| { |
| "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": 0.7850796665996314, |
| "epoch": 0.017803493935684877, |
| "grad_norm": 4.125, |
| "learning_rate": 1.8e-07, |
| "loss": 0.9868, |
| "mean_token_accuracy": 0.7752813071012497, |
| "num_tokens": 259048.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.7656301472336053, |
| "epoch": 0.035606987871369754, |
| "grad_norm": 3.984375, |
| "learning_rate": 3.8e-07, |
| "loss": 0.9514, |
| "mean_token_accuracy": 0.778770961612463, |
| "num_tokens": 532019.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.7800006279721856, |
| "epoch": 0.053410481807054634, |
| "grad_norm": 3.546875, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 0.9645, |
| "mean_token_accuracy": 0.777244720607996, |
| "num_tokens": 807220.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8148440960794687, |
| "epoch": 0.07121397574273951, |
| "grad_norm": 3.578125, |
| "learning_rate": 7.8e-07, |
| "loss": 0.9845, |
| "mean_token_accuracy": 0.7725338555872441, |
| "num_tokens": 1068429.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.7980734214186669, |
| "epoch": 0.08901746967842439, |
| "grad_norm": 3.328125, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 0.9502, |
| "mean_token_accuracy": 0.7774997223168612, |
| "num_tokens": 1336698.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.827414046600461, |
| "epoch": 0.10682096361410927, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 0.969, |
| "mean_token_accuracy": 0.7773337867110968, |
| "num_tokens": 1608574.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8639046084135771, |
| "epoch": 0.12462445754979415, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 0.9913, |
| "mean_token_accuracy": 0.7705458387732506, |
| "num_tokens": 1872747.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8748010536655784, |
| "epoch": 0.14242795148547902, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 0.9868, |
| "mean_token_accuracy": 0.7721451628953219, |
| "num_tokens": 2133173.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8339586811140179, |
| "epoch": 0.1602314454211639, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 0.9473, |
| "mean_token_accuracy": 0.782527657970786, |
| "num_tokens": 2406145.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8617484034970403, |
| "epoch": 0.17803493935684878, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.98e-06, |
| "loss": 0.9705, |
| "mean_token_accuracy": 0.7755408465862275, |
| "num_tokens": 2663709.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_chemistry_entropy": 0.8530604543685913, |
| "eval_chemistry_loss": 0.9879733920097351, |
| "eval_chemistry_mean_token_accuracy": 0.7785031309127808, |
| "eval_chemistry_num_tokens": 2663709.0, |
| "eval_chemistry_runtime": 34.6174, |
| "eval_chemistry_samples_per_second": 14.444, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17803493935684878, |
| "eval_physics_entropy": 0.6429356033802033, |
| "eval_physics_loss": 0.7288391590118408, |
| "eval_physics_mean_token_accuracy": 0.8237355589866638, |
| "eval_physics_num_tokens": 2663709.0, |
| "eval_physics_runtime": 41.2328, |
| "eval_physics_samples_per_second": 12.126, |
| "eval_physics_steps_per_second": 3.032, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8313142383471132, |
| "epoch": 0.19583843329253367, |
| "grad_norm": 2.3125, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 0.931, |
| "mean_token_accuracy": 0.7826357394456863, |
| "num_tokens": 2934980.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8383694937452674, |
| "epoch": 0.21364192722821854, |
| "grad_norm": 2.140625, |
| "learning_rate": 2.38e-06, |
| "loss": 0.9256, |
| "mean_token_accuracy": 0.7815470688045025, |
| "num_tokens": 3208702.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8249719947576523, |
| "epoch": 0.2314454211639034, |
| "grad_norm": 2.3125, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 0.9319, |
| "mean_token_accuracy": 0.7833116251975298, |
| "num_tokens": 3480119.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.8376072259619832, |
| "epoch": 0.2492489150995883, |
| "grad_norm": 2.171875, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 0.9339, |
| "mean_token_accuracy": 0.7807847507297992, |
| "num_tokens": 3746392.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.8562109092250466, |
| "epoch": 0.26705240903527316, |
| "grad_norm": 2.125, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 0.945, |
| "mean_token_accuracy": 0.7766421258449554, |
| "num_tokens": 4019487.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8114860905334353, |
| "epoch": 0.28485590297095803, |
| "grad_norm": 2.3125, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 0.9108, |
| "mean_token_accuracy": 0.7848536405712366, |
| "num_tokens": 4298100.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.8167605808004736, |
| "epoch": 0.30265939690664295, |
| "grad_norm": 1.984375, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 0.896, |
| "mean_token_accuracy": 0.7858319569379091, |
| "num_tokens": 4568116.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.8315373951569199, |
| "epoch": 0.3204628908423278, |
| "grad_norm": 2.359375, |
| "learning_rate": 3.58e-06, |
| "loss": 0.9106, |
| "mean_token_accuracy": 0.7813973911106586, |
| "num_tokens": 4831334.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8178198751062155, |
| "epoch": 0.3382663847780127, |
| "grad_norm": 2.015625, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 0.881, |
| "mean_token_accuracy": 0.7869020272046328, |
| "num_tokens": 5103301.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.8403754932805896, |
| "epoch": 0.35606987871369755, |
| "grad_norm": 1.9921875, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 0.9115, |
| "mean_token_accuracy": 0.7783690184354782, |
| "num_tokens": 5362915.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_chemistry_entropy": 0.8292876372337341, |
| "eval_chemistry_loss": 0.9164837598800659, |
| "eval_chemistry_mean_token_accuracy": 0.7834693474769592, |
| "eval_chemistry_num_tokens": 5362915.0, |
| "eval_chemistry_runtime": 34.57, |
| "eval_chemistry_samples_per_second": 14.463, |
| "eval_chemistry_steps_per_second": 3.616, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.35606987871369755, |
| "eval_physics_entropy": 0.6445197880268096, |
| "eval_physics_loss": 0.6974144577980042, |
| "eval_physics_mean_token_accuracy": 0.8239417452812194, |
| "eval_physics_num_tokens": 5362915.0, |
| "eval_physics_runtime": 41.2388, |
| "eval_physics_samples_per_second": 12.124, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8109689498320222, |
| "epoch": 0.3738733726493824, |
| "grad_norm": 2.078125, |
| "learning_rate": 4.18e-06, |
| "loss": 0.856, |
| "mean_token_accuracy": 0.788726344704628, |
| "num_tokens": 5625455.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.8134721567854285, |
| "epoch": 0.39167686658506734, |
| "grad_norm": 2.0, |
| "learning_rate": 4.38e-06, |
| "loss": 0.8568, |
| "mean_token_accuracy": 0.7867508839815855, |
| "num_tokens": 5890806.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.8066105756908655, |
| "epoch": 0.4094803605207522, |
| "grad_norm": 2.203125, |
| "learning_rate": 4.58e-06, |
| "loss": 0.8275, |
| "mean_token_accuracy": 0.7900486279278993, |
| "num_tokens": 6158751.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.8255353083834052, |
| "epoch": 0.4272838544564371, |
| "grad_norm": 2.125, |
| "learning_rate": 4.78e-06, |
| "loss": 0.8244, |
| "mean_token_accuracy": 0.7881908807903528, |
| "num_tokens": 6423747.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.8402099519968033, |
| "epoch": 0.44508734839212194, |
| "grad_norm": 1.9609375, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 0.8386, |
| "mean_token_accuracy": 0.7834979113191366, |
| "num_tokens": 6684517.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.8010248441249133, |
| "epoch": 0.4628908423278068, |
| "grad_norm": 2.0, |
| "learning_rate": 5.18e-06, |
| "loss": 0.7988, |
| "mean_token_accuracy": 0.7915425561368465, |
| "num_tokens": 6945659.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.814286389760673, |
