| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.6, |
| "eval_steps": 100, |
| "global_step": 9000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.7581225659698247, |
| "epoch": 0.004, |
| "grad_norm": 101.5, |
| "learning_rate": 1.8e-07, |
| "loss": 18.1141, |
| "mean_token_accuracy": 0.7393306585028767, |
| "num_tokens": 72542.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.7520309612154961, |
| "epoch": 0.008, |
| "grad_norm": 162.0, |
| "learning_rate": 3.8e-07, |
| "loss": 18.2441, |
| "mean_token_accuracy": 0.7392390985041857, |
| "num_tokens": 138146.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.741816989146173, |
| "epoch": 0.012, |
| "grad_norm": 92.5, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 17.8539, |
| "mean_token_accuracy": 0.7405363619327545, |
| "num_tokens": 208409.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.7636421315371991, |
| "epoch": 0.016, |
| "grad_norm": 88.0, |
| "learning_rate": 7.8e-07, |
| "loss": 18.142, |
| "mean_token_accuracy": 0.7400334309786558, |
| "num_tokens": 280941.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.7905357209965587, |
| "epoch": 0.02, |
| "grad_norm": 108.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 18.1041, |
| "mean_token_accuracy": 0.7354144401848316, |
| "num_tokens": 349731.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8223027169704438, |
| "epoch": 0.024, |
| "grad_norm": 112.5, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 18.4599, |
| "mean_token_accuracy": 0.731158659607172, |
| "num_tokens": 419127.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.7908559931442142, |
| "epoch": 0.028, |
| "grad_norm": 106.0, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 17.5053, |
| "mean_token_accuracy": 0.7440328601747751, |
| "num_tokens": 489016.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8229548260569572, |
| "epoch": 0.032, |
| "grad_norm": 86.0, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 17.3714, |
| "mean_token_accuracy": 0.7438380591571331, |
| "num_tokens": 558056.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.84512098710984, |
| "epoch": 0.036, |
| "grad_norm": 100.5, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 17.2456, |
| "mean_token_accuracy": 0.7478047309443354, |
| "num_tokens": 627804.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8738276792690158, |
| "epoch": 0.04, |
| "grad_norm": 67.0, |
| "learning_rate": 1.98e-06, |
| "loss": 16.6896, |
| "mean_token_accuracy": 0.748936639726162, |
| "num_tokens": 697556.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_biology_entropy": 1.1379984815716744, |
| "eval_biology_loss": 1.2692387104034424, |
| "eval_biology_mean_token_accuracy": 0.6880569371581078, |
| "eval_biology_num_tokens": 697556.0, |
| "eval_biology_runtime": 40.5573, |
| "eval_biology_samples_per_second": 12.328, |
| "eval_biology_steps_per_second": 12.328, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_math_entropy": 0.9095822716951371, |
| "eval_math_loss": 1.0899367332458496, |
| "eval_math_mean_token_accuracy": 0.7395683622956276, |
| "eval_math_num_tokens": 697556.0, |
| "eval_math_runtime": 48.0541, |
| "eval_math_samples_per_second": 10.405, |
| "eval_math_steps_per_second": 10.405, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.9193563124164938, |
| "epoch": 0.044, |
| "grad_norm": 88.5, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 17.3763, |
| "mean_token_accuracy": 0.7420263335108757, |
| "num_tokens": 765262.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8906641118228436, |
| "epoch": 0.048, |
| "grad_norm": 78.0, |
| "learning_rate": 2.38e-06, |
| "loss": 16.2142, |
| "mean_token_accuracy": 0.7552934620529413, |
| "num_tokens": 836753.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.9178624864667654, |
| "epoch": 0.052, |
| "grad_norm": 70.0, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 16.3594, |
| "mean_token_accuracy": 0.7507667759433388, |
| "num_tokens": 909947.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.9556385111063719, |
| "epoch": 0.056, |
| "grad_norm": 70.0, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 16.6438, |
| "mean_token_accuracy": 0.748611630871892, |
| "num_tokens": 981959.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.9541467605158687, |
| "epoch": 0.06, |
| "grad_norm": 86.0, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 16.5304, |
| "mean_token_accuracy": 0.7486465960741043, |
| "num_tokens": 1053624.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.9960285406559706, |
| "epoch": 0.064, |
| "grad_norm": 78.0, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 16.9435, |
| "mean_token_accuracy": 0.7448480188846588, |
| "num_tokens": 1126446.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.9677659103646874, |
| "epoch": 0.068, |
| "grad_norm": 106.5, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 16.1013, |
| "mean_token_accuracy": 0.7561902595683933, |
| "num_tokens": 1196286.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.9270429328083992, |
| "epoch": 0.072, |
| "grad_norm": 62.0, |
| "learning_rate": 3.58e-06, |
| "loss": 15.3741, |
| "mean_token_accuracy": 0.763100459985435, |
| "num_tokens": 1273905.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.9678281666710973, |
| "epoch": 0.076, |
| "grad_norm": 51.0, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 15.8618, |
| "mean_token_accuracy": 0.7534618094563484, |
| "num_tokens": 1346454.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.9695806819945574, |
| "epoch": 0.08, |
| "grad_norm": 61.25, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 15.7874, |
| "mean_token_accuracy": 0.7560878146439791, |
| "num_tokens": 1413596.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_biology_entropy": 1.214613477051258, |
| "eval_biology_loss": 1.2630934715270996, |
| "eval_biology_mean_token_accuracy": 0.6885634814500808, |
| "eval_biology_num_tokens": 1413596.0, |
| "eval_biology_runtime": 40.4294, |
| "eval_biology_samples_per_second": 12.367, |
| "eval_biology_steps_per_second": 12.367, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_math_entropy": 0.9574190213680267, |
| "eval_math_loss": 0.9871246218681335, |
| "eval_math_mean_token_accuracy": 0.7557044946551323, |
| "eval_math_num_tokens": 1413596.0, |
| "eval_math_runtime": 48.0895, |
| "eval_math_samples_per_second": 10.397, |
| "eval_math_steps_per_second": 10.397, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8888338362798095, |
| "epoch": 0.084, |
| "grad_norm": 61.25, |
| "learning_rate": 4.18e-06, |
| "loss": 14.623, |
| "mean_token_accuracy": 0.7706882227212191, |
| "num_tokens": 1485147.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.906145141646266, |
| "epoch": 0.088, |
| "grad_norm": 53.5, |
| "learning_rate": 4.38e-06, |
| "loss": 14.8128, |
| "mean_token_accuracy": 0.7661576565355063, |
| "num_tokens": 1559450.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.95110107883811, |
| "epoch": 0.092, |
| "grad_norm": 63.25, |
| "learning_rate": 4.58e-06, |
| "loss": 15.2457, |
| "mean_token_accuracy": 0.761219710484147, |
| "num_tokens": 1631535.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.9411580273881555, |
| "epoch": 0.096, |
| "grad_norm": 51.0, |
| "learning_rate": 4.78e-06, |
| "loss": 14.938, |
| "mean_token_accuracy": 0.7646324578672647, |
| "num_tokens": 1701193.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.9254088591784239, |
| "epoch": 0.1, |
| "grad_norm": 56.25, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 14.6029, |
| "mean_token_accuracy": 0.7678503680974245, |
| "num_tokens": 1773596.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.9214777417480946, |
| "epoch": 0.104, |
| "grad_norm": 54.0, |
| "learning_rate": 5.18e-06, |
| "loss": 15.043, |
| "mean_token_accuracy": 0.7618352189660073, |
| "num_tokens": 1840165.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.9363527970388532, |
| "epoch": 0.108, |
| "grad_norm": 111.5, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 15.6052, |
| "mean_token_accuracy": 0.7564919672906398, |
| "num_tokens": 1908149.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9113753678277134, |
| "epoch": 0.112, |
| "grad_norm": 53.0, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 14.4645, |
| "mean_token_accuracy": 0.7736385434865951, |
| "num_tokens": 1979232.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.916975767724216, |
| "epoch": 0.116, |
| "grad_norm": 54.5, |
| "learning_rate": 5.78e-06, |
| "loss": 15.0318, |
| "mean_token_accuracy": 0.7656373519450426, |
| "num_tokens": 2052039.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.915022067911923, |
| "epoch": 0.12, |
| "grad_norm": 46.25, |
| "learning_rate": 5.98e-06, |
| "loss": 14.7236, |
| "mean_token_accuracy": 0.7691222932189703, |
| "num_tokens": 2125815.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_biology_entropy": 1.2286294117569923, |
| "eval_biology_loss": 1.268429160118103, |
| "eval_biology_mean_token_accuracy": 0.6861296361088752, |
| "eval_biology_num_tokens": 2125815.0, |
| "eval_biology_runtime": 40.4544, |
| "eval_biology_samples_per_second": 12.36, |
| "eval_biology_steps_per_second": 12.36, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_math_entropy": 0.9302992832064628, |
| "eval_math_loss": 0.9273415803909302, |
| "eval_math_mean_token_accuracy": 0.7664831226468086, |
| "eval_math_num_tokens": 2125815.0, |
| "eval_math_runtime": 48.047, |
| "eval_math_samples_per_second": 10.406, |
| "eval_math_steps_per_second": 10.406, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9336849415674806, |
| "epoch": 0.124, |
| "grad_norm": 55.25, |
| "learning_rate": 6.18e-06, |
| "loss": 14.9674, |
| "mean_token_accuracy": 0.7661335166543722, |
| "num_tokens": 2201991.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.8756070390343667, |
| "epoch": 0.128, |
| "grad_norm": 52.25, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 14.0, |
| "mean_token_accuracy": 0.7786776568740607, |
| "num_tokens": 2274177.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.892756593041122, |
| "epoch": 0.132, |
| "grad_norm": 45.75, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 14.1896, |
| "mean_token_accuracy": 0.7731969062238931, |
| "num_tokens": 2342173.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.8899674592539668, |
| "epoch": 0.136, |
| "grad_norm": 56.5, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 14.2167, |
| "mean_token_accuracy": 0.7683363553136587, |
| "num_tokens": 2412435.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.9237651886418462, |
| "epoch": 0.14, |
| "grad_norm": 49.5, |
| "learning_rate": 6.98e-06, |
| "loss": 14.7203, |
| "mean_token_accuracy": 0.7684664275497198, |
| "num_tokens": 2481105.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.8494639024138451, |
| "epoch": 0.144, |
| "grad_norm": 49.0, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 13.6956, |
| "mean_token_accuracy": 0.7817885402590037, |
| "num_tokens": 2548445.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.8341110894456506, |
| "epoch": 0.148, |
| "grad_norm": 44.0, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 13.2223, |
| "mean_token_accuracy": 0.789648124948144, |
| "num_tokens": 2621364.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.8608193326741457, |
| "epoch": 0.152, |
| "grad_norm": 52.25, |
| "learning_rate": 7.58e-06, |
| "loss": 13.8861, |
| "mean_token_accuracy": 0.7754677496850491, |
| "num_tokens": 2686462.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.8447565982118249, |
| "epoch": 0.156, |
| "grad_norm": 50.0, |
| "learning_rate": 7.78e-06, |
| "loss": 13.5199, |
| "mean_token_accuracy": 0.7812942665070295, |
| "num_tokens": 2754241.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.8711243279278278, |
| "epoch": 0.16, |
| "grad_norm": 79.0, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 14.1276, |
| "mean_token_accuracy": 0.7742310840636492, |
| "num_tokens": 2824418.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.22758097666502, |
| "eval_biology_loss": 1.2762199640274048, |
| "eval_biology_mean_token_accuracy": 0.6836535354852676, |
| "eval_biology_num_tokens": 2824418.0, |
| "eval_biology_runtime": 40.7281, |
| "eval_biology_samples_per_second": 12.277, |
| "eval_biology_steps_per_second": 12.277, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 0.9047691844701767, |
| "eval_math_loss": 0.8841754198074341, |
| "eval_math_mean_token_accuracy": 0.775034920334816, |
| "eval_math_num_tokens": 2824418.0, |
| "eval_math_runtime": 47.9197, |
| "eval_math_samples_per_second": 10.434, |
| "eval_math_steps_per_second": 10.434, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.8859184484928846, |
| "epoch": 0.164, |
| "grad_norm": 127.0, |
| "learning_rate": 8.18e-06, |
| "loss": 14.142, |
| "mean_token_accuracy": 0.7765266574919224, |
| "num_tokens": 2895667.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.8501209424808621, |
| "epoch": 0.168, |
| "grad_norm": 40.0, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 13.8435, |
| "mean_token_accuracy": 0.7774271417409182, |
| "num_tokens": 2966031.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.8510567891411484, |
| "epoch": 0.172, |
| "grad_norm": 45.75, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 13.3673, |
| "mean_token_accuracy": 0.7858892437070608, |
| "num_tokens": 3039780.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.8355857279151678, |
| "epoch": 0.176, |
| "grad_norm": 44.5, |
| "learning_rate": 8.78e-06, |
| "loss": 13.7383, |
| "mean_token_accuracy": 0.7784004233777523, |
| "num_tokens": 3110313.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.8662491450086236, |
| "epoch": 0.18, |
| "grad_norm": 43.75, |
| "learning_rate": 8.98e-06, |
| "loss": 13.6485, |
| "mean_token_accuracy": 0.7806661173701286, |
| "num_tokens": 3179963.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.8596813324838877, |
| "epoch": 0.184, |
| "grad_norm": 46.0, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 13.7717, |
| "mean_token_accuracy": 0.7796813258901238, |
| "num_tokens": 3251139.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.8454847941175103, |
| "epoch": 0.188, |
| "grad_norm": 56.25, |
| "learning_rate": 9.38e-06, |
| "loss": 13.4682, |
| "mean_token_accuracy": 0.7803339377045632, |
| "num_tokens": 3323519.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.8228314617648721, |
| "epoch": 0.192, |
| "grad_norm": 49.75, |
| "learning_rate": 9.58e-06, |
| "loss": 13.3842, |
| "mean_token_accuracy": 0.7827629748731851, |
| "num_tokens": 3394170.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.8164506537839771, |
| "epoch": 0.196, |
| "grad_norm": 54.25, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 13.1965, |
| "mean_token_accuracy": 0.7878578577190638, |
| "num_tokens": 3465341.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.8838530456647277, |
| "epoch": 0.2, |
| "grad_norm": 51.0, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 13.9771, |
| "mean_token_accuracy": 0.7734297584742308, |
| "num_tokens": 3533571.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_biology_entropy": 1.2344867270588875, |
| "eval_biology_loss": 1.2817195653915405, |
| "eval_biology_mean_token_accuracy": 0.6832386125922203, |
| "eval_biology_num_tokens": 3533571.0, |
| "eval_biology_runtime": 40.4092, |
| "eval_biology_samples_per_second": 12.373, |
| "eval_biology_steps_per_second": 12.373, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_math_entropy": 0.8546719763875008, |
| "eval_math_loss": 0.8547734618186951, |
| "eval_math_mean_token_accuracy": 0.7804626071453095, |
| "eval_math_num_tokens": 3533571.0, |
| "eval_math_runtime": 48.0961, |
| "eval_math_samples_per_second": 10.396, |
| "eval_math_steps_per_second": 10.396, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.8578155474737287, |
| "epoch": 0.204, |
| "grad_norm": 45.75, |
| "learning_rate": 1.018e-05, |
| "loss": 13.8097, |
| "mean_token_accuracy": 0.776990968361497, |
| "num_tokens": 3602983.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.8306718476116657, |
| "epoch": 0.208, |
| "grad_norm": 50.25, |
| "learning_rate": 1.038e-05, |
| "loss": 13.4186, |
| "mean_token_accuracy": 0.7863588180392981, |
| "num_tokens": 3673600.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.8772016573697329, |
| "epoch": 0.212, |
| "grad_norm": 45.75, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 13.9353, |
| "mean_token_accuracy": 0.7783012766391039, |
| "num_tokens": 3743641.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.8009312467649579, |
| "epoch": 0.216, |
| "grad_norm": 55.25, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 12.9356, |
| "mean_token_accuracy": 0.7893661856651306, |
| "num_tokens": 3817397.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.8427618410438299, |
| "epoch": 0.22, |
| "grad_norm": 51.0, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 13.3831, |
| "mean_token_accuracy": 0.7832117900252342, |
| "num_tokens": 3886866.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.8758323887363076, |
| "epoch": 0.224, |
| "grad_norm": 41.5, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 13.8448, |
| "mean_token_accuracy": 0.7745687130838632, |
| "num_tokens": 3953732.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.8352954359725118, |
| "epoch": 0.228, |
| "grad_norm": 41.75, |
| "learning_rate": 1.138e-05, |
| "loss": 13.4861, |
| "mean_token_accuracy": 0.7828061260282994, |
| "num_tokens": 4025795.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.7947553901001811, |
| "epoch": 0.232, |
| "grad_norm": 43.25, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 12.8366, |
| "mean_token_accuracy": 0.7880385994911194, |
| "num_tokens": 4098989.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.8285046070814133, |
| "epoch": 0.236, |
| "grad_norm": 39.25, |
| "learning_rate": 1.178e-05, |
| "loss": 13.0789, |
| "mean_token_accuracy": 0.7872951343655586, |
| "num_tokens": 4170878.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.7863880429416895, |
| "epoch": 0.24, |
| "grad_norm": 45.0, |
| "learning_rate": 1.198e-05, |
| "loss": 12.7479, |
| "mean_token_accuracy": 0.7928395148366689, |
| "num_tokens": 4243655.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_biology_entropy": 1.2514844969511032, |
| "eval_biology_loss": 1.2855682373046875, |
| "eval_biology_mean_token_accuracy": 0.6817903207540512, |
| "eval_biology_num_tokens": 4243655.0, |
| "eval_biology_runtime": 40.7639, |
| "eval_biology_samples_per_second": 12.266, |
| "eval_biology_steps_per_second": 12.266, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_math_entropy": 0.843876126229763, |
| "eval_math_loss": 0.8332411646842957, |
| "eval_math_mean_token_accuracy": 0.783966186761856, |
| "eval_math_num_tokens": 4243655.0, |
| "eval_math_runtime": 48.1965, |
| "eval_math_samples_per_second": 10.374, |
| "eval_math_steps_per_second": 10.374, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.7838549023494125, |
| "epoch": 0.244, |
| "grad_norm": 41.75, |
| "learning_rate": 1.218e-05, |
| "loss": 12.4042, |
| "mean_token_accuracy": 0.7971927735954523, |
| "num_tokens": 4317321.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.7797296568751335, |
| "epoch": 0.248, |
| "grad_norm": 51.75, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 12.5895, |
| "mean_token_accuracy": 0.7916548330336809, |
| "num_tokens": 4387651.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8094759196043014, |
| "epoch": 0.252, |
| "grad_norm": 84.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 12.9446, |
| "mean_token_accuracy": 0.7892590902745724, |
| "num_tokens": 4461909.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.7797345662489533, |
| "epoch": 0.256, |
| "grad_norm": 62.0, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 12.462, |
| "mean_token_accuracy": 0.7935512393712998, |
| "num_tokens": 4531471.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.7911159221082926, |
| "epoch": 0.26, |
| "grad_norm": 40.0, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 12.8515, |
| "mean_token_accuracy": 0.7928617469966411, |
| "num_tokens": 4600757.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8169831104576588, |
| "epoch": 0.264, |
| "grad_norm": 49.5, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 13.117, |
| "mean_token_accuracy": 0.7893466159701348, |
| "num_tokens": 4669389.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.8287884982302784, |
| "epoch": 0.268, |
| "grad_norm": 55.25, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 13.3618, |
| "mean_token_accuracy": 0.7853043630719185, |
| "num_tokens": 4740492.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.7938818972557783, |
| "epoch": 0.272, |
| "grad_norm": 42.5, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 12.6092, |
| "mean_token_accuracy": 0.7941738195717335, |
| "num_tokens": 4810284.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.7964617364108563, |
| "epoch": 0.276, |
| "grad_norm": 37.25, |
| "learning_rate": 1.378e-05, |
| "loss": 12.6495, |
| "mean_token_accuracy": 0.7921258114278317, |
| "num_tokens": 4885701.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.7932526333257556, |
| "epoch": 0.28, |
| "grad_norm": 48.5, |
| "learning_rate": 1.398e-05, |
| "loss": 12.8488, |
| "mean_token_accuracy": 0.7890512771904469, |
| "num_tokens": 4955375.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.28, |
| "eval_biology_entropy": 1.2677810794115068, |
| "eval_biology_loss": 1.2936662435531616, |
| "eval_biology_mean_token_accuracy": 0.6814599596261978, |
| "eval_biology_num_tokens": 4955375.0, |
| "eval_biology_runtime": 40.9028, |
| "eval_biology_samples_per_second": 12.224, |
| "eval_biology_steps_per_second": 12.224, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.28, |
| "eval_math_entropy": 0.8274374225139618, |
| "eval_math_loss": 0.8181362748146057, |
| "eval_math_mean_token_accuracy": 0.7868153600692749, |
| "eval_math_num_tokens": 4955375.0, |
| "eval_math_runtime": 48.2074, |
| "eval_math_samples_per_second": 10.372, |
