| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.5590297095805052, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.1043419890105723, |
| "epoch": 0.017803493935684877, |
| "grad_norm": 10.4375, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 1.8349, |
| "mean_token_accuracy": 0.6576539978384972, |
| "num_tokens": 259048.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.0772266678512097, |
| "epoch": 0.035606987871369754, |
| "grad_norm": 9.125, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 1.7708, |
| "mean_token_accuracy": 0.663632657751441, |
| "num_tokens": 532019.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.1068850204348564, |
| "epoch": 0.053410481807054634, |
| "grad_norm": 8.0, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 1.7509, |
| "mean_token_accuracy": 0.6660936743021011, |
| "num_tokens": 807220.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.1769584748893975, |
| "epoch": 0.07121397574273951, |
| "grad_norm": 6.03125, |
| "learning_rate": 2.6e-06, |
| "loss": 1.7149, |
| "mean_token_accuracy": 0.6657536435872317, |
| "num_tokens": 1068429.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.2056159820407628, |
| "epoch": 0.08901746967842439, |
| "grad_norm": 4.8125, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 1.5855, |
| "mean_token_accuracy": 0.6780260927975178, |
| "num_tokens": 1336698.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.2256702020764352, |
| "epoch": 0.10682096361410927, |
| "grad_norm": 3.171875, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 1.4638, |
| "mean_token_accuracy": 0.6949976272881031, |
| "num_tokens": 1608574.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.2453045435249805, |
| "epoch": 0.12462445754979415, |
| "grad_norm": 2.015625, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 1.3582, |
| "mean_token_accuracy": 0.7114116825163365, |
| "num_tokens": 1872747.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.1628358859568835, |
| "epoch": 0.14242795148547902, |
| "grad_norm": 1.5546875, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 1.259, |
| "mean_token_accuracy": 0.7274526257067919, |
| "num_tokens": 2133173.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.015030711889267, |
| "epoch": 0.1602314454211639, |
| "grad_norm": 1.484375, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 1.141, |
| "mean_token_accuracy": 0.7500751964747906, |
| "num_tokens": 2406145.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.9836219865828753, |
| "epoch": 0.17803493935684878, |
| "grad_norm": 1.40625, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 1.1206, |
| "mean_token_accuracy": 0.7526107206940651, |
| "num_tokens": 2663709.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_biology_entropy": 1.7539519824981689, |
| "eval_biology_loss": 2.223750591278076, |
| "eval_biology_mean_token_accuracy": 0.5860000367164612, |
| "eval_biology_num_tokens": 2663709.0, |
| "eval_biology_runtime": 39.4949, |
| "eval_biology_samples_per_second": 12.66, |
| "eval_biology_steps_per_second": 3.165, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_chemistry_entropy": 0.9367400913238525, |
| "eval_chemistry_loss": 1.1236988306045532, |
| "eval_chemistry_mean_token_accuracy": 0.7581998686790467, |
| "eval_chemistry_num_tokens": 2663709.0, |
| "eval_chemistry_runtime": 49.0416, |
| "eval_chemistry_samples_per_second": 10.195, |
| "eval_chemistry_steps_per_second": 2.549, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_math_entropy": 0.8011676266193389, |
| "eval_math_loss": 1.1609103679656982, |
| "eval_math_mean_token_accuracy": 0.7619935193061829, |
| "eval_math_num_tokens": 2663709.0, |
| "eval_math_runtime": 50.4715, |
| "eval_math_samples_per_second": 9.907, |
| "eval_math_steps_per_second": 2.477, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_physics_entropy": 0.8744560084342956, |
| "eval_physics_loss": 1.0428816080093384, |
| "eval_physics_mean_token_accuracy": 0.7703640584945679, |
| "eval_physics_num_tokens": 2663709.0, |
| "eval_physics_runtime": 58.0003, |
| "eval_physics_samples_per_second": 8.621, |
| "eval_physics_steps_per_second": 2.155, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.9135929390788078, |
| "epoch": 0.19583843329253367, |
| "grad_norm": 1.4140625, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 1.0519, |
| "mean_token_accuracy": 0.764449367672205, |
| "num_tokens": 2934980.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.9051187889650464, |
| "epoch": 0.21364192722821854, |
| "grad_norm": 1.203125, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 1.0212, |
| "mean_token_accuracy": 0.7677608881145715, |
| "num_tokens": 3208702.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8583753045648337, |
| "epoch": 0.2314454211639034, |
| "grad_norm": 1.359375, |
| "learning_rate": 8.6e-06, |
| "loss": 1.008, |
| "mean_token_accuracy": 0.7726358521729708, |
| "num_tokens": 3480119.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.873523542098701, |
| "epoch": 0.2492489150995883, |
| "grad_norm": 1.4296875, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 0.9985, |
| "mean_token_accuracy": 0.7731208562850952, |
| "num_tokens": 3746392.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.8644652301445603, |
| "epoch": 0.26705240903527316, |
| "grad_norm": 1.4921875, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 0.9938, |
| "mean_token_accuracy": 0.7713731925934553, |
| "num_tokens": 4019487.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8324168430641293, |
| "epoch": 0.28485590297095803, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 0.9661, |
| "mean_token_accuracy": 0.7780516821891069, |
| "num_tokens": 4298100.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.8229840690270066, |
| "epoch": 0.30265939690664295, |
| "grad_norm": 1.375, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 0.9472, |
| "mean_token_accuracy": 0.781847920268774, |
| "num_tokens": 4568116.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.8173020841553807, |
| "epoch": 0.3204628908423278, |
| "grad_norm": 1.375, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 0.9483, |
| "mean_token_accuracy": 0.7797277990728617, |
| "num_tokens": 4831334.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8084267199039459, |
| "epoch": 0.3382663847780127, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 0.9229, |
| "mean_token_accuracy": 0.7864153161644936, |
| "num_tokens": 5103301.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.8149868924170732, |
| "epoch": 0.35606987871369755, |
| "grad_norm": 1.125, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 0.9542, |
| "mean_token_accuracy": 0.7797067802399397, |
| "num_tokens": 5362915.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_biology_entropy": 1.6254518747329711, |
| "eval_biology_loss": 1.95707106590271, |
| "eval_biology_mean_token_accuracy": 0.6195439734458923, |
| "eval_biology_num_tokens": 5362915.0, |
| "eval_biology_runtime": 39.4088, |
| "eval_biology_samples_per_second": 12.688, |
| "eval_biology_steps_per_second": 3.172, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_chemistry_entropy": 0.8238337967395782, |
| "eval_chemistry_loss": 0.9596245288848877, |
| "eval_chemistry_mean_token_accuracy": 0.7839277501106262, |
| "eval_chemistry_num_tokens": 5362915.0, |
| "eval_chemistry_runtime": 49.0715, |
| "eval_chemistry_samples_per_second": 10.189, |
| "eval_chemistry_steps_per_second": 2.547, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_math_entropy": 0.7457812407016754, |
| "eval_math_loss": 1.055013656616211, |
| "eval_math_mean_token_accuracy": 0.775778597354889, |
| "eval_math_num_tokens": 5362915.0, |
| "eval_math_runtime": 50.4466, |
| "eval_math_samples_per_second": 9.911, |
| "eval_math_steps_per_second": 2.478, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_physics_entropy": 0.8023798677921296, |
| "eval_physics_loss": 0.9094064235687256, |
| "eval_physics_mean_token_accuracy": 0.7911213874816895, |
| "eval_physics_num_tokens": 5362915.0, |
| "eval_physics_runtime": 58.0142, |
| "eval_physics_samples_per_second": 8.619, |
| "eval_physics_steps_per_second": 2.155, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7771274181082845, |
| "epoch": 0.3738733726493824, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 0.9062, |
| "mean_token_accuracy": 0.7895921036601067, |
| "num_tokens": 5625455.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.7874576786532999, |
| "epoch": 0.39167686658506734, |
| "grad_norm": 1.140625, |
| "learning_rate": 1.46e-05, |
| "loss": 0.9095, |
| "mean_token_accuracy": 0.7898499093949795, |
| "num_tokens": 5890806.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7605824310332536, |
| "epoch": 0.4094803605207522, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 0.8848, |
| "mean_token_accuracy": 0.7932682719081641, |
| "num_tokens": 6158751.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7746684672310948, |
| "epoch": 0.4272838544564371, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 0.8878, |
| "mean_token_accuracy": 0.7916001878678799, |
| "num_tokens": 6423747.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.780491204932332, |
| "epoch": 0.44508734839212194, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.66e-05, |
| "loss": 0.9032, |
| "mean_token_accuracy": 0.7888783302158118, |
| "num_tokens": 6684517.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.753203121572733, |
| "epoch": 0.4628908423278068, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 0.8825, |
| "mean_token_accuracy": 0.7946223251521587, |
| "num_tokens": 6945659.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7606851188465953, |
| "epoch": 0.48069433626349173, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 0.883, |
| "mean_token_accuracy": 0.7942271258682012, |