| "epoch": 0.48069433626349173, |
| "grad_norm": 2.28125, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 0.8029, |
| "mean_token_accuracy": 0.7904389422386885, |
| "num_tokens": 7217797.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.785382822342217, |
| "epoch": 0.4984978301991766, |
| "grad_norm": 2.0, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 0.76, |
| "mean_token_accuracy": 0.7953414976596832, |
| "num_tokens": 7498182.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.8195194514468312, |
| "epoch": 0.5163013241348615, |
| "grad_norm": 1.90625, |
| "learning_rate": 5.78e-06, |
| "loss": 0.7916, |
| "mean_token_accuracy": 0.78877494931221, |
| "num_tokens": 7761686.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.8369470547884703, |
| "epoch": 0.5341048180705463, |
| "grad_norm": 2.375, |
| "learning_rate": 5.98e-06, |
| "loss": 0.8066, |
| "mean_token_accuracy": 0.786370538175106, |
| "num_tokens": 8026923.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_chemistry_entropy": 0.818053927898407, |
| "eval_chemistry_loss": 0.8182278871536255, |
| "eval_chemistry_mean_token_accuracy": 0.7884931750297547, |
| "eval_chemistry_num_tokens": 8026923.0, |
| "eval_chemistry_runtime": 34.5992, |
| "eval_chemistry_samples_per_second": 14.451, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.5341048180705463, |
| "eval_physics_entropy": 0.6585752744674682, |
| "eval_physics_loss": 0.6424084305763245, |
| "eval_physics_mean_token_accuracy": 0.8238574590682983, |
| "eval_physics_num_tokens": 8026923.0, |
| "eval_physics_runtime": 41.2472, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.8153910035267472, |
| "epoch": 0.5519083120062312, |
| "grad_norm": 2.6875, |
| "learning_rate": 6.18e-06, |
| "loss": 0.792, |
| "mean_token_accuracy": 0.7901755437254906, |
| "num_tokens": 8301506.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.8126581776887178, |
| "epoch": 0.5697118059419161, |
| "grad_norm": 1.96875, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 0.7938, |
| "mean_token_accuracy": 0.7923727743327618, |
| "num_tokens": 8559126.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.8185926586389541, |
| "epoch": 0.587515299877601, |
| "grad_norm": 2.375, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 0.7873, |
| "mean_token_accuracy": 0.7911712639033794, |
| "num_tokens": 8821088.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.7809942720457912, |
| "epoch": 0.6053187938132859, |
| "grad_norm": 1.8046875, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 0.75, |
| "mean_token_accuracy": 0.7993631161749363, |
| "num_tokens": 9089866.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.790563733316958, |
| "epoch": 0.6231222877489707, |
| "grad_norm": 2.03125, |
| "learning_rate": 6.98e-06, |
| "loss": 0.7666, |
| "mean_token_accuracy": 0.79404122158885, |
| "num_tokens": 9358866.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.7880068564787507, |
| "epoch": 0.6409257816846556, |
| "grad_norm": 1.8203125, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 0.7467, |
| "mean_token_accuracy": 0.7980206806212664, |
| "num_tokens": 9629977.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.7765095194801688, |
| "epoch": 0.6587292756203404, |
| "grad_norm": 1.921875, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 0.7403, |
| "mean_token_accuracy": 0.799276452511549, |
| "num_tokens": 9897757.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.7657832968980074, |
| "epoch": 0.6765327695560254, |
| "grad_norm": 1.96875, |
| "learning_rate": 7.58e-06, |
| "loss": 0.7391, |
| "mean_token_accuracy": 0.8021788071841002, |
| "num_tokens": 10168367.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.7575604457408189, |
| "epoch": 0.6943362634917103, |
| "grad_norm": 2.03125, |
| "learning_rate": 7.78e-06, |
| "loss": 0.7274, |
| "mean_token_accuracy": 0.803612269833684, |
| "num_tokens": 10439039.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.771031835861504, |
| "epoch": 0.7121397574273951, |
| "grad_norm": 1.890625, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 0.7388, |
| "mean_token_accuracy": 0.8006934400647878, |
| "num_tokens": 10709549.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_chemistry_entropy": 0.8015456414222717, |
| "eval_chemistry_loss": 0.782933235168457, |
| "eval_chemistry_mean_token_accuracy": 0.7955212960243225, |
| "eval_chemistry_num_tokens": 10709549.0, |
| "eval_chemistry_runtime": 34.5635, |
| "eval_chemistry_samples_per_second": 14.466, |
| "eval_chemistry_steps_per_second": 3.617, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.7121397574273951, |
| "eval_physics_entropy": 0.6612502746582031, |
| "eval_physics_loss": 0.6346089839935303, |
| "eval_physics_mean_token_accuracy": 0.8259402804374695, |
| "eval_physics_num_tokens": 10709549.0, |
| "eval_physics_runtime": 41.2249, |
| "eval_physics_samples_per_second": 12.129, |
| "eval_physics_steps_per_second": 3.032, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.7536357572302222, |
| "epoch": 0.72994325136308, |
| "grad_norm": 2.03125, |
| "learning_rate": 8.18e-06, |
| "loss": 0.7226, |
| "mean_token_accuracy": 0.8046278320252895, |
| "num_tokens": 10976011.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.7753360515460372, |
| "epoch": 0.7477467452987648, |
| "grad_norm": 1.8203125, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 0.7386, |
| "mean_token_accuracy": 0.8000223852694035, |
| "num_tokens": 11245251.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.7881368083879352, |
| "epoch": 0.7655502392344498, |
| "grad_norm": 2.140625, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 0.749, |
| "mean_token_accuracy": 0.7968206465244293, |
| "num_tokens": 11518514.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.7809985620900989, |
| "epoch": 0.7833537331701347, |
| "grad_norm": 2.140625, |
| "learning_rate": 8.78e-06, |
| "loss": 0.7485, |
| "mean_token_accuracy": 0.7970930390059948, |
| "num_tokens": 11787117.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.7746961690485478, |
| "epoch": 0.8011572271058195, |
| "grad_norm": 2.109375, |
| "learning_rate": 8.98e-06, |
| "loss": 0.7393, |
| "mean_token_accuracy": 0.8000770397484303, |
| "num_tokens": 12050993.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.7727582400664688, |
| "epoch": 0.8189607210415044, |
| "grad_norm": 2.046875, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 0.7349, |
| "mean_token_accuracy": 0.7991833999752999, |
| "num_tokens": 12311688.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.7381363824009896, |
| "epoch": 0.8367642149771892, |
| "grad_norm": 1.96875, |
| "learning_rate": 9.38e-06, |
| "loss": 0.7018, |
| "mean_token_accuracy": 0.8056569200009107, |
| "num_tokens": 12598408.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.7458288084715605, |
| "epoch": 0.8545677089128741, |
| "grad_norm": 1.765625, |
| "learning_rate": 9.58e-06, |
| "loss": 0.7163, |
| "mean_token_accuracy": 0.8052155710756779, |
| "num_tokens": 12869642.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.7388540361076593, |
| "epoch": 0.8723712028485591, |
| "grad_norm": 1.7578125, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 0.702, |
| "mean_token_accuracy": 0.8063301011919976, |
| "num_tokens": 13139555.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.7744078584015369, |
| "epoch": 0.8901746967842439, |
| "grad_norm": 1.921875, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 0.7401, |
| "mean_token_accuracy": 0.7977699730545282, |
| "num_tokens": 13396545.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_chemistry_entropy": 0.7673972373008728, |
| "eval_chemistry_loss": 0.7539037466049194, |
| "eval_chemistry_mean_token_accuracy": 0.7999647192955017, |
| "eval_chemistry_num_tokens": 13396545.0, |
| "eval_chemistry_runtime": 34.5998, |
| "eval_chemistry_samples_per_second": 14.451, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8901746967842439, |
| "eval_physics_entropy": 0.6430810136795044, |
| "eval_physics_loss": 0.6303399205207825, |
| "eval_physics_mean_token_accuracy": 0.8259163227081299, |
