": 0.0957, |
| "mean_token_accuracy": 0.9697885576635599, |
| "num_tokens": 9429519.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.12928434419445695, |
| "epoch": 0.544, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.920888888888889e-05, |
| "loss": 0.0975, |
| "mean_token_accuracy": 0.9691570393741131, |
| "num_tokens": 9718247.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.1348258295096457, |
| "epoch": 0.56, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.912e-05, |
| "loss": 0.1009, |
| "mean_token_accuracy": 0.9687925592064858, |
| "num_tokens": 9991100.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.13326651249080895, |
| "epoch": 0.576, |
| "grad_norm": 3.03125, |
| "learning_rate": 1.9031111111111114e-05, |
| "loss": 0.0989, |
| "mean_token_accuracy": 0.9687243364751339, |
| "num_tokens": 10276090.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.13166287271305918, |
| "epoch": 0.592, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.8942222222222224e-05, |
| "loss": 0.0985, |
| "mean_token_accuracy": 0.9695811975747347, |
| "num_tokens": 10566762.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.13076401995494963, |
| "epoch": 0.608, |
| "grad_norm": 2.75, |
| "learning_rate": 1.8853333333333335e-05, |
| "loss": 0.0961, |
| "mean_token_accuracy": 0.9704167168587446, |
| "num_tokens": 10857002.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.13047253522090613, |
| "epoch": 0.624, |
| "grad_norm": 2.71875, |
| "learning_rate": 1.8764444444444445e-05, |
| "loss": 0.0967, |
| "mean_token_accuracy": 0.9699896775186062, |
| "num_tokens": 11147326.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.12835560292005538, |
| "epoch": 0.64, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.867555555555556e-05, |
| "loss": 0.0947, |
| "mean_token_accuracy": 0.9701224651187659, |
| "num_tokens": 11437710.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.26942539191246034, |
| "eval_chemistry_loss": 1.3352558612823486, |
| "eval_chemistry_mean_token_accuracy": 0.7756590819358826, |
| "eval_chemistry_num_tokens": 11437710.0, |
| "eval_chemistry_runtime": 34.7257, |
| "eval_chemistry_samples_per_second": 14.399, |
| "eval_chemistry_steps_per_second": 3.6, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.22599203491210937, |
| "eval_math_loss": 1.1578930616378784, |
| "eval_math_mean_token_accuracy": 0.8086360969543457, |
| "eval_math_num_tokens": 11437710.0, |
| "eval_math_runtime": 35.7376, |
| "eval_math_samples_per_second": 13.991, |
| "eval_math_steps_per_second": 3.498, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.12649631341919304, |
| "epoch": 0.656, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.858666666666667e-05, |
| "loss": 0.095, |
| "mean_token_accuracy": 0.9704382967203855, |
| "num_tokens": 11727268.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.12803141153417527, |
| "epoch": 0.672, |
| "grad_norm": 2.828125, |
| "learning_rate": 1.849777777777778e-05, |
| "loss": 0.0939, |
| "mean_token_accuracy": 0.9711931616067886, |
| "num_tokens": 12020478.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.13106235223822296, |
| "epoch": 0.688, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.840888888888889e-05, |
| "loss": 0.0975, |
| "mean_token_accuracy": 0.9697022248059511, |
| "num_tokens": 12300380.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.12510091881267726, |
| "epoch": 0.704, |
| "grad_norm": 2.90625, |
| "learning_rate": 1.832e-05, |
| "loss": 0.0918, |
| "mean_token_accuracy": 0.9710133023560047, |
| "num_tokens": 12591366.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.1280051540117711, |
| "epoch": 0.72, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.823111111111111e-05, |
| "loss": 0.0932, |
| "mean_token_accuracy": 0.9706473097205162, |
| "num_tokens": 12869618.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.1262424793560058, |
| "epoch": 0.736, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.814222222222222e-05, |
| "loss": 0.0911, |
| "mean_token_accuracy": 0.9721917398273945, |
| "num_tokens": 13156311.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.1231601320207119, |
| "epoch": 0.752, |
| "grad_norm": 2.609375, |
| "learning_rate": 1.8053333333333332e-05, |
| "loss": 0.0899, |
| "mean_token_accuracy": 0.9719577796757222, |
| "num_tokens": 13438911.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.12007905230857432, |
| "epoch": 0.768, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.7964444444444446e-05, |
| "loss": 0.0877, |
| "mean_token_accuracy": 0.9723689883947373, |
| "num_tokens": 13735397.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.1257410950958729, |
| "epoch": 0.784, |
| "grad_norm": 2.75, |
| "learning_rate": 1.7875555555555556e-05, |
| "loss": 0.0898, |
| "mean_token_accuracy": 0.9727259453386068, |
| "num_tokens": 14024497.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.12430990342982114, |
| "epoch": 0.8, |
| "grad_norm": 2.828125, |
| "learning_rate": 1.7786666666666667e-05, |
| "loss": 0.0894, |
| "mean_token_accuracy": 0.9722163360565901, |
| "num_tokens": 14316427.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.2670133001804352, |
| "eval_chemistry_loss": 1.3451496362686157, |
| "eval_chemistry_mean_token_accuracy": 0.7761107358932495, |
| "eval_chemistry_num_tokens": 14316427.0, |
| "eval_chemistry_runtime": 34.7622, |
| "eval_chemistry_samples_per_second": 14.383, |
| "eval_chemistry_steps_per_second": 3.596, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.223143694460392, |
| "eval_math_loss": 1.1661651134490967, |
| "eval_math_mean_token_accuracy": 0.8089712624549865, |
| "eval_math_num_tokens": 14316427.0, |
| "eval_math_runtime": 35.4897, |
| "eval_math_samples_per_second": 14.089, |
| "eval_math_steps_per_second": 3.522, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.12252137046307325, |
| "epoch": 0.816, |
| "grad_norm": 2.9375, |
| "learning_rate": 1.7697777777777777e-05, |
| "loss": 0.0878, |
| "mean_token_accuracy": 0.9729147344827652, |
| "num_tokens": 14607136.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.12182142208330334, |
| "epoch": 0.832, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.760888888888889e-05, |
| "loss": 0.0867, |
| "mean_token_accuracy": 0.973680478706956, |
| "num_tokens": 14894084.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.1252059457823634, |
| "epoch": 0.848, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.752e-05, |
| "loss": 0.0886, |
| "mean_token_accuracy": 0.9728491563349962, |
| "num_tokens": 15177039.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.12408317467197776, |
| "epoch": 0.864, |
| "grad_norm": 2.828125, |
| "learning_rate": 1.7431111111111112e-05, |
| "loss": 0.0873, |
| "mean_token_accuracy": 0.9731103349477053, |
| "num_tokens": 15463393.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.12185636130161584, |