| "eval_math_steps_per_second": 10.372, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.7877025598660111, |
| "epoch": 0.284, |
| "grad_norm": 39.25, |
| "learning_rate": 1.418e-05, |
| "loss": 12.6293, |
| "mean_token_accuracy": 0.7944098394364119, |
| "num_tokens": 5025638.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.7564179511740804, |
| "epoch": 0.288, |
| "grad_norm": 38.75, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 12.0294, |
| "mean_token_accuracy": 0.8003660906106234, |
| "num_tokens": 5095104.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.7647898152470589, |
| "epoch": 0.292, |
| "grad_norm": 56.0, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 12.4713, |
| "mean_token_accuracy": 0.7947716873139143, |
| "num_tokens": 5167049.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.8119000235572458, |
| "epoch": 0.296, |
| "grad_norm": 36.75, |
| "learning_rate": 1.478e-05, |
| "loss": 13.0034, |
| "mean_token_accuracy": 0.7876484204083681, |
| "num_tokens": 5240988.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.7802535065449774, |
| "epoch": 0.3, |
| "grad_norm": 40.5, |
| "learning_rate": 1.498e-05, |
| "loss": 12.4371, |
| "mean_token_accuracy": 0.795463678240776, |
| "num_tokens": 5312258.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8087782407179475, |
| "epoch": 0.304, |
| "grad_norm": 44.75, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 12.9661, |
| "mean_token_accuracy": 0.7857579983770847, |
| "num_tokens": 5383732.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8243673056364059, |
| "epoch": 0.308, |
| "grad_norm": 39.5, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 13.1334, |
| "mean_token_accuracy": 0.7867084331810474, |
| "num_tokens": 5456650.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.7901528097689152, |
| "epoch": 0.312, |
| "grad_norm": 40.0, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 12.6608, |
| "mean_token_accuracy": 0.7905277730897069, |
| "num_tokens": 5530666.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.7515985878184438, |
| "epoch": 0.316, |
| "grad_norm": 44.75, |
| "learning_rate": 1.578e-05, |
| "loss": 11.9655, |
| "mean_token_accuracy": 0.8011779848486185, |
| "num_tokens": 5605269.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.7314818985760212, |
| "epoch": 0.32, |
| "grad_norm": 42.5, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 11.7244, |
| "mean_token_accuracy": 0.8059452049434185, |
| "num_tokens": 5676334.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.2663928886651994, |
| "eval_biology_loss": 1.2994052171707153, |
| "eval_biology_mean_token_accuracy": 0.6803199350833893, |
| "eval_biology_num_tokens": 5676334.0, |
| "eval_biology_runtime": 40.7632, |
| "eval_biology_samples_per_second": 12.266, |
| "eval_biology_steps_per_second": 12.266, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 0.7805578476190567, |
| "eval_math_loss": 0.8004927635192871, |
| "eval_math_mean_token_accuracy": 0.7903167794942856, |
| "eval_math_num_tokens": 5676334.0, |
| "eval_math_runtime": 48.1783, |
| "eval_math_samples_per_second": 10.378, |
| "eval_math_steps_per_second": 10.378, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.7661854576319456, |
| "epoch": 0.324, |
| "grad_norm": 40.5, |
| "learning_rate": 1.618e-05, |
| "loss": 12.5767, |
| "mean_token_accuracy": 0.7944763712584972, |
| "num_tokens": 5747779.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.803387301787734, |
| "epoch": 0.328, |
| "grad_norm": 42.0, |
| "learning_rate": 1.638e-05, |
| "loss": 12.825, |
| "mean_token_accuracy": 0.7907922245562077, |
| "num_tokens": 5816488.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8049221463501454, |
| "epoch": 0.332, |
| "grad_norm": 44.0, |
| "learning_rate": 1.658e-05, |
| "loss": 12.9669, |
| "mean_token_accuracy": 0.7854703675955534, |
| "num_tokens": 5887180.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8102014064788818, |
| "epoch": 0.336, |
| "grad_norm": 38.5, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 12.9814, |
| "mean_token_accuracy": 0.7886964596807957, |
| "num_tokens": 5958480.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.7344874950125814, |
| "epoch": 0.34, |
| "grad_norm": 50.25, |
| "learning_rate": 1.698e-05, |
| "loss": 11.7609, |
| "mean_token_accuracy": 0.8062219381332397, |
| "num_tokens": 6031020.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.802255941182375, |
| "epoch": 0.344, |
| "grad_norm": 39.0, |
| "learning_rate": 1.718e-05, |
| "loss": 12.8439, |
| "mean_token_accuracy": 0.7877463910728693, |
| "num_tokens": 6102081.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.7980229431763292, |
| "epoch": 0.348, |
| "grad_norm": 36.25, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 12.7774, |
| "mean_token_accuracy": 0.7900870595127344, |
| "num_tokens": 6172871.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.7401213280856609, |
| "epoch": 0.352, |
| "grad_norm": 40.75, |
| "learning_rate": 1.758e-05, |
| "loss": 11.9409, |
| "mean_token_accuracy": 0.801040057092905, |
| "num_tokens": 6242161.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.8095649138092995, |
| "epoch": 0.356, |
| "grad_norm": 46.5, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 12.9134, |
| "mean_token_accuracy": 0.7872339155524969, |
| "num_tokens": 6312187.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.7853369496762752, |
| "epoch": 0.36, |
| "grad_norm": 37.75, |
| "learning_rate": 1.798e-05, |
| "loss": 12.4062, |
| "mean_token_accuracy": 0.7960268270224333, |
| "num_tokens": 6377327.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_biology_entropy": 1.250610340833664, |
| "eval_biology_loss": 1.3047173023223877, |
| "eval_biology_mean_token_accuracy": 0.6794998321533203, |
| "eval_biology_num_tokens": 6377327.0, |
| "eval_biology_runtime": 40.5105, |
| "eval_biology_samples_per_second": 12.342, |
| "eval_biology_steps_per_second": 12.342, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_math_entropy": 0.773322255730629, |
| "eval_math_loss": 0.789590060710907, |
| "eval_math_mean_token_accuracy": 0.7914780566692352, |
| "eval_math_num_tokens": 6377327.0, |
| "eval_math_runtime": 48.04, |
| "eval_math_samples_per_second": 10.408, |
| "eval_math_steps_per_second": 10.408, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.763472201116383, |
| "epoch": 0.364, |
| "grad_norm": 41.5, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 12.3288, |
| "mean_token_accuracy": 0.7959023587405681, |
| "num_tokens": 6453254.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.7712945869192481, |
| "epoch": 0.368, |
| "grad_norm": 40.25, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 12.2854, |
| "mean_token_accuracy": 0.795868669450283, |
| "num_tokens": 6523679.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.7641221083700657, |
| "epoch": 0.372, |
| "grad_norm": 37.25, |
| "learning_rate": 1.858e-05, |
| "loss": 12.2867, |
| "mean_token_accuracy": 0.7949323236942292, |
| "num_tokens": 6594651.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.7518373979255557, |
| "epoch": 0.376, |
| "grad_norm": 42.0, |
| "learning_rate": 1.878e-05, |
| "loss": 12.1179, |
| "mean_token_accuracy": 0.798585458099842, |
| "num_tokens": 6669278.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.7543134400621057, |
| "epoch": 0.38, |
| "grad_norm": 33.75, |
| "learning_rate": 1.898e-05, |
| "loss": 12.023, |
| "mean_token_accuracy": 0.7991646058857441, |
| "num_tokens": 6740491.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.7203706834465265, |
| "epoch": 0.384, |
| "grad_norm": 35.5, |
| "learning_rate": 1.918e-05, |
| "loss": 11.6634, |
| "mean_token_accuracy": 0.8036353968083858, |
| "num_tokens": 6810978.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.7761434537358582, |
| "epoch": 0.388, |
| "grad_norm": 39.5, |
| "learning_rate": 1.938e-05, |
| "loss": 12.4387, |
| "mean_token_accuracy": 0.7944561909884215, |
| "num_tokens": 6882077.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.7785217678174376, |
| "epoch": 0.392, |
| "grad_norm": 33.75, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 12.607, |
| "mean_token_accuracy": 0.7944069147109986, |
| "num_tokens": 6953929.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.8195465696975589, |
| "epoch": 0.396, |
| "grad_norm": 65.5, |
| "learning_rate": 1.978e-05, |
| "loss": 13.2202, |
| "mean_token_accuracy": 0.7838506288826466, |
| "num_tokens": 7023835.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8217529645189643, |
| "epoch": 0.4, |
| "grad_norm": 39.75, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 13.1698, |
| "mean_token_accuracy": 0.7848756689578295, |
| "num_tokens": 7096903.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_biology_entropy": 1.266352553486824, |
| "eval_biology_loss": 1.312301516532898, |
| "eval_biology_mean_token_accuracy": 0.6775026069879532, |
| "eval_biology_num_tokens": 7096903.0, |
| "eval_biology_runtime": 40.7343, |
| "eval_biology_samples_per_second": 12.275, |
| "eval_biology_steps_per_second": 12.275, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_math_entropy": 0.7719556384682655, |
| "eval_math_loss": 0.7821868062019348, |
| "eval_math_mean_token_accuracy": 0.7932658619880676, |
| "eval_math_num_tokens": 7096903.0, |
| "eval_math_runtime": 47.8946, |
| "eval_math_samples_per_second": 10.44, |
| "eval_math_steps_per_second": 10.44, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.7681734091602266, |
| "epoch": 0.404, |
| "grad_norm": 50.25, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 12.3128, |
| "mean_token_accuracy": 0.7956426542252302, |
| "num_tokens": 7165970.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.7375578137114644, |
| "epoch": 0.408, |
| "grad_norm": 39.5, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 12.0172, |
| "mean_token_accuracy": 0.7997171815484763, |
| "num_tokens": 7239071.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.7803538812324404, |
| "epoch": 0.412, |
| "grad_norm": 35.75, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 12.3805, |
| "mean_token_accuracy": 0.7939646884799003, |
| "num_tokens": 7308430.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.7741465013474226, |
| "epoch": 0.416, |
| "grad_norm": 35.5, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 12.3671, |
| "mean_token_accuracy": 0.7934583045542241, |
| "num_tokens": 7377181.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.7638113724067808, |
| "epoch": 0.42, |
| "grad_norm": 90.5, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 12.3505, |
| "mean_token_accuracy": 0.7913536671549082, |
| "num_tokens": 7443045.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.763846804574132, |
| "epoch": 0.424, |
| "grad_norm": 30.5, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 12.2045, |
| "mean_token_accuracy": 0.7992881424725056, |
| "num_tokens": 7513035.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.7383330492302775, |
| "epoch": 0.428, |
| "grad_norm": 34.0, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 11.9274, |
| "mean_token_accuracy": 0.80244723521173, |
| "num_tokens": 7580676.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.7183827603235841, |
| "epoch": 0.432, |
| "grad_norm": 37.25, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 11.5217, |
| "mean_token_accuracy": 0.8065667234361171, |
| "num_tokens": 7650523.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.784936741553247, |
| "epoch": 0.436, |
| "grad_norm": 60.5, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 12.7959, |
| "mean_token_accuracy": 0.7918864544481039, |
| "num_tokens": 7725087.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.7581901378929615, |
| "epoch": 0.44, |
| "grad_norm": 35.75, |
| "learning_rate": 1.978e-05, |
| "loss": 12.0708, |
| "mean_token_accuracy": 0.7991419725120068, |
| "num_tokens": 7794006.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_biology_entropy": 1.3096306604146957, |
| "eval_biology_loss": 1.3199082612991333, |
| "eval_biology_mean_token_accuracy": 0.6769449328184127, |
| "eval_biology_num_tokens": 7794006.0, |
| "eval_biology_runtime": 41.8088, |
| "eval_biology_samples_per_second": 11.959, |
| "eval_biology_steps_per_second": 11.959, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_math_entropy": 0.7593404217660427, |
| "eval_math_loss": 0.7711546421051025, |
| "eval_math_mean_token_accuracy": 0.7947946254014969, |
| "eval_math_num_tokens": 7794006.0, |
| "eval_math_runtime": 48.5633, |
| "eval_math_samples_per_second": 10.296, |
| "eval_math_steps_per_second": 10.296, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.7571122448891401, |
| "epoch": 0.444, |
| "grad_norm": 35.0, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 12.1206, |
| "mean_token_accuracy": 0.7960348557680845, |
| "num_tokens": 7869588.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.7115281330421567, |
| "epoch": 0.448, |
| "grad_norm": 39.75, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 11.603, |
| "mean_token_accuracy": 0.8045209631323814, |
| "num_tokens": 7936788.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.7751916225999593, |
| "epoch": 0.452, |
| "grad_norm": 44.5, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 12.2553, |
| "mean_token_accuracy": 0.7963129185140133, |
| "num_tokens": 8007062.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.7358326924964785, |
| "epoch": 0.456, |
| "grad_norm": 35.75, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 11.8682, |
| "mean_token_accuracy": 0.8016852129250764, |
| "num_tokens": 8081035.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.748681915551424, |
| "epoch": 0.46, |
| "grad_norm": 38.75, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 12.0865, |
| "mean_token_accuracy": 0.8016125675290823, |
| "num_tokens": 8152005.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.7732457166537643, |
| "epoch": 0.464, |
| "grad_norm": 30.625, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 12.3834, |
| "mean_token_accuracy": 0.7960082966834306, |
| "num_tokens": 8223147.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.7652292996644974, |
| "epoch": 0.468, |
| "grad_norm": 47.0, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 12.5645, |
| "mean_token_accuracy": 0.7940644320100546, |
| "num_tokens": 8296261.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.7421515263617039, |
| "epoch": 0.472, |
| "grad_norm": 45.75, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 11.76, |
| "mean_token_accuracy": 0.8024631194770336, |
| "num_tokens": 8367380.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.74504639133811, |
| "epoch": 0.476, |
| "grad_norm": 40.25, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 11.8789, |
| "mean_token_accuracy": 0.798084694892168, |
| "num_tokens": 8438675.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.7464414104819298, |
| "epoch": 0.48, |
| "grad_norm": 34.75, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 12.21, |
| "mean_token_accuracy": 0.7978132009506226, |
| "num_tokens": 8506030.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.3110252878069877, |
| "eval_biology_loss": 1.324716567993164, |
| "eval_biology_mean_token_accuracy": 0.6763759871721268, |
| "eval_biology_num_tokens": 8506030.0, |
| "eval_biology_runtime": 67.8224, |
| "eval_biology_samples_per_second": 7.372, |
| "eval_biology_steps_per_second": 7.372, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.7597530280947685, |
| "eval_math_loss": 0.7603512406349182, |
| "eval_math_mean_token_accuracy": 0.7970666073560715, |
| "eval_math_num_tokens": 8506030.0, |
| "eval_math_runtime": 49.7375, |
| "eval_math_samples_per_second": 10.053, |
| "eval_math_steps_per_second": 10.053, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.7492084631696343, |
| "epoch": 0.484, |
| "grad_norm": 35.25, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 11.91, |
| "mean_token_accuracy": 0.8018037732690573, |
| "num_tokens": 8576915.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.7045143865048885, |
| "epoch": 0.488, |
| "grad_norm": 33.5, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 11.3318, |
| "mean_token_accuracy": 0.8072403989732265, |
| "num_tokens": 8646384.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.7107147205621004, |
| "epoch": 0.492, |
| "grad_norm": 34.5, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 11.4898, |
| "mean_token_accuracy": 0.8078423041850329, |
| "num_tokens": 8718547.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.7382461177185178, |
| "epoch": 0.496, |
| "grad_norm": 34.0, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 11.8211, |
| "mean_token_accuracy": 0.8013716965913773, |
| "num_tokens": 8788726.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.7180610377341509, |
| "epoch": 0.5, |
| "grad_norm": 32.0, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 11.6359, |
| "mean_token_accuracy": 0.8073719006031752, |
| "num_tokens": 8861341.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.7594908589497209, |
| "epoch": 0.504, |
| "grad_norm": 33.25, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 12.1494, |
| "mean_token_accuracy": 0.7983963489532471, |
| "num_tokens": 8934959.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.6865442413836718, |
| "epoch": 0.508, |
| "grad_norm": 47.0, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 11.0336, |
| "mean_token_accuracy": 0.8123888235539198, |
| "num_tokens": 9008794.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.736488389596343, |
| "epoch": 0.512, |
| "grad_norm": 36.0, |
| "learning_rate": 1.938e-05, |
| "loss": 11.8236, |
| "mean_token_accuracy": 0.80280033480376, |
| "num_tokens": 9078039.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.7221866660751403, |
| "epoch": 0.516, |
| "grad_norm": 38.5, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 11.5549, |
| "mean_token_accuracy": 0.8055313553661108, |
| "num_tokens": 9148271.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.7004529038444162, |
| "epoch": 0.52, |
| "grad_norm": 35.5, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 11.4288, |
| "mean_token_accuracy": 0.8069275438785553, |
| "num_tokens": 9220298.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.52, |
| "eval_biology_entropy": 1.2943904809951783, |
| "eval_biology_loss": 1.3324984312057495, |
| "eval_biology_mean_token_accuracy": 0.67458989828825, |
| "eval_biology_num_tokens": 9220298.0, |
| "eval_biology_runtime": 40.8416, |
| "eval_biology_samples_per_second": 12.242, |
| "eval_biology_steps_per_second": 12.242, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.52, |
| "eval_math_entropy": 0.7494015108048916, |
| "eval_math_loss": 0.7516281604766846, |
| "eval_math_mean_token_accuracy": 0.799598550081253, |
| "eval_math_num_tokens": 9220298.0, |
| "eval_math_runtime": 48.3706, |
| "eval_math_samples_per_second": 10.337, |
| "eval_math_steps_per_second": 10.337, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.7174371790140868, |
| "epoch": 0.524, |
| "grad_norm": 28.375, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 11.5, |
| "mean_token_accuracy": 0.8083903137594461, |
| "num_tokens": 9299112.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.7080350153148174, |
| "epoch": 0.528, |
| "grad_norm": 30.75, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 11.26, |
| "mean_token_accuracy": 0.8113009247928857, |
| "num_tokens": 9373860.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.7306012082844973, |
| "epoch": 0.532, |
| "grad_norm": 44.5, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 11.8475, |
| "mean_token_accuracy": 0.8006449308246374, |
| "num_tokens": 9446235.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.7229157039895654, |
| "epoch": 0.536, |
| "grad_norm": 37.25, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 11.6284, |
| "mean_token_accuracy": 0.8080755949020386, |
| "num_tokens": 9515962.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.7307992668822407, |
| "epoch": 0.54, |
| "grad_norm": 41.25, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 11.6765, |
| "mean_token_accuracy": 0.8042573977261782, |
| "num_tokens": 9587715.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.7034009215421975, |
| "epoch": 0.544, |
| "grad_norm": 35.75, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 11.7562, |
| "mean_token_accuracy": 0.8090484531596303, |
| "num_tokens": 9660940.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.7848182378336788, |
| "epoch": 0.548, |
| "grad_norm": 36.25, |
| "learning_rate": 1.918e-05, |
| "loss": 12.6939, |
| "mean_token_accuracy": 0.7890455313026905, |
| "num_tokens": 9727806.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.7604333622381091, |
| "epoch": 0.552, |
| "grad_norm": 39.75, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 12.152, |
| "mean_token_accuracy": 0.7959740109741688, |
| "num_tokens": 9793170.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.7220884849317372, |
| "epoch": 0.556, |
| "grad_norm": 37.0, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 11.6033, |
| "mean_token_accuracy": 0.8036061096936464, |
| "num_tokens": 9860484.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.704200503975153, |
| "epoch": 0.56, |
| "grad_norm": 29.375, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 11.1518, |
| "mean_token_accuracy": 0.8113750666379929, |
| "num_tokens": 9932302.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_biology_entropy": 1.2828473488092422, |
| "eval_biology_loss": 1.3331190347671509, |
| "eval_biology_mean_token_accuracy": 0.6747415412068367, |
| "eval_biology_num_tokens": 9932302.0, |
| "eval_biology_runtime": 45.9485, |
| "eval_biology_samples_per_second": 10.882, |
| "eval_biology_steps_per_second": 10.882, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_math_entropy": 0.7451034058630467, |
| "eval_math_loss": 0.7451240420341492, |
| "eval_math_mean_token_accuracy": 0.8007836003303528, |
| "eval_math_num_tokens": 9932302.0, |
| "eval_math_runtime": 56.801, |
| "eval_math_samples_per_second": 8.803, |