| "eval_physics_num_tokens": 13396545.0, |
| "eval_physics_runtime": 41.284, |
| "eval_physics_samples_per_second": 12.111, |
| "eval_physics_steps_per_second": 3.028, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.7333443723618984, |
| "epoch": 0.9079781907199288, |
| "grad_norm": 1.625, |
| "learning_rate": 1.018e-05, |
| "loss": 0.6952, |
| "mean_token_accuracy": 0.8094376768916846, |
| "num_tokens": 13667857.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.7600304771214723, |
| "epoch": 0.9257816846556136, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.038e-05, |
| "loss": 0.7179, |
| "mean_token_accuracy": 0.8037680003792047, |
| "num_tokens": 13932771.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.7434984143823385, |
| "epoch": 0.9435851785912985, |
| "grad_norm": 1.7578125, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 0.7058, |
| "mean_token_accuracy": 0.8062188472598791, |
| "num_tokens": 14194632.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.7611551772803068, |
| "epoch": 0.9613886725269835, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 0.7268, |
| "mean_token_accuracy": 0.8000180393457412, |
| "num_tokens": 14457650.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.7252845833078027, |
| "epoch": 0.9791921664626683, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 0.6969, |
| "mean_token_accuracy": 0.8071364350616932, |
| "num_tokens": 14724967.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.7578858522698283, |
| "epoch": 0.9969956603983532, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 0.7274, |
| "mean_token_accuracy": 0.8026934862136841, |
| "num_tokens": 14993272.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.6899186813062237, |
| "epoch": 1.014242795148548, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.138e-05, |
| "loss": 0.6595, |
| "mean_token_accuracy": 0.8154077602970985, |
| "num_tokens": 15254936.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.7271509811282157, |
| "epoch": 1.0320462890842328, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 0.6948, |
| "mean_token_accuracy": 0.804929868131876, |
| "num_tokens": 15515007.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.6891195146366954, |
| "epoch": 1.0498497830199176, |
| "grad_norm": 1.875, |
| "learning_rate": 1.178e-05, |
| "loss": 0.6619, |
| "mean_token_accuracy": 0.8155979864299298, |
| "num_tokens": 15794203.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.7109181759878993, |
| "epoch": 1.0676532769556026, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.198e-05, |
| "loss": 0.6773, |
| "mean_token_accuracy": 0.8111642371863127, |
| "num_tokens": 16063664.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_chemistry_entropy": 0.7275329558849335, |
| "eval_chemistry_loss": 0.7331510186195374, |
| "eval_chemistry_mean_token_accuracy": 0.8037505540847778, |
| "eval_chemistry_num_tokens": 16063664.0, |
| "eval_chemistry_runtime": 34.6124, |
| "eval_chemistry_samples_per_second": 14.446, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.0676532769556026, |
| "eval_physics_entropy": 0.6278075368404389, |
| "eval_physics_loss": 0.6282901763916016, |
| "eval_physics_mean_token_accuracy": 0.825420051574707, |
| "eval_physics_num_tokens": 16063664.0, |
| "eval_physics_runtime": 41.2841, |
| "eval_physics_samples_per_second": 12.111, |
| "eval_physics_steps_per_second": 3.028, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6886052008718252, |
| "epoch": 1.0854567708912874, |
| "grad_norm": 1.9296875, |
| "learning_rate": 1.218e-05, |
| "loss": 0.664, |
| "mean_token_accuracy": 0.8155276246368885, |
| "num_tokens": 16338303.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.7301610294729471, |
| "epoch": 1.1032602648269723, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 0.69, |
| "mean_token_accuracy": 0.8072479218244553, |
| "num_tokens": 16606302.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.7009955205023288, |
| "epoch": 1.121063758762657, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 0.6662, |
| "mean_token_accuracy": 0.8098525639623404, |
| "num_tokens": 16877071.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.6957393456250429, |
| "epoch": 1.138867252698342, |
| "grad_norm": 1.7421875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 0.6634, |
| "mean_token_accuracy": 0.8121555853635073, |
| "num_tokens": 17145775.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.7081588903442025, |
| "epoch": 1.156670746634027, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 0.6716, |
| "mean_token_accuracy": 0.8104519549757242, |
| "num_tokens": 17410775.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.7330411653965712, |
| "epoch": 1.1744742405697117, |
| "grad_norm": 2.015625, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 0.6949, |
| "mean_token_accuracy": 0.8045349054038524, |
| "num_tokens": 17677791.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.7073222378268837, |
| "epoch": 1.1922777345053968, |
| "grad_norm": 1.8046875, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 0.666, |
| "mean_token_accuracy": 0.8106517169624567, |
| "num_tokens": 17949441.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.697087805531919, |
| "epoch": 1.2100812284410816, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 0.6644, |
| "mean_token_accuracy": 0.8117371387779713, |
| "num_tokens": 18217742.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.7058945974335075, |
| "epoch": 1.2278847223767664, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.378e-05, |
| "loss": 0.6693, |
| "mean_token_accuracy": 0.8107333112508058, |
| "num_tokens": 18487769.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6924632461741567, |
| "epoch": 1.2456882163124514, |
| "grad_norm": 1.859375, |
| "learning_rate": 1.398e-05, |
| "loss": 0.6628, |
| "mean_token_accuracy": 0.811711684986949, |
| "num_tokens": 18760296.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_chemistry_entropy": 0.7344678246974945, |
| "eval_chemistry_loss": 0.7194931507110596, |
| "eval_chemistry_mean_token_accuracy": 0.8056631999015809, |
| "eval_chemistry_num_tokens": 18760296.0, |
| "eval_chemistry_runtime": 34.6183, |
| "eval_chemistry_samples_per_second": 14.443, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.2456882163124514, |
| "eval_physics_entropy": 0.6411541821956634, |
| "eval_physics_loss": 0.6283924579620361, |
| "eval_physics_mean_token_accuracy": 0.8252067227363586, |
| "eval_physics_num_tokens": 18760296.0, |
| "eval_physics_runtime": 41.2883, |
| "eval_physics_samples_per_second": 12.11, |
| "eval_physics_steps_per_second": 3.027, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.6927230576053262, |
| "epoch": 1.2634917102481362, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.418e-05, |
| "loss": 0.6577, |
| "mean_token_accuracy": 0.8144037745893001, |
| "num_tokens": 19035733.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.7126594386994839, |
| "epoch": 1.281295204183821, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 0.681, |
| "mean_token_accuracy": 0.80786834359169, |
| "num_tokens": 19311432.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6811089560389518, |
| "epoch": 1.299098698119506, |
| "grad_norm": 1.6640625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 0.65, |
| "mean_token_accuracy": 0.8166949838399887, |
| "num_tokens": 19593016.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6871400082483887, |
| "epoch": 1.3169021920551909, |
| "grad_norm": 1.8359375, |
| "learning_rate": 1.478e-05, |
| "loss": 0.6507, |
| "mean_token_accuracy": 0.8143899407237768, |
| "num_tokens": 19860476.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6881214780732989, |
| "epoch": 1.3347056859908757, |
| "grad_norm": 1.9375, |
| "learning_rate": 1.498e-05, |
| "loss": 0.6517, |
| "mean_token_accuracy": 0.8132891681045293, |
| "num_tokens": 20124436.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6930879963561892, |
| "epoch": 1.3525091799265607, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 0.6664, |
| "mean_token_accuracy": 0.8128124658018351, |