| "epoch": 0.88, |
| "grad_norm": 2.78125, |
| "learning_rate": 1.7342222222222226e-05, |
| "loss": 0.0864, |
| "mean_token_accuracy": 0.9735086120665073, |
| "num_tokens": 15747943.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.12166528352536261, |
| "epoch": 0.896, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.7253333333333336e-05, |
| "loss": 0.0861, |
| "mean_token_accuracy": 0.9738527856767177, |
| "num_tokens": 16027931.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.12183338818140328, |
| "epoch": 0.912, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.7164444444444447e-05, |
| "loss": 0.0849, |
| "mean_token_accuracy": 0.9736077032983304, |
| "num_tokens": 16317127.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.11985594518482685, |
| "epoch": 0.928, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.7075555555555557e-05, |
| "loss": 0.0845, |
| "mean_token_accuracy": 0.9738747924566269, |
| "num_tokens": 16598757.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.12089673141017557, |
| "epoch": 0.944, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.6986666666666668e-05, |
| "loss": 0.0853, |
| "mean_token_accuracy": 0.974429277703166, |
| "num_tokens": 16882236.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.117778305336833, |
| "epoch": 0.96, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.6897777777777778e-05, |
| "loss": 0.0818, |
| "mean_token_accuracy": 0.9753529716283083, |
| "num_tokens": 17169495.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.2685954006910324, |
| "eval_chemistry_loss": 1.3377221822738647, |
| "eval_chemistry_mean_token_accuracy": 0.7758646550178527, |
| "eval_chemistry_num_tokens": 17169495.0, |
| "eval_chemistry_runtime": 34.7493, |
| "eval_chemistry_samples_per_second": 14.389, |
| "eval_chemistry_steps_per_second": 3.597, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.21977827924489976, |
| "eval_math_loss": 1.166742205619812, |
| "eval_math_mean_token_accuracy": 0.8097256078720093, |
| "eval_math_num_tokens": 17169495.0, |
| "eval_math_runtime": 35.8222, |
| "eval_math_samples_per_second": 13.958, |
| "eval_math_steps_per_second": 3.489, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.11901968852616847, |
| "epoch": 0.976, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.680888888888889e-05, |
| "loss": 0.0819, |
| "mean_token_accuracy": 0.9747418876737356, |
| "num_tokens": 17454147.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.11983131496235729, |
| "epoch": 0.992, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.672e-05, |
| "loss": 0.0825, |
| "mean_token_accuracy": 0.9750407934188843, |
| "num_tokens": 17742596.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.10401241672225296, |
| "epoch": 1.008, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.6631111111111113e-05, |
| "loss": 0.0702, |
| "mean_token_accuracy": 0.9788199871778488, |
| "num_tokens": 18033540.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.08530816063284874, |
| "epoch": 1.024, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.6542222222222223e-05, |
| "loss": 0.059, |
| "mean_token_accuracy": 0.982539826259017, |
| "num_tokens": 18316346.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.08670682646334171, |
| "epoch": 1.04, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.6453333333333334e-05, |
| "loss": 0.0605, |
| "mean_token_accuracy": 0.9820570342242718, |
| "num_tokens": 18595230.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.08672361080534756, |
| "epoch": 1.056, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.6364444444444444e-05, |
| "loss": 0.0606, |
| "mean_token_accuracy": 0.9823420338332653, |
| "num_tokens": 18885547.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.08972970175091177, |
| "epoch": 1.072, |
| "grad_norm": 2.1875, |
| "learning_rate": 1.6275555555555558e-05, |
| "loss": 0.0619, |
| "mean_token_accuracy": 0.9819000914692879, |
| "num_tokens": 19169083.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.08532073406968266, |
| "epoch": 1.088, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.618666666666667e-05, |
| "loss": 0.0586, |
| "mean_token_accuracy": 0.9825958292931318, |
| "num_tokens": 19461465.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.08605421562679112, |
| "epoch": 1.104, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.609777777777778e-05, |
| "loss": 0.0594, |
| "mean_token_accuracy": 0.9826772827655077, |
| "num_tokens": 19755027.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.084381258697249, |
| "epoch": 1.12, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.600888888888889e-05, |
| "loss": 0.0584, |
| "mean_token_accuracy": 0.9828958626836538, |
| "num_tokens": 20043293.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.2597039748430252, |
| "eval_chemistry_loss": 1.3768759965896606, |
| "eval_chemistry_mean_token_accuracy": 0.776004008769989, |
| "eval_chemistry_num_tokens": 20043293.0, |
| "eval_chemistry_runtime": 34.7425, |
| "eval_chemistry_samples_per_second": 14.392, |
| "eval_chemistry_steps_per_second": 3.598, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.2076559336781502, |
| "eval_math_loss": 1.2019093036651611, |
| "eval_math_mean_token_accuracy": 0.8111752443313599, |
| "eval_math_num_tokens": 20043293.0, |
| "eval_math_runtime": 35.7886, |
| "eval_math_samples_per_second": 13.971, |
| "eval_math_steps_per_second": 3.493, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.09029958308674395, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.5920000000000003e-05, |
| "loss": 0.0646, |
| "mean_token_accuracy": 0.9808140520006419, |
| "num_tokens": 20323010.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.08711256731767207, |
| "epoch": 1.152, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.5831111111111114e-05, |
| "loss": 0.0607, |
| "mean_token_accuracy": 0.9819911230355501, |
| "num_tokens": 20613266.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.08424482597038149, |
| "epoch": 1.168, |
| "grad_norm": 1.8828125, |
| "learning_rate": 1.5742222222222224e-05, |
| "loss": 0.0583, |
| "mean_token_accuracy": 0.9828614890575409, |
| "num_tokens": 20910404.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.086951145529747, |
| "epoch": 1.184, |
| "grad_norm": 2.015625, |
| "learning_rate": 1.5653333333333335e-05, |
| "loss": 0.0596, |
| "mean_token_accuracy": 0.9822407346218824, |
| "num_tokens": 21200197.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.08553398705553264, |
| "epoch": 1.2, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.5564444444444445e-05, |
| "loss": 0.0595, |
| "mean_token_accuracy": 0.9823605198413133, |
| "num_tokens": 21496441.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.08706241557374597, |
| "epoch": 1.216, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.5475555555555556e-05, |