| "eval_math_steps_per_second": 8.803, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.7115965198725462, |
| "epoch": 0.564, |
| "grad_norm": 34.25, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 11.5991, |
| "mean_token_accuracy": 0.8047812730073929, |
| "num_tokens": 10001923.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.7081295600160956, |
| "epoch": 0.568, |
| "grad_norm": 32.75, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 11.2206, |
| "mean_token_accuracy": 0.8095470562577247, |
| "num_tokens": 10072362.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.7141612969338894, |
| "epoch": 0.572, |
| "grad_norm": 30.375, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 11.6816, |
| "mean_token_accuracy": 0.8019938107579947, |
| "num_tokens": 10140236.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.7410996645689011, |
| "epoch": 0.576, |
| "grad_norm": 32.5, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 11.7841, |
| "mean_token_accuracy": 0.8034846507012844, |
| "num_tokens": 10215462.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.7485123924911022, |
| "epoch": 0.58, |
| "grad_norm": 43.25, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 11.9787, |
| "mean_token_accuracy": 0.798300775885582, |
| "num_tokens": 10286386.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.7154224492609501, |
| "epoch": 0.584, |
| "grad_norm": 32.5, |
| "learning_rate": 1.898e-05, |
| "loss": 11.5298, |
| "mean_token_accuracy": 0.8057492181658745, |
| "num_tokens": 10354952.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.7272289376705885, |
| "epoch": 0.588, |
| "grad_norm": 40.25, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 11.652, |
| "mean_token_accuracy": 0.8025094602257014, |
| "num_tokens": 10425215.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.7166844455525279, |
| "epoch": 0.592, |
| "grad_norm": 30.625, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 11.4116, |
| "mean_token_accuracy": 0.8082372698932886, |
| "num_tokens": 10504396.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.7413169213570654, |
| "epoch": 0.596, |
| "grad_norm": 32.0, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 11.959, |
| "mean_token_accuracy": 0.7982309013605118, |
| "num_tokens": 10576150.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.7089840611442924, |
| "epoch": 0.6, |
| "grad_norm": 35.0, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 11.6029, |
| "mean_token_accuracy": 0.8050658291205763, |
| "num_tokens": 10644641.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_biology_entropy": 1.301250998735428, |
| "eval_biology_loss": 1.3366436958312988, |
| "eval_biology_mean_token_accuracy": 0.6739367877840996, |
| "eval_biology_num_tokens": 10644641.0, |
| "eval_biology_runtime": 40.559, |
| "eval_biology_samples_per_second": 12.328, |
| "eval_biology_steps_per_second": 12.328, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_math_entropy": 0.7238052343726158, |
| "eval_math_loss": 0.7381152510643005, |
| "eval_math_mean_token_accuracy": 0.8017449014186859, |
| "eval_math_num_tokens": 10644641.0, |
| "eval_math_runtime": 48.6353, |
| "eval_math_samples_per_second": 10.281, |
| "eval_math_steps_per_second": 10.281, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.7384224113076925, |
| "epoch": 0.604, |
| "grad_norm": 31.375, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 11.6601, |
| "mean_token_accuracy": 0.8054828774183989, |
| "num_tokens": 10718879.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.7157115396112204, |
| "epoch": 0.608, |
| "grad_norm": 38.0, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 11.7253, |
| "mean_token_accuracy": 0.805310570448637, |
| "num_tokens": 10793126.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.7220246642827988, |
| "epoch": 0.612, |
| "grad_norm": 31.875, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 11.4871, |
| "mean_token_accuracy": 0.8036613553762436, |
| "num_tokens": 10868501.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.7019952710717916, |
| "epoch": 0.616, |
| "grad_norm": 37.75, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 11.3501, |
| "mean_token_accuracy": 0.8096903506666422, |
| "num_tokens": 10941399.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.7124224284663796, |
| "epoch": 0.62, |
| "grad_norm": 28.0, |
| "learning_rate": 1.878e-05, |
| "loss": 11.2568, |
| "mean_token_accuracy": 0.8094121634960174, |
| "num_tokens": 11013227.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.6985575210303068, |
| "epoch": 0.624, |
| "grad_norm": 41.0, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 11.3067, |
| "mean_token_accuracy": 0.8099827248603105, |
| "num_tokens": 11081768.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.7057704029604792, |
| "epoch": 0.628, |
| "grad_norm": 35.5, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 11.2109, |
| "mean_token_accuracy": 0.8099541310220957, |
| "num_tokens": 11155362.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.6665219226852059, |
| "epoch": 0.632, |
| "grad_norm": 33.5, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 10.7041, |
| "mean_token_accuracy": 0.8133308228105307, |
| "num_tokens": 11226419.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.6918639421463013, |
| "epoch": 0.636, |
| "grad_norm": 28.0, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 11.1789, |
| "mean_token_accuracy": 0.810427363216877, |
| "num_tokens": 11298800.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.7074929376132786, |
| "epoch": 0.64, |
| "grad_norm": 32.0, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 11.4364, |
| "mean_token_accuracy": 0.8062872476875782, |
| "num_tokens": 11370320.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.2752835777401925, |
| "eval_biology_loss": 1.3371055126190186, |
| "eval_biology_mean_token_accuracy": 0.6743674901127815, |
| "eval_biology_num_tokens": 11370320.0, |
| "eval_biology_runtime": 41.1859, |
| "eval_biology_samples_per_second": 12.14, |
| "eval_biology_steps_per_second": 12.14, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.7216429207324981, |
| "eval_math_loss": 0.7321795225143433, |
| "eval_math_mean_token_accuracy": 0.8027968648672104, |
| "eval_math_num_tokens": 11370320.0, |
| "eval_math_runtime": 50.051, |
| "eval_math_samples_per_second": 9.99, |
| "eval_math_steps_per_second": 9.99, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.7110869940370321, |
| "epoch": 0.644, |
| "grad_norm": 33.5, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 11.4826, |
| "mean_token_accuracy": 0.8066194530576467, |
| "num_tokens": 11445577.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.6680682392790913, |
| "epoch": 0.648, |
| "grad_norm": 41.75, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 10.6001, |
| "mean_token_accuracy": 0.8180687021464109, |
| "num_tokens": 11515332.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.6972724428400398, |
| "epoch": 0.652, |
| "grad_norm": 28.375, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 11.2625, |
| "mean_token_accuracy": 0.8118615668267012, |
| "num_tokens": 11589344.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.7117041515186429, |
| "epoch": 0.656, |
| "grad_norm": 30.125, |
| "learning_rate": 1.858e-05, |
| "loss": 11.2332, |
| "mean_token_accuracy": 0.8057843204587698, |
| "num_tokens": 11657835.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.7337123662233352, |
| "epoch": 0.66, |
| "grad_norm": 35.0, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 11.9959, |
| "mean_token_accuracy": 0.7986002914607525, |
| "num_tokens": 11728761.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.6996571996249259, |
| "epoch": 0.664, |
| "grad_norm": 36.0, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 11.225, |
| "mean_token_accuracy": 0.807379187270999, |
| "num_tokens": 11799325.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.7201200375333429, |
| "epoch": 0.668, |
| "grad_norm": 26.625, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 11.4655, |
| "mean_token_accuracy": 0.8065472077578306, |
| "num_tokens": 11875761.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.6706034051254391, |
| "epoch": 0.672, |
| "grad_norm": 29.5, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 11.0248, |
| "mean_token_accuracy": 0.8108973000198603, |
| "num_tokens": 11949262.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.7067649567499756, |
| "epoch": 0.676, |
| "grad_norm": 39.0, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 11.2029, |
| "mean_token_accuracy": 0.8082552354782819, |
| "num_tokens": 12017021.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.7115105454809963, |
| "epoch": 0.68, |
| "grad_norm": 28.375, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 11.5299, |
| "mean_token_accuracy": 0.8070931307971477, |
| "num_tokens": 12090029.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.68, |
| "eval_biology_entropy": 1.3190836807489394, |
| "eval_biology_loss": 1.3399405479431152, |
| "eval_biology_mean_token_accuracy": 0.6731415511965751, |
| "eval_biology_num_tokens": 12090029.0, |
| "eval_biology_runtime": 41.5684, |
| "eval_biology_samples_per_second": 12.028, |
| "eval_biology_steps_per_second": 12.028, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.68, |
| "eval_math_entropy": 0.7365337611436844, |
| "eval_math_loss": 0.7248461842536926, |
| "eval_math_mean_token_accuracy": 0.8049273964166641, |
| "eval_math_num_tokens": 12090029.0, |
| "eval_math_runtime": 48.658, |
| "eval_math_samples_per_second": 10.276, |
| "eval_math_steps_per_second": 10.276, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.7276539742946625, |
| "epoch": 0.684, |
| "grad_norm": 45.75, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 11.6271, |
| "mean_token_accuracy": 0.8031581114977598, |
| "num_tokens": 12157148.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.692380516603589, |
| "epoch": 0.688, |
| "grad_norm": 32.75, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 11.0596, |
| "mean_token_accuracy": 0.8091447809711099, |
| "num_tokens": 12227640.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.7150112164206803, |
| "epoch": 0.692, |
| "grad_norm": 31.5, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 11.4079, |
| "mean_token_accuracy": 0.8083063390105962, |
| "num_tokens": 12300485.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.707249708659947, |
| "epoch": 0.696, |
| "grad_norm": 44.5, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 11.4663, |
| "mean_token_accuracy": 0.8062439143657685, |
| "num_tokens": 12371560.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.6466306034475565, |
| "epoch": 0.7, |
| "grad_norm": 31.5, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 10.4753, |
| "mean_token_accuracy": 0.8210283190011978, |
| "num_tokens": 12444485.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.7072455288842321, |
| "epoch": 0.704, |
| "grad_norm": 28.625, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 11.2585, |
| "mean_token_accuracy": 0.8080747354775667, |
| "num_tokens": 12516641.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.6972917837090791, |
| "epoch": 0.708, |
| "grad_norm": 31.875, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 11.1016, |
| "mean_token_accuracy": 0.8106624346226454, |
| "num_tokens": 12586969.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.733701079338789, |
| "epoch": 0.712, |
| "grad_norm": 33.0, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 11.9791, |
| "mean_token_accuracy": 0.8003675427287817, |
| "num_tokens": 12656623.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.7007612373679877, |
| "epoch": 0.716, |
| "grad_norm": 32.5, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 11.3258, |
| "mean_token_accuracy": 0.8079633958637714, |
| "num_tokens": 12724402.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.7202634891495109, |
| "epoch": 0.72, |
| "grad_norm": 45.25, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 11.4906, |
| "mean_token_accuracy": 0.8054792225360871, |
| "num_tokens": 12793343.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_biology_entropy": 1.2490641915798186, |
| "eval_biology_loss": 1.3441540002822876, |
| "eval_biology_mean_token_accuracy": 0.6730483353734017, |
| "eval_biology_num_tokens": 12793343.0, |
| "eval_biology_runtime": 41.3736, |
| "eval_biology_samples_per_second": 12.085, |
| "eval_biology_steps_per_second": 12.085, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_math_entropy": 0.6915989356935024, |
| "eval_math_loss": 0.7225330471992493, |
| "eval_math_mean_token_accuracy": 0.8046257610321045, |
| "eval_math_num_tokens": 12793343.0, |
| "eval_math_runtime": 48.3235, |
| "eval_math_samples_per_second": 10.347, |
| "eval_math_steps_per_second": 10.347, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.689653181284666, |
| "epoch": 0.724, |
| "grad_norm": 30.0, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 11.095, |
| "mean_token_accuracy": 0.8116659399121999, |
| "num_tokens": 12862261.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.6818234785459936, |
| "epoch": 0.728, |
| "grad_norm": 33.25, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 11.0911, |
| "mean_token_accuracy": 0.812104906514287, |
| "num_tokens": 12932479.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.7217532116919756, |
| "epoch": 0.732, |
| "grad_norm": 39.0, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 11.5997, |
| "mean_token_accuracy": 0.80367177054286, |
| "num_tokens": 13005640.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.678720659762621, |
| "epoch": 0.736, |
| "grad_norm": 30.25, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 10.8064, |
| "mean_token_accuracy": 0.817896718531847, |
| "num_tokens": 13077981.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.6725755118764937, |
| "epoch": 0.74, |
| "grad_norm": 32.25, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 10.8668, |
| "mean_token_accuracy": 0.8124937634915114, |
| "num_tokens": 13147662.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.6931444002315402, |
| "epoch": 0.744, |
| "grad_norm": 29.75, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 10.9939, |
| "mean_token_accuracy": 0.8126753181219101, |
| "num_tokens": 13217888.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.6799735228531063, |
| "epoch": 0.748, |
| "grad_norm": 30.625, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 11.1209, |
| "mean_token_accuracy": 0.8113108247518539, |
| "num_tokens": 13290215.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.6741954877972602, |
| "epoch": 0.752, |
| "grad_norm": 37.25, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 10.68, |
| "mean_token_accuracy": 0.8178396210074425, |
| "num_tokens": 13358957.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.6795736599713564, |
| "epoch": 0.756, |
| "grad_norm": 27.5, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 10.7669, |
| "mean_token_accuracy": 0.8167448993772268, |
| "num_tokens": 13435490.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.6350338911637664, |
| "epoch": 0.76, |
| "grad_norm": 33.75, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 10.293, |
| "mean_token_accuracy": 0.8226836711168289, |
| "num_tokens": 13510850.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.76, |
| "eval_biology_entropy": 1.2716083939671516, |
| "eval_biology_loss": 1.343294620513916, |
| "eval_biology_mean_token_accuracy": 0.6733359169363976, |
| "eval_biology_num_tokens": 13510850.0, |
| "eval_biology_runtime": 40.8464, |
| "eval_biology_samples_per_second": 12.241, |
| "eval_biology_steps_per_second": 12.241, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.76, |
| "eval_math_entropy": 0.7108331805765629, |
| "eval_math_loss": 0.7166345715522766, |
| "eval_math_mean_token_accuracy": 0.8057290604114532, |
| "eval_math_num_tokens": 13510850.0, |
| "eval_math_runtime": 47.8127, |
| "eval_math_samples_per_second": 10.457, |
| "eval_math_steps_per_second": 10.457, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.6852852246724069, |
| "epoch": 0.764, |
| "grad_norm": 26.0, |
| "learning_rate": 1.798e-05, |
| "loss": 10.9532, |
| "mean_token_accuracy": 0.8128071770071983, |
| "num_tokens": 13581848.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.6663387446664274, |
| "epoch": 0.768, |
| "grad_norm": 29.125, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 10.8527, |
| "mean_token_accuracy": 0.8174721624702215, |
| "num_tokens": 13653412.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7165529853664339, |
| "epoch": 0.772, |
| "grad_norm": 31.125, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 11.2507, |
| "mean_token_accuracy": 0.8098738107830286, |
| "num_tokens": 13721740.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.6845185202546418, |
| "epoch": 0.776, |
| "grad_norm": 27.25, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 11.1313, |
| "mean_token_accuracy": 0.810427762567997, |
| "num_tokens": 13794252.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.693409715499729, |
| "epoch": 0.78, |
| "grad_norm": 37.25, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 11.1187, |
| "mean_token_accuracy": 0.8099165614694357, |
| "num_tokens": 13865839.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.707060054782778, |
| "epoch": 0.784, |
| "grad_norm": 81.5, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 11.428, |
| "mean_token_accuracy": 0.8070406079292297, |
| "num_tokens": 13940856.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.7144587634131312, |
| "epoch": 0.788, |
| "grad_norm": 34.0, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 11.4499, |
| "mean_token_accuracy": 0.8067739550024271, |
| "num_tokens": 14014568.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.6812865046784282, |
| "epoch": 0.792, |
| "grad_norm": 29.75, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 10.9827, |
| "mean_token_accuracy": 0.8135220106691122, |
| "num_tokens": 14086117.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7018590277060867, |
| "epoch": 0.796, |
| "grad_norm": 28.875, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 11.4353, |
| "mean_token_accuracy": 0.8069973647594452, |
| "num_tokens": 14160050.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7179917026311159, |
| "epoch": 0.8, |
| "grad_norm": 33.0, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 11.3243, |
| "mean_token_accuracy": 0.8104559704661369, |
| "num_tokens": 14230754.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.2935789892077445, |
| "eval_biology_loss": 1.3476186990737915, |
| "eval_biology_mean_token_accuracy": 0.672012362420559, |
| "eval_biology_num_tokens": 14230754.0, |
| "eval_biology_runtime": 44.053, |
| "eval_biology_samples_per_second": 11.35, |
| "eval_biology_steps_per_second": 11.35, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.6968656893670558, |
| "eval_math_loss": 0.7122905850410461, |
| "eval_math_mean_token_accuracy": 0.8068100419044495, |
| "eval_math_num_tokens": 14230754.0, |
| "eval_math_runtime": 48.0745, |
| "eval_math_samples_per_second": 10.401, |
| "eval_math_steps_per_second": 10.401, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.6868740394711494, |
| "epoch": 0.804, |
| "grad_norm": 35.75, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 11.1864, |
| "mean_token_accuracy": 0.8112314622849226, |
| "num_tokens": 14301887.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.6712423311546445, |
| "epoch": 0.808, |
| "grad_norm": 28.5, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 10.6354, |
| "mean_token_accuracy": 0.8163831599056721, |
| "num_tokens": 14374548.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.7165202729403972, |
| "epoch": 0.812, |
| "grad_norm": 31.625, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 11.6733, |
| "mean_token_accuracy": 0.8052945740520954, |
| "num_tokens": 14451445.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.663957865908742, |
| "epoch": 0.816, |
| "grad_norm": 28.625, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 10.5899, |
| "mean_token_accuracy": 0.8158421032130718, |
| "num_tokens": 14519893.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.7052116150036454, |
| "epoch": 0.82, |
| "grad_norm": 36.5, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 11.6679, |
| "mean_token_accuracy": 0.8059296425431967, |
| "num_tokens": 14594443.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.675534434337169, |
| "epoch": 0.824, |
| "grad_norm": 39.5, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 10.7127, |
| "mean_token_accuracy": 0.8169379521161317, |
| "num_tokens": 14664170.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.6607738406397402, |
| "epoch": 0.828, |
| "grad_norm": 30.375, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 10.8303, |
| "mean_token_accuracy": 0.8149679381400347, |
| "num_tokens": 14735227.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.686363017372787, |
| "epoch": 0.832, |
| "grad_norm": 28.875, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 11.0101, |
| "mean_token_accuracy": 0.8137267474085093, |
| "num_tokens": 14805088.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.6817418249323964, |
| "epoch": 0.836, |
| "grad_norm": 31.125, |
| "learning_rate": 1.758e-05, |
| "loss": 11.1016, |
| "mean_token_accuracy": 0.811983397975564, |
| "num_tokens": 14875740.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.698756986297667, |
| "epoch": 0.84, |
| "grad_norm": 46.25, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 11.3359, |
| "mean_token_accuracy": 0.8067192178219557, |
| "num_tokens": 14944711.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.84, |
| "eval_biology_entropy": 1.268996791422367, |
| "eval_biology_loss": 1.3460168838500977, |
| "eval_biology_mean_token_accuracy": 0.673141638636589, |
| "eval_biology_num_tokens": 14944711.0, |
| "eval_biology_runtime": 41.4449, |
| "eval_biology_samples_per_second": 12.064, |
| "eval_biology_steps_per_second": 12.064, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.84, |