| "num_tokens": 20388027.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6938662787899375, |
| "epoch": 1.3703126738622455, |
| "grad_norm": 1.8828125, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 0.6592, |
| "mean_token_accuracy": 0.812091863155365, |
| "num_tokens": 20656076.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.7249738791957497, |
| "epoch": 1.3881161677979303, |
| "grad_norm": 1.875, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.6947, |
| "mean_token_accuracy": 0.8059011038392783, |
| "num_tokens": 20911633.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.7112141275778413, |
| "epoch": 1.4059196617336152, |
| "grad_norm": 1.6640625, |
| "learning_rate": 1.578e-05, |
| "loss": 0.6839, |
| "mean_token_accuracy": 0.8064889151602983, |
| "num_tokens": 21176227.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6923007164150476, |
| "epoch": 1.4237231556693, |
| "grad_norm": 1.8828125, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.6619, |
| "mean_token_accuracy": 0.8126598935574293, |
| "num_tokens": 21445242.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_chemistry_entropy": 0.7049469864368438, |
| "eval_chemistry_loss": 0.7038090229034424, |
| "eval_chemistry_mean_token_accuracy": 0.8079238352775574, |
| "eval_chemistry_num_tokens": 21445242.0, |
| "eval_chemistry_runtime": 34.5889, |
| "eval_chemistry_samples_per_second": 14.456, |
| "eval_chemistry_steps_per_second": 3.614, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.4237231556693, |
| "eval_physics_entropy": 0.6302921390533447, |
| "eval_physics_loss": 0.6273708343505859, |
| "eval_physics_mean_token_accuracy": 0.824629822731018, |
| "eval_physics_num_tokens": 21445242.0, |
| "eval_physics_runtime": 41.2241, |
| "eval_physics_samples_per_second": 12.129, |
| "eval_physics_steps_per_second": 3.032, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.6938302194699645, |
| "epoch": 1.441526649604985, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.618e-05, |
| "loss": 0.6643, |
| "mean_token_accuracy": 0.8114070001989603, |
| "num_tokens": 21718455.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6958711674436927, |
| "epoch": 1.4593301435406698, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.638e-05, |
| "loss": 0.6619, |
| "mean_token_accuracy": 0.8139785166829825, |
| "num_tokens": 21974076.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.6980619359761476, |
| "epoch": 1.4771336374763546, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.658e-05, |
| "loss": 0.6687, |
| "mean_token_accuracy": 0.8110807090997696, |
| "num_tokens": 22241599.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.6977157359942794, |
| "epoch": 1.4949371314120397, |
| "grad_norm": 1.9921875, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.6706, |
| "mean_token_accuracy": 0.8098580442368984, |
| "num_tokens": 22510725.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6765838697552681, |
| "epoch": 1.5127406253477245, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.698e-05, |
| "loss": 0.6549, |
| "mean_token_accuracy": 0.8153976205736398, |
| "num_tokens": 22778514.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6760151350870729, |
| "epoch": 1.5305441192834093, |
| "grad_norm": 1.859375, |
| "learning_rate": 1.718e-05, |
| "loss": 0.6375, |
| "mean_token_accuracy": 0.8156527172774076, |
| "num_tokens": 23051230.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.6908003646880388, |
| "epoch": 1.5483476132190943, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.658, |
| "mean_token_accuracy": 0.8122621558606624, |
| "num_tokens": 23306327.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6964091071859002, |
| "epoch": 1.5661511071547791, |
| "grad_norm": 1.7578125, |
| "learning_rate": 1.758e-05, |
| "loss": 0.6597, |
| "mean_token_accuracy": 0.811262060701847, |
| "num_tokens": 23573127.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6847964959219098, |
| "epoch": 1.583954601090464, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.6532, |
| "mean_token_accuracy": 0.8122437499463558, |
| "num_tokens": 23841914.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.6890485780313611, |
| "epoch": 1.601758095026149, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.798e-05, |
| "loss": 0.6587, |
| "mean_token_accuracy": 0.8129192098975182, |
| "num_tokens": 24105114.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_chemistry_entropy": 0.6863252129554749, |
| "eval_chemistry_loss": 0.6930021643638611, |
| "eval_chemistry_mean_token_accuracy": 0.8097150478363037, |
| "eval_chemistry_num_tokens": 24105114.0, |
| "eval_chemistry_runtime": 34.6005, |
| "eval_chemistry_samples_per_second": 14.451, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.601758095026149, |
| "eval_physics_entropy": 0.6244586195945739, |
| "eval_physics_loss": 0.6287086606025696, |
| "eval_physics_mean_token_accuracy": 0.8245933861732483, |
| "eval_physics_num_tokens": 24105114.0, |
| "eval_physics_runtime": 41.2389, |
| "eval_physics_samples_per_second": 12.124, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.6585973870009184, |
| "epoch": 1.6195615889618338, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.6354, |
| "mean_token_accuracy": 0.8186014011502266, |
| "num_tokens": 24384008.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.693634419888258, |
| "epoch": 1.6373650828975186, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.6526, |
| "mean_token_accuracy": 0.812073552981019, |
| "num_tokens": 24638205.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.6443359967321157, |
| "epoch": 1.6551685768332036, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.858e-05, |
| "loss": 0.6138, |
| "mean_token_accuracy": 0.8221632152795791, |
| "num_tokens": 24910104.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.6681255014613271, |
| "epoch": 1.6729720707688884, |
| "grad_norm": 1.8828125, |
| "learning_rate": 1.878e-05, |
| "loss": 0.6349, |
| "mean_token_accuracy": 0.8162749476730824, |
| "num_tokens": 25181306.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.676227905228734, |
| "epoch": 1.6907755647045732, |
| "grad_norm": 1.75, |
| "learning_rate": 1.898e-05, |
| "loss": 0.6452, |
| "mean_token_accuracy": 0.8150841273367405, |
| "num_tokens": 25449138.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.6844975538551807, |
| "epoch": 1.7085790586402583, |
| "grad_norm": 1.703125, |
| "learning_rate": 1.918e-05, |
| "loss": 0.6493, |
| "mean_token_accuracy": 0.8142464257776737, |
| "num_tokens": 25709030.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.684933065995574, |
| "epoch": 1.7263825525759429, |
| "grad_norm": 1.828125, |
| "learning_rate": 1.938e-05, |
| "loss": 0.6539, |
| "mean_token_accuracy": 0.8128440115600825, |
| "num_tokens": 25972807.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.6664420425891876, |
| "epoch": 1.744186046511628, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 0.632, |
| "mean_token_accuracy": 0.8175933144986629, |
| "num_tokens": 26244955.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.639144916087389, |
| "epoch": 1.761989540447313, |
| "grad_norm": 1.515625, |
| "learning_rate": 1.978e-05, |
| "loss": 0.6137, |
| "mean_token_accuracy": 0.8227979384362698, |
| "num_tokens": 26515548.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.6666794860735535, |
| "epoch": 1.7797930343829975, |
| "grad_norm": 1.578125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 0.6268, |
| "mean_token_accuracy": 0.8173129990696907, |
| "num_tokens": 26782376.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_chemistry_entropy": 0.6736977963447571, |
| "eval_chemistry_loss": 0.6848514676094055, |
| "eval_chemistry_mean_token_accuracy": 0.8112705006599427, |
| "eval_chemistry_num_tokens": 26782376.0, |
| "eval_chemistry_runtime": 34.6023, |
| "eval_chemistry_samples_per_second": 14.45, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.7797930343829975, |
| "eval_physics_entropy": 0.6228616139888763, |
| "eval_physics_loss": 0.6303527355194092, |
| "eval_physics_mean_token_accuracy": 0.8230352792739868, |