| "loss": 0.0604, |
| "mean_token_accuracy": 0.9823201231658458, |
| "num_tokens": 21778030.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.08789820885285735, |
| "epoch": 1.232, |
| "grad_norm": 2.0, |
| "learning_rate": 1.5386666666666666e-05, |
| "loss": 0.0619, |
| "mean_token_accuracy": 0.9814675074070692, |
| "num_tokens": 22060851.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.08552719466388226, |
| "epoch": 1.248, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.5297777777777777e-05, |
| "loss": 0.0594, |
| "mean_token_accuracy": 0.9823314778506755, |
| "num_tokens": 22348738.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.08782941014505923, |
| "epoch": 1.264, |
| "grad_norm": 1.7578125, |
| "learning_rate": 1.520888888888889e-05, |
| "loss": 0.0609, |
| "mean_token_accuracy": 0.9820897091180086, |
| "num_tokens": 22635039.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.08776575655210764, |
| "epoch": 1.28, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.5120000000000001e-05, |
| "loss": 0.062, |
| "mean_token_accuracy": 0.9818732112646102, |
| "num_tokens": 22915190.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.262844596862793, |
| "eval_chemistry_loss": 1.3627640008926392, |
| "eval_chemistry_mean_token_accuracy": 0.7757310657501221, |
| "eval_chemistry_num_tokens": 22915190.0, |
| "eval_chemistry_runtime": 34.756, |
| "eval_chemistry_samples_per_second": 14.386, |
| "eval_chemistry_steps_per_second": 3.596, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.20891815394163132, |
| "eval_math_loss": 1.1918283700942993, |
| "eval_math_mean_token_accuracy": 0.8114760699272155, |
| "eval_math_num_tokens": 22915190.0, |
| "eval_math_runtime": 35.7894, |
| "eval_math_samples_per_second": 13.971, |
| "eval_math_steps_per_second": 3.493, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.08705279752612113, |
| "epoch": 1.296, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.5031111111111111e-05, |
| "loss": 0.0603, |
| "mean_token_accuracy": 0.982403326779604, |
| "num_tokens": 23194947.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.08580190755892544, |
| "epoch": 1.312, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.4942222222222222e-05, |
| "loss": 0.058, |
| "mean_token_accuracy": 0.9832350175827742, |
| "num_tokens": 23483496.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.08680400990415364, |
| "epoch": 1.328, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.4853333333333336e-05, |
| "loss": 0.0601, |
| "mean_token_accuracy": 0.9825425524264574, |
| "num_tokens": 23762589.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.08539955569431186, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 1.875, |
| "learning_rate": 1.4764444444444446e-05, |
| "loss": 0.0582, |
| "mean_token_accuracy": 0.983034698292613, |
| "num_tokens": 24041497.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.08369267631787806, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 1.8046875, |
| "learning_rate": 1.4675555555555557e-05, |
| "loss": 0.0594, |
| "mean_token_accuracy": 0.9825323965400458, |
| "num_tokens": 24330591.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.0880416841711849, |
| "epoch": 1.376, |
| "grad_norm": 1.890625, |
| "learning_rate": 1.4586666666666667e-05, |
| "loss": 0.0595, |
| "mean_token_accuracy": 0.982681792974472, |
| "num_tokens": 24611420.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.08504764605313539, |
| "epoch": 1.392, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.449777777777778e-05, |
| "loss": 0.0584, |
| "mean_token_accuracy": 0.9826866809278727, |
| "num_tokens": 24894694.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.083847014605999, |
| "epoch": 1.408, |
| "grad_norm": 2.0, |
| "learning_rate": 1.440888888888889e-05, |
| "loss": 0.057, |
| "mean_token_accuracy": 0.9832074366509914, |
| "num_tokens": 25184946.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.08376108810771257, |
| "epoch": 1.424, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.432e-05, |
| "loss": 0.0582, |
| "mean_token_accuracy": 0.9830348044633865, |
| "num_tokens": 25474746.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.0831807145383209, |
| "epoch": 1.44, |
| "grad_norm": 2.125, |
| "learning_rate": 1.4231111111111114e-05, |
| "loss": 0.0573, |
| "mean_token_accuracy": 0.9832527209073305, |
| "num_tokens": 25758550.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.2588958297967911, |
| "eval_chemistry_loss": 1.3800510168075562, |
| "eval_chemistry_mean_token_accuracy": 0.7761167316436768, |
| "eval_chemistry_num_tokens": 25758550.0, |
| "eval_chemistry_runtime": 34.746, |
| "eval_chemistry_samples_per_second": 14.39, |
| "eval_chemistry_steps_per_second": 3.598, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.20648883867263795, |
| "eval_math_loss": 1.1997566223144531, |
| "eval_math_mean_token_accuracy": 0.8119877796173096, |
| "eval_math_num_tokens": 25758550.0, |
| "eval_math_runtime": 35.7996, |
| "eval_math_samples_per_second": 13.967, |
| "eval_math_steps_per_second": 3.492, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.08563981833867729, |
| "epoch": 1.456, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.4142222222222224e-05, |
| "loss": 0.0572, |
| "mean_token_accuracy": 0.9828862383961677, |
| "num_tokens": 26039862.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.0852160410489887, |
| "epoch": 1.472, |
| "grad_norm": 1.8125, |
| "learning_rate": 1.4053333333333335e-05, |
| "loss": 0.0589, |
| "mean_token_accuracy": 0.9824321672320366, |
| "num_tokens": 26318666.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.08569062508177012, |
| "epoch": 1.488, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.3964444444444445e-05, |
| "loss": 0.0585, |
| "mean_token_accuracy": 0.9830207195132971, |
| "num_tokens": 26594867.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.08032872283365577, |
| "epoch": 1.504, |
| "grad_norm": 1.765625, |
| "learning_rate": 1.3875555555555557e-05, |
| "loss": 0.0554, |
| "mean_token_accuracy": 0.9839188922196627, |
| "num_tokens": 26887233.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.0836705738445744, |
| "epoch": 1.52, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.3786666666666668e-05, |
| "loss": 0.0568, |
| "mean_token_accuracy": 0.9833197060972452, |
| "num_tokens": 27177448.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.08208613845054061, |
| "epoch": 1.536, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.3697777777777778e-05, |
| "loss": 0.0553, |
| "mean_token_accuracy": 0.9837426465004683, |
| "num_tokens": 27467857.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.08092109204735606, |
| "epoch": 1.552, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.3608888888888889e-05, |
| "loss": 0.0554, |
| "mean_token_accuracy": 0.9836731895804405, |