| "num_tokens": 7217797.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7277190953493118, |
| "epoch": 0.4984978301991766, |
| "grad_norm": 1.1328125, |
| "learning_rate": 1.86e-05, |
| "loss": 0.8361, |
| "mean_token_accuracy": 0.8006175145506859, |
| "num_tokens": 7498182.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.7577250253409147, |
| "epoch": 0.5163013241348615, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 0.8779, |
| "mean_token_accuracy": 0.7946350980550051, |
| "num_tokens": 7761686.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7584944000467658, |
| "epoch": 0.5341048180705463, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 0.8804, |
| "mean_token_accuracy": 0.7928748667240143, |
| "num_tokens": 8026923.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_biology_entropy": 1.5056169233322143, |
| "eval_biology_loss": 1.85715913772583, |
| "eval_biology_mean_token_accuracy": 0.63224751329422, |
| "eval_biology_num_tokens": 8026923.0, |
| "eval_biology_runtime": 39.4557, |
| "eval_biology_samples_per_second": 12.672, |
| "eval_biology_steps_per_second": 3.168, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_chemistry_entropy": 0.7292679617404938, |
| "eval_chemistry_loss": 0.897660493850708, |
| "eval_chemistry_mean_token_accuracy": 0.7952092924118042, |
| "eval_chemistry_num_tokens": 8026923.0, |
| "eval_chemistry_runtime": 49.0708, |
| "eval_chemistry_samples_per_second": 10.189, |
| "eval_chemistry_steps_per_second": 2.547, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_math_entropy": 0.6854351389408112, |
| "eval_math_loss": 1.022861361503601, |
| "eval_math_mean_token_accuracy": 0.782105106830597, |
| "eval_math_num_tokens": 8026923.0, |
| "eval_math_runtime": 50.5137, |
| "eval_math_samples_per_second": 9.898, |
| "eval_math_steps_per_second": 2.475, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_physics_entropy": 0.7186625280380249, |
| "eval_physics_loss": 0.8633946180343628, |
| "eval_physics_mean_token_accuracy": 0.7989518413543701, |
| "eval_physics_num_tokens": 8026923.0, |
| "eval_physics_runtime": 58.023, |
| "eval_physics_samples_per_second": 8.617, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.739030777104199, |
| "epoch": 0.5519083120062312, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 0.8602, |
| "mean_token_accuracy": 0.7965835016220808, |
| "num_tokens": 8301506.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.7407136702910065, |
| "epoch": 0.5697118059419161, |
| "grad_norm": 0.98828125, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 0.8705, |
| "mean_token_accuracy": 0.7978284440934658, |
| "num_tokens": 8559126.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.7342625459656119, |
| "epoch": 0.587515299877601, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 0.8624, |
| "mean_token_accuracy": 0.7968722756952047, |
| "num_tokens": 8821088.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.7211404331028461, |
| "epoch": 0.6053187938132859, |
| "grad_norm": 1.1640625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 0.8405, |
| "mean_token_accuracy": 0.8016587898135186, |
| "num_tokens": 9089866.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.7180757647380233, |
| "epoch": 0.6231222877489707, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 0.8443, |
| "mean_token_accuracy": 0.7990191083401441, |
| "num_tokens": 9358866.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.7139529844745993, |
| "epoch": 0.6409257816846556, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 0.828, |
| "mean_token_accuracy": 0.80230086222291, |
| "num_tokens": 9629977.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.7079251240938902, |
| "epoch": 0.6587292756203404, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 0.8268, |
| "mean_token_accuracy": 0.8037587519735098, |
| "num_tokens": 9897757.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.692839514836669, |
| "epoch": 0.6765327695560254, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 0.8141, |
| "mean_token_accuracy": 0.8078057061880827, |
| "num_tokens": 10168367.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.6842571076005697, |
| "epoch": 0.6943362634917103, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 0.8014, |
| "mean_token_accuracy": 0.8098865192383528, |
| "num_tokens": 10439039.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.6961875386536122, |
| "epoch": 0.7121397574273951, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 0.8155, |
| "mean_token_accuracy": 0.80610904507339, |
| "num_tokens": 10709549.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_biology_entropy": 1.507101933479309, |
| "eval_biology_loss": 1.8014453649520874, |
| "eval_biology_mean_token_accuracy": 0.6401848444938659, |
| "eval_biology_num_tokens": 10709549.0, |
| "eval_biology_runtime": 39.406, |
| "eval_biology_samples_per_second": 12.688, |
| "eval_biology_steps_per_second": 3.172, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_chemistry_entropy": 0.7235584797859191, |
| "eval_chemistry_loss": 0.8613554239273071, |
| "eval_chemistry_mean_token_accuracy": 0.8018820552825928, |
| "eval_chemistry_num_tokens": 10709549.0, |
| "eval_chemistry_runtime": 49.0945, |
| "eval_chemistry_samples_per_second": 10.184, |
| "eval_chemistry_steps_per_second": 2.546, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_math_entropy": 0.6896943941116332, |
| "eval_math_loss": 1.000999927520752, |
| "eval_math_mean_token_accuracy": 0.7846557745933532, |
| "eval_math_num_tokens": 10709549.0, |
| "eval_math_runtime": 50.5019, |
| "eval_math_samples_per_second": 9.901, |
| "eval_math_steps_per_second": 2.475, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_physics_entropy": 0.7171106839179993, |
| "eval_physics_loss": 0.8377377986907959, |
| "eval_physics_mean_token_accuracy": 0.8029740490913391, |
| "eval_physics_num_tokens": 10709549.0, |
| "eval_physics_runtime": 58.0325, |
| "eval_physics_samples_per_second": 8.616, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.6906447313725949, |
| "epoch": 0.72994325136308, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 0.8102, |
| "mean_token_accuracy": 0.8085145801305771, |
| "num_tokens": 10976011.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.69708113335073, |
| "epoch": 0.7477467452987648, |
| "grad_norm": 0.9765625, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 0.8109, |
| "mean_token_accuracy": 0.806747506186366, |
| "num_tokens": 11245251.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.7070492304861545, |
| "epoch": 0.7655502392344498, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 0.8228, |
| "mean_token_accuracy": 0.8028202462941408, |
| "num_tokens": 11518514.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.7102027812972664, |
| "epoch": 0.7833537331701347, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 0.8262, |
| "mean_token_accuracy": 0.8028789341449738, |
| "num_tokens": 11787117.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.7030388584360481, |
| "epoch": 0.8011572271058195, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 0.8174, |
| "mean_token_accuracy": 0.8053972851485014, |
| "num_tokens": 12050993.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.7048017742112279, |
| "epoch": 0.8189607210415044, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 0.8205, |
| "mean_token_accuracy": 0.8050019450485706, |
| "num_tokens": 12311688.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.6749309128150344, |
| "epoch": 0.8367642149771892, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 0.7839, |
| "mean_token_accuracy": 0.8101571343839169, |
| "num_tokens": 12598408.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.680501715093851, |
| "epoch": 0.8545677089128741, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 0.8005, |
| "mean_token_accuracy": 0.8100913222879171, |
| "num_tokens": 12869642.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.6734585866332055, |
| "epoch": 0.8723712028485591, |
| "grad_norm": 0.92578125, |
| "learning_rate": 1.86e-05, |
| "loss": 0.787, |
| "mean_token_accuracy": 0.8114359531551599, |
| "num_tokens": 13139555.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.7050674421712756, |
| "epoch": 0.8901746967842439, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 0.8282, |
| "mean_token_accuracy": 0.8043184716254472, |
| "num_tokens": 13396545.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_biology_entropy": 1.4803909873962402, |
| "eval_biology_loss": 1.7652114629745483, |
| "eval_biology_mean_token_accuracy": 0.6454687314033508, |
| "eval_biology_num_tokens": 13396545.0, |
| "eval_biology_runtime": 39.3799, |
| "eval_biology_samples_per_second": 12.697, |
| "eval_biology_steps_per_second": 3.174, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_chemistry_entropy": 0.7009948241710663, |
| "eval_chemistry_loss": 0.8403192758560181, |
| "eval_chemistry_mean_token_accuracy": 0.8064807448387146, |
| "eval_chemistry_num_tokens": 13396545.0, |
| "eval_chemistry_runtime": 49.2607, |
| "eval_chemistry_samples_per_second": 10.15, |
| "eval_chemistry_steps_per_second": 2.538, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_math_entropy": 0.6736877272129059, |
| "eval_math_loss": 0.9955361485481262, |
| "eval_math_mean_token_accuracy": 0.7851687602996826, |
| "eval_math_num_tokens": 13396545.0, |
| "eval_math_runtime": 50.5877, |
| "eval_math_samples_per_second": 9.884, |
| "eval_math_steps_per_second": 2.471, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_physics_entropy": 0.6975537137985229, |
| "eval_physics_loss": 0.8254684209823608, |
| "eval_physics_mean_token_accuracy": 0.8055949606895447, |
| "eval_physics_num_tokens": 13396545.0, |
| "eval_physics_runtime": 58.1535, |
| "eval_physics_samples_per_second": 8.598, |
| "eval_physics_steps_per_second": 2.149, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.6699989970773459, |