| "eval_math_entropy": 0.6983661242723465, |
| "eval_math_loss": 0.7085059881210327, |
| "eval_math_mean_token_accuracy": 0.8069341617822647, |
| "eval_math_num_tokens": 14944711.0, |
| "eval_math_runtime": 48.3919, |
| "eval_math_samples_per_second": 10.332, |
| "eval_math_steps_per_second": 10.332, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.6973433203995227, |
| "epoch": 0.844, |
| "grad_norm": 26.5, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 11.1804, |
| "mean_token_accuracy": 0.8072816133499146, |
| "num_tokens": 15014711.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.6738097196444869, |
| "epoch": 0.848, |
| "grad_norm": 31.125, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 10.8837, |
| "mean_token_accuracy": 0.8131649781018495, |
| "num_tokens": 15086306.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.7231233287602663, |
| "epoch": 0.852, |
| "grad_norm": 34.75, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 11.505, |
| "mean_token_accuracy": 0.8036668822169304, |
| "num_tokens": 15155311.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.7016849497333169, |
| "epoch": 0.856, |
| "grad_norm": 35.5, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 11.5304, |
| "mean_token_accuracy": 0.8056407071650028, |
| "num_tokens": 15228872.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.6963578762486577, |
| "epoch": 0.86, |
| "grad_norm": 34.0, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 11.1786, |
| "mean_token_accuracy": 0.8113155175000429, |
| "num_tokens": 15301140.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.6930005135014653, |
| "epoch": 0.864, |
| "grad_norm": 29.25, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 11.0449, |
| "mean_token_accuracy": 0.810348815843463, |
| "num_tokens": 15370985.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.6782323656603694, |
| "epoch": 0.868, |
| "grad_norm": 33.75, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 11.0558, |
| "mean_token_accuracy": 0.8125830505043268, |
| "num_tokens": 15440581.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.681379459425807, |
| "epoch": 0.872, |
| "grad_norm": 32.5, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 11.028, |
| "mean_token_accuracy": 0.8101276684552431, |
| "num_tokens": 15513360.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.7348810566589237, |
| "epoch": 0.876, |
| "grad_norm": 40.75, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 11.6674, |
| "mean_token_accuracy": 0.8031103067100048, |
| "num_tokens": 15585619.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.7099354591220617, |
| "epoch": 0.88, |
| "grad_norm": 36.25, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 11.585, |
| "mean_token_accuracy": 0.8052990060299635, |
| "num_tokens": 15653836.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_biology_entropy": 1.3010711609721184, |
| "eval_biology_loss": 1.3462131023406982, |
| "eval_biology_mean_token_accuracy": 0.6750594420433045, |
| "eval_biology_num_tokens": 15653836.0, |
| "eval_biology_runtime": 42.0081, |
| "eval_biology_samples_per_second": 11.902, |
| "eval_biology_steps_per_second": 11.902, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_math_entropy": 0.7088503445386887, |
| "eval_math_loss": 0.7035099864006042, |
| "eval_math_mean_token_accuracy": 0.8091315562725068, |
| "eval_math_num_tokens": 15653836.0, |
| "eval_math_runtime": 52.4596, |
| "eval_math_samples_per_second": 9.531, |
| "eval_math_steps_per_second": 9.531, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.6961297648027539, |
| "epoch": 0.884, |
| "grad_norm": 38.5, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 11.055, |
| "mean_token_accuracy": 0.811257503926754, |
| "num_tokens": 15721058.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.6642772014252841, |
| "epoch": 0.888, |
| "grad_norm": 28.5, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 10.7968, |
| "mean_token_accuracy": 0.8171007815748453, |
| "num_tokens": 15791102.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.7127458658069372, |
| "epoch": 0.892, |
| "grad_norm": 28.375, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 11.5136, |
| "mean_token_accuracy": 0.803728886321187, |
| "num_tokens": 15861638.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.6892058169469237, |
| "epoch": 0.896, |
| "grad_norm": 39.75, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 11.2148, |
| "mean_token_accuracy": 0.8115568786859513, |
| "num_tokens": 15932179.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.6717539247125387, |
| "epoch": 0.9, |
| "grad_norm": 32.75, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 10.7604, |
| "mean_token_accuracy": 0.8124691691249609, |
| "num_tokens": 16005581.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.6672627380117774, |
| "epoch": 0.904, |
| "grad_norm": 28.75, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 10.6646, |
| "mean_token_accuracy": 0.8165866218507289, |
| "num_tokens": 16074409.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.7063423937186599, |
| "epoch": 0.908, |
| "grad_norm": 31.5, |
| "learning_rate": 1.718e-05, |
| "loss": 11.431, |
| "mean_token_accuracy": 0.8058984253555537, |
| "num_tokens": 16147328.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.6730604227632284, |
| "epoch": 0.912, |
| "grad_norm": 32.75, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 10.9602, |
| "mean_token_accuracy": 0.8139959491789341, |
| "num_tokens": 16219842.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.6916197570040822, |
| "epoch": 0.916, |
| "grad_norm": 34.0, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 11.2951, |
| "mean_token_accuracy": 0.81037244386971, |
| "num_tokens": 16287996.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.6452676621265709, |
| "epoch": 0.92, |
| "grad_norm": 32.75, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 10.48, |
| "mean_token_accuracy": 0.8181278888136149, |
| "num_tokens": 16359249.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.92, |
| "eval_biology_entropy": 1.3079236669540406, |
| "eval_biology_loss": 1.3488103151321411, |
| "eval_biology_mean_token_accuracy": 0.6741236344575882, |
| "eval_biology_num_tokens": 16359249.0, |
| "eval_biology_runtime": 41.0049, |
| "eval_biology_samples_per_second": 12.194, |
| "eval_biology_steps_per_second": 12.194, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.92, |
| "eval_math_entropy": 0.7039950062930584, |
| "eval_math_loss": 0.7003913521766663, |
| "eval_math_mean_token_accuracy": 0.8092928317785263, |
| "eval_math_num_tokens": 16359249.0, |
| "eval_math_runtime": 48.3023, |
| "eval_math_samples_per_second": 10.351, |
| "eval_math_steps_per_second": 10.351, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.6840571843087673, |
| "epoch": 0.924, |
| "grad_norm": 30.375, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 11.0281, |
| "mean_token_accuracy": 0.8124604672193527, |
| "num_tokens": 16430183.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.6504545219242572, |
| "epoch": 0.928, |
| "grad_norm": 40.75, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 10.4019, |
| "mean_token_accuracy": 0.8224281292408705, |
| "num_tokens": 16499842.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.6886620251461864, |
| "epoch": 0.932, |
| "grad_norm": 35.5, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 11.2413, |
| "mean_token_accuracy": 0.8109501354396343, |
| "num_tokens": 16569436.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.6863971030339598, |
| "epoch": 0.936, |
| "grad_norm": 37.5, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 10.976, |
| "mean_token_accuracy": 0.812617215141654, |
| "num_tokens": 16640499.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.6811121774837374, |
| "epoch": 0.94, |
| "grad_norm": 29.0, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 10.9927, |
| "mean_token_accuracy": 0.812858996540308, |
| "num_tokens": 16708143.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.6965519505552947, |
| "epoch": 0.944, |
| "grad_norm": 29.625, |
| "learning_rate": 1.698e-05, |
| "loss": 11.2976, |
| "mean_token_accuracy": 0.8083064600825309, |
| "num_tokens": 16781882.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.6884551353752613, |
| "epoch": 0.948, |
| "grad_norm": 34.0, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 11.1246, |
| "mean_token_accuracy": 0.8140901360660792, |
| "num_tokens": 16855942.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.682703816704452, |
| "epoch": 0.952, |
| "grad_norm": 37.5, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 10.9073, |
| "mean_token_accuracy": 0.8152716889977455, |
| "num_tokens": 16922913.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.7374881992116571, |
| "epoch": 0.956, |
| "grad_norm": 42.25, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 11.9923, |
| "mean_token_accuracy": 0.7996022880077363, |
| "num_tokens": 16991940.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.679273992497474, |
| "epoch": 0.96, |
| "grad_norm": 26.75, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 10.8099, |
| "mean_token_accuracy": 0.8152121867984533, |
| "num_tokens": 17067407.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.2861202019453049, |
| "eval_biology_loss": 1.348455786705017, |
| "eval_biology_mean_token_accuracy": 0.6744914083480835, |
| "eval_biology_num_tokens": 17067407.0, |
| "eval_biology_runtime": 40.276, |
| "eval_biology_samples_per_second": 12.414, |
| "eval_biology_steps_per_second": 12.414, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.6839256008565426, |
| "eval_math_loss": 0.6964585185050964, |
| "eval_math_mean_token_accuracy": 0.8100001140832901, |
| "eval_math_num_tokens": 17067407.0, |
| "eval_math_runtime": 48.0956, |
| "eval_math_samples_per_second": 10.396, |
| "eval_math_steps_per_second": 10.396, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.6621893520466984, |
| "epoch": 0.964, |
| "grad_norm": 31.625, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 10.733, |
| "mean_token_accuracy": 0.8197611160576344, |
| "num_tokens": 17135614.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.6837352143600584, |
| "epoch": 0.968, |
| "grad_norm": 32.0, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 10.9487, |
| "mean_token_accuracy": 0.811904390156269, |
| "num_tokens": 17211153.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.6816597860306501, |
| "epoch": 0.972, |
| "grad_norm": 33.5, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 11.1352, |
| "mean_token_accuracy": 0.8097947511821986, |
| "num_tokens": 17280414.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.6748472723178566, |
| "epoch": 0.976, |
| "grad_norm": 41.5, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 10.8497, |
| "mean_token_accuracy": 0.8141761176288128, |
| "num_tokens": 17350994.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.6920185225084424, |
| "epoch": 0.98, |
| "grad_norm": 37.75, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 11.0203, |
| "mean_token_accuracy": 0.8110847111791373, |
| "num_tokens": 17422905.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.6656286349520087, |
| "epoch": 0.984, |
| "grad_norm": 32.0, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 10.8741, |
| "mean_token_accuracy": 0.8118232063949108, |
| "num_tokens": 17492960.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.7282194011844695, |
| "epoch": 0.988, |
| "grad_norm": 29.0, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 11.4374, |
| "mean_token_accuracy": 0.8076034974306822, |
| "num_tokens": 17564626.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.656241605617106, |
| "epoch": 0.992, |
| "grad_norm": 37.25, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 10.7226, |
| "mean_token_accuracy": 0.814904909953475, |
| "num_tokens": 17637514.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.7101638291031123, |
| "epoch": 0.996, |
| "grad_norm": 26.75, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 11.4156, |
| "mean_token_accuracy": 0.8066222406923771, |
| "num_tokens": 17708953.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.6990812286734581, |
| "epoch": 1.0, |
| "grad_norm": 28.5, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 11.3146, |
| "mean_token_accuracy": 0.8074303966015577, |
| "num_tokens": 17780432.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_biology_entropy": 1.278903669476509, |
| "eval_biology_loss": 1.3472909927368164, |
| "eval_biology_mean_token_accuracy": 0.6741317790150643, |
| "eval_biology_num_tokens": 17780432.0, |
| "eval_biology_runtime": 41.8672, |
| "eval_biology_samples_per_second": 11.943, |
| "eval_biology_steps_per_second": 11.943, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_math_entropy": 0.6809976233839988, |
| "eval_math_loss": 0.6932579278945923, |
| "eval_math_mean_token_accuracy": 0.8099925496578216, |
| "eval_math_num_tokens": 17780432.0, |
| "eval_math_runtime": 51.1798, |
| "eval_math_samples_per_second": 9.769, |
| "eval_math_steps_per_second": 9.769, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.6249620960094034, |
| "epoch": 1.004, |
| "grad_norm": 30.25, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 9.5977, |
| "mean_token_accuracy": 0.828125762566924, |
| "num_tokens": 17850416.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.5866982389241457, |
| "epoch": 1.008, |
| "grad_norm": 28.25, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 9.6873, |
| "mean_token_accuracy": 0.8302390404045582, |
| "num_tokens": 17926570.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.5911122627556324, |
| "epoch": 1.012, |
| "grad_norm": 29.25, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 9.5876, |
| "mean_token_accuracy": 0.8311092354357242, |
| "num_tokens": 18004785.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.6059481381438673, |
| "epoch": 1.016, |
| "grad_norm": 35.25, |
| "learning_rate": 1.658e-05, |
| "loss": 9.5013, |
| "mean_token_accuracy": 0.8301099374890327, |
| "num_tokens": 18070722.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.5719698538072407, |
| "epoch": 1.02, |
| "grad_norm": 37.75, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 9.265, |
| "mean_token_accuracy": 0.8315337430685759, |
| "num_tokens": 18135626.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.6137038953602314, |
| "epoch": 1.024, |
| "grad_norm": 30.625, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 9.8276, |
| "mean_token_accuracy": 0.8274661310017108, |
| "num_tokens": 18207652.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.6035456160083413, |
| "epoch": 1.028, |
| "grad_norm": 34.75, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 9.7604, |
| "mean_token_accuracy": 0.8305842254310847, |
| "num_tokens": 18276316.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.6081588315777481, |
| "epoch": 1.032, |
| "grad_norm": 35.0, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 9.842, |
| "mean_token_accuracy": 0.8259528614580631, |
| "num_tokens": 18349030.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.5707021232694387, |
| "epoch": 1.036, |
| "grad_norm": 37.5, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 9.1612, |
| "mean_token_accuracy": 0.8354629080742597, |
| "num_tokens": 18416740.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.5813462669961155, |
| "epoch": 1.04, |
| "grad_norm": 31.75, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 9.3108, |
| "mean_token_accuracy": 0.8341508328914642, |
| "num_tokens": 18484931.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_biology_entropy": 1.0625684696435929, |
| "eval_biology_loss": 1.382550835609436, |
| "eval_biology_mean_token_accuracy": 0.671364296078682, |
| "eval_biology_num_tokens": 18484931.0, |
| "eval_biology_runtime": 42.7958, |
| "eval_biology_samples_per_second": 11.683, |
| "eval_biology_steps_per_second": 11.683, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_math_entropy": 0.6263263171613216, |
| "eval_math_loss": 0.696963369846344, |
| "eval_math_mean_token_accuracy": 0.8110898947715759, |
| "eval_math_num_tokens": 18484931.0, |
| "eval_math_runtime": 50.6161, |
| "eval_math_samples_per_second": 9.878, |
| "eval_math_steps_per_second": 9.878, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.6172513328492641, |
| "epoch": 1.044, |
| "grad_norm": 38.0, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 9.8933, |
| "mean_token_accuracy": 0.8242699548602104, |
| "num_tokens": 18556670.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.6109676019288599, |
| "epoch": 1.048, |
| "grad_norm": 40.0, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 9.7651, |
| "mean_token_accuracy": 0.8267323866486549, |
| "num_tokens": 18630034.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.5600854479707778, |
| "epoch": 1.052, |
| "grad_norm": 31.125, |
| "learning_rate": 1.638e-05, |
| "loss": 8.9362, |
| "mean_token_accuracy": 0.8408110480755567, |
| "num_tokens": 18704870.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.5959546828642488, |
| "epoch": 1.056, |
| "grad_norm": 33.75, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 9.8347, |
| "mean_token_accuracy": 0.82447290122509, |
| "num_tokens": 18773457.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.6250034215860069, |
| "epoch": 1.06, |
| "grad_norm": 32.0, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 9.978, |
| "mean_token_accuracy": 0.8218581341207027, |
| "num_tokens": 18843237.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.5990947030484677, |
| "epoch": 1.064, |
| "grad_norm": 27.375, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 9.7507, |
| "mean_token_accuracy": 0.8283481735736131, |
| "num_tokens": 18918389.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.594149179942906, |
| "epoch": 1.068, |
| "grad_norm": 39.25, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 9.5644, |
| "mean_token_accuracy": 0.8269389532506466, |
| "num_tokens": 18988408.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.585951535589993, |
| "epoch": 1.072, |
| "grad_norm": 30.625, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 9.3202, |
| "mean_token_accuracy": 0.8317288246005774, |
| "num_tokens": 19055365.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.5888129183091223, |
| "epoch": 1.076, |
| "grad_norm": 42.25, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 9.6561, |
| "mean_token_accuracy": 0.8286302089691162, |
| "num_tokens": 19130316.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.6106328693218529, |
| "epoch": 1.08, |
| "grad_norm": 31.0, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 9.7639, |
| "mean_token_accuracy": 0.8254039615392685, |
| "num_tokens": 19202006.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.08, |
| "eval_biology_entropy": 1.0181471843123435, |
| "eval_biology_loss": 1.4018951654434204, |
| "eval_biology_mean_token_accuracy": 0.6708607443571091, |
| "eval_biology_num_tokens": 19202006.0, |
| "eval_biology_runtime": 41.0743, |
| "eval_biology_samples_per_second": 12.173, |
| "eval_biology_steps_per_second": 12.173, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.08, |
| "eval_math_entropy": 0.6143972256779671, |
| "eval_math_loss": 0.6992253661155701, |
| "eval_math_mean_token_accuracy": 0.8098398085832595, |
| "eval_math_num_tokens": 19202006.0, |
| "eval_math_runtime": 48.3437, |
| "eval_math_samples_per_second": 10.343, |
| "eval_math_steps_per_second": 10.343, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.5660794250667095, |
| "epoch": 1.084, |
| "grad_norm": 32.25, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 9.0991, |
| "mean_token_accuracy": 0.8397791922092438, |
| "num_tokens": 19273389.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.5959568014368415, |
| "epoch": 1.088, |
| "grad_norm": 33.0, |
| "learning_rate": 1.618e-05, |
| "loss": 9.6922, |
| "mean_token_accuracy": 0.8286801934242248, |
| "num_tokens": 19345730.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.6252739569172263, |
| "epoch": 1.092, |
| "grad_norm": 33.0, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 9.8698, |
| "mean_token_accuracy": 0.8247292656451464, |
| "num_tokens": 19418032.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.5888181790709496, |
| "epoch": 1.096, |
| "grad_norm": 33.75, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 9.5177, |
| "mean_token_accuracy": 0.8303285989910364, |
| "num_tokens": 19494084.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.5961418231017888, |
| "epoch": 1.1, |
| "grad_norm": 28.125, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 9.7765, |
| "mean_token_accuracy": 0.827918016910553, |
| "num_tokens": 19566904.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.6118787584826351, |
| "epoch": 1.104, |
| "grad_norm": 37.0, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 9.6539, |
| "mean_token_accuracy": 0.8265355542302132, |
| "num_tokens": 19637390.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.5835532064549624, |
| "epoch": 1.108, |
| "grad_norm": 41.0, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 9.4517, |
| "mean_token_accuracy": 0.8296675242483615, |
| "num_tokens": 19704730.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.548611044511199, |
| "epoch": 1.112, |
| "grad_norm": 29.25, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 8.8514, |
| "mean_token_accuracy": 0.8405190438032151, |
| "num_tokens": 19777388.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.5902225982397795, |
| "epoch": 1.116, |
| "grad_norm": 33.25, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 9.4995, |
| "mean_token_accuracy": 0.8318900916725397, |
| "num_tokens": 19847340.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.5956728520803154, |
| "epoch": 1.12, |
| "grad_norm": 33.25, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 9.626, |
| "mean_token_accuracy": 0.8307756468653679, |
| "num_tokens": 19923914.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.015951538324356, |
| "eval_biology_loss": 1.4020975828170776, |
| "eval_biology_mean_token_accuracy": 0.6706568294167519, |