| "num_tokens": 27755138.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.07861132090911269, |
| "epoch": 1.568, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.3520000000000003e-05, |
| "loss": 0.0537, |
| "mean_token_accuracy": 0.9842242918908596, |
| "num_tokens": 28049363.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.08059575033839791, |
| "epoch": 1.584, |
| "grad_norm": 1.9453125, |
| "learning_rate": 1.3431111111111113e-05, |
| "loss": 0.0548, |
| "mean_token_accuracy": 0.9839956350624561, |
| "num_tokens": 28337890.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.08031310953665524, |
| "epoch": 1.6, |
| "grad_norm": 1.84375, |
| "learning_rate": 1.3342222222222224e-05, |
| "loss": 0.0544, |
| "mean_token_accuracy": 0.983878904953599, |
| "num_tokens": 28626068.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.2602477210760117, |
| "eval_chemistry_loss": 1.3820487260818481, |
| "eval_chemistry_mean_token_accuracy": 0.7756126351356506, |
| "eval_chemistry_num_tokens": 28626068.0, |
| "eval_chemistry_runtime": 34.7493, |
| "eval_chemistry_samples_per_second": 14.389, |
| "eval_chemistry_steps_per_second": 3.597, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.20556720286607744, |
| "eval_math_loss": 1.207412600517273, |
| "eval_math_mean_token_accuracy": 0.8115635991096497, |
| "eval_math_num_tokens": 28626068.0, |
| "eval_math_runtime": 35.4675, |
| "eval_math_samples_per_second": 14.097, |
| "eval_math_steps_per_second": 3.524, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.08080549829173833, |
| "epoch": 1.616, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.3253333333333334e-05, |
| "loss": 0.0542, |
| "mean_token_accuracy": 0.9843609642237425, |
| "num_tokens": 28913208.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.08292383486405015, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.3164444444444446e-05, |
| "loss": 0.0575, |
| "mean_token_accuracy": 0.9831736672669649, |
| "num_tokens": 29188917.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.07944189475383609, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.3075555555555557e-05, |
| "loss": 0.0531, |
| "mean_token_accuracy": 0.9844527587294578, |
| "num_tokens": 29477292.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.08005691035650671, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.2986666666666667e-05, |
| "loss": 0.0532, |
| "mean_token_accuracy": 0.9843937639147043, |
| "num_tokens": 29772480.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.07776331030763686, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.2897777777777778e-05, |
| "loss": 0.053, |
| "mean_token_accuracy": 0.9847775120288134, |
| "num_tokens": 30061007.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.0810230050003156, |
| "epoch": 1.696, |
| "grad_norm": 1.6875, |
| "learning_rate": 1.2808888888888891e-05, |
| "loss": 0.0541, |
| "mean_token_accuracy": 0.984524767100811, |
| "num_tokens": 30344415.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.0787459987681359, |
| "epoch": 1.712, |
| "grad_norm": 1.7421875, |
| "learning_rate": 1.2720000000000002e-05, |
| "loss": 0.0534, |
| "mean_token_accuracy": 0.9842950701713562, |
| "num_tokens": 30640891.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.0789328703423962, |
| "epoch": 1.728, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.2631111111111112e-05, |
| "loss": 0.0531, |
| "mean_token_accuracy": 0.9845080733299255, |
| "num_tokens": 30921295.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.0788962548598647, |
| "epoch": 1.744, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.2542222222222223e-05, |
| "loss": 0.0534, |
| "mean_token_accuracy": 0.9843788214027882, |
| "num_tokens": 31206375.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.08012886063661426, |
| "epoch": 1.76, |
| "grad_norm": 1.9453125, |
| "learning_rate": 1.2453333333333335e-05, |
| "loss": 0.0536, |
| "mean_token_accuracy": 0.9843036912381649, |
| "num_tokens": 31491856.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.2553063322305679, |
| "eval_chemistry_loss": 1.3884915113449097, |
| "eval_chemistry_mean_token_accuracy": 0.7761442866325379, |
| "eval_chemistry_num_tokens": 31491856.0, |
| "eval_chemistry_runtime": 34.7224, |
| "eval_chemistry_samples_per_second": 14.4, |
| "eval_chemistry_steps_per_second": 3.6, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.20213844722509383, |
| "eval_math_loss": 1.2131967544555664, |
| "eval_math_mean_token_accuracy": 0.811974012374878, |
| "eval_math_num_tokens": 31491856.0, |
| "eval_math_runtime": 35.7761, |
| "eval_math_samples_per_second": 13.976, |
| "eval_math_steps_per_second": 3.494, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.07542885933071375, |
| "epoch": 1.776, |
| "grad_norm": 1.765625, |
| "learning_rate": 1.2364444444444445e-05, |
| "loss": 0.051, |
| "mean_token_accuracy": 0.9852005202323199, |
| "num_tokens": 31782829.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.07717105301562696, |
| "epoch": 1.792, |
| "grad_norm": 1.6640625, |
| "learning_rate": 1.2275555555555556e-05, |
| "loss": 0.0519, |
| "mean_token_accuracy": 0.9847183488309383, |
| "num_tokens": 32070187.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.07923196398187429, |
| "epoch": 1.808, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.2186666666666666e-05, |
| "loss": 0.0524, |
| "mean_token_accuracy": 0.9846949424594641, |
| "num_tokens": 32355072.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.07861853984650224, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 1.6640625, |
| "learning_rate": 1.209777777777778e-05, |
| "loss": 0.0531, |
| "mean_token_accuracy": 0.9844680838286877, |
| "num_tokens": 32634013.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.07815332850441337, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 1.609375, |
| "learning_rate": 1.200888888888889e-05, |
| "loss": 0.0524, |
| "mean_token_accuracy": 0.9846453677862883, |
| "num_tokens": 32917222.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.0771151206921786, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 1.75, |
| "learning_rate": 1.1920000000000001e-05, |
| "loss": 0.0515, |
| "mean_token_accuracy": 0.9850007418543101, |
| "num_tokens": 33207147.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.07652715451549738, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.1831111111111112e-05, |
| "loss": 0.0519, |
| "mean_token_accuracy": 0.9849152982234954, |
| "num_tokens": 33496572.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.07685239620041102, |
| "epoch": 1.888, |
| "grad_norm": 1.7421875, |
| "learning_rate": 1.1742222222222224e-05, |
| "loss": 0.0511, |
| "mean_token_accuracy": 0.9850867554545403, |
| "num_tokens": 33775788.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.07776038108859211, |
| "epoch": 1.904, |
| "grad_norm": 1.7890625, |