| "epoch": 0.9079781907199288, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 0.7858, |
| "mean_token_accuracy": 0.8136365339159966, |
| "num_tokens": 13667857.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.6965480221435427, |
| "epoch": 0.9257816846556136, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 0.8068, |
| "mean_token_accuracy": 0.8083094954490662, |
| "num_tokens": 13932771.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.6731918448582291, |
| "epoch": 0.9435851785912985, |
| "grad_norm": 0.9609375, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 0.7957, |
| "mean_token_accuracy": 0.81204720325768, |
| "num_tokens": 14194632.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.6959722986444831, |
| "epoch": 0.9613886725269835, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 0.8154, |
| "mean_token_accuracy": 0.8060157299041748, |
| "num_tokens": 14457650.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.6704609870910645, |
| "epoch": 0.9791921664626683, |
| "grad_norm": 0.9375, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 0.793, |
| "mean_token_accuracy": 0.8112935304641724, |
| "num_tokens": 14724967.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.693532545492053, |
| "epoch": 0.9969956603983532, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 0.8139, |
| "mean_token_accuracy": 0.8083073332905769, |
| "num_tokens": 14993272.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.6279453858252495, |
| "epoch": 1.014242795148548, |
| "grad_norm": 0.9453125, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 0.745, |
| "mean_token_accuracy": 0.8219076448871243, |
| "num_tokens": 15254936.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.65065772626549, |
| "epoch": 1.0320462890842328, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 0.7716, |
| "mean_token_accuracy": 0.8141439683735371, |
| "num_tokens": 15515007.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.6248815247789026, |
| "epoch": 1.0498497830199176, |
| "grad_norm": 1.1015625, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 0.7409, |
| "mean_token_accuracy": 0.8221572887152433, |
| "num_tokens": 15794203.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6451754650101066, |
| "epoch": 1.0676532769556026, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 0.7597, |
| "mean_token_accuracy": 0.8173915591090918, |
| "num_tokens": 16063664.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_biology_entropy": 1.3715826230049133, |
| "eval_biology_loss": 1.7631210088729858, |
| "eval_biology_mean_token_accuracy": 0.6465545001029969, |
| "eval_biology_num_tokens": 16063664.0, |
| "eval_biology_runtime": 39.4085, |
| "eval_biology_samples_per_second": 12.688, |
| "eval_biology_steps_per_second": 3.172, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_chemistry_entropy": 0.6441384816169738, |
| "eval_chemistry_loss": 0.8298478722572327, |
| "eval_chemistry_mean_token_accuracy": 0.8090244836807251, |
| "eval_chemistry_num_tokens": 16063664.0, |
| "eval_chemistry_runtime": 49.0682, |
| "eval_chemistry_samples_per_second": 10.19, |
| "eval_chemistry_steps_per_second": 2.547, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_math_entropy": 0.6348585722446441, |
| "eval_math_loss": 0.9978653788566589, |
| "eval_math_mean_token_accuracy": 0.7868981838226319, |
| "eval_math_num_tokens": 16063664.0, |
| "eval_math_runtime": 50.4477, |
| "eval_math_samples_per_second": 9.911, |
| "eval_math_steps_per_second": 2.478, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_physics_entropy": 0.6496791577339173, |
| "eval_physics_loss": 0.8208206295967102, |
| "eval_physics_mean_token_accuracy": 0.8067752327919007, |
| "eval_physics_num_tokens": 16063664.0, |
| "eval_physics_runtime": 58.0277, |
| "eval_physics_samples_per_second": 8.617, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6230961512774229, |
| "epoch": 1.0854567708912874, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 0.745, |
| "mean_token_accuracy": 0.8219233963638544, |
| "num_tokens": 16338303.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.6581493113189936, |
| "epoch": 1.1032602648269723, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 0.7699, |
| "mean_token_accuracy": 0.8144335735589265, |
| "num_tokens": 16606302.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.6356614841148257, |
| "epoch": 1.121063758762657, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 0.7491, |
| "mean_token_accuracy": 0.8173594359308481, |
| "num_tokens": 16877071.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.6287711111828684, |
| "epoch": 1.138867252698342, |
| "grad_norm": 1.0546875, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 0.7491, |
| "mean_token_accuracy": 0.8200664207339287, |
| "num_tokens": 17145775.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6477484388276935, |
| "epoch": 1.156670746634027, |
| "grad_norm": 1.0546875, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 0.7566, |
| "mean_token_accuracy": 0.8180534493178129, |
| "num_tokens": 17410775.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.6610335078090429, |
| "epoch": 1.1744742405697117, |
| "grad_norm": 1.1640625, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 0.7746, |
| "mean_token_accuracy": 0.8117418963462114, |
| "num_tokens": 17677791.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.6399538004770875, |
| "epoch": 1.1922777345053968, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 0.748, |
| "mean_token_accuracy": 0.8180656544864178, |
| "num_tokens": 17949441.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6249047636985778, |
| "epoch": 1.2100812284410816, |
| "grad_norm": 1.0, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 0.7445, |
| "mean_token_accuracy": 0.819851553067565, |
| "num_tokens": 18217742.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6427486116066575, |
| "epoch": 1.2278847223767664, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 0.7565, |
| "mean_token_accuracy": 0.8179482214152813, |
| "num_tokens": 18487769.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6315130151808261, |
| "epoch": 1.2456882163124514, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 0.7467, |
| "mean_token_accuracy": 0.8189041160047055, |
| "num_tokens": 18760296.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_biology_entropy": 1.3494874267578125, |
| "eval_biology_loss": 1.7597026824951172, |
| "eval_biology_mean_token_accuracy": 0.6474382462501526, |
| "eval_biology_num_tokens": 18760296.0, |
| "eval_biology_runtime": 39.4416, |
| "eval_biology_samples_per_second": 12.677, |
| "eval_biology_steps_per_second": 3.169, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_chemistry_entropy": 0.6433191192150116, |
| "eval_chemistry_loss": 0.8230161070823669, |
| "eval_chemistry_mean_token_accuracy": 0.8107681026458741, |
| "eval_chemistry_num_tokens": 18760296.0, |
| "eval_chemistry_runtime": 49.1024, |
| "eval_chemistry_samples_per_second": 10.183, |
| "eval_chemistry_steps_per_second": 2.546, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_math_entropy": 0.6405739262104034, |
| "eval_math_loss": 0.9985885620117188, |
| "eval_math_mean_token_accuracy": 0.7867832021713257, |
| "eval_math_num_tokens": 18760296.0, |
| "eval_math_runtime": 50.4916, |
| "eval_math_samples_per_second": 9.903, |
| "eval_math_steps_per_second": 2.476, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_physics_entropy": 0.6501080214977264, |
| "eval_physics_loss": 0.8173370361328125, |
| "eval_physics_mean_token_accuracy": 0.8081451225280761, |
| "eval_physics_num_tokens": 18760296.0, |
| "eval_physics_runtime": 58.0347, |
| "eval_physics_samples_per_second": 8.616, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.6334251884371043, |
| "epoch": 1.2634917102481362, |
| "grad_norm": 0.9765625, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 0.7436, |
| "mean_token_accuracy": 0.8211747828871012, |
| "num_tokens": 19035733.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6478446906432509, |
| "epoch": 1.281295204183821, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 0.7619, |
| "mean_token_accuracy": 0.8154366932809353, |
| "num_tokens": 19311432.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6194965062662959, |
| "epoch": 1.299098698119506, |
| "grad_norm": 0.953125, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 0.7348, |
| "mean_token_accuracy": 0.8232949875295162, |
| "num_tokens": 19593016.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6300735978409648, |
| "epoch": 1.3169021920551909, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 0.7414, |
| "mean_token_accuracy": 0.8206213727593422, |
| "num_tokens": 19860476.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6315095463767648, |
| "epoch": 1.3347056859908757, |
| "grad_norm": 1.171875, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 0.7454, |
| "mean_token_accuracy": 0.8194679979234933, |
| "num_tokens": 20124436.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6350213142111898, |
| "epoch": 1.3525091799265607, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.66e-05, |
| "loss": 0.7551, |
| "mean_token_accuracy": 0.8202624648809433, |
| "num_tokens": 20388027.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6325125509873033, |
| "epoch": 1.3703126738622455, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 0.7501, |
| "mean_token_accuracy": 0.8188550252467394, |
| "num_tokens": 20656076.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.6585695873945951, |
| "epoch": 1.3881161677979303, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 0.7837, |
| "mean_token_accuracy": 0.8139019794762135, |
| "num_tokens": 20911633.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.6532254891470075, |
| "epoch": 1.4059196617336152, |