| "eval_biology_num_tokens": 19923914.0, |
| "eval_biology_runtime": 41.0194, |
| "eval_biology_samples_per_second": 12.189, |
| "eval_biology_steps_per_second": 12.189, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.627720057964325, |
| "eval_math_loss": 0.6943842172622681, |
| "eval_math_mean_token_accuracy": 0.8108133491277695, |
| "eval_math_num_tokens": 19923914.0, |
| "eval_math_runtime": 48.3784, |
| "eval_math_samples_per_second": 10.335, |
| "eval_math_steps_per_second": 10.335, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.559093719534576, |
| "epoch": 1.124, |
| "grad_norm": 27.625, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 9.0395, |
| "mean_token_accuracy": 0.8391204752027989, |
| "num_tokens": 19994212.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.5992768860422075, |
| "epoch": 1.1280000000000001, |
| "grad_norm": 40.5, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 9.5995, |
| "mean_token_accuracy": 0.8290460936725139, |
| "num_tokens": 20061623.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.5645275509916245, |
| "epoch": 1.1320000000000001, |
| "grad_norm": 32.5, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 9.0703, |
| "mean_token_accuracy": 0.8376249179244042, |
| "num_tokens": 20131887.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.6051725963130593, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 38.5, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 9.9314, |
| "mean_token_accuracy": 0.8219255816191435, |
| "num_tokens": 20201892.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.5845890076830983, |
| "epoch": 1.1400000000000001, |
| "grad_norm": 41.25, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 9.277, |
| "mean_token_accuracy": 0.83502765186131, |
| "num_tokens": 20272508.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.6243995492346585, |
| "epoch": 1.144, |
| "grad_norm": 31.375, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 10.0605, |
| "mean_token_accuracy": 0.8231050688773394, |
| "num_tokens": 20343948.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.6145558839663863, |
| "epoch": 1.148, |
| "grad_norm": 30.125, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 10.1073, |
| "mean_token_accuracy": 0.8245069060474635, |
| "num_tokens": 20418622.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.5770234601572156, |
| "epoch": 1.152, |
| "grad_norm": 34.75, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 9.2963, |
| "mean_token_accuracy": 0.8351634360849858, |
| "num_tokens": 20490282.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.5991640514694154, |
| "epoch": 1.156, |
| "grad_norm": 38.25, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 9.6991, |
| "mean_token_accuracy": 0.8288435827940702, |
| "num_tokens": 20565841.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.5759193726815284, |
| "epoch": 1.16, |
| "grad_norm": 34.5, |
| "learning_rate": 1.578e-05, |
| "loss": 9.1954, |
| "mean_token_accuracy": 0.8349741615355015, |
| "num_tokens": 20641737.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.16, |
| "eval_biology_entropy": 0.9992075580358505, |
| "eval_biology_loss": 1.4079357385635376, |
| "eval_biology_mean_token_accuracy": 0.6715812242031097, |
| "eval_biology_num_tokens": 20641737.0, |
| "eval_biology_runtime": 40.562, |
| "eval_biology_samples_per_second": 12.327, |
| "eval_biology_steps_per_second": 12.327, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.16, |
| "eval_math_entropy": 0.5956115295290947, |
| "eval_math_loss": 0.6947958469390869, |
| "eval_math_mean_token_accuracy": 0.811661752462387, |
| "eval_math_num_tokens": 20641737.0, |
| "eval_math_runtime": 48.7191, |
| "eval_math_samples_per_second": 10.263, |
| "eval_math_steps_per_second": 10.263, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.6053885708563029, |
| "epoch": 1.164, |
| "grad_norm": 30.875, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 9.5988, |
| "mean_token_accuracy": 0.8267572112381458, |
| "num_tokens": 20714608.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.5695252175442874, |
| "epoch": 1.168, |
| "grad_norm": 32.5, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 9.2606, |
| "mean_token_accuracy": 0.8332584574818611, |
| "num_tokens": 20785786.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.5976280993781984, |
| "epoch": 1.172, |
| "grad_norm": 33.5, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 9.7345, |
| "mean_token_accuracy": 0.8283390998840332, |
| "num_tokens": 20854639.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.5922115191817283, |
| "epoch": 1.176, |
| "grad_norm": 29.625, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 9.5617, |
| "mean_token_accuracy": 0.8294767212122679, |
| "num_tokens": 20928880.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.6270347313024104, |
| "epoch": 1.18, |
| "grad_norm": 29.125, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 10.0052, |
| "mean_token_accuracy": 0.8231472406536341, |
| "num_tokens": 21006747.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.5596231662668287, |
| "epoch": 1.184, |
| "grad_norm": 29.75, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 9.1299, |
| "mean_token_accuracy": 0.8357201408594847, |
| "num_tokens": 21074205.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.5571073008701205, |
| "epoch": 1.188, |
| "grad_norm": 34.75, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 9.0309, |
| "mean_token_accuracy": 0.8378865152597428, |
| "num_tokens": 21149284.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.5892432911321521, |
| "epoch": 1.192, |
| "grad_norm": 34.25, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 9.3704, |
| "mean_token_accuracy": 0.8327233593910932, |
| "num_tokens": 21222288.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.5967930853366852, |
| "epoch": 1.196, |
| "grad_norm": 32.5, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 9.7825, |
| "mean_token_accuracy": 0.8243178736418486, |
| "num_tokens": 21295514.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.5806791403330862, |
| "epoch": 1.2, |
| "grad_norm": 33.25, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 9.1916, |
| "mean_token_accuracy": 0.8367064792662859, |
| "num_tokens": 21369159.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_biology_entropy": 0.9886249232292176, |
| "eval_biology_loss": 1.4157956838607788, |
| "eval_biology_mean_token_accuracy": 0.6691707735061645, |
| "eval_biology_num_tokens": 21369159.0, |
| "eval_biology_runtime": 41.807, |
| "eval_biology_samples_per_second": 11.96, |
| "eval_biology_steps_per_second": 11.96, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_math_entropy": 0.6074744529724121, |
| "eval_math_loss": 0.6924204230308533, |
| "eval_math_mean_token_accuracy": 0.8120252940654754, |
| "eval_math_num_tokens": 21369159.0, |
| "eval_math_runtime": 48.5264, |
| "eval_math_samples_per_second": 10.304, |
| "eval_math_steps_per_second": 10.304, |
| "step": 3000 |
| }, |
| { |
| "entropy": 0.6160458439961076, |
| "epoch": 1.204, |
| "grad_norm": 38.5, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 9.9057, |
| "mean_token_accuracy": 0.8249781485646963, |
| "num_tokens": 21439140.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 0.572793152462691, |
| "epoch": 1.208, |
| "grad_norm": 46.5, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 9.3677, |
| "mean_token_accuracy": 0.8327320765703916, |
| "num_tokens": 21511015.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 0.5832694384269417, |
| "epoch": 1.212, |
| "grad_norm": 34.25, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 9.3447, |
| "mean_token_accuracy": 0.8320426400750875, |
| "num_tokens": 21581182.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 0.5971815595403314, |
| "epoch": 1.216, |
| "grad_norm": 35.0, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 9.5621, |
| "mean_token_accuracy": 0.8280133757740259, |
| "num_tokens": 21649178.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 0.5768723101355135, |
| "epoch": 1.22, |
| "grad_norm": 37.0, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 9.4209, |
| "mean_token_accuracy": 0.8305264201015234, |
| "num_tokens": 21721516.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 0.5728980491869151, |
| "epoch": 1.224, |
| "grad_norm": 30.125, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 9.0729, |
| "mean_token_accuracy": 0.8389325402677059, |
| "num_tokens": 21790802.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 0.5569340754300356, |
| "epoch": 1.228, |
| "grad_norm": 33.0, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 9.0697, |
| "mean_token_accuracy": 0.8380256276577711, |
| "num_tokens": 21864401.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 0.5946331669576466, |
| "epoch": 1.232, |
| "grad_norm": 40.5, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 9.5008, |
| "mean_token_accuracy": 0.8295462980866433, |
| "num_tokens": 21930239.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 0.5537866359576583, |
| "epoch": 1.236, |
| "grad_norm": 35.25, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 8.9583, |
| "mean_token_accuracy": 0.8384150017052889, |
| "num_tokens": 22002260.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 0.5869318909943104, |
| "epoch": 1.24, |
| "grad_norm": 34.5, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 9.5525, |
| "mean_token_accuracy": 0.8298884276300669, |
| "num_tokens": 22075270.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_biology_entropy": 0.9955327052474022, |
| "eval_biology_loss": 1.4185529947280884, |
| "eval_biology_mean_token_accuracy": 0.6690302005410195, |
| "eval_biology_num_tokens": 22075270.0, |
| "eval_biology_runtime": 40.5339, |
| "eval_biology_samples_per_second": 12.335, |
| "eval_biology_steps_per_second": 12.335, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_math_entropy": 0.6026208091974259, |
| "eval_math_loss": 0.6936736106872559, |
| "eval_math_mean_token_accuracy": 0.8121141378879547, |
| "eval_math_num_tokens": 22075270.0, |
| "eval_math_runtime": 47.9825, |
| "eval_math_samples_per_second": 10.42, |
| "eval_math_steps_per_second": 10.42, |
| "step": 3100 |
| }, |
| { |
| "entropy": 0.5803977906703949, |
| "epoch": 1.244, |
| "grad_norm": 29.125, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 9.394, |
| "mean_token_accuracy": 0.8314684689044952, |
| "num_tokens": 22144779.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 0.5916906374506652, |
| "epoch": 1.248, |
| "grad_norm": 29.25, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 9.6453, |
| "mean_token_accuracy": 0.8284844689071178, |
| "num_tokens": 22216387.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 0.6286893571726978, |
| "epoch": 1.252, |
| "grad_norm": 40.75, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 9.9522, |
| "mean_token_accuracy": 0.8251317605376244, |
| "num_tokens": 22288125.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 0.5737162339501083, |
| "epoch": 1.256, |
| "grad_norm": 34.5, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 9.3693, |
| "mean_token_accuracy": 0.8332099460065365, |
| "num_tokens": 22359813.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 0.5986535185016691, |
| "epoch": 1.26, |
| "grad_norm": 45.25, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 9.5895, |
| "mean_token_accuracy": 0.8297188606113195, |
| "num_tokens": 22428601.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 0.615026047732681, |
| "epoch": 1.264, |
| "grad_norm": 37.5, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 9.6387, |
| "mean_token_accuracy": 0.827248927205801, |
| "num_tokens": 22501002.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 0.5916862264275551, |
| "epoch": 1.268, |
| "grad_norm": 33.25, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 9.8238, |
| "mean_token_accuracy": 0.8254994552582502, |
| "num_tokens": 22568447.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 0.5594694192521275, |
| "epoch": 1.272, |
| "grad_norm": 38.75, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 8.8557, |
| "mean_token_accuracy": 0.8411485534161329, |
| "num_tokens": 22634652.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 0.5982500876300036, |
| "epoch": 1.276, |
| "grad_norm": 34.25, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 9.8455, |
| "mean_token_accuracy": 0.8254418596625328, |
| "num_tokens": 22707514.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 0.585130360070616, |
| "epoch": 1.28, |
| "grad_norm": 38.25, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 9.3764, |
| "mean_token_accuracy": 0.833708243817091, |
| "num_tokens": 22779682.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 0.9487634599804878, |
| "eval_biology_loss": 1.432287573814392, |
| "eval_biology_mean_token_accuracy": 0.6715161613225937, |
| "eval_biology_num_tokens": 22779682.0, |
| "eval_biology_runtime": 41.9511, |
| "eval_biology_samples_per_second": 11.919, |
| "eval_biology_steps_per_second": 11.919, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.6086316105127335, |
| "eval_math_loss": 0.6877865195274353, |
| "eval_math_mean_token_accuracy": 0.8134469233751297, |
| "eval_math_num_tokens": 22779682.0, |
| "eval_math_runtime": 48.7756, |
| "eval_math_samples_per_second": 10.251, |
| "eval_math_steps_per_second": 10.251, |
| "step": 3200 |
| }, |
| { |
| "entropy": 0.584856943693012, |
| "epoch": 1.284, |
| "grad_norm": 52.75, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 9.2731, |
| "mean_token_accuracy": 0.8305226769298315, |
| "num_tokens": 22849488.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 0.5926172870211304, |
| "epoch": 1.288, |
| "grad_norm": 31.75, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 9.6435, |
| "mean_token_accuracy": 0.8274250105023384, |
| "num_tokens": 22919374.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 0.5529075180180371, |
| "epoch": 1.292, |
| "grad_norm": 41.75, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 9.0033, |
| "mean_token_accuracy": 0.8389334350824356, |
| "num_tokens": 22987319.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 0.5644551729783416, |
| "epoch": 1.296, |
| "grad_norm": 35.25, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 8.9124, |
| "mean_token_accuracy": 0.8382747415453196, |
| "num_tokens": 23057744.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 0.5793901988305151, |
| "epoch": 1.3, |
| "grad_norm": 31.125, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 9.3239, |
| "mean_token_accuracy": 0.8315478291362524, |
| "num_tokens": 23133675.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 0.5897874507121742, |
| "epoch": 1.304, |
| "grad_norm": 37.25, |
| "learning_rate": 1.498e-05, |
| "loss": 9.4533, |
| "mean_token_accuracy": 0.8307483058422804, |
| "num_tokens": 23201590.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 0.6082971251569689, |
| "epoch": 1.308, |
| "grad_norm": 34.75, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 9.8766, |
| "mean_token_accuracy": 0.8264070238918066, |
| "num_tokens": 23275697.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 0.577689622156322, |
| "epoch": 1.312, |
| "grad_norm": 42.5, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 9.3941, |
| "mean_token_accuracy": 0.8330640692263842, |
| "num_tokens": 23344644.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 0.6009934969246388, |
| "epoch": 1.316, |
| "grad_norm": 35.0, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 9.4744, |
| "mean_token_accuracy": 0.8295592527836562, |
| "num_tokens": 23414468.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 0.5623171862214804, |
| "epoch": 1.32, |
| "grad_norm": 33.5, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 9.2539, |
| "mean_token_accuracy": 0.8360460609197616, |
| "num_tokens": 23485056.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_biology_entropy": 0.9134255470633507, |
| "eval_biology_loss": 1.4568630456924438, |
| "eval_biology_mean_token_accuracy": 0.6691547855138779, |
| "eval_biology_num_tokens": 23485056.0, |
| "eval_biology_runtime": 41.6663, |
| "eval_biology_samples_per_second": 12.0, |
| "eval_biology_steps_per_second": 12.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_math_entropy": 0.5871166962385178, |
| "eval_math_loss": 0.6909680366516113, |
| "eval_math_mean_token_accuracy": 0.8131083092689514, |
| "eval_math_num_tokens": 23485056.0, |
| "eval_math_runtime": 48.702, |
| "eval_math_samples_per_second": 10.267, |
| "eval_math_steps_per_second": 10.267, |
| "step": 3300 |
| }, |
| { |
| "entropy": 0.5966707519255579, |
| "epoch": 1.324, |
| "grad_norm": 37.75, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 9.5778, |
| "mean_token_accuracy": 0.8291139416396618, |
| "num_tokens": 23554135.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 0.6284289790317417, |
| "epoch": 1.328, |
| "grad_norm": 33.25, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 9.9966, |
| "mean_token_accuracy": 0.8251899681985378, |
| "num_tokens": 23622405.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 0.5863202163018286, |
| "epoch": 1.332, |
| "grad_norm": 27.875, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 9.5443, |
| "mean_token_accuracy": 0.8295067902654409, |
| "num_tokens": 23690429.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 0.598145164269954, |
| "epoch": 1.336, |
| "grad_norm": 36.0, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 9.5305, |
| "mean_token_accuracy": 0.8304523423314094, |
| "num_tokens": 23760641.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 0.5735043300315738, |
| "epoch": 1.34, |
| "grad_norm": 32.75, |
| "learning_rate": 1.478e-05, |
| "loss": 9.397, |
| "mean_token_accuracy": 0.8326968785375357, |
| "num_tokens": 23829902.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 0.5639217297546566, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 35.5, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 9.0614, |
| "mean_token_accuracy": 0.837357560545206, |
| "num_tokens": 23899771.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 0.5979444002732635, |
| "epoch": 1.3479999999999999, |
| "grad_norm": 32.25, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 9.6483, |
| "mean_token_accuracy": 0.830116517841816, |
| "num_tokens": 23970398.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 0.5784852374345064, |
| "epoch": 1.3519999999999999, |
| "grad_norm": 32.75, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 9.268, |
| "mean_token_accuracy": 0.8342861253768206, |
| "num_tokens": 24041995.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 0.5973453964106739, |
| "epoch": 1.3559999999999999, |
| "grad_norm": 43.5, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 9.6308, |
| "mean_token_accuracy": 0.8302569206804037, |
| "num_tokens": 24112838.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 0.5741045112721622, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 40.0, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 9.1876, |
| "mean_token_accuracy": 0.8341640867292881, |
| "num_tokens": 24187023.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_biology_entropy": 0.9914887221455574, |
| "eval_biology_loss": 1.4215466976165771, |
| "eval_biology_mean_token_accuracy": 0.6697278326749802, |
| "eval_biology_num_tokens": 24187023.0, |
| "eval_biology_runtime": 40.0744, |
| "eval_biology_samples_per_second": 12.477, |
| "eval_biology_steps_per_second": 12.477, |
| "step": 3400 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_math_entropy": 0.5978010043203831, |
| "eval_math_loss": 0.6856448650360107, |
| "eval_math_mean_token_accuracy": 0.8142528500556946, |
| "eval_math_num_tokens": 24187023.0, |
| "eval_math_runtime": 47.8711, |
| "eval_math_samples_per_second": 10.445, |
| "eval_math_steps_per_second": 10.445, |
| "step": 3400 |
| }, |
| { |
| "entropy": 0.605070736259222, |
| "epoch": 1.3639999999999999, |
| "grad_norm": 41.5, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 9.8716, |
| "mean_token_accuracy": 0.8238684110343456, |
| "num_tokens": 24255699.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 0.5953077180311084, |
| "epoch": 1.3679999999999999, |
| "grad_norm": 34.5, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 9.4636, |
| "mean_token_accuracy": 0.8314888432621956, |
| "num_tokens": 24325844.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 0.6159492513164878, |
| "epoch": 1.3719999999999999, |
| "grad_norm": 27.375, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 10.0754, |
| "mean_token_accuracy": 0.819038163498044, |
| "num_tokens": 24396138.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 0.6475082284770906, |
| "epoch": 1.376, |
| "grad_norm": 26.25, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 10.3142, |
| "mean_token_accuracy": 0.8197898093611002, |
| "num_tokens": 24466132.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 0.5750578778795898, |
| "epoch": 1.38, |
| "grad_norm": 31.0, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 9.4634, |
| "mean_token_accuracy": 0.8325258079916239, |
| "num_tokens": 24538051.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 0.5948665856383741, |
| "epoch": 1.384, |
| "grad_norm": 38.0, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 9.4067, |
| "mean_token_accuracy": 0.8300522934645415, |
| "num_tokens": 24606147.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 0.5827950645238161, |
| "epoch": 1.388, |
| "grad_norm": 33.25, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 9.4662, |
| "mean_token_accuracy": 0.8301668491214513, |
| "num_tokens": 24677337.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 0.5718853862024844, |
| "epoch": 1.392, |
| "grad_norm": 32.75, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 9.2656, |
| "mean_token_accuracy": 0.8346508856862783, |
| "num_tokens": 24748043.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 0.5869543513283133, |
| "epoch": 1.396, |
| "grad_norm": 30.875, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 9.4427, |
| "mean_token_accuracy": 0.8317004144191742, |