| "eval_physics_num_tokens": 26782376.0, |
| "eval_physics_runtime": 40.9503, |
| "eval_physics_samples_per_second": 12.21, |
| "eval_physics_steps_per_second": 3.052, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.6698539135977626, |
| "epoch": 1.7975965283186826, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 0.64, |
| "mean_token_accuracy": 0.815592122823, |
| "num_tokens": 27049648.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.6951070323586463, |
| "epoch": 1.8154000222543676, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 0.6636, |
| "mean_token_accuracy": 0.810206351801753, |
| "num_tokens": 27315061.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.6561093555763364, |
| "epoch": 1.8332035161900522, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 0.6193, |
| "mean_token_accuracy": 0.8209650807082653, |
| "num_tokens": 27587650.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.673011284135282, |
| "epoch": 1.8510070101257372, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 0.6476, |
| "mean_token_accuracy": 0.8167991150170565, |
| "num_tokens": 27855458.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.6599440133199096, |
| "epoch": 1.868810504061422, |
| "grad_norm": 1.578125, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 0.6309, |
| "mean_token_accuracy": 0.82068553827703, |
| "num_tokens": 28127703.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.6387615172192455, |
| "epoch": 1.8866139979971068, |
| "grad_norm": 1.375, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 0.6027, |
| "mean_token_accuracy": 0.8241243101656437, |
| "num_tokens": 28395128.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.661963646300137, |
| "epoch": 1.9044174919327919, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 0.6311, |
| "mean_token_accuracy": 0.817503708600998, |
| "num_tokens": 28662344.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.6643290909007191, |
| "epoch": 1.9222209858684767, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 0.6355, |
| "mean_token_accuracy": 0.8175244953483343, |
| "num_tokens": 28928769.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.663678290694952, |
| "epoch": 1.9400244798041615, |
| "grad_norm": 1.8359375, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 0.6318, |
| "mean_token_accuracy": 0.8165966872125864, |
| "num_tokens": 29186214.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.6729184232652188, |
| "epoch": 1.9578279737398465, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.978e-05, |
| "loss": 0.6437, |
| "mean_token_accuracy": 0.8148789752274752, |
| "num_tokens": 29446445.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_chemistry_entropy": 0.667749724149704, |
| "eval_chemistry_loss": 0.670352041721344, |
| "eval_chemistry_mean_token_accuracy": 0.8144329328536988, |
| "eval_chemistry_num_tokens": 29446445.0, |
| "eval_chemistry_runtime": 34.5967, |
| "eval_chemistry_samples_per_second": 14.452, |
| "eval_chemistry_steps_per_second": 3.613, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.9578279737398465, |
| "eval_physics_entropy": 0.6239387395381928, |
| "eval_physics_loss": 0.6269068121910095, |
| "eval_physics_mean_token_accuracy": 0.8240941405296326, |
| "eval_physics_num_tokens": 29446445.0, |
| "eval_physics_runtime": 41.2535, |
| "eval_physics_samples_per_second": 12.12, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.6582659302279353, |
| "epoch": 1.9756314676755313, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 0.6207, |
| "mean_token_accuracy": 0.8203308090567589, |
| "num_tokens": 29711771.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.6604506926611066, |
| "epoch": 1.9934349616112161, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 0.6317, |
| "mean_token_accuracy": 0.8193308029323816, |
| "num_tokens": 29979994.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.604119231431715, |
| "epoch": 2.010682096361411, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 0.5701, |
| "mean_token_accuracy": 0.83187814950943, |
| "num_tokens": 30240793.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.5444270031526685, |
| "epoch": 2.028485590297096, |
| "grad_norm": 1.671875, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 0.5188, |
| "mean_token_accuracy": 0.8410136397928, |
| "num_tokens": 30503988.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.5468037134036422, |
| "epoch": 2.0462890842327806, |
| "grad_norm": 1.8125, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 0.5163, |
| "mean_token_accuracy": 0.8412827380001545, |
| "num_tokens": 30771841.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.5360686406493187, |
| "epoch": 2.0640925781684656, |
| "grad_norm": 1.8203125, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 0.5083, |
| "mean_token_accuracy": 0.8440209306776524, |
| "num_tokens": 31045037.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.5755055965855718, |
| "epoch": 2.0818960721041506, |
| "grad_norm": 1.7890625, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 0.5422, |
| "mean_token_accuracy": 0.8334469784051179, |
| "num_tokens": 31299575.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.5716140467673541, |
| "epoch": 2.099699566039835, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 0.5423, |
| "mean_token_accuracy": 0.8336410760879517, |
| "num_tokens": 31565963.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.5382599098607898, |
| "epoch": 2.1175030599755202, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 0.5168, |
| "mean_token_accuracy": 0.8428799293935298, |
| "num_tokens": 31834212.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.5634422322735191, |
| "epoch": 2.1353065539112053, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 0.5327, |
| "mean_token_accuracy": 0.8359647855162621, |
| "num_tokens": 32103356.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_chemistry_entropy": 0.5813326196670532, |
| "eval_chemistry_loss": 0.6811516284942627, |
| "eval_chemistry_mean_token_accuracy": 0.8141688752174377, |
| "eval_chemistry_num_tokens": 32103356.0, |
| "eval_chemistry_runtime": 34.6182, |
| "eval_chemistry_samples_per_second": 14.443, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.1353065539112053, |
| "eval_physics_entropy": 0.5421837236881256, |
| "eval_physics_loss": 0.6482248306274414, |
| "eval_physics_mean_token_accuracy": 0.8212747263908386, |
| "eval_physics_num_tokens": 32103356.0, |
| "eval_physics_runtime": 41.2474, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.5539132442325354, |
| "epoch": 2.15311004784689, |
| "grad_norm": 1.6171875, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 0.519, |
| "mean_token_accuracy": 0.8395270679146052, |
| "num_tokens": 32374830.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5340574728325009, |
| "epoch": 2.170913541782575, |
| "grad_norm": 1.671875, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 0.5083, |
| "mean_token_accuracy": 0.8442022494971753, |
| "num_tokens": 32643479.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.540911185927689, |
| "epoch": 2.18871703571826, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 0.5126, |
| "mean_token_accuracy": 0.8427833508700132, |
| "num_tokens": 32908761.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.5624004138633609, |
| "epoch": 2.2065205296539445, |
| "grad_norm": 1.5625, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 0.5288, |
| "mean_token_accuracy": 0.8391110297292471, |
| "num_tokens": 33166413.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.5559011250734329, |
| "epoch": 2.2243240235896296, |
| "grad_norm": 1.53125, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 0.5314, |
| "mean_token_accuracy": 0.8375788435339928, |
| "num_tokens": 33432686.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.5524005191400647, |
| "epoch": 2.242127517525314, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 0.5252, |
| "mean_token_accuracy": 0.8396644528955222, |
| "num_tokens": 33693242.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.5266111794859171, |
| "epoch": 2.259931011460999, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 0.4966, |