| "grad_norm": 0.9921875, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 0.7764, |
| "mean_token_accuracy": 0.8136138409376145, |
| "num_tokens": 21176227.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6332234110683203, |
| "epoch": 1.4237231556693, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 0.7509, |
| "mean_token_accuracy": 0.8198547769337893, |
| "num_tokens": 21445242.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_biology_entropy": 1.3427690668106078, |
| "eval_biology_loss": 1.7495665550231934, |
| "eval_biology_mean_token_accuracy": 0.6489976191520691, |
| "eval_biology_num_tokens": 21445242.0, |
| "eval_biology_runtime": 39.4019, |
| "eval_biology_samples_per_second": 12.69, |
| "eval_biology_steps_per_second": 3.172, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_chemistry_entropy": 0.6382337064743042, |
| "eval_chemistry_loss": 0.8153977394104004, |
| "eval_chemistry_mean_token_accuracy": 0.8124490218162537, |
| "eval_chemistry_num_tokens": 21445242.0, |
| "eval_chemistry_runtime": 49.0471, |
| "eval_chemistry_samples_per_second": 10.194, |
| "eval_chemistry_steps_per_second": 2.549, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_math_entropy": 0.635095400094986, |
| "eval_math_loss": 0.9956026673316956, |
| "eval_math_mean_token_accuracy": 0.7870233664512635, |
| "eval_math_num_tokens": 21445242.0, |
| "eval_math_runtime": 50.7413, |
| "eval_math_samples_per_second": 9.854, |
| "eval_math_steps_per_second": 2.463, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_physics_entropy": 0.6485776097774506, |
| "eval_physics_loss": 0.8143745064735413, |
| "eval_physics_mean_token_accuracy": 0.8084757113456726, |
| "eval_physics_num_tokens": 21445242.0, |
| "eval_physics_runtime": 58.0381, |
| "eval_physics_samples_per_second": 8.615, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.6331835286691785, |
| "epoch": 1.441526649604985, |
| "grad_norm": 1.140625, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 0.7506, |
| "mean_token_accuracy": 0.8180669218301773, |
| "num_tokens": 21718455.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6370328038930893, |
| "epoch": 1.4593301435406698, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 0.7572, |
| "mean_token_accuracy": 0.8202964432537556, |
| "num_tokens": 21974076.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.6381731679663062, |
| "epoch": 1.4771336374763546, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 0.7579, |
| "mean_token_accuracy": 0.8184547819197178, |
| "num_tokens": 22241599.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.6486698495224118, |
| "epoch": 1.4949371314120397, |
| "grad_norm": 1.140625, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 0.7624, |
| "mean_token_accuracy": 0.8160121306777001, |
| "num_tokens": 22510725.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6194259492680431, |
| "epoch": 1.5127406253477245, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 0.7469, |
| "mean_token_accuracy": 0.8220851149410009, |
| "num_tokens": 22778514.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6234903791919351, |
| "epoch": 1.5305441192834093, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 0.7359, |
| "mean_token_accuracy": 0.8212783958762884, |
| "num_tokens": 23051230.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.6352918868884444, |
| "epoch": 1.5483476132190943, |
| "grad_norm": 1.0546875, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 0.7591, |
| "mean_token_accuracy": 0.8191397253423929, |
| "num_tokens": 23306327.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6375154785811901, |
| "epoch": 1.5661511071547791, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 0.7523, |
| "mean_token_accuracy": 0.8182045802474022, |
| "num_tokens": 23573127.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6325192734599113, |
| "epoch": 1.583954601090464, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 0.7446, |
| "mean_token_accuracy": 0.8195050850510597, |
| "num_tokens": 23841914.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.633665832132101, |
| "epoch": 1.601758095026149, |
| "grad_norm": 1.1015625, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 0.7591, |
| "mean_token_accuracy": 0.8187031075358391, |
| "num_tokens": 24105114.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_biology_entropy": 1.3434553322792053, |
| "eval_biology_loss": 1.7432700395584106, |
| "eval_biology_mean_token_accuracy": 0.6507211275100708, |
| "eval_biology_num_tokens": 24105114.0, |
| "eval_biology_runtime": 39.4357, |
| "eval_biology_samples_per_second": 12.679, |
| "eval_biology_steps_per_second": 3.17, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_chemistry_entropy": 0.6393454251289368, |
| "eval_chemistry_loss": 0.8100669980049133, |
| "eval_chemistry_mean_token_accuracy": 0.8131857676506042, |
| "eval_chemistry_num_tokens": 24105114.0, |
| "eval_chemistry_runtime": 49.0586, |
| "eval_chemistry_samples_per_second": 10.192, |
| "eval_chemistry_steps_per_second": 2.548, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_math_entropy": 0.6370618710517884, |
| "eval_math_loss": 0.9941329956054688, |
| "eval_math_mean_token_accuracy": 0.7876229991912842, |
| "eval_math_num_tokens": 24105114.0, |
| "eval_math_runtime": 50.5321, |
| "eval_math_samples_per_second": 9.895, |
| "eval_math_steps_per_second": 2.474, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_physics_entropy": 0.6476399068832397, |
| "eval_physics_loss": 0.8108195662498474, |
| "eval_physics_mean_token_accuracy": 0.8089202756881714, |
| "eval_physics_num_tokens": 24105114.0, |
| "eval_physics_runtime": 58.0639, |
| "eval_physics_samples_per_second": 8.611, |
| "eval_physics_steps_per_second": 2.153, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.6105097897350789, |
| "epoch": 1.6195615889618338, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 0.7282, |
| "mean_token_accuracy": 0.8247725363820791, |
| "num_tokens": 24384008.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.6406213022768498, |
| "epoch": 1.6373650828975186, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 0.7526, |
| "mean_token_accuracy": 0.8185552686452866, |
| "num_tokens": 24638205.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.591418044641614, |
| "epoch": 1.6551685768332036, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 0.7101, |
| "mean_token_accuracy": 0.8291535805910826, |
| "num_tokens": 24910104.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.6204024739563465, |
| "epoch": 1.6729720707688884, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 0.7309, |
| "mean_token_accuracy": 0.8225251983851194, |
| "num_tokens": 25181306.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.6261829130351544, |
| "epoch": 1.6907755647045732, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 0.743, |
| "mean_token_accuracy": 0.8214005049318075, |
| "num_tokens": 25449138.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.6261117428541183, |
| "epoch": 1.7085790586402583, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 0.7499, |
| "mean_token_accuracy": 0.8206599947065115, |
| "num_tokens": 25709030.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.636249510012567, |
| "epoch": 1.7263825525759429, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 0.7537, |
| "mean_token_accuracy": 0.8185036793351174, |
| "num_tokens": 25972807.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.6218453321605921, |
| "epoch": 1.744186046511628, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 0.7321, |
| "mean_token_accuracy": 0.8228784658014774, |
| "num_tokens": 26244955.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.5934561615809798, |
| "epoch": 1.761989540447313, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 0.7147, |
| "mean_token_accuracy": 0.8287193164229393, |
| "num_tokens": 26515548.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.6157584773376584, |
| "epoch": 1.7797930343829975, |
| "grad_norm": 1.1015625, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 0.7263, |
| "mean_token_accuracy": 0.8242408398538827, |
| "num_tokens": 26782376.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_biology_entropy": 1.3182690114974975, |
| "eval_biology_loss": 1.7408480644226074, |
| "eval_biology_mean_token_accuracy": 0.650810341835022, |
| "eval_biology_num_tokens": 26782376.0, |
| "eval_biology_runtime": 39.4221, |
| "eval_biology_samples_per_second": 12.683, |
| "eval_biology_steps_per_second": 3.171, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_chemistry_entropy": 0.6269300398826599, |
| "eval_chemistry_loss": 0.8059195876121521, |
| "eval_chemistry_mean_token_accuracy": 0.8141151933670044, |
| "eval_chemistry_num_tokens": 26782376.0, |
| "eval_chemistry_runtime": 49.1405, |
| "eval_chemistry_samples_per_second": 10.175, |
| "eval_chemistry_steps_per_second": 2.544, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_math_entropy": 0.6211173584461213, |
| "eval_math_loss": 0.9962060451507568, |
| "eval_math_mean_token_accuracy": 0.787696545124054, |
| "eval_math_num_tokens": 26782376.0, |
| "eval_math_runtime": 50.557, |
| "eval_math_samples_per_second": 9.89, |
| "eval_math_steps_per_second": 2.472, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_physics_entropy": 0.6375918807983398, |
| "eval_physics_loss": 0.8093506097793579, |
| "eval_physics_mean_token_accuracy": 0.8095529170036316, |
| "eval_physics_num_tokens": 26782376.0, |
| "eval_physics_runtime": 58.1306, |
| "eval_physics_samples_per_second": 8.601, |
| "eval_physics_steps_per_second": 2.15, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.6244443429633975, |
| "epoch": 1.7975965283186826, |
| "grad_norm": 1.15625, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 0.7408, |
| "mean_token_accuracy": 0.8216248027980327, |
| "num_tokens": 27049648.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.6443131631240249, |