| "num_tokens": 24820876.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 0.562530672736466, |
| "epoch": 1.4, |
| "grad_norm": 42.75, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 9.0566, |
| "mean_token_accuracy": 0.8376737374812364, |
| "num_tokens": 24894293.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_biology_entropy": 0.9792917854189873, |
| "eval_biology_loss": 1.430548071861267, |
| "eval_biology_mean_token_accuracy": 0.6689909261465072, |
| "eval_biology_num_tokens": 24894293.0, |
| "eval_biology_runtime": 590.2328, |
| "eval_biology_samples_per_second": 0.847, |
| "eval_biology_steps_per_second": 0.847, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_math_entropy": 0.6067364411056042, |
| "eval_math_loss": 0.6825031042098999, |
| "eval_math_mean_token_accuracy": 0.8138161517381668, |
| "eval_math_num_tokens": 24894293.0, |
| "eval_math_runtime": 47.8922, |
| "eval_math_samples_per_second": 10.44, |
| "eval_math_steps_per_second": 10.44, |
| "step": 3500 |
| }, |
| { |
| "entropy": 0.5819358373992145, |
| "epoch": 1.404, |
| "grad_norm": 29.75, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 9.358, |
| "mean_token_accuracy": 0.8343918733298779, |
| "num_tokens": 24963721.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 0.5887822126969695, |
| "epoch": 1.408, |
| "grad_norm": 39.0, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 9.4385, |
| "mean_token_accuracy": 0.8313043463975192, |
| "num_tokens": 25036674.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 0.600373360607773, |
| "epoch": 1.412, |
| "grad_norm": 27.875, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 9.6251, |
| "mean_token_accuracy": 0.8280442032963038, |
| "num_tokens": 25107532.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 0.5769751055166126, |
| "epoch": 1.416, |
| "grad_norm": 39.75, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 9.3623, |
| "mean_token_accuracy": 0.8321974739432335, |
| "num_tokens": 25178935.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 0.5816602191887796, |
| "epoch": 1.42, |
| "grad_norm": 44.25, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 9.3983, |
| "mean_token_accuracy": 0.8320797413587571, |
| "num_tokens": 25251580.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 0.5431325471028685, |
| "epoch": 1.424, |
| "grad_norm": 35.75, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 8.7863, |
| "mean_token_accuracy": 0.8444295462220908, |
| "num_tokens": 25324579.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 0.5805144936777651, |
| "epoch": 1.428, |
| "grad_norm": 34.5, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 9.1213, |
| "mean_token_accuracy": 0.8340296439826489, |
| "num_tokens": 25393906.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 0.5574665242806077, |
| "epoch": 1.432, |
| "grad_norm": 31.0, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 9.1778, |
| "mean_token_accuracy": 0.8357073605060578, |
| "num_tokens": 25466485.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 0.5930931678041815, |
| "epoch": 1.436, |
| "grad_norm": 30.375, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 9.4113, |
| "mean_token_accuracy": 0.8343671444803477, |
| "num_tokens": 25533565.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 0.6089992932043969, |
| "epoch": 1.44, |
| "grad_norm": 32.0, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 9.7831, |
| "mean_token_accuracy": 0.8265475559979677, |
| "num_tokens": 25606824.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 0.9797519612908363, |
| "eval_biology_loss": 1.4330155849456787, |
| "eval_biology_mean_token_accuracy": 0.6687032113671303, |
| "eval_biology_num_tokens": 25606824.0, |
| "eval_biology_runtime": 41.1629, |
| "eval_biology_samples_per_second": 12.147, |
| "eval_biology_steps_per_second": 12.147, |
| "step": 3600 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.6119256466031074, |
| "eval_math_loss": 0.6814781427383423, |
| "eval_math_mean_token_accuracy": 0.813846978187561, |
| "eval_math_num_tokens": 25606824.0, |
| "eval_math_runtime": 47.8257, |
| "eval_math_samples_per_second": 10.455, |
| "eval_math_steps_per_second": 10.455, |
| "step": 3600 |
| }, |
| { |
| "entropy": 0.6298922318965197, |
| "epoch": 1.444, |
| "grad_norm": 35.25, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 10.115, |
| "mean_token_accuracy": 0.8214882526546716, |
| "num_tokens": 25676784.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 0.5483139703050256, |
| "epoch": 1.448, |
| "grad_norm": 36.5, |
| "learning_rate": 1.418e-05, |
| "loss": 8.6914, |
| "mean_token_accuracy": 0.8437648601830006, |
| "num_tokens": 25745729.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 0.5973785981535912, |
| "epoch": 1.452, |
| "grad_norm": 32.75, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 9.5693, |
| "mean_token_accuracy": 0.8276977069675923, |
| "num_tokens": 25818154.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 0.5748674402944743, |
| "epoch": 1.456, |
| "grad_norm": 34.5, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 9.1328, |
| "mean_token_accuracy": 0.8347029034048319, |
| "num_tokens": 25886396.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 0.5753889610059559, |
| "epoch": 1.46, |
| "grad_norm": 31.0, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 9.5318, |
| "mean_token_accuracy": 0.8270909361541271, |
| "num_tokens": 25957363.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 0.5866841897368431, |
| "epoch": 1.464, |
| "grad_norm": 29.375, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 9.0008, |
| "mean_token_accuracy": 0.8385462883859873, |
| "num_tokens": 26026227.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 0.5919315716251731, |
| "epoch": 1.468, |
| "grad_norm": 30.375, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 9.9865, |
| "mean_token_accuracy": 0.8225046474486589, |
| "num_tokens": 26095693.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 0.5718209708109498, |
| "epoch": 1.472, |
| "grad_norm": 38.0, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 8.9535, |
| "mean_token_accuracy": 0.8386616911739111, |
| "num_tokens": 26163618.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 0.5722383395768702, |
| "epoch": 1.476, |
| "grad_norm": 36.0, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 9.233, |
| "mean_token_accuracy": 0.8336343917995691, |
| "num_tokens": 26231385.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 0.6007269184105098, |
| "epoch": 1.48, |
| "grad_norm": 39.25, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 9.5961, |
| "mean_token_accuracy": 0.8279270879924298, |
| "num_tokens": 26299086.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 1.48, |
| "eval_biology_entropy": 0.9609544697999954, |
| "eval_biology_loss": 1.442490816116333, |
| "eval_biology_mean_token_accuracy": 0.6690095273256302, |
| "eval_biology_num_tokens": 26299086.0, |
| "eval_biology_runtime": 40.2031, |
| "eval_biology_samples_per_second": 12.437, |
| "eval_biology_steps_per_second": 12.437, |
| "step": 3700 |
| }, |
| { |
| "epoch": 1.48, |
| "eval_math_entropy": 0.6104013387560845, |
| "eval_math_loss": 0.6809260845184326, |
| "eval_math_mean_token_accuracy": 0.8146877212524414, |
| "eval_math_num_tokens": 26299086.0, |
| "eval_math_runtime": 47.8582, |
| "eval_math_samples_per_second": 10.448, |
| "eval_math_steps_per_second": 10.448, |
| "step": 3700 |
| }, |
| { |
| "entropy": 0.5897867341525853, |
| "epoch": 1.484, |
| "grad_norm": 37.25, |
| "learning_rate": 1.398e-05, |
| "loss": 9.2958, |
| "mean_token_accuracy": 0.8317970305681228, |
| "num_tokens": 26371404.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 0.6106340009719133, |
| "epoch": 1.488, |
| "grad_norm": 38.25, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 10.2332, |
| "mean_token_accuracy": 0.8225075386464595, |
| "num_tokens": 26438338.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 0.6055848128162324, |
| "epoch": 1.492, |
| "grad_norm": 29.5, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 9.5273, |
| "mean_token_accuracy": 0.8313061073422432, |
| "num_tokens": 26509237.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 0.5675995083525777, |
| "epoch": 1.496, |
| "grad_norm": 36.25, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 9.1118, |
| "mean_token_accuracy": 0.835762781277299, |
| "num_tokens": 26583001.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 0.5545772364363074, |
| "epoch": 1.5, |
| "grad_norm": 41.0, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 9.0405, |
| "mean_token_accuracy": 0.8398870058357716, |
| "num_tokens": 26656022.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 0.578260760102421, |
| "epoch": 1.504, |
| "grad_norm": 42.0, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 9.3141, |
| "mean_token_accuracy": 0.832975946739316, |
| "num_tokens": 26729255.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 0.6024700452573597, |
| "epoch": 1.508, |
| "grad_norm": 33.25, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 9.6424, |
| "mean_token_accuracy": 0.8306707143783569, |
| "num_tokens": 26799507.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 0.581959372665733, |
| "epoch": 1.512, |
| "grad_norm": 34.0, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 9.3967, |
| "mean_token_accuracy": 0.8320085100829602, |
| "num_tokens": 26870139.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 0.61700435038656, |
| "epoch": 1.516, |
| "grad_norm": 29.375, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 9.7896, |
| "mean_token_accuracy": 0.8244428005069494, |
| "num_tokens": 26943128.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 0.5605267263017595, |
| "epoch": 1.52, |
| "grad_norm": 37.25, |
| "learning_rate": 1.378e-05, |
| "loss": 9.1249, |
| "mean_token_accuracy": 0.8354942105710507, |
| "num_tokens": 27017935.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_biology_entropy": 0.9649589394330979, |
| "eval_biology_loss": 1.4455039501190186, |
| "eval_biology_mean_token_accuracy": 0.6683246765732765, |
| "eval_biology_num_tokens": 27017935.0, |
| "eval_biology_runtime": 40.1053, |
| "eval_biology_samples_per_second": 12.467, |
| "eval_biology_steps_per_second": 12.467, |
| "step": 3800 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_math_entropy": 0.6081940614283085, |
| "eval_math_loss": 0.6789872050285339, |
| "eval_math_mean_token_accuracy": 0.8143288016319274, |
| "eval_math_num_tokens": 27017935.0, |
| "eval_math_runtime": 47.8241, |
| "eval_math_samples_per_second": 10.455, |
| "eval_math_steps_per_second": 10.455, |
| "step": 3800 |
| }, |
| { |
| "entropy": 0.5895396873354912, |
| "epoch": 1.524, |
| "grad_norm": 37.5, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 9.2776, |
| "mean_token_accuracy": 0.8315344247967005, |
| "num_tokens": 27088191.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 0.5761679332703352, |
| "epoch": 1.528, |
| "grad_norm": 36.75, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 9.3957, |
| "mean_token_accuracy": 0.8313839174807072, |
| "num_tokens": 27159322.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 0.5651649198494851, |
| "epoch": 1.532, |
| "grad_norm": 38.0, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 9.1094, |
| "mean_token_accuracy": 0.8352541666477918, |
| "num_tokens": 27231437.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 0.5857988853938878, |
| "epoch": 1.536, |
| "grad_norm": 30.75, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 9.3565, |
| "mean_token_accuracy": 0.8327815044671297, |
| "num_tokens": 27306339.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 0.5975470501929522, |
| "epoch": 1.54, |
| "grad_norm": 31.75, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 9.7524, |
| "mean_token_accuracy": 0.8263407997786999, |
| "num_tokens": 27375431.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 0.5421086225658656, |
| "epoch": 1.544, |
| "grad_norm": 33.75, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 8.557, |
| "mean_token_accuracy": 0.8459726881235838, |
| "num_tokens": 27445653.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 0.5725512441247702, |
| "epoch": 1.548, |
| "grad_norm": 30.75, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 9.3179, |
| "mean_token_accuracy": 0.832388847693801, |
| "num_tokens": 27516941.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 0.5819174693897367, |
| "epoch": 1.552, |
| "grad_norm": 33.75, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 9.3819, |
| "mean_token_accuracy": 0.8336057595908641, |
| "num_tokens": 27591959.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 0.5708610557019711, |
| "epoch": 1.556, |
| "grad_norm": 36.25, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 9.1628, |
| "mean_token_accuracy": 0.8358365040272474, |
| "num_tokens": 27666099.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 0.5568031224422156, |
| "epoch": 1.56, |
| "grad_norm": 35.0, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 8.96, |
| "mean_token_accuracy": 0.8376407705247402, |
| "num_tokens": 27736783.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 1.56, |
| "eval_biology_entropy": 0.9510138745307922, |
| "eval_biology_loss": 1.4489119052886963, |
| "eval_biology_mean_token_accuracy": 0.6683122003674508, |
| "eval_biology_num_tokens": 27736783.0, |
| "eval_biology_runtime": 40.0454, |
| "eval_biology_samples_per_second": 12.486, |
| "eval_biology_steps_per_second": 12.486, |
| "step": 3900 |
| }, |
| { |
| "epoch": 1.56, |
| "eval_math_entropy": 0.6076814251840115, |
| "eval_math_loss": 0.6762118935585022, |
| "eval_math_mean_token_accuracy": 0.8148132631778717, |
| "eval_math_num_tokens": 27736783.0, |
| "eval_math_runtime": 47.8162, |
| "eval_math_samples_per_second": 10.457, |
| "eval_math_steps_per_second": 10.457, |
| "step": 3900 |
| }, |
| { |
| "entropy": 0.5877414913848042, |
| "epoch": 1.564, |
| "grad_norm": 33.75, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 9.5262, |
| "mean_token_accuracy": 0.8329178016632796, |
| "num_tokens": 27812042.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 0.5690148934721947, |
| "epoch": 1.568, |
| "grad_norm": 37.75, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 9.1384, |
| "mean_token_accuracy": 0.8348321747034788, |
| "num_tokens": 27884398.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 0.5602716447785496, |
| "epoch": 1.572, |
| "grad_norm": 31.125, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 8.948, |
| "mean_token_accuracy": 0.8383640520274639, |
| "num_tokens": 27952140.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 0.6017616359516978, |
| "epoch": 1.576, |
| "grad_norm": 42.25, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 9.7858, |
| "mean_token_accuracy": 0.8279835518449545, |
| "num_tokens": 28025205.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 0.5878816215321422, |
| "epoch": 1.58, |
| "grad_norm": 33.75, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 9.3884, |
| "mean_token_accuracy": 0.8303833037614823, |
| "num_tokens": 28099027.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 0.5508759981021285, |
| "epoch": 1.584, |
| "grad_norm": 30.5, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 8.8698, |
| "mean_token_accuracy": 0.8383103083819151, |
| "num_tokens": 28171274.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 0.5493643004447222, |
| "epoch": 1.588, |
| "grad_norm": 28.5, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 8.808, |
| "mean_token_accuracy": 0.8376678515225648, |
| "num_tokens": 28240500.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 0.5919014382176101, |
| "epoch": 1.592, |
| "grad_norm": 34.0, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 9.4445, |
| "mean_token_accuracy": 0.8296855770051479, |
| "num_tokens": 28311058.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 0.5845738126896322, |
| "epoch": 1.596, |
| "grad_norm": 35.25, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 9.4886, |
| "mean_token_accuracy": 0.8311159454286099, |
| "num_tokens": 28383083.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 0.5855710312724114, |
| "epoch": 1.6, |
| "grad_norm": 28.25, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 9.3343, |
| "mean_token_accuracy": 0.8316084563732147, |
| "num_tokens": 28457624.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 0.9375748844742775, |
| "eval_biology_loss": 1.4528971910476685, |
| "eval_biology_mean_token_accuracy": 0.6690746158957481, |
| "eval_biology_num_tokens": 28457624.0, |
| "eval_biology_runtime": 39.9114, |
| "eval_biology_samples_per_second": 12.528, |
| "eval_biology_steps_per_second": 12.528, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.5905984618067741, |
| "eval_math_loss": 0.675506591796875, |
| "eval_math_mean_token_accuracy": 0.8157378882169724, |
| "eval_math_num_tokens": 28457624.0, |
| "eval_math_runtime": 47.889, |
| "eval_math_samples_per_second": 10.441, |
| "eval_math_steps_per_second": 10.441, |
| "step": 4000 |
| }, |
| { |
| "entropy": 0.5616078768856824, |
| "epoch": 1.604, |
| "grad_norm": 34.25, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 9.0526, |
| "mean_token_accuracy": 0.8392556302249432, |
| "num_tokens": 28529193.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 0.5796514172106981, |
| "epoch": 1.608, |
| "grad_norm": 42.0, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 9.244, |
| "mean_token_accuracy": 0.8326216947287322, |
| "num_tokens": 28596873.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 0.5706561678089201, |
| "epoch": 1.612, |
| "grad_norm": 46.25, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 9.4557, |
| "mean_token_accuracy": 0.8322120282799006, |
| "num_tokens": 28669050.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 0.6126288353465498, |
| "epoch": 1.616, |
| "grad_norm": 33.25, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 9.6829, |
| "mean_token_accuracy": 0.8284514036029578, |
| "num_tokens": 28743099.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 0.589196318667382, |
| "epoch": 1.62, |
| "grad_norm": 40.0, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 9.3436, |
| "mean_token_accuracy": 0.8315168414264917, |
| "num_tokens": 28812135.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 0.5632121779024601, |
| "epoch": 1.624, |
| "grad_norm": 31.5, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 9.2014, |
| "mean_token_accuracy": 0.8347886327654124, |
| "num_tokens": 28879957.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 0.5824346755631268, |
| "epoch": 1.6280000000000001, |
| "grad_norm": 38.0, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 9.2832, |
| "mean_token_accuracy": 0.8342964049428702, |
| "num_tokens": 28947145.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 0.5937619049102068, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 32.75, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 9.5875, |
| "mean_token_accuracy": 0.8263148795813322, |
| "num_tokens": 29017297.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 0.5855198019184172, |
| "epoch": 1.6360000000000001, |
| "grad_norm": 32.25, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 9.4701, |
| "mean_token_accuracy": 0.833595710620284, |
| "num_tokens": 29089253.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 0.5570460225455462, |
| "epoch": 1.6400000000000001, |
| "grad_norm": 38.0, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 8.9507, |
| "mean_token_accuracy": 0.8396992761641741, |
| "num_tokens": 29157749.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "eval_biology_entropy": 0.9786186684370041, |
| "eval_biology_loss": 1.4336193799972534, |
| "eval_biology_mean_token_accuracy": 0.6692190272808075, |
| "eval_biology_num_tokens": 29157749.0, |
| "eval_biology_runtime": 40.367, |
| "eval_biology_samples_per_second": 12.386, |
| "eval_biology_steps_per_second": 12.386, |
| "step": 4100 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "eval_math_entropy": 0.6095308331847191, |
| "eval_math_loss": 0.6744098663330078, |
| "eval_math_mean_token_accuracy": 0.8152628104686737, |
| "eval_math_num_tokens": 29157749.0, |
| "eval_math_runtime": 47.9142, |
| "eval_math_samples_per_second": 10.435, |
| "eval_math_steps_per_second": 10.435, |
| "step": 4100 |
| }, |
| { |
| "entropy": 0.5602362494915724, |
| "epoch": 1.6440000000000001, |
| "grad_norm": 35.5, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 9.0045, |
| "mean_token_accuracy": 0.8362956717610359, |
| "num_tokens": 29226918.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 0.596904530748725, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 39.5, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 9.7331, |
| "mean_token_accuracy": 0.830748264491558, |
| "num_tokens": 29303707.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 0.603045157995075, |
| "epoch": 1.6520000000000001, |
| "grad_norm": 30.625, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 9.4955, |
| "mean_token_accuracy": 0.8272947661578656, |
| "num_tokens": 29376847.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 0.5914746409282088, |
| "epoch": 1.6560000000000001, |
| "grad_norm": 37.25, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 9.6769, |
| "mean_token_accuracy": 0.8277469739317894, |
| "num_tokens": 29451704.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 0.5825162298977375, |
| "epoch": 1.6600000000000001, |
| "grad_norm": 33.0, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 9.2127, |
| "mean_token_accuracy": 0.8340028449892998, |
| "num_tokens": 29525739.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 0.5832555149681866, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 40.0, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 9.2916, |
| "mean_token_accuracy": 0.8325723119080066, |
| "num_tokens": 29597156.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 0.6144870173186063, |
| "epoch": 1.6680000000000001, |
| "grad_norm": 36.0, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 9.9071, |