| "learning_rate": 1.1653333333333334e-05, |
| "loss": 0.0519, |
| "mean_token_accuracy": 0.984935637190938, |
| "num_tokens": 34052322.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.07653295549098402, |
| "epoch": 1.92, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.1564444444444445e-05, |
| "loss": 0.051, |
| "mean_token_accuracy": 0.9852206721901894, |
| "num_tokens": 34331468.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.2549130402803421, |
| "eval_chemistry_loss": 1.392774224281311, |
| "eval_chemistry_mean_token_accuracy": 0.7761725673675537, |
| "eval_chemistry_num_tokens": 34331468.0, |
| "eval_chemistry_runtime": 34.7035, |
| "eval_chemistry_samples_per_second": 14.408, |
| "eval_chemistry_steps_per_second": 3.602, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.19969486129283906, |
| "eval_math_loss": 1.219051480293274, |
| "eval_math_mean_token_accuracy": 0.8130026779174805, |
| "eval_math_num_tokens": 34331468.0, |
| "eval_math_runtime": 35.7413, |
| "eval_math_samples_per_second": 13.989, |
| "eval_math_steps_per_second": 3.497, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.07580579959321768, |
| "epoch": 1.936, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.1475555555555557e-05, |
| "loss": 0.0515, |
| "mean_token_accuracy": 0.9851022522896529, |
| "num_tokens": 34612743.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.07751624898519367, |
| "epoch": 1.952, |
| "grad_norm": 1.6328125, |
| "learning_rate": 1.1386666666666669e-05, |
| "loss": 0.0513, |
| "mean_token_accuracy": 0.9850031793117523, |
| "num_tokens": 34891115.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.07615254581905902, |
| "epoch": 1.968, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.129777777777778e-05, |
| "loss": 0.0506, |
| "mean_token_accuracy": 0.9852058101445437, |
| "num_tokens": 35179505.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.07313250387087464, |
| "epoch": 1.984, |
| "grad_norm": 1.6171875, |
| "learning_rate": 1.120888888888889e-05, |
| "loss": 0.0488, |
| "mean_token_accuracy": 0.9859140012413263, |
| "num_tokens": 35472596.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.0761550744296983, |
| "epoch": 2.0, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.1120000000000002e-05, |
| "loss": 0.0504, |
| "mean_token_accuracy": 0.9853185825049877, |
| "num_tokens": 35772848.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.056211807671934365, |
| "epoch": 2.016, |
| "grad_norm": 0.69921875, |
| "learning_rate": 1.1031111111111112e-05, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.9902306616306304, |
| "num_tokens": 36060330.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.05393635639920831, |
| "epoch": 2.032, |
| "grad_norm": 0.80078125, |
| "learning_rate": 1.0942222222222223e-05, |
| "loss": 0.0347, |
| "mean_token_accuracy": 0.989858265221119, |
| "num_tokens": 36343894.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.05242305672727525, |
| "epoch": 2.048, |
| "grad_norm": 0.82421875, |
| "learning_rate": 1.0853333333333333e-05, |
| "loss": 0.0336, |
| "mean_token_accuracy": 0.9902104564011097, |
| "num_tokens": 36636410.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.054460834711790085, |
| "epoch": 2.064, |
| "grad_norm": 0.7109375, |
| "learning_rate": 1.0764444444444445e-05, |
| "loss": 0.0339, |
| "mean_token_accuracy": 0.989987451210618, |
| "num_tokens": 36925382.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.05415360638871789, |
| "epoch": 2.08, |
| "grad_norm": 0.84375, |
| "learning_rate": 1.0675555555555558e-05, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9898186340928078, |
| "num_tokens": 37208672.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.24465568888187408, |
| "eval_chemistry_loss": 1.4367998838424683, |
| "eval_chemistry_mean_token_accuracy": 0.7765562734603881, |
| "eval_chemistry_num_tokens": 37208672.0, |
| "eval_chemistry_runtime": 34.6804, |
| "eval_chemistry_samples_per_second": 14.417, |
| "eval_chemistry_steps_per_second": 3.604, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.1864414402246475, |
| "eval_math_loss": 1.2687431573867798, |
| "eval_math_mean_token_accuracy": 0.814744426727295, |
| "eval_math_num_tokens": 37208672.0, |
| "eval_math_runtime": 35.7331, |
| "eval_math_samples_per_second": 13.993, |
| "eval_math_steps_per_second": 3.498, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.053227876662276685, |
| "epoch": 2.096, |
| "grad_norm": 0.7265625, |
| "learning_rate": 1.0586666666666668e-05, |
| "loss": 0.0347, |
| "mean_token_accuracy": 0.989788818731904, |
| "num_tokens": 37491954.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.05289595597423613, |
| "epoch": 2.112, |
| "grad_norm": 0.71875, |
| "learning_rate": 1.0497777777777779e-05, |
| "loss": 0.0336, |
| "mean_token_accuracy": 0.9899541862308979, |
| "num_tokens": 37787033.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.05384734331164509, |
| "epoch": 2.128, |
| "grad_norm": 0.64453125, |
| "learning_rate": 1.040888888888889e-05, |
| "loss": 0.0347, |
| "mean_token_accuracy": 0.9895545594394207, |
| "num_tokens": 38068955.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.052285452221985905, |
| "epoch": 2.144, |
| "grad_norm": 0.8359375, |
| "learning_rate": 1.0320000000000001e-05, |
| "loss": 0.034, |
| "mean_token_accuracy": 0.9899950183928012, |
| "num_tokens": 38359914.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.053004868398420515, |
| "epoch": 2.16, |
| "grad_norm": 0.76171875, |
| "learning_rate": 1.0231111111111112e-05, |
| "loss": 0.0343, |
| "mean_token_accuracy": 0.9898937761783599, |
| "num_tokens": 38649684.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.05320889897411689, |
| "epoch": 2.176, |
| "grad_norm": 0.5703125, |
| "learning_rate": 1.0142222222222222e-05, |
| "loss": 0.0342, |
| "mean_token_accuracy": 0.9898546732962131, |
| "num_tokens": 38938155.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.05214473644737154, |
| "epoch": 2.192, |
| "grad_norm": 0.61328125, |
| "learning_rate": 1.0053333333333334e-05, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9901005830615759, |
| "num_tokens": 39228800.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.053439665515907106, |
| "epoch": 2.208, |
| "grad_norm": 0.6171875, |
| "learning_rate": 9.964444444444445e-06, |
| "loss": 0.0343, |
| "mean_token_accuracy": 0.989708437025547, |
| "num_tokens": 39513118.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.052510200347751376, |
| "epoch": 2.224, |
| "grad_norm": 0.7734375, |
| "learning_rate": 9.875555555555557e-06, |
| "loss": 0.0338, |
| "mean_token_accuracy": 0.9900678258389235, |
| "num_tokens": 39804212.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.05290528133045882, |
| "epoch": 2.24, |
| "grad_norm": 0.64453125, |
| "learning_rate": 9.786666666666667e-06, |
| "loss": 0.0344, |