| "mean_token_accuracy": 0.847085589542985, |
| "num_tokens": 33970820.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.5435060489922762, |
| "epoch": 2.277734505396684, |
| "grad_norm": 1.75, |
| "learning_rate": 1.938e-05, |
| "loss": 0.5191, |
| "mean_token_accuracy": 0.8409085746854543, |
| "num_tokens": 34243042.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.5497656201943755, |
| "epoch": 2.295537999332369, |
| "grad_norm": 1.6875, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 0.5245, |
| "mean_token_accuracy": 0.8400060147047043, |
| "num_tokens": 34515236.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.5251918986439705, |
| "epoch": 2.313341493268054, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 0.4979, |
| "mean_token_accuracy": 0.848014684766531, |
| "num_tokens": 34785068.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_chemistry_entropy": 0.5687024412155152, |
| "eval_chemistry_loss": 0.6814095377922058, |
| "eval_chemistry_mean_token_accuracy": 0.8144371585845948, |
| "eval_chemistry_num_tokens": 34785068.0, |
| "eval_chemistry_runtime": 34.6075, |
| "eval_chemistry_samples_per_second": 14.448, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.313341493268054, |
| "eval_physics_entropy": 0.5373544907569885, |
| "eval_physics_loss": 0.6523818373680115, |
| "eval_physics_mean_token_accuracy": 0.8211935892105102, |
| "eval_physics_num_tokens": 34785068.0, |
| "eval_physics_runtime": 41.2549, |
| "eval_physics_samples_per_second": 12.12, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.5557622062042356, |
| "epoch": 2.331144987203739, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 0.5275, |
| "mean_token_accuracy": 0.8374897003173828, |
| "num_tokens": 35055845.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5468069275841116, |
| "epoch": 2.3489484811394234, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.5189, |
| "mean_token_accuracy": 0.8414008297026158, |
| "num_tokens": 35327481.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.5578970231115818, |
| "epoch": 2.3667519750751085, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 0.5327, |
| "mean_token_accuracy": 0.837207018584013, |
| "num_tokens": 35596362.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5461308034136891, |
| "epoch": 2.3845554690107935, |
| "grad_norm": 1.6328125, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 0.5144, |
| "mean_token_accuracy": 0.8420327302068472, |
| "num_tokens": 35869585.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5518748119473458, |
| "epoch": 2.402358962946478, |
| "grad_norm": 1.765625, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 0.5271, |
| "mean_token_accuracy": 0.8386039260774851, |
| "num_tokens": 36134072.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.5443382592871785, |
| "epoch": 2.420162456882163, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 0.5177, |
| "mean_token_accuracy": 0.8425142876803875, |
| "num_tokens": 36406263.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.5503683429211378, |
| "epoch": 2.437965950817848, |
| "grad_norm": 1.71875, |
| "learning_rate": 1.918e-05, |
| "loss": 0.5193, |
| "mean_token_accuracy": 0.8414091795682908, |
| "num_tokens": 36676822.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.5444278137758374, |
| "epoch": 2.4557694447535328, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 0.5187, |
| "mean_token_accuracy": 0.841403117403388, |
| "num_tokens": 36942837.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.5489616293460131, |
| "epoch": 2.473572938689218, |
| "grad_norm": 1.8046875, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 0.523, |
| "mean_token_accuracy": 0.8393812999129295, |
| "num_tokens": 37210791.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.5314275240525603, |
| "epoch": 2.491376432624903, |
| "grad_norm": 1.7265625, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.5119, |
| "mean_token_accuracy": 0.8442874882370234, |
| "num_tokens": 37485822.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_chemistry_entropy": 0.5652910959720612, |
| "eval_chemistry_loss": 0.6705389618873596, |
| "eval_chemistry_mean_token_accuracy": 0.8175577726364136, |
| "eval_chemistry_num_tokens": 37485822.0, |
| "eval_chemistry_runtime": 34.6247, |
| "eval_chemistry_samples_per_second": 14.441, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.491376432624903, |
| "eval_physics_entropy": 0.5328873710632325, |
| "eval_physics_loss": 0.6484245657920837, |
| "eval_physics_mean_token_accuracy": 0.8219989495277404, |
| "eval_physics_num_tokens": 37485822.0, |
| "eval_physics_runtime": 41.2499, |
| "eval_physics_samples_per_second": 12.121, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.5436741337180138, |
| "epoch": 2.5091799265605874, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.5135, |
| "mean_token_accuracy": 0.8424118865281344, |
| "num_tokens": 37746150.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.5462065914645791, |
| "epoch": 2.5269834204962724, |
| "grad_norm": 1.609375, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.5158, |
| "mean_token_accuracy": 0.8402331713587046, |
| "num_tokens": 38016863.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.56711411960423, |
| "epoch": 2.544786914431957, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 0.5349, |
| "mean_token_accuracy": 0.8344442717730999, |
| "num_tokens": 38277754.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.5571039380505681, |
| "epoch": 2.562590408367642, |
| "grad_norm": 1.734375, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 0.529, |
| "mean_token_accuracy": 0.8378622565418482, |
| "num_tokens": 38542761.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.5426763394847512, |
| "epoch": 2.580393902303327, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.5159, |
| "mean_token_accuracy": 0.8411751441657543, |
| "num_tokens": 38820999.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.5453944187611341, |
| "epoch": 2.598197396239012, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.898e-05, |
| "loss": 0.5238, |
| "mean_token_accuracy": 0.8403888408094644, |
| "num_tokens": 39080016.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5451987128704786, |
| "epoch": 2.6160008901746967, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 0.5123, |
| "mean_token_accuracy": 0.8426350578665733, |
| "num_tokens": 39345434.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.5478348048403859, |
| "epoch": 2.6338043841103818, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 0.5212, |
| "mean_token_accuracy": 0.8402178969234229, |
| "num_tokens": 39610041.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.5249082477763295, |
| "epoch": 2.6516078780460663, |
| "grad_norm": 1.7890625, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 0.4993, |
| "mean_token_accuracy": 0.8461574167013168, |
| "num_tokens": 39884185.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.5606990542262793, |
| "epoch": 2.6694113719817514, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.533, |
| "mean_token_accuracy": 0.8361463028937578, |
| "num_tokens": 40143543.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_chemistry_entropy": 0.5683894844055176, |
| "eval_chemistry_loss": 0.6641016602516174, |
| "eval_chemistry_mean_token_accuracy": 0.817899118423462, |
| "eval_chemistry_num_tokens": 40143543.0, |
| "eval_chemistry_runtime": 34.613, |
| "eval_chemistry_samples_per_second": 14.445, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.6694113719817514, |
| "eval_physics_entropy": 0.5443970229625702, |
| "eval_physics_loss": 0.6480051279067993, |
| "eval_physics_mean_token_accuracy": 0.8217208013534546, |
| "eval_physics_num_tokens": 40143543.0, |
| "eval_physics_runtime": 40.9767, |
| "eval_physics_samples_per_second": 12.202, |
| "eval_physics_steps_per_second": 3.051, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.5430334324017168, |
| "epoch": 2.6872148659174364, |
| "grad_norm": 1.7421875, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 0.5152, |