| "epoch": 1.8154000222543676, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 0.7608, |
| "mean_token_accuracy": 0.8165751941502094, |
| "num_tokens": 27315061.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.6104631774127484, |
| "epoch": 1.8332035161900522, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.46e-05, |
| "loss": 0.7215, |
| "mean_token_accuracy": 0.8252159222960472, |
| "num_tokens": 27587650.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.6240769168362021, |
| "epoch": 1.8510070101257372, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 0.7475, |
| "mean_token_accuracy": 0.8225702874362468, |
| "num_tokens": 27855458.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.6136018102988601, |
| "epoch": 1.868810504061422, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 0.7309, |
| "mean_token_accuracy": 0.8256569776684046, |
| "num_tokens": 28127703.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.5999570606276393, |
| "epoch": 1.8866139979971068, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 0.712, |
| "mean_token_accuracy": 0.828136184066534, |
| "num_tokens": 28395128.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.6200173202902078, |
| "epoch": 1.9044174919327919, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 0.7373, |
| "mean_token_accuracy": 0.8221532251685858, |
| "num_tokens": 28662344.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.6189345275983215, |
| "epoch": 1.9222209858684767, |
| "grad_norm": 1.140625, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 0.7395, |
| "mean_token_accuracy": 0.8227261707186699, |
| "num_tokens": 28928769.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.6273608855903149, |
| "epoch": 1.9400244798041615, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 0.7448, |
| "mean_token_accuracy": 0.8211823914200067, |
| "num_tokens": 29186214.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.6261450372636318, |
| "epoch": 1.9578279737398465, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 0.7531, |
| "mean_token_accuracy": 0.8194734916090965, |
| "num_tokens": 29446445.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_biology_entropy": 1.346300323009491, |
| "eval_biology_loss": 1.7372664213180542, |
| "eval_biology_mean_token_accuracy": 0.6519479279518128, |
| "eval_biology_num_tokens": 29446445.0, |
| "eval_biology_runtime": 39.4226, |
| "eval_biology_samples_per_second": 12.683, |
| "eval_biology_steps_per_second": 3.171, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_chemistry_entropy": 0.6323098564147949, |
| "eval_chemistry_loss": 0.8011805415153503, |
| "eval_chemistry_mean_token_accuracy": 0.8155805091857911, |
| "eval_chemistry_num_tokens": 29446445.0, |
| "eval_chemistry_runtime": 49.0772, |
| "eval_chemistry_samples_per_second": 10.188, |
| "eval_chemistry_steps_per_second": 2.547, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_math_entropy": 0.630736531496048, |
| "eval_math_loss": 0.9944341778755188, |
| "eval_math_mean_token_accuracy": 0.7879194793701172, |
| "eval_math_num_tokens": 29446445.0, |
| "eval_math_runtime": 50.5369, |
| "eval_math_samples_per_second": 9.894, |
| "eval_math_steps_per_second": 2.473, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_physics_entropy": 0.6453085649013519, |
| "eval_physics_loss": 0.8070665597915649, |
| "eval_physics_mean_token_accuracy": 0.8097171154022217, |
| "eval_physics_num_tokens": 29446445.0, |
| "eval_physics_runtime": 58.1008, |
| "eval_physics_samples_per_second": 8.606, |
| "eval_physics_steps_per_second": 2.151, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.6213501699268817, |
| "epoch": 1.9756314676755313, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 0.7301, |
| "mean_token_accuracy": 0.8246351405978203, |
| "num_tokens": 29711771.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.6162075458094478, |
| "epoch": 1.9934349616112161, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 0.737, |
| "mean_token_accuracy": 0.8235828243196011, |
| "num_tokens": 29979994.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.6002126078451834, |
| "epoch": 2.010682096361411, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 0.7089, |
| "mean_token_accuracy": 0.8300682148625774, |
| "num_tokens": 30240793.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.5583542298525572, |
| "epoch": 2.028485590297096, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 0.6862, |
| "mean_token_accuracy": 0.8347798395901919, |
| "num_tokens": 30503988.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.5757273212075233, |
| "epoch": 2.0462890842327806, |
| "grad_norm": 1.125, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 0.6898, |
| "mean_token_accuracy": 0.8326508197933435, |
| "num_tokens": 30771841.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.5637224279344082, |
| "epoch": 2.0640925781684656, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 0.6795, |
| "mean_token_accuracy": 0.8356141887605191, |
| "num_tokens": 31045037.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.6050490867346525, |
| "epoch": 2.0818960721041506, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 0.7224, |
| "mean_token_accuracy": 0.8252181001007557, |
| "num_tokens": 31299575.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.5962092559784651, |
| "epoch": 2.099699566039835, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 0.7131, |
| "mean_token_accuracy": 0.8261703923344612, |
| "num_tokens": 31565963.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.5632721507921815, |
| "epoch": 2.1175030599755202, |
| "grad_norm": 1.1328125, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 0.6914, |
| "mean_token_accuracy": 0.8345137506723403, |
| "num_tokens": 31834212.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.5934858426451683, |
| "epoch": 2.1353065539112053, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 0.7027, |
| "mean_token_accuracy": 0.8284218475222588, |
| "num_tokens": 32103356.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_biology_entropy": 1.2053067927360535, |
| "eval_biology_loss": 1.7820520401000977, |
| "eval_biology_mean_token_accuracy": 0.650058352470398, |
| "eval_biology_num_tokens": 32103356.0, |
| "eval_biology_runtime": 39.4397, |
| "eval_biology_samples_per_second": 12.678, |
| "eval_biology_steps_per_second": 3.169, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_chemistry_entropy": 0.5787506141662597, |
| "eval_chemistry_loss": 0.8066306114196777, |
| "eval_chemistry_mean_token_accuracy": 0.8150252356529236, |
| "eval_chemistry_num_tokens": 32103356.0, |
| "eval_chemistry_runtime": 49.0527, |
| "eval_chemistry_samples_per_second": 10.193, |
| "eval_chemistry_steps_per_second": 2.548, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_math_entropy": 0.5850761234760284, |
| "eval_math_loss": 1.0127291679382324, |
| "eval_math_mean_token_accuracy": 0.7874734072685242, |
| "eval_math_num_tokens": 32103356.0, |
| "eval_math_runtime": 50.4025, |
| "eval_math_samples_per_second": 9.92, |
| "eval_math_steps_per_second": 2.48, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_physics_entropy": 0.5898979375362396, |
| "eval_physics_loss": 0.8185544610023499, |
| "eval_physics_mean_token_accuracy": 0.8093464169502258, |
| "eval_physics_num_tokens": 32103356.0, |
| "eval_physics_runtime": 57.9316, |
| "eval_physics_samples_per_second": 8.631, |
| "eval_physics_steps_per_second": 2.158, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.581636905297637, |
| "epoch": 2.15311004784689, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 0.6917, |
| "mean_token_accuracy": 0.8308463472872972, |
| "num_tokens": 32374830.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5604663521051407, |
| "epoch": 2.170913541782575, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 0.6806, |
| "mean_token_accuracy": 0.8363849833607674, |
| "num_tokens": 32643479.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.5652653275057673, |
| "epoch": 2.18871703571826, |
| "grad_norm": 1.15625, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 0.6847, |
| "mean_token_accuracy": 0.8347171179950237, |
| "num_tokens": 32908761.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.5932779083028435, |
| "epoch": 2.2065205296539445, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 0.7123, |
| "mean_token_accuracy": 0.8300972994416952, |
| "num_tokens": 33166413.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.5816807165741921, |
| "epoch": 2.2243240235896296, |
| "grad_norm": 1.0546875, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 0.7029, |
| "mean_token_accuracy": 0.8300608668476344, |
| "num_tokens": 33432686.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.5771228864789009, |
| "epoch": 2.242127517525314, |
| "grad_norm": 1.2421875, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 0.6995, |
| "mean_token_accuracy": 0.8321005046367645, |
| "num_tokens": 33693242.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.5507659334689379, |
| "epoch": 2.259931011460999, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 0.662, |
| "mean_token_accuracy": 0.8395966865122318, |
| "num_tokens": 33970820.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.56566769964993, |
| "epoch": 2.277734505396684, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 0.6859, |
| "mean_token_accuracy": 0.8336932327598333, |
| "num_tokens": 34243042.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.5772348834201694, |
| "epoch": 2.295537999332369, |
| "grad_norm": 1.125, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 0.6912, |
| "mean_token_accuracy": 0.8332745991647243, |
| "num_tokens": 34515236.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.5464396445080638, |
| "epoch": 2.313341493268054, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 0.6666, |