| "mean_token_accuracy": 0.9897066332399845, |
| "num_tokens": 40092012.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.24463952136039735, |
| "eval_chemistry_loss": 1.4411935806274414, |
| "eval_chemistry_mean_token_accuracy": 0.7765587282180786, |
| "eval_chemistry_num_tokens": 40092012.0, |
| "eval_chemistry_runtime": 34.6754, |
| "eval_chemistry_samples_per_second": 14.419, |
| "eval_chemistry_steps_per_second": 3.605, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.18511718076467515, |
| "eval_math_loss": 1.2752846479415894, |
| "eval_math_mean_token_accuracy": 0.8149898409843445, |
| "eval_math_num_tokens": 40092012.0, |
| "eval_math_runtime": 35.713, |
| "eval_math_samples_per_second": 14.0, |
| "eval_math_steps_per_second": 3.5, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.05300161670893431, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 0.80859375, |
| "learning_rate": 9.697777777777778e-06, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.9895025338977576, |
| "num_tokens": 40380069.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.054382482054643334, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 0.8203125, |
| "learning_rate": 9.60888888888889e-06, |
| "loss": 0.0358, |
| "mean_token_accuracy": 0.9892595294862986, |
| "num_tokens": 40655903.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.05288956081494689, |
| "epoch": 2.288, |
| "grad_norm": 0.8359375, |
| "learning_rate": 9.52e-06, |
| "loss": 0.0338, |
| "mean_token_accuracy": 0.9897118430584669, |
| "num_tokens": 40944791.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.052792912826407704, |
| "epoch": 2.304, |
| "grad_norm": 0.7578125, |
| "learning_rate": 9.431111111111112e-06, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.989751597866416, |
| "num_tokens": 41227146.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.053943155938759445, |
| "epoch": 2.32, |
| "grad_norm": 0.7890625, |
| "learning_rate": 9.342222222222223e-06, |
| "loss": 0.0351, |
| "mean_token_accuracy": 0.9896157145500183, |
| "num_tokens": 41509322.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.05290579678257927, |
| "epoch": 2.336, |
| "grad_norm": 0.7109375, |
| "learning_rate": 9.253333333333333e-06, |
| "loss": 0.0349, |
| "mean_token_accuracy": 0.9897964365780354, |
| "num_tokens": 41792799.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.05324005091097206, |
| "epoch": 2.352, |
| "grad_norm": 0.69140625, |
| "learning_rate": 9.164444444444446e-06, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.989767849072814, |
| "num_tokens": 42083080.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.052101924282032995, |
| "epoch": 2.368, |
| "grad_norm": 0.60546875, |
| "learning_rate": 9.075555555555556e-06, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9897124897688627, |
| "num_tokens": 42368442.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.05193082329351455, |
| "epoch": 2.384, |
| "grad_norm": 0.77734375, |
| "learning_rate": 8.986666666666666e-06, |
| "loss": 0.0337, |
| "mean_token_accuracy": 0.99007040373981, |
| "num_tokens": 42663926.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.05398317989893257, |
| "epoch": 2.4, |
| "grad_norm": 0.8046875, |
| "learning_rate": 8.897777777777779e-06, |
| "loss": 0.035, |
| "mean_token_accuracy": 0.989552128314972, |
| "num_tokens": 42946351.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.2435439591407776, |
| "eval_chemistry_loss": 1.4468469619750977, |
| "eval_chemistry_mean_token_accuracy": 0.7765617952346802, |
| "eval_chemistry_num_tokens": 42946351.0, |
| "eval_chemistry_runtime": 34.6793, |
| "eval_chemistry_samples_per_second": 14.418, |
| "eval_chemistry_steps_per_second": 3.604, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.18436058980226516, |
| "eval_math_loss": 1.2825661897659302, |
| "eval_math_mean_token_accuracy": 0.8150607142448425, |
| "eval_math_num_tokens": 42946351.0, |
| "eval_math_runtime": 35.4927, |
| "eval_math_samples_per_second": 14.087, |
| "eval_math_steps_per_second": 3.522, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.05408466912340373, |
| "epoch": 2.416, |
| "grad_norm": 0.9296875, |
| "learning_rate": 8.80888888888889e-06, |
| "loss": 0.0349, |
| "mean_token_accuracy": 0.9895286500453949, |
| "num_tokens": 43227782.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.052965849777683614, |
| "epoch": 2.432, |
| "grad_norm": 1.0625, |
| "learning_rate": 8.720000000000001e-06, |
| "loss": 0.0353, |
| "mean_token_accuracy": 0.9892889097332954, |
| "num_tokens": 43511939.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.051178892608731985, |
| "epoch": 2.448, |
| "grad_norm": 0.6875, |
| "learning_rate": 8.631111111111112e-06, |
| "loss": 0.0329, |
| "mean_token_accuracy": 0.9901369903236628, |
| "num_tokens": 43811005.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.05202908001374453, |
| "epoch": 2.464, |
| "grad_norm": 0.72265625, |
| "learning_rate": 8.542222222222222e-06, |
| "loss": 0.0339, |
| "mean_token_accuracy": 0.9899697575718165, |
| "num_tokens": 44106223.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.05558811663649976, |
| "epoch": 2.48, |
| "grad_norm": 0.76171875, |
| "learning_rate": 8.453333333333334e-06, |
| "loss": 0.0361, |
| "mean_token_accuracy": 0.9893935658037663, |
| "num_tokens": 44379095.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.05247082086279988, |
| "epoch": 2.496, |
| "grad_norm": 0.69921875, |
| "learning_rate": 8.364444444444445e-06, |
| "loss": 0.0346, |
| "mean_token_accuracy": 0.9898388765752315, |
| "num_tokens": 44670316.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.05363391661085189, |
| "epoch": 2.512, |
| "grad_norm": 0.81640625, |
| "learning_rate": 8.275555555555557e-06, |
| "loss": 0.0352, |
| "mean_token_accuracy": 0.9896232932806015, |
| "num_tokens": 44952877.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.05272995062405243, |
| "epoch": 2.528, |
| "grad_norm": 0.68359375, |
| "learning_rate": 8.186666666666667e-06, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9896573759615421, |
| "num_tokens": 45237653.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.05119302978273481, |
| "epoch": 2.544, |
| "grad_norm": 0.75390625, |
| "learning_rate": 8.09777777777778e-06, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9897664703428746, |
| "num_tokens": 45524866.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.05278779338113963, |
| "epoch": 2.56, |
| "grad_norm": 0.6796875, |
| "learning_rate": 8.00888888888889e-06, |
| "loss": 0.0356, |
| "mean_token_accuracy": 0.9893198136240244, |
| "num_tokens": 45803783.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 0.24356829929351806, |
| "eval_chemistry_loss": 1.449495792388916, |
| "eval_chemistry_mean_token_accuracy": 0.7762749667167663, |
| "eval_chemistry_num_tokens": 45803783.0, |