| "mean_token_accuracy": 0.8234692554920912, |
| "num_tokens": 29670204.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 0.5920791003853083, |
| "epoch": 1.6720000000000002, |
| "grad_norm": 36.25, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 9.7153, |
| "mean_token_accuracy": 0.829657818377018, |
| "num_tokens": 29742548.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 0.5664140341803432, |
| "epoch": 1.6760000000000002, |
| "grad_norm": 34.5, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 8.9313, |
| "mean_token_accuracy": 0.8403172269463539, |
| "num_tokens": 29811812.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 0.560033879801631, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 33.75, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 8.9422, |
| "mean_token_accuracy": 0.8400102704763412, |
| "num_tokens": 29883879.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_biology_entropy": 0.9669315983057022, |
| "eval_biology_loss": 1.4416576623916626, |
| "eval_biology_mean_token_accuracy": 0.668340895652771, |
| "eval_biology_num_tokens": 29883879.0, |
| "eval_biology_runtime": 40.0153, |
| "eval_biology_samples_per_second": 12.495, |
| "eval_biology_steps_per_second": 12.495, |
| "step": 4200 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_math_entropy": 0.5963365880548954, |
| "eval_math_loss": 0.6737173795700073, |
| "eval_math_mean_token_accuracy": 0.8158558534383774, |
| "eval_math_num_tokens": 29883879.0, |
| "eval_math_runtime": 47.884, |
| "eval_math_samples_per_second": 10.442, |
| "eval_math_steps_per_second": 10.442, |
| "step": 4200 |
| }, |
| { |
| "entropy": 0.6051275508478284, |
| "epoch": 1.6840000000000002, |
| "grad_norm": 36.0, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 9.7011, |
| "mean_token_accuracy": 0.8268253810703754, |
| "num_tokens": 29953177.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 0.6047606054693461, |
| "epoch": 1.688, |
| "grad_norm": 32.25, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 9.6414, |
| "mean_token_accuracy": 0.8279032737016678, |
| "num_tokens": 30021432.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 0.6012007250450552, |
| "epoch": 1.692, |
| "grad_norm": 30.75, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 9.7068, |
| "mean_token_accuracy": 0.8285441923886537, |
| "num_tokens": 30093248.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 0.5650927022099494, |
| "epoch": 1.696, |
| "grad_norm": 36.75, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 9.1103, |
| "mean_token_accuracy": 0.8372129604220391, |
| "num_tokens": 30165760.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 0.5502642083913087, |
| "epoch": 1.7, |
| "grad_norm": 32.75, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 8.7677, |
| "mean_token_accuracy": 0.8388985089957715, |
| "num_tokens": 30239253.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 0.5787820436060429, |
| "epoch": 1.704, |
| "grad_norm": 38.0, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 9.4901, |
| "mean_token_accuracy": 0.8329074073582887, |
| "num_tokens": 30310423.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 0.5839776087552309, |
| "epoch": 1.708, |
| "grad_norm": 38.25, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 9.3397, |
| "mean_token_accuracy": 0.8316182408481836, |
| "num_tokens": 30381505.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 0.6065620223991572, |
| "epoch": 1.712, |
| "grad_norm": 30.25, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 9.8124, |
| "mean_token_accuracy": 0.8275521472096443, |
| "num_tokens": 30460367.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 0.577207105141133, |
| "epoch": 1.716, |
| "grad_norm": 29.875, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 9.0792, |
| "mean_token_accuracy": 0.8350558109581471, |
| "num_tokens": 30531161.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 0.5455361174419522, |
| "epoch": 1.72, |
| "grad_norm": 35.0, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 8.8997, |
| "mean_token_accuracy": 0.8397724449634552, |
| "num_tokens": 30599087.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 1.72, |
| "eval_biology_entropy": 0.9735766390562057, |
| "eval_biology_loss": 1.4352941513061523, |
| "eval_biology_mean_token_accuracy": 0.6690446574687958, |
| "eval_biology_num_tokens": 30599087.0, |
| "eval_biology_runtime": 40.7617, |
| "eval_biology_samples_per_second": 12.266, |
| "eval_biology_steps_per_second": 12.266, |
| "step": 4300 |
| }, |
| { |
| "epoch": 1.72, |
| "eval_math_entropy": 0.603808888912201, |
| "eval_math_loss": 0.6710309982299805, |
| "eval_math_mean_token_accuracy": 0.8162526323795318, |
| "eval_math_num_tokens": 30599087.0, |
| "eval_math_runtime": 48.5086, |
| "eval_math_samples_per_second": 10.307, |
| "eval_math_steps_per_second": 10.307, |
| "step": 4300 |
| }, |
| { |
| "entropy": 0.552053046785295, |
| "epoch": 1.724, |
| "grad_norm": 26.125, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 8.8287, |
| "mean_token_accuracy": 0.8419120132923126, |
| "num_tokens": 30668818.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 0.589231385756284, |
| "epoch": 1.728, |
| "grad_norm": 37.0, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 9.5031, |
| "mean_token_accuracy": 0.8326994139701128, |
| "num_tokens": 30739137.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 0.56565668201074, |
| "epoch": 1.732, |
| "grad_norm": 47.5, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 9.244, |
| "mean_token_accuracy": 0.8360852379351854, |
| "num_tokens": 30808925.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 0.5672437489032746, |
| "epoch": 1.736, |
| "grad_norm": 36.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 9.0226, |
| "mean_token_accuracy": 0.837062469124794, |
| "num_tokens": 30879704.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 0.5930971786379814, |
| "epoch": 1.74, |
| "grad_norm": 31.625, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 9.4625, |
| "mean_token_accuracy": 0.8336025845259428, |
| "num_tokens": 30951718.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 0.5807823000475765, |
| "epoch": 1.744, |
| "grad_norm": 34.5, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 9.4083, |
| "mean_token_accuracy": 0.8329890601336956, |
| "num_tokens": 31022663.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 0.5925798653624952, |
| "epoch": 1.748, |
| "grad_norm": 32.5, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 9.3963, |
| "mean_token_accuracy": 0.8316555388271809, |
| "num_tokens": 31094975.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 0.5572940752841532, |
| "epoch": 1.752, |
| "grad_norm": 41.5, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 8.8119, |
| "mean_token_accuracy": 0.8397982127964496, |
| "num_tokens": 31165247.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 0.5578285950236023, |
| "epoch": 1.756, |
| "grad_norm": 34.75, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 9.0406, |
| "mean_token_accuracy": 0.8383508753031492, |
| "num_tokens": 31234511.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 0.591203158441931, |
| "epoch": 1.76, |
| "grad_norm": 36.75, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 9.4542, |
| "mean_token_accuracy": 0.8299999993294478, |
| "num_tokens": 31306494.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 0.960493447959423, |
| "eval_biology_loss": 1.444394826889038, |
| "eval_biology_mean_token_accuracy": 0.6679441928863525, |
| "eval_biology_num_tokens": 31306494.0, |
| "eval_biology_runtime": 40.5394, |
| "eval_biology_samples_per_second": 12.334, |
| "eval_biology_steps_per_second": 12.334, |
| "step": 4400 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.5978306273221969, |
| "eval_math_loss": 0.6700568795204163, |
| "eval_math_mean_token_accuracy": 0.8162602570056915, |
| "eval_math_num_tokens": 31306494.0, |
| "eval_math_runtime": 47.9419, |
| "eval_math_samples_per_second": 10.429, |
| "eval_math_steps_per_second": 10.429, |
| "step": 4400 |
| }, |
| { |
| "entropy": 0.5447422393597663, |
| "epoch": 1.764, |
| "grad_norm": 36.0, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 8.8576, |
| "mean_token_accuracy": 0.8383992154151201, |
| "num_tokens": 31379700.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 0.5872387504205108, |
| "epoch": 1.768, |
| "grad_norm": 32.0, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 9.4327, |
| "mean_token_accuracy": 0.8300140161067248, |
| "num_tokens": 31453005.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 0.5547977220267057, |
| "epoch": 1.772, |
| "grad_norm": 46.5, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 8.7781, |
| "mean_token_accuracy": 0.8429825089871883, |
| "num_tokens": 31525548.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 0.5874564352445304, |
| "epoch": 1.776, |
| "grad_norm": 32.75, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 9.5026, |
| "mean_token_accuracy": 0.8296657670289278, |
| "num_tokens": 31595542.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 0.5846576142124832, |
| "epoch": 1.78, |
| "grad_norm": 36.5, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 9.4931, |
| "mean_token_accuracy": 0.8304748021066188, |
| "num_tokens": 31668404.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 0.5975283981300891, |
| "epoch": 1.784, |
| "grad_norm": 42.75, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 9.588, |
| "mean_token_accuracy": 0.830332250893116, |
| "num_tokens": 31739430.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 0.5766318107023836, |
| "epoch": 1.788, |
| "grad_norm": 30.375, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 9.2683, |
| "mean_token_accuracy": 0.8339407432824373, |
| "num_tokens": 31811558.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 0.5880326750688255, |
| "epoch": 1.792, |
| "grad_norm": 37.0, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 9.4081, |
| "mean_token_accuracy": 0.8317685853689909, |
| "num_tokens": 31881189.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 0.6076588155701756, |
| "epoch": 1.796, |
| "grad_norm": 32.5, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 9.7964, |
| "mean_token_accuracy": 0.8275455288589001, |
| "num_tokens": 31950347.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 0.5874856126494705, |
| "epoch": 1.8, |
| "grad_norm": 41.25, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 9.5288, |
| "mean_token_accuracy": 0.8309772223234176, |
| "num_tokens": 32019870.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.8, |
| "eval_biology_entropy": 0.9439696848988534, |
| "eval_biology_loss": 1.4482781887054443, |
| "eval_biology_mean_token_accuracy": 0.6687484840154648, |
| "eval_biology_num_tokens": 32019870.0, |
| "eval_biology_runtime": 40.1071, |
| "eval_biology_samples_per_second": 12.467, |
| "eval_biology_steps_per_second": 12.467, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.8, |
| "eval_math_entropy": 0.5939250442683697, |
| "eval_math_loss": 0.670048177242279, |
| "eval_math_mean_token_accuracy": 0.81644309425354, |
| "eval_math_num_tokens": 32019870.0, |
| "eval_math_runtime": 47.9268, |
| "eval_math_samples_per_second": 10.433, |
| "eval_math_steps_per_second": 10.433, |
| "step": 4500 |
| }, |
| { |
| "entropy": 0.591592953633517, |
| "epoch": 1.804, |
| "grad_norm": 37.5, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 9.4663, |
| "mean_token_accuracy": 0.8314048469066619, |
| "num_tokens": 32090411.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 0.5729503126814961, |
| "epoch": 1.808, |
| "grad_norm": 30.5, |
| "learning_rate": 1.218e-05, |
| "loss": 9.1846, |
| "mean_token_accuracy": 0.8344620518386364, |
| "num_tokens": 32164196.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 0.5401923522353173, |
| "epoch": 1.812, |
| "grad_norm": 40.0, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 8.6224, |
| "mean_token_accuracy": 0.8445983592420816, |
| "num_tokens": 32231361.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 0.5625739673152566, |
| "epoch": 1.8159999999999998, |
| "grad_norm": 36.75, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 9.2104, |
| "mean_token_accuracy": 0.8329465672373771, |
| "num_tokens": 32301902.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 0.5721864491701126, |
| "epoch": 1.8199999999999998, |
| "grad_norm": 40.0, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 9.1753, |
| "mean_token_accuracy": 0.8361610863357782, |
| "num_tokens": 32371851.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 0.552773086540401, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 34.75, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 8.6725, |
| "mean_token_accuracy": 0.8419431153684854, |
| "num_tokens": 32441530.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 0.5879080731421709, |
| "epoch": 1.8279999999999998, |
| "grad_norm": 34.25, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 9.7656, |
| "mean_token_accuracy": 0.8273586433380842, |
| "num_tokens": 32511421.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 0.5778103187680245, |
| "epoch": 1.8319999999999999, |
| "grad_norm": 28.875, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 9.0902, |
| "mean_token_accuracy": 0.8377371799200773, |
| "num_tokens": 32585977.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 0.5578692473471165, |
| "epoch": 1.8359999999999999, |
| "grad_norm": 35.5, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 9.1372, |
| "mean_token_accuracy": 0.8364483464509249, |
| "num_tokens": 32651479.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 0.5636928272433579, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 38.0, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 8.8979, |
| "mean_token_accuracy": 0.8396005164831877, |
| "num_tokens": 32723145.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_biology_entropy": 0.9701221086978913, |
| "eval_biology_loss": 1.4401638507843018, |
| "eval_biology_mean_token_accuracy": 0.6682176177501679, |
| "eval_biology_num_tokens": 32723145.0, |
| "eval_biology_runtime": 40.6836, |
| "eval_biology_samples_per_second": 12.29, |
| "eval_biology_steps_per_second": 12.29, |
| "step": 4600 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_math_entropy": 0.6022572484612465, |
| "eval_math_loss": 0.667454719543457, |
| "eval_math_mean_token_accuracy": 0.8170107330083847, |
| "eval_math_num_tokens": 32723145.0, |
| "eval_math_runtime": 48.0409, |
| "eval_math_samples_per_second": 10.408, |
| "eval_math_steps_per_second": 10.408, |
| "step": 4600 |
| }, |
| { |
| "entropy": 0.5732459939084947, |
| "epoch": 1.8439999999999999, |
| "grad_norm": 45.75, |
| "learning_rate": 1.198e-05, |
| "loss": 9.3036, |
| "mean_token_accuracy": 0.8351278465241194, |
| "num_tokens": 32794586.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 0.6367456924170256, |
| "epoch": 1.8479999999999999, |
| "grad_norm": 35.5, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 10.2297, |
| "mean_token_accuracy": 0.8200139585882426, |
| "num_tokens": 32863234.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 0.5722073020413518, |
| "epoch": 1.8519999999999999, |
| "grad_norm": 36.5, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 9.3578, |
| "mean_token_accuracy": 0.832675900682807, |
| "num_tokens": 32935208.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 0.5556033413857222, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 39.75, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 8.8935, |
| "mean_token_accuracy": 0.8411820638924837, |
| "num_tokens": 33011263.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 0.5753098588436842, |
| "epoch": 1.8599999999999999, |
| "grad_norm": 31.375, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 9.4306, |
| "mean_token_accuracy": 0.8335670538246631, |
| "num_tokens": 33081603.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 0.5721217789687216, |
| "epoch": 1.8639999999999999, |
| "grad_norm": 29.375, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 8.9636, |
| "mean_token_accuracy": 0.837339187040925, |
| "num_tokens": 33154530.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 0.5496943547390402, |
| "epoch": 1.8679999999999999, |
| "grad_norm": 36.75, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 8.9544, |
| "mean_token_accuracy": 0.8386598479002714, |
| "num_tokens": 33229616.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 0.5827247424051165, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 33.75, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 9.4018, |
| "mean_token_accuracy": 0.8328843250870704, |
| "num_tokens": 33298921.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 0.592114740703255, |
| "epoch": 1.876, |
| "grad_norm": 35.5, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 9.4461, |
| "mean_token_accuracy": 0.8303261484950781, |
| "num_tokens": 33368289.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 0.5575213777832687, |
| "epoch": 1.88, |
| "grad_norm": 34.25, |
| "learning_rate": 1.178e-05, |
| "loss": 8.95, |
| "mean_token_accuracy": 0.8379732754081488, |
| "num_tokens": 33439162.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 1.88, |
| "eval_biology_entropy": 0.9647734404802323, |
| "eval_biology_loss": 1.4424110651016235, |
| "eval_biology_mean_token_accuracy": 0.6681531444787979, |
| "eval_biology_num_tokens": 33439162.0, |
| "eval_biology_runtime": 40.2881, |
| "eval_biology_samples_per_second": 12.411, |
| "eval_biology_steps_per_second": 12.411, |
| "step": 4700 |
| }, |
| { |
| "epoch": 1.88, |
| "eval_math_entropy": 0.5872527770996093, |
| "eval_math_loss": 0.6669490933418274, |
| "eval_math_mean_token_accuracy": 0.8173821663856506, |
| "eval_math_num_tokens": 33439162.0, |
| "eval_math_runtime": 47.9029, |
| "eval_math_samples_per_second": 10.438, |
| "eval_math_steps_per_second": 10.438, |
| "step": 4700 |
| }, |
| { |
| "entropy": 0.572116751037538, |
| "epoch": 1.884, |
| "grad_norm": 42.75, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 9.2337, |
| "mean_token_accuracy": 0.8340180922299624, |
| "num_tokens": 33507952.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 0.5745996570214629, |
| "epoch": 1.888, |
| "grad_norm": 40.0, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 9.3428, |
| "mean_token_accuracy": 0.8338315512984991, |
| "num_tokens": 33576243.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 0.5917730777524411, |
| "epoch": 1.892, |
| "grad_norm": 37.0, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 9.3915, |
| "mean_token_accuracy": 0.8309908647090196, |
| "num_tokens": 33644378.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 0.5671752511523664, |
| "epoch": 1.896, |
| "grad_norm": 37.0, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 9.2054, |
| "mean_token_accuracy": 0.8340408805757761, |
| "num_tokens": 33713332.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 0.5596568941138684, |
| "epoch": 1.9, |
| "grad_norm": 30.5, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 8.9831, |
| "mean_token_accuracy": 0.8362199828028679, |
| "num_tokens": 33779501.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 0.5808821063488722, |
| "epoch": 1.904, |
| "grad_norm": 40.0, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 9.4328, |
| "mean_token_accuracy": 0.8312843669205904, |
| "num_tokens": 33850968.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 0.5848106031306088, |
| "epoch": 1.908, |
| "grad_norm": 53.0, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 9.4357, |
| "mean_token_accuracy": 0.8321265209466219, |
| "num_tokens": 33922000.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 0.5309702676720918, |
| "epoch": 1.912, |
| "grad_norm": 33.0, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 8.4926, |
| "mean_token_accuracy": 0.8452226776629687, |
| "num_tokens": 33991124.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 0.5600500871427357, |
| "epoch": 1.916, |
| "grad_norm": 33.25, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 8.9772, |
| "mean_token_accuracy": 0.8372367139905691, |
| "num_tokens": 34058829.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 0.5772063042037189, |
| "epoch": 1.92, |
| "grad_norm": 28.25, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 9.3403, |
| "mean_token_accuracy": 0.8328812483698129, |
| "num_tokens": 34128558.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 0.9833846768736839, |
| "eval_biology_loss": 1.4343079328536987, |
| "eval_biology_mean_token_accuracy": 0.6688793112039566, |
| "eval_biology_num_tokens": 34128558.0, |
| "eval_biology_runtime": 40.4822, |
| "eval_biology_samples_per_second": 12.351, |
| "eval_biology_steps_per_second": 12.351, |
| "step": 4800 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.5965022592246533, |
| "eval_math_loss": 0.6646488904953003, |
| "eval_math_mean_token_accuracy": 0.8175751775503158, |
| "eval_math_num_tokens": 34128558.0, |
| "eval_math_runtime": 47.8781, |
| "eval_math_samples_per_second": 10.443, |
| "eval_math_steps_per_second": 10.443, |
| "step": 4800 |
| }, |
| { |
| "entropy": 0.581888607982546, |
| "epoch": 1.924, |
| "grad_norm": 42.5, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 9.3822, |
| "mean_token_accuracy": 0.8320084352046251, |
| "num_tokens": 34196696.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 0.538437579292804, |
| "epoch": 1.928, |
| "grad_norm": 34.0, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 8.5328, |
| "mean_token_accuracy": 0.8442497037351131, |
| "num_tokens": 34270120.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 0.593761433660984, |
| "epoch": 1.932, |
| "grad_norm": 31.0, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 9.8251, |
| "mean_token_accuracy": 0.8250451762229204, |
| "num_tokens": 34341459.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 0.5615032879635692, |
| "epoch": 1.936, |
| "grad_norm": 42.25, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 9.0915, |
| "mean_token_accuracy": 0.8376366276293993, |
| "num_tokens": 34408056.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 0.5424795900471509, |