| "mean_token_accuracy": 0.8425797671079636, |
| "num_tokens": 40406812.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.5591320911422372, |
| "epoch": 2.7050183598531214, |
| "grad_norm": 1.8359375, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.5348, |
| "mean_token_accuracy": 0.8362568471580744, |
| "num_tokens": 40668543.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.5530628427863121, |
| "epoch": 2.722821853788806, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 0.518, |
| "mean_token_accuracy": 0.8411050379276276, |
| "num_tokens": 40942603.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.556857710890472, |
| "epoch": 2.740625347724491, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.5243, |
| "mean_token_accuracy": 0.838795681297779, |
| "num_tokens": 41208093.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.5422420864924788, |
| "epoch": 2.7584288416601757, |
| "grad_norm": 1.609375, |
| "learning_rate": 1.878e-05, |
| "loss": 0.5204, |
| "mean_token_accuracy": 0.8388970494270325, |
| "num_tokens": 41477562.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.5319694015197456, |
| "epoch": 2.7762323355958607, |
| "grad_norm": 1.8359375, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.5078, |
| "mean_token_accuracy": 0.8438673295080662, |
| "num_tokens": 41746571.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5427202990278601, |
| "epoch": 2.7940358295315457, |
| "grad_norm": 1.703125, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 0.5141, |
| "mean_token_accuracy": 0.842416912689805, |
| "num_tokens": 42011633.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.5518008042126894, |
| "epoch": 2.8118393234672303, |
| "grad_norm": 1.546875, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.5238, |
| "mean_token_accuracy": 0.8391041114926339, |
| "num_tokens": 42283087.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.5269314305856824, |
| "epoch": 2.8296428174029153, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 0.5, |
| "mean_token_accuracy": 0.8462934419512749, |
| "num_tokens": 42561534.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.54100974611938, |
| "epoch": 2.8474463113386, |
| "grad_norm": 1.6328125, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 0.5206, |
| "mean_token_accuracy": 0.8429192952811718, |
| "num_tokens": 42834266.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_chemistry_entropy": 0.5678292090892791, |
| "eval_chemistry_loss": 0.6582387089729309, |
| "eval_chemistry_mean_token_accuracy": 0.8187746982574463, |
| "eval_chemistry_num_tokens": 42834266.0, |
| "eval_chemistry_runtime": 34.6036, |
| "eval_chemistry_samples_per_second": 14.449, |
| "eval_chemistry_steps_per_second": 3.612, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.8474463113386, |
| "eval_physics_entropy": 0.5397046103477477, |
| "eval_physics_loss": 0.6487213969230652, |
| "eval_physics_mean_token_accuracy": 0.8217604222297669, |
| "eval_physics_num_tokens": 42834266.0, |
| "eval_physics_runtime": 41.2474, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.03, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.5642041997984052, |
| "epoch": 2.865249805274285, |
| "grad_norm": 1.796875, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 0.5371, |
| "mean_token_accuracy": 0.8362252287566662, |
| "num_tokens": 43100513.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.553196731954813, |
| "epoch": 2.88305329920997, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 0.5214, |
| "mean_token_accuracy": 0.8403059497475625, |
| "num_tokens": 43375632.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.5369223712012172, |
| "epoch": 2.900856793145655, |
| "grad_norm": 1.640625, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 0.5167, |
| "mean_token_accuracy": 0.8430781438946724, |
| "num_tokens": 43644341.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.5357730442658066, |
| "epoch": 2.9186602870813396, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.858e-05, |
| "loss": 0.5061, |
| "mean_token_accuracy": 0.8430858284235001, |
| "num_tokens": 43917580.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.5374868368729949, |
| "epoch": 2.9364637810170247, |
| "grad_norm": 1.7578125, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 0.5116, |
| "mean_token_accuracy": 0.8447067823261023, |
| "num_tokens": 44177325.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.5514863247051836, |
| "epoch": 2.9542672749527092, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.5211, |
| "mean_token_accuracy": 0.8395764529705048, |
| "num_tokens": 44442947.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5531215896829963, |
| "epoch": 2.9720707688883943, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.5287, |
| "mean_token_accuracy": 0.8382051270455122, |
| "num_tokens": 44700601.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.5407275201752781, |
| "epoch": 2.9898742628240793, |
| "grad_norm": 1.59375, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.51, |
| "mean_token_accuracy": 0.844194658845663, |
| "num_tokens": 44971479.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.49610343229386117, |
| "epoch": 3.007121397574274, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 0.4379, |
| "mean_token_accuracy": 0.8622554202233591, |
| "num_tokens": 45231796.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.3639660466462374, |
| "epoch": 3.0249248915099587, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.3489, |
| "mean_token_accuracy": 0.8842393532395363, |
| "num_tokens": 45500317.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_chemistry_entropy": 0.40514632713794707, |
| "eval_chemistry_loss": 0.756369411945343, |
| "eval_chemistry_mean_token_accuracy": 0.815176130771637, |
| "eval_chemistry_num_tokens": 45500317.0, |
| "eval_chemistry_runtime": 34.6159, |
| "eval_chemistry_samples_per_second": 14.444, |
| "eval_chemistry_steps_per_second": 3.611, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.0249248915099587, |
| "eval_physics_entropy": 0.4021349341869354, |
| "eval_physics_loss": 0.7458917498588562, |
| "eval_physics_mean_token_accuracy": 0.8158441076278686, |
| "eval_physics_num_tokens": 45500317.0, |
| "eval_physics_runtime": 41.2473, |
| "eval_physics_samples_per_second": 12.122, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.35829943269491193, |
| "epoch": 3.0427283854456437, |
| "grad_norm": 2.125, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 0.3155, |
| "mean_token_accuracy": 0.8941227950155735, |
| "num_tokens": 45762680.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.3386535131372511, |
| "epoch": 3.0605318793813288, |
| "grad_norm": 2.296875, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 0.3128, |
| "mean_token_accuracy": 0.8955158289521933, |
| "num_tokens": 46029923.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.3458150915801525, |
| "epoch": 3.0783353733170133, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.3192, |
| "mean_token_accuracy": 0.8931533485651016, |
| "num_tokens": 46305156.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.3377646486274898, |
| "epoch": 3.0961388672526984, |
| "grad_norm": 2.1875, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.3127, |
| "mean_token_accuracy": 0.8952449705451727, |
| "num_tokens": 46582874.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.35565020963549615, |
| "epoch": 3.1139423611883834, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.327, |
| "mean_token_accuracy": 0.8901556842029095, |
| "num_tokens": 46850175.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.3427689325064421, |
| "epoch": 3.131745855124068, |
| "grad_norm": 2.0625, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.3146, |
| "mean_token_accuracy": 0.8948938645422458, |
| "num_tokens": 47125899.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.345877396594733, |
| "epoch": 3.149549349059753, |
| "grad_norm": 2.421875, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 0.3207, |
| "mean_token_accuracy": 0.8928927041590213, |
| "num_tokens": 47390194.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.3544696198776364, |