| "eval_chemistry_runtime": 34.6626, |
| "eval_chemistry_samples_per_second": 14.425, |
| "eval_chemistry_steps_per_second": 3.606, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.18325595593452454, |
| "eval_math_loss": 1.2849191427230835, |
| "eval_math_mean_token_accuracy": 0.8152834696769714, |
| "eval_math_num_tokens": 45803783.0, |
| "eval_math_runtime": 35.702, |
| "eval_math_samples_per_second": 14.005, |
| "eval_math_steps_per_second": 3.501, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.05134590875823051, |
| "epoch": 2.576, |
| "grad_norm": 0.7265625, |
| "learning_rate": 7.92e-06, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9896847944706678, |
| "num_tokens": 46090713.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.05221337205730379, |
| "epoch": 2.592, |
| "grad_norm": 0.7109375, |
| "learning_rate": 7.831111111111111e-06, |
| "loss": 0.0344, |
| "mean_token_accuracy": 0.9897220823913813, |
| "num_tokens": 46384084.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.0540282168192789, |
| "epoch": 2.608, |
| "grad_norm": 0.79296875, |
| "learning_rate": 7.742222222222223e-06, |
| "loss": 0.0349, |
| "mean_token_accuracy": 0.9895514253526926, |
| "num_tokens": 46667552.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.05048422114923597, |
| "epoch": 2.624, |
| "grad_norm": 0.70703125, |
| "learning_rate": 7.653333333333333e-06, |
| "loss": 0.0331, |
| "mean_token_accuracy": 0.989990346878767, |
| "num_tokens": 46961124.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.05211569468956441, |
| "epoch": 2.64, |
| "grad_norm": 0.6796875, |
| "learning_rate": 7.564444444444446e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.9895898640155792, |
| "num_tokens": 47245326.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.05267454255372286, |
| "epoch": 2.656, |
| "grad_norm": 0.68359375, |
| "learning_rate": 7.475555555555556e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.989424579590559, |
| "num_tokens": 47527347.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.05313762115547434, |
| "epoch": 2.672, |
| "grad_norm": 0.95703125, |
| "learning_rate": 7.386666666666667e-06, |
| "loss": 0.0354, |
| "mean_token_accuracy": 0.9894337601959705, |
| "num_tokens": 47808167.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.05315034419763833, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 0.62890625, |
| "learning_rate": 7.297777777777778e-06, |
| "loss": 0.0349, |
| "mean_token_accuracy": 0.9893308505415916, |
| "num_tokens": 48092035.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.050815221259836105, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 0.73828125, |
| "learning_rate": 7.20888888888889e-06, |
| "loss": 0.0335, |
| "mean_token_accuracy": 0.9900240320712328, |
| "num_tokens": 48386231.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.05376783083193004, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 0.76953125, |
| "learning_rate": 7.1200000000000004e-06, |
| "loss": 0.036, |
| "mean_token_accuracy": 0.9892230879515409, |
| "num_tokens": 48663262.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 0.24340871345996856, |
| "eval_chemistry_loss": 1.45062255859375, |
| "eval_chemistry_mean_token_accuracy": 0.7765913362503052, |
| "eval_chemistry_num_tokens": 48663262.0, |
| "eval_chemistry_runtime": 34.6595, |
| "eval_chemistry_samples_per_second": 14.426, |
| "eval_chemistry_steps_per_second": 3.607, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.18342817682027818, |
| "eval_math_loss": 1.286737322807312, |
| "eval_math_mean_token_accuracy": 0.8153419017791748, |
| "eval_math_num_tokens": 48663262.0, |
| "eval_math_runtime": 35.697, |
| "eval_math_samples_per_second": 14.007, |
| "eval_math_steps_per_second": 3.502, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.05238793781027198, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 0.73046875, |
| "learning_rate": 7.031111111111112e-06, |
| "loss": 0.0343, |
| "mean_token_accuracy": 0.9897935677319765, |
| "num_tokens": 48949456.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.05296056693186983, |
| "epoch": 2.752, |
| "grad_norm": 0.75390625, |
| "learning_rate": 6.942222222222222e-06, |
| "loss": 0.0345, |
| "mean_token_accuracy": 0.9895687259733676, |
| "num_tokens": 49235751.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.051222407980822025, |
| "epoch": 2.768, |
| "grad_norm": 0.62109375, |
| "learning_rate": 6.853333333333334e-06, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9898298654705286, |
| "num_tokens": 49524628.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.05459372613113374, |
| "epoch": 2.784, |
| "grad_norm": 0.69140625, |
| "learning_rate": 6.764444444444445e-06, |
| "loss": 0.0358, |
| "mean_token_accuracy": 0.9893500994890928, |
| "num_tokens": 49799604.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.05316627144347876, |
| "epoch": 2.8, |
| "grad_norm": 0.578125, |
| "learning_rate": 6.675555555555556e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.9896351356059313, |
| "num_tokens": 50079307.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.05220057033002377, |
| "epoch": 2.816, |
| "grad_norm": 0.6796875, |
| "learning_rate": 6.5866666666666666e-06, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9897652886807918, |
| "num_tokens": 50372216.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.05233512815320864, |
| "epoch": 2.832, |
| "grad_norm": 0.6875, |
| "learning_rate": 6.497777777777779e-06, |
| "loss": 0.0343, |
| "mean_token_accuracy": 0.9895316615700722, |
| "num_tokens": 50657211.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.0521509436191991, |
| "epoch": 2.848, |
| "grad_norm": 0.640625, |
| "learning_rate": 6.408888888888889e-06, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.989656625688076, |
| "num_tokens": 50946503.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.05292182555422187, |
| "epoch": 2.864, |
| "grad_norm": 0.81640625, |
| "learning_rate": 6.3200000000000005e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.9893360774964094, |
| "num_tokens": 51229298.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.05223811571486294, |
| "epoch": 2.88, |
| "grad_norm": 0.57421875, |
| "learning_rate": 6.231111111111111e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.989601906761527, |
| "num_tokens": 51509622.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 0.24320586144924164, |
| "eval_chemistry_loss": 1.4516633749008179, |
| "eval_chemistry_mean_token_accuracy": 0.776555242061615, |
| "eval_chemistry_num_tokens": 51509622.0, |
| "eval_chemistry_runtime": 34.6698, |
| "eval_chemistry_samples_per_second": 14.422, |
| "eval_chemistry_steps_per_second": 3.605, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.18338554829359055, |
| "eval_math_loss": 1.28714120388031, |
| "eval_math_mean_token_accuracy": 0.8153921165466309, |