| "mean_token_accuracy": 0.8404746253043414, |
| "num_tokens": 34785068.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_biology_entropy": 1.2253249354362488, |
| "eval_biology_loss": 1.7809605598449707, |
| "eval_biology_mean_token_accuracy": 0.6500100207328796, |
| "eval_biology_num_tokens": 34785068.0, |
| "eval_biology_runtime": 39.4361, |
| "eval_biology_samples_per_second": 12.679, |
| "eval_biology_steps_per_second": 3.17, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_chemistry_entropy": 0.5928636178970337, |
| "eval_chemistry_loss": 0.8039361238479614, |
| "eval_chemistry_mean_token_accuracy": 0.8150280647277832, |
| "eval_chemistry_num_tokens": 34785068.0, |
| "eval_chemistry_runtime": 49.0994, |
| "eval_chemistry_samples_per_second": 10.183, |
| "eval_chemistry_steps_per_second": 2.546, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_math_entropy": 0.6018223831653595, |
| "eval_math_loss": 1.0096815824508667, |
| "eval_math_mean_token_accuracy": 0.7871740417480468, |
| "eval_math_num_tokens": 34785068.0, |
| "eval_math_runtime": 50.7658, |
| "eval_math_samples_per_second": 9.849, |
| "eval_math_steps_per_second": 2.462, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_physics_entropy": 0.6088585712909699, |
| "eval_physics_loss": 0.8172869086265564, |
| "eval_physics_mean_token_accuracy": 0.8088493304252624, |
| "eval_physics_num_tokens": 34785068.0, |
| "eval_physics_runtime": 58.0306, |
| "eval_physics_samples_per_second": 8.616, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.5890513943508268, |
| "epoch": 2.331144987203739, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 0.6998, |
| "mean_token_accuracy": 0.8297090869396925, |
| "num_tokens": 35055845.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5705585546791554, |
| "epoch": 2.3489484811394234, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 0.6906, |
| "mean_token_accuracy": 0.8329575475305319, |
| "num_tokens": 35327481.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.5891954906284809, |
| "epoch": 2.3667519750751085, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 0.7044, |
| "mean_token_accuracy": 0.8297746371477842, |
| "num_tokens": 35596362.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5661238888278604, |
| "epoch": 2.3845554690107935, |
| "grad_norm": 1.1796875, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 0.6833, |
| "mean_token_accuracy": 0.8340384036302566, |
| "num_tokens": 35869585.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5856554383412004, |
| "epoch": 2.402358962946478, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 0.6975, |
| "mean_token_accuracy": 0.8312610980123282, |
| "num_tokens": 36134072.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.5633181067183614, |
| "epoch": 2.420162456882163, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 0.6856, |
| "mean_token_accuracy": 0.8359689902514219, |
| "num_tokens": 36406263.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.5795387465506792, |
| "epoch": 2.437965950817848, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 0.6909, |
| "mean_token_accuracy": 0.8334782596677541, |
| "num_tokens": 36676822.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.5749526681378484, |
| "epoch": 2.4557694447535328, |
| "grad_norm": 1.25, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 0.6926, |
| "mean_token_accuracy": 0.832967896386981, |
| "num_tokens": 36942837.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.5685159346088767, |
| "epoch": 2.473572938689218, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 0.6926, |
| "mean_token_accuracy": 0.8322753962129354, |
| "num_tokens": 37210791.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.5581249188631773, |
| "epoch": 2.491376432624903, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 0.6748, |
| "mean_token_accuracy": 0.8378939881920815, |
| "num_tokens": 37485822.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_biology_entropy": 1.2020832428932189, |
| "eval_biology_loss": 1.7927825450897217, |
| "eval_biology_mean_token_accuracy": 0.6494763836860656, |
| "eval_biology_num_tokens": 37485822.0, |
| "eval_biology_runtime": 39.4, |
| "eval_biology_samples_per_second": 12.69, |
| "eval_biology_steps_per_second": 3.173, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_chemistry_entropy": 0.5806494352817535, |
| "eval_chemistry_loss": 0.8034472465515137, |
| "eval_chemistry_mean_token_accuracy": 0.8160974597930908, |
| "eval_chemistry_num_tokens": 37485822.0, |
| "eval_chemistry_runtime": 49.0478, |
| "eval_chemistry_samples_per_second": 10.194, |
| "eval_chemistry_steps_per_second": 2.549, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_math_entropy": 0.5932310831546783, |
| "eval_math_loss": 1.0124026536941528, |
| "eval_math_mean_token_accuracy": 0.7870475478172302, |
| "eval_math_num_tokens": 37485822.0, |
| "eval_math_runtime": 50.4556, |
| "eval_math_samples_per_second": 9.91, |
| "eval_math_steps_per_second": 2.477, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_physics_entropy": 0.5949395530223847, |
| "eval_physics_loss": 0.8190144300460815, |
| "eval_physics_mean_token_accuracy": 0.8086953344345092, |
| "eval_physics_num_tokens": 37485822.0, |
| "eval_physics_runtime": 58.0046, |
| "eval_physics_samples_per_second": 8.62, |
| "eval_physics_steps_per_second": 2.155, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.5696863148361444, |
| "epoch": 2.5091799265605874, |
| "grad_norm": 1.1328125, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 0.6901, |
| "mean_token_accuracy": 0.8347679734230041, |
| "num_tokens": 37746150.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.5688033113256097, |
| "epoch": 2.5269834204962724, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 0.6844, |
| "mean_token_accuracy": 0.8332536488771438, |
| "num_tokens": 38016863.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.5956829225644469, |
| "epoch": 2.544786914431957, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 0.7092, |
| "mean_token_accuracy": 0.8267782922834158, |
| "num_tokens": 38277754.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.5805349120870232, |
| "epoch": 2.562590408367642, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 0.7032, |
| "mean_token_accuracy": 0.8302845597267151, |
| "num_tokens": 38542761.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.5706984421238304, |
| "epoch": 2.580393902303327, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 0.6784, |
| "mean_token_accuracy": 0.8334123123437166, |
| "num_tokens": 38820999.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.5705496910959482, |
| "epoch": 2.598197396239012, |
| "grad_norm": 1.1640625, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 0.7012, |
| "mean_token_accuracy": 0.8325878925621509, |
| "num_tokens": 39080016.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5722655635327101, |
| "epoch": 2.6160008901746967, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 0.6843, |
| "mean_token_accuracy": 0.8355795852839947, |
| "num_tokens": 39345434.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.5695728290826082, |
| "epoch": 2.6338043841103818, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 0.6919, |
| "mean_token_accuracy": 0.8327361304312945, |
| "num_tokens": 39610041.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.5567968433722854, |
| "epoch": 2.6516078780460663, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 0.6679, |
| "mean_token_accuracy": 0.8385143429040909, |
| "num_tokens": 39884185.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.5825007958337665, |
| "epoch": 2.6694113719817514, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 0.709, |
| "mean_token_accuracy": 0.8283659152686595, |
| "num_tokens": 40143543.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_biology_entropy": 1.2247914123535155, |
| "eval_biology_loss": 1.781731367111206, |
| "eval_biology_mean_token_accuracy": 0.6498326325416565, |
| "eval_biology_num_tokens": 40143543.0, |
| "eval_biology_runtime": 39.4523, |
| "eval_biology_samples_per_second": 12.674, |
| "eval_biology_steps_per_second": 3.168, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_chemistry_entropy": 0.593181202173233, |
| "eval_chemistry_loss": 0.8005113005638123, |
| "eval_chemistry_mean_token_accuracy": 0.816295175075531, |
| "eval_chemistry_num_tokens": 40143543.0, |
| "eval_chemistry_runtime": 48.8453, |
| "eval_chemistry_samples_per_second": 10.236, |
| "eval_chemistry_steps_per_second": 2.559, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_math_entropy": 0.6026733157634735, |
| "eval_math_loss": 1.0098049640655518, |
| "eval_math_mean_token_accuracy": 0.787514422416687, |
| "eval_math_num_tokens": 40143543.0, |
| "eval_math_runtime": 50.407, |
| "eval_math_samples_per_second": 9.919, |
| "eval_math_steps_per_second": 2.48, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_physics_entropy": 0.6053323314189911, |
| "eval_physics_loss": 0.8167455792427063, |
| "eval_physics_mean_token_accuracy": 0.8090441026687623, |
| "eval_physics_num_tokens": 40143543.0, |
| "eval_physics_runtime": 57.9918, |
| "eval_physics_samples_per_second": 8.622, |
| "eval_physics_steps_per_second": 2.155, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.5760323867201805, |
| "epoch": 2.6872148659174364, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 0.6904, |
| "mean_token_accuracy": 0.8340807158499957, |
| "num_tokens": 40406812.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.5874979956075549, |
| "epoch": 2.7050183598531214, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 0.7135, |
| "mean_token_accuracy": 0.8278962858021259, |
| "num_tokens": 40668543.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.5766199560835957, |
| "epoch": 2.722821853788806, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 0.687, |