| "epoch": 3.167352842995438, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 0.324, |
| "mean_token_accuracy": 0.8916583750396967, |
| "num_tokens": 47654784.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.345364089962095, |
| "epoch": 3.1851563369311227, |
| "grad_norm": 2.25, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 0.32, |
| "mean_token_accuracy": 0.8935606256127357, |
| "num_tokens": 47919400.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.3512003905139863, |
| "epoch": 3.2029598308668077, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 0.3262, |
| "mean_token_accuracy": 0.8913596641272307, |
| "num_tokens": 48187796.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_chemistry_entropy": 0.39868619096279145, |
| "eval_chemistry_loss": 0.7814938426017761, |
| "eval_chemistry_mean_token_accuracy": 0.8135026731491088, |
| "eval_chemistry_num_tokens": 48187796.0, |
| "eval_chemistry_runtime": 34.5848, |
| "eval_chemistry_samples_per_second": 14.457, |
| "eval_chemistry_steps_per_second": 3.614, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.2029598308668077, |
| "eval_physics_entropy": 0.38585546863079073, |
| "eval_physics_loss": 0.785114049911499, |
| "eval_physics_mean_token_accuracy": 0.8136380562782287, |
| "eval_physics_num_tokens": 48187796.0, |
| "eval_physics_runtime": 41.2383, |
| "eval_physics_samples_per_second": 12.125, |
| "eval_physics_steps_per_second": 3.031, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.34991011349484324, |
| "epoch": 3.2207633248024923, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 0.3223, |
| "mean_token_accuracy": 0.8920448690652847, |
| "num_tokens": 48457428.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.3759195095859468, |
| "epoch": 3.2385668187381773, |
| "grad_norm": 2.359375, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.341, |
| "mean_token_accuracy": 0.8847057934850454, |
| "num_tokens": 48716417.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.3503100727684796, |
| "epoch": 3.2563703126738623, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 0.3211, |
| "mean_token_accuracy": 0.8925649128854275, |
| "num_tokens": 48977306.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.34978028303012254, |
| "epoch": 3.274173806609547, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 0.3206, |
| "mean_token_accuracy": 0.8920259241014719, |
| "num_tokens": 49249667.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.35568869626149535, |
| "epoch": 3.291977300545232, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.3283, |
| "mean_token_accuracy": 0.8901085384190083, |
| "num_tokens": 49520900.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.35994688216596843, |
| "epoch": 3.309780794480917, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.3309, |
| "mean_token_accuracy": 0.8887050859630108, |
| "num_tokens": 49786411.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.34956529131159186, |
| "epoch": 3.3275842884166016, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 0.3194, |
| "mean_token_accuracy": 0.8924664918333292, |
| "num_tokens": 50056207.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.35508423103019593, |
| "epoch": 3.3453877823522866, |
| "grad_norm": 2.078125, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.3273, |
| "mean_token_accuracy": 0.8904457587748766, |
| "num_tokens": 50330876.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.35782372783869504, |
| "epoch": 3.3631912762879717, |
| "grad_norm": 2.4375, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.3313, |
| "mean_token_accuracy": 0.8893947552889585, |
| "num_tokens": 50602004.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.36222805520519613, |
| "epoch": 3.3809947702236562, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.3328, |
| "mean_token_accuracy": 0.8882043663412332, |
| "num_tokens": 50873444.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_chemistry_entropy": 0.3968628585338593, |
| "eval_chemistry_loss": 0.7823912501335144, |
| "eval_chemistry_mean_token_accuracy": 0.8138084063529968, |
| "eval_chemistry_num_tokens": 50873444.0, |
| "eval_chemistry_runtime": 34.5801, |
| "eval_chemistry_samples_per_second": 14.459, |
| "eval_chemistry_steps_per_second": 3.615, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.3809947702236562, |
| "eval_physics_entropy": 0.3825342948436737, |
| "eval_physics_loss": 0.789954423904419, |
| "eval_physics_mean_token_accuracy": 0.8131929535865784, |
| "eval_physics_num_tokens": 50873444.0, |
| "eval_physics_runtime": 41.2311, |
| "eval_physics_samples_per_second": 12.127, |
| "eval_physics_steps_per_second": 3.032, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.3620292558334768, |
| "epoch": 3.3987982641593413, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.798e-05, |
| "loss": 0.3342, |
| "mean_token_accuracy": 0.8878430653363466, |
| "num_tokens": 51141077.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.3582372093573213, |
| "epoch": 3.4166017580950263, |
| "grad_norm": 2.375, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.3297, |
| "mean_token_accuracy": 0.8890867009758949, |
| "num_tokens": 51406483.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.3618839686736465, |
| "epoch": 3.434405252030711, |
| "grad_norm": 2.4375, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.3301, |
| "mean_token_accuracy": 0.8890988543629647, |
| "num_tokens": 51668458.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.36373381447046993, |
| "epoch": 3.452208745966396, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.3361, |
| "mean_token_accuracy": 0.8878721829503775, |
| "num_tokens": 51928199.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.35968201477080586, |
| "epoch": 3.470012239902081, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.3309, |
| "mean_token_accuracy": 0.8883917711675167, |
| "num_tokens": 52199244.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.35602816781029106, |
| "epoch": 3.4878157338377656, |
| "grad_norm": 2.375, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.3274, |
| "mean_token_accuracy": 0.8896724134683609, |
| "num_tokens": 52473476.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.3581710479222238, |
| "epoch": 3.5056192277734506, |
| "grad_norm": 1.9921875, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.3307, |
| "mean_token_accuracy": 0.8886904548853636, |
| "num_tokens": 52743275.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.3580263523384929, |
| "epoch": 3.523422721709135, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.328, |
| "mean_token_accuracy": 0.8894063163548708, |
| "num_tokens": 53014827.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.3583196923136711, |
| "epoch": 3.54122621564482, |
| "grad_norm": 2.46875, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.3328, |
| "mean_token_accuracy": 0.8892181769013405, |
| "num_tokens": 53285577.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.35019520940259097, |
| "epoch": 3.5590297095805052, |
| "grad_norm": 2.359375, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.3253, |
| "mean_token_accuracy": 0.8909101441502572, |
| "num_tokens": 53553391.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_chemistry_entropy": 0.40147681057453155, |
| "eval_chemistry_loss": 0.778447151184082, |
| "eval_chemistry_mean_token_accuracy": 0.8141673741340637, |
| "eval_chemistry_num_tokens": 53553391.0, |
| "eval_chemistry_runtime": 34.6216, |
| "eval_chemistry_samples_per_second": 14.442, |
| "eval_chemistry_steps_per_second": 3.61, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.5590297095805052, |
| "eval_physics_entropy": 0.3869015793800354, |
| "eval_physics_loss": 0.7880970239639282, |
| "eval_physics_mean_token_accuracy": 0.813429744720459, |
| "eval_physics_num_tokens": 53553391.0, |
| "eval_physics_runtime": 40.9274, |
| "eval_physics_samples_per_second": 12.217, |
| "eval_physics_steps_per_second": 3.054, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 18, |
| "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 |
| } |
|
|