| "eval_math_num_tokens": 51509622.0, |
| "eval_math_runtime": 35.7007, |
| "eval_math_samples_per_second": 14.005, |
| "eval_math_steps_per_second": 3.501, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.053149703633971514, |
| "epoch": 2.896, |
| "grad_norm": 0.56640625, |
| "learning_rate": 6.142222222222223e-06, |
| "loss": 0.0348, |
| "mean_token_accuracy": 0.9895217914134264, |
| "num_tokens": 51790856.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.05247251965338364, |
| "epoch": 2.912, |
| "grad_norm": 0.67578125, |
| "learning_rate": 6.0533333333333335e-06, |
| "loss": 0.0344, |
| "mean_token_accuracy": 0.9898622412234545, |
| "num_tokens": 52074393.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.05151027952088043, |
| "epoch": 2.928, |
| "grad_norm": 0.76953125, |
| "learning_rate": 5.964444444444445e-06, |
| "loss": 0.034, |
| "mean_token_accuracy": 0.9896991837769746, |
| "num_tokens": 52366467.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.051106683211401105, |
| "epoch": 2.944, |
| "grad_norm": 0.56640625, |
| "learning_rate": 5.875555555555556e-06, |
| "loss": 0.0341, |
| "mean_token_accuracy": 0.9898365925997495, |
| "num_tokens": 52657863.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.0515750922029838, |
| "epoch": 2.96, |
| "grad_norm": 0.7734375, |
| "learning_rate": 5.7866666666666674e-06, |
| "loss": 0.0343, |
| "mean_token_accuracy": 0.9897553384304046, |
| "num_tokens": 52947183.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.05394409007858485, |
| "epoch": 2.976, |
| "grad_norm": 0.734375, |
| "learning_rate": 5.697777777777779e-06, |
| "loss": 0.035, |
| "mean_token_accuracy": 0.989257312566042, |
| "num_tokens": 53226984.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.051608621410559864, |
| "epoch": 2.992, |
| "grad_norm": 0.66796875, |
| "learning_rate": 5.608888888888889e-06, |
| "loss": 0.0339, |
| "mean_token_accuracy": 0.9897487934678793, |
| "num_tokens": 53515204.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.050469014700502156, |
| "epoch": 3.008, |
| "grad_norm": 0.30078125, |
| "learning_rate": 5.5200000000000005e-06, |
| "loss": 0.0328, |
| "mean_token_accuracy": 0.9901252970099449, |
| "num_tokens": 53800592.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.04760575578548014, |
| "epoch": 3.024, |
| "grad_norm": 0.263671875, |
| "learning_rate": 5.431111111111112e-06, |
| "loss": 0.0301, |
| "mean_token_accuracy": 0.9910080127418042, |
| "num_tokens": 54092017.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.04652785587823018, |
| "epoch": 3.04, |
| "grad_norm": 0.279296875, |
| "learning_rate": 5.342222222222223e-06, |
| "loss": 0.0301, |
| "mean_token_accuracy": 0.9910020593553781, |
| "num_tokens": 54380158.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 0.24179766488075255, |
| "eval_chemistry_loss": 1.4621965885162354, |
| "eval_chemistry_mean_token_accuracy": 0.7764029130935669, |
| "eval_chemistry_num_tokens": 54380158.0, |
| "eval_chemistry_runtime": 34.6754, |
| "eval_chemistry_samples_per_second": 14.419, |
| "eval_chemistry_steps_per_second": 3.605, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.18078194934129715, |
| "eval_math_loss": 1.2997863292694092, |
| "eval_math_mean_token_accuracy": 0.8158343658447266, |
| "eval_math_num_tokens": 54380158.0, |
| "eval_math_runtime": 35.7053, |
| "eval_math_samples_per_second": 14.004, |
| "eval_math_steps_per_second": 3.501, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.04787755131255835, |
| "epoch": 3.056, |
| "grad_norm": 0.3671875, |
| "learning_rate": 5.2533333333333336e-06, |
| "loss": 0.0313, |
| "mean_token_accuracy": 0.9905039332807064, |
| "num_tokens": 54658093.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.04592872656648979, |
| "epoch": 3.072, |
| "grad_norm": 0.30859375, |
| "learning_rate": 5.164444444444445e-06, |
| "loss": 0.0299, |
| "mean_token_accuracy": 0.9909734159708024, |
| "num_tokens": 54949938.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.04627904891967773, |
| "epoch": 3.088, |
| "grad_norm": 0.33984375, |
| "learning_rate": 5.075555555555556e-06, |
| "loss": 0.0305, |
| "mean_token_accuracy": 0.9907639529556036, |
| "num_tokens": 55234643.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.047028734674677254, |
| "epoch": 3.104, |
| "grad_norm": 0.3828125, |
| "learning_rate": 4.986666666666667e-06, |
| "loss": 0.0307, |
| "mean_token_accuracy": 0.9904829286038875, |
| "num_tokens": 55519679.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.04583920445293188, |
| "epoch": 3.12, |
| "grad_norm": 0.3203125, |
| "learning_rate": 4.897777777777778e-06, |
| "loss": 0.0297, |
| "mean_token_accuracy": 0.9909133236855269, |
| "num_tokens": 55812397.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.046013057266827674, |
| "epoch": 3.136, |
| "grad_norm": 0.94140625, |
| "learning_rate": 4.808888888888889e-06, |
| "loss": 0.03, |
| "mean_token_accuracy": 0.9908631198108196, |
| "num_tokens": 56101550.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.04605525400256738, |
| "epoch": 3.152, |
| "grad_norm": 0.384765625, |
| "learning_rate": 4.7200000000000005e-06, |
| "loss": 0.0304, |
| "mean_token_accuracy": 0.9907284393906594, |
| "num_tokens": 56387662.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.045735701138619335, |
| "epoch": 3.168, |
| "grad_norm": 0.369140625, |
| "learning_rate": 4.631111111111111e-06, |
| "loss": 0.0305, |
| "mean_token_accuracy": 0.990792403742671, |
| "num_tokens": 56673426.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.04651033441768959, |
| "epoch": 3.184, |
| "grad_norm": 0.408203125, |
| "learning_rate": 4.542222222222223e-06, |
| "loss": 0.0308, |
| "mean_token_accuracy": 0.9905624877661466, |
| "num_tokens": 56957044.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.046006057260092346, |
| "epoch": 3.2, |
| "grad_norm": 0.390625, |
| "learning_rate": 4.453333333333334e-06, |
| "loss": 0.0298, |
| "mean_token_accuracy": 0.990779859945178, |
| "num_tokens": 57250097.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 0.24037811040878296, |
| "eval_chemistry_loss": 1.4743051528930664, |
| "eval_chemistry_mean_token_accuracy": 0.7761513013839721, |
| "eval_chemistry_num_tokens": 57250097.0, |
| "eval_chemistry_runtime": 34.67, |
| "eval_chemistry_samples_per_second": 14.422, |
| "eval_chemistry_steps_per_second": 3.605, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.17890368604660034, |
| "eval_math_loss": 1.3124377727508545, |
| "eval_math_mean_token_accuracy": 0.8159590802192688, |
| "eval_math_num_tokens": 57250097.0, |
| "eval_math_runtime": 35.7969, |
| "eval_math_samples_per_second": 13.968, |
| "eval_math_steps_per_second": 3.492, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 2500, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.343629994565034e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