| "mean_token_accuracy": 0.8324626468122005, |
| "num_tokens": 40942603.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.5831139018759132, |
| "epoch": 2.740625347724491, |
| "grad_norm": 1.1796875, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 0.6943, |
| "mean_token_accuracy": 0.8322464641183615, |
| "num_tokens": 41208093.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.5711008545011282, |
| "epoch": 2.7584288416601757, |
| "grad_norm": 1.1484375, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 0.6939, |
| "mean_token_accuracy": 0.8310312990099191, |
| "num_tokens": 41477562.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.5629550171084702, |
| "epoch": 2.7762323355958607, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 0.6823, |
| "mean_token_accuracy": 0.8355346731841564, |
| "num_tokens": 41746571.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5724150063470006, |
| "epoch": 2.7940358295315457, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 0.6916, |
| "mean_token_accuracy": 0.8337334085255861, |
| "num_tokens": 42011633.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.5808822834864259, |
| "epoch": 2.8118393234672303, |
| "grad_norm": 1.1328125, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 0.6945, |
| "mean_token_accuracy": 0.8315359558910131, |
| "num_tokens": 42283087.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.5478847596794367, |
| "epoch": 2.8296428174029153, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 0.662, |
| "mean_token_accuracy": 0.8392819732427597, |
| "num_tokens": 42561534.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.572634200938046, |
| "epoch": 2.8474463113386, |
| "grad_norm": 1.1640625, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 0.6931, |
| "mean_token_accuracy": 0.8341519974172116, |
| "num_tokens": 42834266.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_biology_entropy": 1.2264355325698852, |
| "eval_biology_loss": 1.7825937271118164, |
| "eval_biology_mean_token_accuracy": 0.6504359455108643, |
| "eval_biology_num_tokens": 42834266.0, |
| "eval_biology_runtime": 39.3682, |
| "eval_biology_samples_per_second": 12.701, |
| "eval_biology_steps_per_second": 3.175, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_chemistry_entropy": 0.5885767588615417, |
| "eval_chemistry_loss": 0.7991457581520081, |
| "eval_chemistry_mean_token_accuracy": 0.8166323800086975, |
| "eval_chemistry_num_tokens": 42834266.0, |
| "eval_chemistry_runtime": 49.0263, |
| "eval_chemistry_samples_per_second": 10.199, |
| "eval_chemistry_steps_per_second": 2.55, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_math_entropy": 0.5984120781421661, |
| "eval_math_loss": 1.0110392570495605, |
| "eval_math_mean_token_accuracy": 0.7872469811439514, |
| "eval_math_num_tokens": 42834266.0, |
| "eval_math_runtime": 50.4589, |
| "eval_math_samples_per_second": 9.909, |
| "eval_math_steps_per_second": 2.477, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_physics_entropy": 0.6037993173599243, |
| "eval_physics_loss": 0.8166781067848206, |
| "eval_physics_mean_token_accuracy": 0.8091938343048096, |
| "eval_physics_num_tokens": 42834266.0, |
| "eval_physics_runtime": 58.0039, |
| "eval_physics_samples_per_second": 8.62, |
| "eval_physics_steps_per_second": 2.155, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.5951185580343008, |
| "epoch": 2.865249805274285, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 0.7092, |
| "mean_token_accuracy": 0.8289202846586704, |
| "num_tokens": 43100513.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.5747599832713604, |
| "epoch": 2.88305329920997, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 0.6868, |
| "mean_token_accuracy": 0.8322439193725586, |
| "num_tokens": 43375632.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.5638955064117909, |
| "epoch": 2.900856793145655, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 0.6846, |
| "mean_token_accuracy": 0.8354361861944198, |
| "num_tokens": 43644341.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.5602986583486199, |
| "epoch": 2.9186602870813396, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 0.672, |
| "mean_token_accuracy": 0.8361254766583442, |
| "num_tokens": 43917580.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.564799927175045, |
| "epoch": 2.9364637810170247, |
| "grad_norm": 1.1796875, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 0.6892, |
| "mean_token_accuracy": 0.8363475266844034, |
| "num_tokens": 44177325.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.5784864582121372, |
| "epoch": 2.9542672749527092, |
| "grad_norm": 1.25, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 0.6976, |
| "mean_token_accuracy": 0.8306768286973238, |
| "num_tokens": 44442947.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5877037810161709, |
| "epoch": 2.9720707688883943, |
| "grad_norm": 1.2890625, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 0.7063, |
| "mean_token_accuracy": 0.8293527953326703, |
| "num_tokens": 44700601.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.5639561323449016, |
| "epoch": 2.9898742628240793, |
| "grad_norm": 1.1796875, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 0.6834, |
| "mean_token_accuracy": 0.8358524903655052, |
| "num_tokens": 44971479.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.5595516356729692, |
| "epoch": 3.007121397574274, |
| "grad_norm": 1.1171875, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 0.6598, |
| "mean_token_accuracy": 0.8396484417300071, |
| "num_tokens": 45231796.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.5708899799734354, |
| "epoch": 3.0249248915099587, |
| "grad_norm": 1.3125, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 0.6911, |
| "mean_token_accuracy": 0.8313625615090132, |
| "num_tokens": 45500317.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_biology_entropy": 1.1691610884666443, |
| "eval_biology_loss": 1.8092824220657349, |
| "eval_biology_mean_token_accuracy": 0.6491630167961121, |
| "eval_biology_num_tokens": 45500317.0, |
| "eval_biology_runtime": 39.4629, |
| "eval_biology_samples_per_second": 12.67, |
| "eval_biology_steps_per_second": 3.168, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_chemistry_entropy": 0.5587203588485717, |
| "eval_chemistry_loss": 0.8038747310638428, |
| "eval_chemistry_mean_token_accuracy": 0.8160518984794617, |
| "eval_chemistry_num_tokens": 45500317.0, |
| "eval_chemistry_runtime": 49.1073, |
| "eval_chemistry_samples_per_second": 10.182, |
| "eval_chemistry_steps_per_second": 2.545, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_math_entropy": 0.5764329619407654, |
| "eval_math_loss": 1.0242400169372559, |
| "eval_math_mean_token_accuracy": 0.7866667547225952, |
| "eval_math_num_tokens": 45500317.0, |
| "eval_math_runtime": 50.5138, |
| "eval_math_samples_per_second": 9.898, |
| "eval_math_steps_per_second": 2.475, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_physics_entropy": 0.5760543973445892, |
| "eval_physics_loss": 0.8250848650932312, |
| "eval_physics_mean_token_accuracy": 0.808276960849762, |
| "eval_physics_num_tokens": 45500317.0, |
| "eval_physics_runtime": 58.0509, |
| "eval_physics_samples_per_second": 8.613, |
| "eval_physics_steps_per_second": 2.153, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.5444689901545644, |
| "epoch": 3.0427283854456437, |
| "grad_norm": 1.2265625, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 0.654, |
| "mean_token_accuracy": 0.8422593709081412, |
| "num_tokens": 45762680.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.5309807119891048, |
| "epoch": 3.0605318793813288, |
| "grad_norm": 1.4296875, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 0.6511, |
| "mean_token_accuracy": 0.8431796990334988, |
| "num_tokens": 46029923.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.5383216420188546, |
| "epoch": 3.0783353733170133, |
| "grad_norm": 1.1875, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 0.6527, |
| "mean_token_accuracy": 0.8410400524735451, |
| "num_tokens": 46305156.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.5262485228478908, |
| "epoch": 3.0961388672526984, |
| "grad_norm": 1.3046875, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 0.6407, |
| "mean_token_accuracy": 0.8441985324025154, |
| "num_tokens": 46582874.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5557232910767198, |
| "epoch": 3.1139423611883834, |
| "grad_norm": 1.4609375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 0.6726, |
| "mean_token_accuracy": 0.8360112313181162, |
| "num_tokens": 46850175.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.532409735955298, |
| "epoch": 3.131745855124068, |
| "grad_norm": 1.234375, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 0.6465, |
| "mean_token_accuracy": 0.8444639045745135, |
| "num_tokens": 47125899.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.5422864174470305, |
| "epoch": 3.149549349059753, |
| "grad_norm": 1.359375, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 0.6651, |
| "mean_token_accuracy": 0.8404106464236974, |
| "num_tokens": 47390194.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.54858369352296, |
| "epoch": 3.167352842995438, |
| "grad_norm": 1.234375, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 0.6649, |
| "mean_token_accuracy": 0.8392045438289643, |
| "num_tokens": 47654784.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.5342581951990724, |
| "epoch": 3.1851563369311227, |
| "grad_norm": 1.328125, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 0.6588, |
| "mean_token_accuracy": 0.8428513869643212, |
| "num_tokens": 47919400.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.5475335421040655, |
| "epoch": 3.2029598308668077, |
| "grad_norm": 1.34375, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 0.6658, |
| "mean_token_accuracy": 0.8400416314601898, |
| "num_tokens": 48187796.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_biology_entropy": 1.1653422470092774, |
| "eval_biology_loss": 1.8214138746261597, |
| "eval_biology_mean_token_accuracy": 0.6483545289039612, |
| "eval_biology_num_tokens": 48187796.0, |
| "eval_biology_runtime": 39.3568, |
| "eval_biology_samples_per_second": 12.704, |
| "eval_biology_steps_per_second": 3.176, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_chemistry_entropy": 0.5646139283180237, |
| "eval_chemistry_loss": 0.8045586943626404, |
| "eval_chemistry_mean_token_accuracy": 0.8160006055831909, |
| "eval_chemistry_num_tokens": 48187796.0, |
| "eval_chemistry_runtime": 48.9899, |
| "eval_chemistry_samples_per_second": 10.206, |
| "eval_chemistry_steps_per_second": 2.552, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_math_entropy": 0.5786230795383454, |
| "eval_math_loss": 1.0251545906066895, |
| "eval_math_mean_token_accuracy": 0.7865431332588195, |
| "eval_math_num_tokens": 48187796.0, |
| "eval_math_runtime": 50.3994, |
| "eval_math_samples_per_second": 9.921, |
| "eval_math_steps_per_second": 2.48, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_physics_entropy": 0.5797749905586242, |
| "eval_physics_loss": 0.8271687030792236, |
| "eval_physics_mean_token_accuracy": 0.8078161134719849, |
| "eval_physics_num_tokens": 48187796.0, |
| "eval_physics_runtime": 58.028, |
| "eval_physics_samples_per_second": 8.617, |
| "eval_physics_steps_per_second": 2.154, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.5395594524219632, |
| "epoch": 3.2207633248024923, |
| "grad_norm": 1.2421875, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 0.6567, |
| "mean_token_accuracy": 0.841542175412178, |
| "num_tokens": 48457428.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.5826782539486886, |
| "epoch": 3.2385668187381773, |
| "grad_norm": 1.34375, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 0.6937, |
| "mean_token_accuracy": 0.8312644015997648, |
| "num_tokens": 48716417.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.5456481980159879, |
| "epoch": 3.2563703126738623, |
| "grad_norm": 1.28125, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 0.6639, |
| "mean_token_accuracy": 0.8407398212701083, |
| "num_tokens": 48977306.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.5435102425515652, |
| "epoch": 3.274173806609547, |
| "grad_norm": 1.1640625, |
| "learning_rate": 8.6e-06, |
| "loss": 0.657, |
| "mean_token_accuracy": 0.8410529471933842, |
| "num_tokens": 49249667.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.5419310368597507, |
| "epoch": 3.291977300545232, |
| "grad_norm": 1.2265625, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 0.6545, |
| "mean_token_accuracy": 0.8413508605211973, |
| "num_tokens": 49520900.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.549009719118476, |
| "epoch": 3.309780794480917, |
| "grad_norm": 1.296875, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 0.6708, |
| "mean_token_accuracy": 0.8386994738131761, |
| "num_tokens": 49786411.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.542225575260818, |
| "epoch": 3.3275842884166016, |
| "grad_norm": 1.2578125, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 0.6497, |
| "mean_token_accuracy": 0.8425345428287982, |
| "num_tokens": 50056207.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.5501431416720152, |
| "epoch": 3.3453877823522866, |
| "grad_norm": 1.2265625, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 0.6645, |
| "mean_token_accuracy": 0.8394666783511638, |
| "num_tokens": 50330876.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.5492377877235413, |
| "epoch": 3.3631912762879717, |
| "grad_norm": 1.421875, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 0.6658, |
| "mean_token_accuracy": 0.8383394341915846, |
| "num_tokens": 50602004.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.5501461183652282, |
| "epoch": 3.3809947702236562, |
| "grad_norm": 1.265625, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 0.6634, |
| "mean_token_accuracy": 0.8376425668597222, |
| "num_tokens": 50873444.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_biology_entropy": 1.1502807393074035, |
| "eval_biology_loss": 1.829302191734314, |
| "eval_biology_mean_token_accuracy": 0.647935733795166, |
| "eval_biology_num_tokens": 50873444.0, |
| "eval_biology_runtime": 39.2037, |
| "eval_biology_samples_per_second": 12.754, |
| "eval_biology_steps_per_second": 3.188, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_chemistry_entropy": 0.5574937677383422, |
| "eval_chemistry_loss": 0.8056073188781738, |
| "eval_chemistry_mean_token_accuracy": 0.8162664136886597, |
| "eval_chemistry_num_tokens": 50873444.0, |
| "eval_chemistry_runtime": 48.8236, |
| "eval_chemistry_samples_per_second": 10.241, |
| "eval_chemistry_steps_per_second": 2.56, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_math_entropy": 0.5737685420513153, |
| "eval_math_loss": 1.0286157131195068, |
| "eval_math_mean_token_accuracy": 0.7861198062896728, |
| "eval_math_num_tokens": 50873444.0, |
| "eval_math_runtime": 50.2275, |
| "eval_math_samples_per_second": 9.955, |
| "eval_math_steps_per_second": 2.489, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_physics_entropy": 0.5730379042625428, |
| "eval_physics_loss": 0.8303305506706238, |
| "eval_physics_mean_token_accuracy": 0.8077737083435058, |
| "eval_physics_num_tokens": 50873444.0, |
| "eval_physics_runtime": 57.7614, |
| "eval_physics_samples_per_second": 8.656, |
| "eval_physics_steps_per_second": 2.164, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.5610906140878796, |
| "epoch": 3.3987982641593413, |
| "grad_norm": 1.3515625, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 0.6761, |
| "mean_token_accuracy": 0.8352380134165287, |
| "num_tokens": 51141077.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.5438909037038684, |
| "epoch": 3.4166017580950263, |
| "grad_norm": 1.40625, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 0.6611, |
| "mean_token_accuracy": 0.8404773697257042, |
| "num_tokens": 51406483.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.5512320777401328, |
| "epoch": 3.434405252030711, |
| "grad_norm": 1.359375, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 0.6666, |
| "mean_token_accuracy": 0.8385514669120312, |
| "num_tokens": 51668458.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.5570000663399697, |
| "epoch": 3.452208745966396, |
| "grad_norm": 1.3359375, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 0.6807, |
| "mean_token_accuracy": 0.8363874733448029, |
| "num_tokens": 51928199.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.5494465444236993, |
| "epoch": 3.470012239902081, |
| "grad_norm": 1.2578125, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 0.6571, |
| "mean_token_accuracy": 0.8388365376740694, |
| "num_tokens": 52199244.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.5377742035314441, |
| "epoch": 3.4878157338377656, |
| "grad_norm": 1.3515625, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 0.6513, |
| "mean_token_accuracy": 0.8415070977061987, |
| "num_tokens": 52473476.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.5498209903016686, |
| "epoch": 3.5056192277734506, |
| "grad_norm": 1.1484375, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 0.6659, |
| "mean_token_accuracy": 0.8391986701637506, |
| "num_tokens": 52743275.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.5453953156247735, |
| "epoch": 3.523422721709135, |
| "grad_norm": 1.2734375, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 0.6549, |
| "mean_token_accuracy": 0.8408893909305334, |
| "num_tokens": 53014827.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.5506702629849315, |
| "epoch": 3.54122621564482, |
| "grad_norm": 1.390625, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 0.6686, |
| "mean_token_accuracy": 0.83898391649127, |
| "num_tokens": 53285577.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.5308271149173379, |
| "epoch": 3.5590297095805052, |
| "grad_norm": 1.3046875, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 0.6485, |
| "mean_token_accuracy": 0.8444319948554039, |
| "num_tokens": 53553391.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_biology_entropy": 1.1442560811042786, |
| "eval_biology_loss": 1.8303390741348267, |
| "eval_biology_mean_token_accuracy": 0.6481823086738586, |
| "eval_biology_num_tokens": 53553391.0, |
| "eval_biology_runtime": 39.2289, |
| "eval_biology_samples_per_second": 12.746, |
| "eval_biology_steps_per_second": 3.186, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_chemistry_entropy": 0.5585863585472107, |
| "eval_chemistry_loss": 0.8046336770057678, |
| "eval_chemistry_mean_token_accuracy": 0.8162187285423279, |
| "eval_chemistry_num_tokens": 53553391.0, |
| "eval_chemistry_runtime": 48.6434, |
| "eval_chemistry_samples_per_second": 10.279, |
| "eval_chemistry_steps_per_second": 2.57, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_math_entropy": 0.5739703712463379, |
| "eval_math_loss": 1.0285587310791016, |
| "eval_math_mean_token_accuracy": 0.7863604693412781, |
| "eval_math_num_tokens": 53553391.0, |
| "eval_math_runtime": 50.3001, |
| "eval_math_samples_per_second": 9.94, |
| "eval_math_steps_per_second": 2.485, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_physics_entropy": 0.5745766685009003, |
| "eval_physics_loss": 0.8294666409492493, |
| "eval_physics_mean_token_accuracy": 0.8078997702598572, |
| "eval_physics_num_tokens": 53553391.0, |
| "eval_physics_runtime": 57.9219, |
| "eval_physics_samples_per_second": 8.632, |
| "eval_physics_steps_per_second": 2.158, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 6, |
| "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": 4.16464083921255e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