| "epoch": 1.94, |
| "grad_norm": 30.875, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 8.7544, |
| "mean_token_accuracy": 0.8421198908239603, |
| "num_tokens": 34475741.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 0.6079878654330969, |
| "epoch": 1.944, |
| "grad_norm": 38.25, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 9.725, |
| "mean_token_accuracy": 0.8278125166893006, |
| "num_tokens": 34545839.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 0.5549586090259254, |
| "epoch": 1.948, |
| "grad_norm": 35.0, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 8.9287, |
| "mean_token_accuracy": 0.8391751620918513, |
| "num_tokens": 34615650.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 0.5337929665111005, |
| "epoch": 1.952, |
| "grad_norm": 28.375, |
| "learning_rate": 1.138e-05, |
| "loss": 8.4922, |
| "mean_token_accuracy": 0.8454725973308086, |
| "num_tokens": 34684679.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 0.5776936015114188, |
| "epoch": 1.956, |
| "grad_norm": 37.0, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 9.13, |
| "mean_token_accuracy": 0.8355145722627639, |
| "num_tokens": 34755218.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 0.5653772335499525, |
| "epoch": 1.96, |
| "grad_norm": 31.75, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 9.1018, |
| "mean_token_accuracy": 0.8351291462779045, |
| "num_tokens": 34827507.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 1.96, |
| "eval_biology_entropy": 0.944690758883953, |
| "eval_biology_loss": 1.4469478130340576, |
| "eval_biology_mean_token_accuracy": 0.6691858900785446, |
| "eval_biology_num_tokens": 34827507.0, |
| "eval_biology_runtime": 40.466, |
| "eval_biology_samples_per_second": 12.356, |
| "eval_biology_steps_per_second": 12.356, |
| "step": 4900 |
| }, |
| { |
| "epoch": 1.96, |
| "eval_math_entropy": 0.5791578129529953, |
| "eval_math_loss": 0.6653723120689392, |
| "eval_math_mean_token_accuracy": 0.8178791055679321, |
| "eval_math_num_tokens": 34827507.0, |
| "eval_math_runtime": 47.9041, |
| "eval_math_samples_per_second": 10.438, |
| "eval_math_steps_per_second": 10.438, |
| "step": 4900 |
| }, |
| { |
| "entropy": 0.573746531456709, |
| "epoch": 1.964, |
| "grad_norm": 32.0, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 9.4628, |
| "mean_token_accuracy": 0.8325379803776741, |
| "num_tokens": 34901067.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 0.6074263927526772, |
| "epoch": 1.968, |
| "grad_norm": 39.75, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 9.5241, |
| "mean_token_accuracy": 0.8318448815494776, |
| "num_tokens": 34971038.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 0.5576337265782059, |
| "epoch": 1.972, |
| "grad_norm": 32.25, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 9.3257, |
| "mean_token_accuracy": 0.8344093907624484, |
| "num_tokens": 35043493.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 0.5668115041218698, |
| "epoch": 1.976, |
| "grad_norm": 45.75, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 9.0377, |
| "mean_token_accuracy": 0.8393813017755747, |
| "num_tokens": 35117690.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 0.5835123412311077, |
| "epoch": 1.98, |
| "grad_norm": 36.75, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 9.388, |
| "mean_token_accuracy": 0.8327907927334308, |
| "num_tokens": 35189434.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 0.5391603717580438, |
| "epoch": 1.984, |
| "grad_norm": 35.25, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 8.5631, |
| "mean_token_accuracy": 0.8431579664349556, |
| "num_tokens": 35262281.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 0.5445182221010327, |
| "epoch": 1.988, |
| "grad_norm": 28.75, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 8.9242, |
| "mean_token_accuracy": 0.8386421050876379, |
| "num_tokens": 35338715.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 0.5806571534834802, |
| "epoch": 1.992, |
| "grad_norm": 40.5, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 9.1673, |
| "mean_token_accuracy": 0.8352854270488024, |
| "num_tokens": 35417691.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 0.5637283084914089, |
| "epoch": 1.996, |
| "grad_norm": 41.0, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 9.1498, |
| "mean_token_accuracy": 0.8383650720119477, |
| "num_tokens": 35490048.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 0.5771158320829273, |
| "epoch": 2.0, |
| "grad_norm": 29.0, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 9.3271, |
| "mean_token_accuracy": 0.8334357690066099, |
| "num_tokens": 35560864.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_biology_entropy": 0.9617877964973449, |
| "eval_biology_loss": 1.4474459886550903, |
| "eval_biology_mean_token_accuracy": 0.668456871509552, |
| "eval_biology_num_tokens": 35560864.0, |
| "eval_biology_runtime": 40.0579, |
| "eval_biology_samples_per_second": 12.482, |
| "eval_biology_steps_per_second": 12.482, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_math_entropy": 0.6012033255696296, |
| "eval_math_loss": 0.6616828441619873, |
| "eval_math_mean_token_accuracy": 0.8183047770261764, |
| "eval_math_num_tokens": 35560864.0, |
| "eval_math_runtime": 48.0389, |
| "eval_math_samples_per_second": 10.408, |
| "eval_math_steps_per_second": 10.408, |
| "step": 5000 |
| }, |
| { |
| "entropy": 0.48333557937294247, |
| "epoch": 2.004, |
| "grad_norm": 57.0, |
| "learning_rate": 1.1091111111111112e-05, |
| "loss": 7.2018, |
| "mean_token_accuracy": 0.8657604165375232, |
| "num_tokens": 35632501.0, |
| "step": 5010 |
| }, |
| { |
| "entropy": 0.44603592399507763, |
| "epoch": 2.008, |
| "grad_norm": 51.25, |
| "learning_rate": 1.106888888888889e-05, |
| "loss": 7.1893, |
| "mean_token_accuracy": 0.8661236446350813, |
| "num_tokens": 35704045.0, |
| "step": 5020 |
| }, |
| { |
| "entropy": 0.4656199609860778, |
| "epoch": 2.012, |
| "grad_norm": 50.5, |
| "learning_rate": 1.1046666666666667e-05, |
| "loss": 7.1912, |
| "mean_token_accuracy": 0.865003102645278, |
| "num_tokens": 35776863.0, |
| "step": 5030 |
| }, |
| { |
| "entropy": 0.4289406555704772, |
| "epoch": 2.016, |
| "grad_norm": 47.75, |
| "learning_rate": 1.1024444444444445e-05, |
| "loss": 6.8621, |
| "mean_token_accuracy": 0.8691702011972666, |
| "num_tokens": 35846704.0, |
| "step": 5040 |
| }, |
| { |
| "entropy": 0.4225952158682048, |
| "epoch": 2.02, |
| "grad_norm": 51.75, |
| "learning_rate": 1.1002222222222222e-05, |
| "loss": 6.5534, |
| "mean_token_accuracy": 0.875089630484581, |
| "num_tokens": 35915069.0, |
| "step": 5050 |
| }, |
| { |
| "entropy": 0.4090152862481773, |
| "epoch": 2.024, |
| "grad_norm": 49.0, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 6.7162, |
| "mean_token_accuracy": 0.8718582101166248, |
| "num_tokens": 35981996.0, |
| "step": 5060 |
| }, |
| { |
| "entropy": 0.47538163838908076, |
| "epoch": 2.028, |
| "grad_norm": 46.5, |
| "learning_rate": 1.0957777777777778e-05, |
| "loss": 7.3422, |
| "mean_token_accuracy": 0.8629732731729746, |
| "num_tokens": 36055284.0, |
| "step": 5070 |
| }, |
| { |
| "entropy": 0.4147109162993729, |
| "epoch": 2.032, |
| "grad_norm": 45.75, |
| "learning_rate": 1.0935555555555557e-05, |
| "loss": 6.8187, |
| "mean_token_accuracy": 0.8697662387043238, |
| "num_tokens": 36128775.0, |
| "step": 5080 |
| }, |
| { |
| "entropy": 0.4318197165615857, |
| "epoch": 2.036, |
| "grad_norm": 42.75, |
| "learning_rate": 1.0913333333333333e-05, |
| "loss": 6.7415, |
| "mean_token_accuracy": 0.8734662868082523, |
| "num_tokens": 36200156.0, |
| "step": 5090 |
| }, |
| { |
| "entropy": 0.4198383717797697, |
| "epoch": 2.04, |
| "grad_norm": 46.5, |
| "learning_rate": 1.0891111111111112e-05, |
| "loss": 6.9036, |
| "mean_token_accuracy": 0.8698579989373684, |
| "num_tokens": 36275139.0, |
| "step": 5100 |
| }, |
| { |
| "epoch": 2.04, |
| "eval_biology_entropy": 0.6692150049209594, |
| "eval_biology_loss": 1.8137288093566895, |
| "eval_biology_mean_token_accuracy": 0.6568104530572891, |
| "eval_biology_num_tokens": 36275139.0, |
| "eval_biology_runtime": 40.3935, |
| "eval_biology_samples_per_second": 12.378, |
| "eval_biology_steps_per_second": 12.378, |
| "step": 5100 |
| }, |
| { |
| "epoch": 2.04, |
| "eval_math_entropy": 0.4824463215172291, |
| "eval_math_loss": 0.727523922920227, |
| "eval_math_mean_token_accuracy": 0.8140114243030548, |
| "eval_math_num_tokens": 36275139.0, |
| "eval_math_runtime": 48.0008, |
| "eval_math_samples_per_second": 10.417, |
| "eval_math_steps_per_second": 10.417, |
| "step": 5100 |
| }, |
| { |
| "entropy": 0.4135431839153171, |
| "epoch": 2.044, |
| "grad_norm": 41.75, |
| "learning_rate": 1.0868888888888888e-05, |
| "loss": 6.5728, |
| "mean_token_accuracy": 0.8758707977831364, |
| "num_tokens": 36347742.0, |
| "step": 5110 |
| }, |
| { |
| "entropy": 0.4344665420241654, |
| "epoch": 2.048, |
| "grad_norm": 65.0, |
| "learning_rate": 1.0846666666666667e-05, |
| "loss": 6.908, |
| "mean_token_accuracy": 0.8692439094185829, |
| "num_tokens": 36419504.0, |
| "step": 5120 |
| }, |
| { |
| "entropy": 0.4054893317632377, |
| "epoch": 2.052, |
| "grad_norm": 48.0, |
| "learning_rate": 1.0824444444444447e-05, |
| "loss": 6.3364, |
| "mean_token_accuracy": 0.879656569287181, |
| "num_tokens": 36492520.0, |
| "step": 5130 |
| }, |
| { |
| "entropy": 0.4304758722893894, |
| "epoch": 2.056, |
| "grad_norm": 50.75, |
| "learning_rate": 1.0802222222222223e-05, |
| "loss": 6.9645, |
| "mean_token_accuracy": 0.8667810395359993, |
| "num_tokens": 36562175.0, |
| "step": 5140 |
| }, |
| { |
| "entropy": 0.4236756752245128, |
| "epoch": 2.06, |
| "grad_norm": 51.5, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 6.6688, |
| "mean_token_accuracy": 0.8728464279323817, |
| "num_tokens": 36630924.0, |
| "step": 5150 |
| }, |
| { |
| "entropy": 0.4411948885302991, |
| "epoch": 2.064, |
| "grad_norm": 40.5, |
| "learning_rate": 1.0757777777777778e-05, |
| "loss": 6.9957, |
| "mean_token_accuracy": 0.8682134177535772, |
| "num_tokens": 36706816.0, |
| "step": 5160 |
| }, |
| { |
| "entropy": 0.42259727269411085, |
| "epoch": 2.068, |
| "grad_norm": 51.75, |
| "learning_rate": 1.0735555555555557e-05, |
| "loss": 6.712, |
| "mean_token_accuracy": 0.8712298650294542, |
| "num_tokens": 36778853.0, |
| "step": 5170 |
| }, |
| { |
| "entropy": 0.45333392871543765, |
| "epoch": 2.072, |
| "grad_norm": 51.25, |
| "learning_rate": 1.0713333333333333e-05, |
| "loss": 7.2448, |
| "mean_token_accuracy": 0.8623873520642519, |
| "num_tokens": 36852961.0, |
| "step": 5180 |
| }, |
| { |
| "entropy": 0.396093035582453, |
| "epoch": 2.076, |
| "grad_norm": 42.25, |
| "learning_rate": 1.0691111111111112e-05, |
| "loss": 6.2931, |
| "mean_token_accuracy": 0.8800598863512278, |
| "num_tokens": 36921060.0, |
| "step": 5190 |
| }, |
| { |
| "entropy": 0.4133703854866326, |
| "epoch": 2.08, |
| "grad_norm": 47.5, |
| "learning_rate": 1.0668888888888892e-05, |
| "loss": 6.5781, |
| "mean_token_accuracy": 0.8726602621376515, |
| "num_tokens": 36988475.0, |
| "step": 5200 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 0.6288086797595024, |
| "eval_biology_loss": 1.8924697637557983, |
| "eval_biology_mean_token_accuracy": 0.6553987149596214, |
| "eval_biology_num_tokens": 36988475.0, |
| "eval_biology_runtime": 40.0324, |
| "eval_biology_samples_per_second": 12.49, |
| "eval_biology_steps_per_second": 12.49, |
| "step": 5200 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.466902303814888, |
| "eval_math_loss": 0.7450554966926575, |
| "eval_math_mean_token_accuracy": 0.8134832118749619, |
| "eval_math_num_tokens": 36988475.0, |
| "eval_math_runtime": 47.9376, |
| "eval_math_samples_per_second": 10.43, |
| "eval_math_steps_per_second": 10.43, |
| "step": 5200 |
| }, |
| { |
| "entropy": 0.4273119827732444, |
| "epoch": 2.084, |
| "grad_norm": 53.75, |
| "learning_rate": 1.0646666666666668e-05, |
| "loss": 6.7974, |
| "mean_token_accuracy": 0.8704006802290678, |
| "num_tokens": 37061454.0, |
| "step": 5210 |
| }, |
| { |
| "entropy": 0.40666443668305874, |
| "epoch": 2.088, |
| "grad_norm": 51.0, |
| "learning_rate": 1.0624444444444447e-05, |
| "loss": 6.563, |
| "mean_token_accuracy": 0.875283233076334, |
| "num_tokens": 37130283.0, |
| "step": 5220 |
| }, |
| { |
| "entropy": 0.42788051180541514, |
| "epoch": 2.092, |
| "grad_norm": 40.25, |
| "learning_rate": 1.0602222222222223e-05, |
| "loss": 6.7324, |
| "mean_token_accuracy": 0.8710669506341219, |
| "num_tokens": 37198149.0, |
| "step": 5230 |
| }, |
| { |
| "entropy": 0.439534904062748, |
| "epoch": 2.096, |
| "grad_norm": 57.0, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 7.1844, |
| "mean_token_accuracy": 0.8663857374340296, |
| "num_tokens": 37270131.0, |
| "step": 5240 |
| }, |
| { |
| "entropy": 0.418748477846384, |
| "epoch": 2.1, |
| "grad_norm": 43.75, |
| "learning_rate": 1.0557777777777778e-05, |
| "loss": 6.5874, |
| "mean_token_accuracy": 0.8749070011079312, |
| "num_tokens": 37341541.0, |
| "step": 5250 |
| }, |
| { |
| "entropy": 0.41906254151836037, |
| "epoch": 2.104, |
| "grad_norm": 41.75, |
| "learning_rate": 1.0535555555555557e-05, |
| "loss": 6.7278, |
| "mean_token_accuracy": 0.871961610764265, |
| "num_tokens": 37417483.0, |
| "step": 5260 |
| }, |
| { |
| "entropy": 0.4073872173205018, |
| "epoch": 2.108, |
| "grad_norm": 51.25, |
| "learning_rate": 1.0513333333333333e-05, |
| "loss": 6.47, |
| "mean_token_accuracy": 0.875134440138936, |
| "num_tokens": 37490431.0, |
| "step": 5270 |
| }, |
| { |
| "entropy": 0.4316187701188028, |
| "epoch": 2.112, |
| "grad_norm": 44.5, |
| "learning_rate": 1.0491111111111112e-05, |
| "loss": 7.0635, |
| "mean_token_accuracy": 0.8656156584620476, |
| "num_tokens": 37563537.0, |
| "step": 5280 |
| }, |
| { |
| "entropy": 0.414285259693861, |
| "epoch": 2.116, |
| "grad_norm": 41.25, |
| "learning_rate": 1.046888888888889e-05, |
| "loss": 6.5058, |
| "mean_token_accuracy": 0.8762889303267002, |
| "num_tokens": 37630624.0, |
| "step": 5290 |
| }, |
| { |
| "entropy": 0.4244235543999821, |
| "epoch": 2.12, |
| "grad_norm": 42.25, |
| "learning_rate": 1.0446666666666668e-05, |
| "loss": 6.7214, |
| "mean_token_accuracy": 0.8719112060964107, |
| "num_tokens": 37700425.0, |
| "step": 5300 |
| }, |
| { |
| "epoch": 2.12, |
| "eval_biology_entropy": 0.6237461242079735, |
| "eval_biology_loss": 1.9198321104049683, |
| "eval_biology_mean_token_accuracy": 0.6545260121226311, |
| "eval_biology_num_tokens": 37700425.0, |
| "eval_biology_runtime": 39.9563, |
| "eval_biology_samples_per_second": 12.514, |
| "eval_biology_steps_per_second": 12.514, |
| "step": 5300 |
| }, |
| { |
| "epoch": 2.12, |
| "eval_math_entropy": 0.47117885762453077, |
| "eval_math_loss": 0.7463982105255127, |
| "eval_math_mean_token_accuracy": 0.8131656917333603, |
| "eval_math_num_tokens": 37700425.0, |
| "eval_math_runtime": 47.8274, |
| "eval_math_samples_per_second": 10.454, |
| "eval_math_steps_per_second": 10.454, |
| "step": 5300 |
| }, |
| { |
| "entropy": 0.4058587061241269, |
| "epoch": 2.124, |
| "grad_norm": 62.5, |
| "learning_rate": 1.0424444444444445e-05, |
| "loss": 6.5157, |
| "mean_token_accuracy": 0.8754384856671095, |
| "num_tokens": 37773222.0, |
| "step": 5310 |
| }, |
| { |
| "entropy": 0.4109736021142453, |
| "epoch": 2.128, |
| "grad_norm": 43.5, |
| "learning_rate": 1.0402222222222223e-05, |
| "loss": 6.4708, |
| "mean_token_accuracy": 0.876412882655859, |
| "num_tokens": 37843959.0, |
| "step": 5320 |
| }, |
| { |
| "entropy": 0.3838260056450963, |
| "epoch": 2.132, |
| "grad_norm": 45.75, |
| "learning_rate": 1.038e-05, |
| "loss": 6.3446, |
| "mean_token_accuracy": 0.8816161844879389, |
| "num_tokens": 37914076.0, |
| "step": 5330 |
| }, |
| { |
| "entropy": 0.4285282687284052, |
| "epoch": 2.136, |
| "grad_norm": 49.5, |
| "learning_rate": 1.0357777777777778e-05, |
| "loss": 6.8685, |
| "mean_token_accuracy": 0.8690095663070678, |
| "num_tokens": 37987862.0, |
| "step": 5340 |
| }, |
| { |
| "entropy": 0.4520016681402922, |
| "epoch": 2.14, |
| "grad_norm": 49.25, |
| "learning_rate": 1.0335555555555556e-05, |
| "loss": 7.1454, |
| "mean_token_accuracy": 0.8653109017759562, |
| "num_tokens": 38057937.0, |
| "step": 5350 |
| }, |
| { |
| "entropy": 0.42330782283097507, |
| "epoch": 2.144, |
| "grad_norm": 48.5, |
| "learning_rate": 1.0313333333333335e-05, |
| "loss": 6.7141, |
| "mean_token_accuracy": 0.8715841066092253, |
| "num_tokens": 38133092.0, |
| "step": 5360 |
| }, |
| { |
| "entropy": 0.4245350930839777, |
| "epoch": 2.148, |
| "grad_norm": 53.0, |
| "learning_rate": 1.0291111111111111e-05, |
| "loss": 6.7435, |
| "mean_token_accuracy": 0.8724341690540314, |
| "num_tokens": 38200544.0, |
| "step": 5370 |
| }, |
| { |
| "entropy": 0.4385092446580529, |
| "epoch": 2.152, |
| "grad_norm": 90.5, |
| "learning_rate": 1.026888888888889e-05, |
| "loss": 6.9722, |
| "mean_token_accuracy": 0.8670150700956583, |
| "num_tokens": 38269938.0, |
| "step": 5380 |
| }, |
| { |
| "entropy": 0.4299027158878744, |
| "epoch": 2.156, |
| "grad_norm": 46.25, |
| "learning_rate": 1.0246666666666666e-05, |
| "loss": 7.0385, |
| "mean_token_accuracy": 0.8668250177055598, |
| "num_tokens": 38345608.0, |
| "step": 5390 |
| }, |
| { |
| "entropy": 0.4040191687643528, |
| "epoch": 2.16, |
| "grad_norm": 51.75, |
| "learning_rate": 1.0224444444444446e-05, |
| "loss": 6.3397, |
| "mean_token_accuracy": 0.8804498553276062, |
| "num_tokens": 38421229.0, |
| "step": 5400 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_biology_entropy": 0.6098170667290688, |
| "eval_biology_loss": 1.939279317855835, |
| "eval_biology_mean_token_accuracy": 0.6549080154895782, |
| "eval_biology_num_tokens": 38421229.0, |
| "eval_biology_runtime": 39.896, |
| "eval_biology_samples_per_second": 12.533, |
| "eval_biology_steps_per_second": 12.533, |
| "step": 5400 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_math_entropy": 0.4659426206201315, |
| "eval_math_loss": 0.7484189867973328, |
| "eval_math_mean_token_accuracy": 0.8140275942087174, |
| "eval_math_num_tokens": 38421229.0, |
| "eval_math_runtime": 47.832, |
| "eval_math_samples_per_second": 10.453, |
| "eval_math_steps_per_second": 10.453, |
| "step": 5400 |
| }, |
| { |
| "entropy": 0.40074054813012483, |
| "epoch": 2.164, |
| "grad_norm": 49.75, |
| "learning_rate": 1.0202222222222221e-05, |
| "loss": 6.526, |
| "mean_token_accuracy": 0.875863891094923, |
| "num_tokens": 38487269.0, |
| "step": 5410 |
| }, |
| { |
| "entropy": 0.42428737026639285, |
| "epoch": 2.168, |
| "grad_norm": 47.25, |
| "learning_rate": 1.018e-05, |
| "loss": 6.6164, |
| "mean_token_accuracy": 0.8728472393006086, |
| "num_tokens": 38560848.0, |
| "step": 5420 |
| }, |
| { |
| "entropy": 0.39232771513052284, |
| "epoch": 2.172, |
| "grad_norm": 51.25, |
| "learning_rate": 1.0157777777777777e-05, |
| "loss": 6.3707, |
| "mean_token_accuracy": 0.8788433004170656, |
| "num_tokens": 38633193.0, |
| "step": 5430 |
| }, |
| { |
| "entropy": 0.41454373560845853, |
| "epoch": 2.176, |
| "grad_norm": 50.5, |
| "learning_rate": 1.0135555555555556e-05, |
| "loss": 6.5018, |
| "mean_token_accuracy": 0.877242972329259, |
| "num_tokens": 38708043.0, |
| "step": 5440 |
| }, |
| { |
| "entropy": 0.4186694511212409, |
| "epoch": 2.18, |
| "grad_norm": 48.75, |
| "learning_rate": 1.0113333333333335e-05, |
| "loss": 6.7404, |
| "mean_token_accuracy": 0.8722327191382646, |
| "num_tokens": 38777814.0, |
| "step": 5450 |
| }, |
| { |
| "entropy": 0.4334541825577617, |
| "epoch": 2.184, |
| "grad_norm": 45.0, |
| "learning_rate": 1.0091111111111111e-05, |
| "loss": 6.8788, |
| "mean_token_accuracy": 0.8696263659745455, |
| "num_tokens": 38853032.0, |
| "step": 5460 |
| }, |
| { |
| "entropy": 0.4310882013291121, |
| "epoch": 2.188, |
| "grad_norm": 41.25, |
| "learning_rate": 1.006888888888889e-05, |
| "loss": 6.9125, |
| "mean_token_accuracy": 0.8697678282856941, |
| "num_tokens": 38923341.0, |
| "step": 5470 |
| }, |
| { |
| "entropy": 0.4318785283714533, |
| "epoch": 2.192, |
| "grad_norm": 49.5, |
| "learning_rate": 1.0046666666666666e-05, |
| "loss": 6.9133, |
| "mean_token_accuracy": 0.8704640254378319, |
| "num_tokens": 38996930.0, |
| "step": 5480 |
| }, |
| { |
| "entropy": 0.417129055922851, |
| "epoch": 2.196, |
| "grad_norm": 51.25, |
| "learning_rate": 1.0024444444444446e-05, |
| "loss": 6.623, |
| "mean_token_accuracy": 0.8735948640853166, |
| "num_tokens": 39065367.0, |
| "step": 5490 |
| }, |
| { |
| "entropy": 0.43122339341789484, |
| "epoch": 2.2, |
| "grad_norm": 47.25, |
| "learning_rate": 1.0002222222222222e-05, |
| "loss": 6.9899, |
| "mean_token_accuracy": 0.8681222185492515, |
| "num_tokens": 39134848.0, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.2, |
| "eval_biology_entropy": 0.6017167692184449, |
| "eval_biology_loss": 1.9638246297836304, |
| "eval_biology_mean_token_accuracy": 0.6543927209377289, |
| "eval_biology_num_tokens": 39134848.0, |
| "eval_biology_runtime": 39.9488, |
| "eval_biology_samples_per_second": 12.516, |
| "eval_biology_steps_per_second": 12.516, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.2, |
| "eval_math_entropy": 0.46603856950998307, |
| "eval_math_loss": 0.7466142177581787, |
| "eval_math_mean_token_accuracy": 0.8139623812437058, |
| "eval_math_num_tokens": 39134848.0, |
| "eval_math_runtime": 48.2258, |
| "eval_math_samples_per_second": 10.368, |
| "eval_math_steps_per_second": 10.368, |
| "step": 5500 |
| }, |
| { |
| "entropy": 0.44710859870538117, |
| "epoch": 2.204, |
| "grad_norm": 43.0, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 7.0922, |
| "mean_token_accuracy": 0.8686847653239965, |
| "num_tokens": 39210579.0, |
| "step": 5510 |
| }, |
| { |
| "entropy": 0.40868861358612774, |
| "epoch": 2.208, |
| "grad_norm": 50.0, |
| "learning_rate": 9.957777777777779e-06, |
| "loss": 6.5547, |
| "mean_token_accuracy": 0.8753906391561032, |
| "num_tokens": 39279481.0, |
| "step": 5520 |
| }, |
| { |
| "entropy": 0.4289680261164904, |
| "epoch": 2.212, |
| "grad_norm": 55.75, |
| "learning_rate": 9.935555555555556e-06, |
| "loss": 6.9122, |
| "mean_token_accuracy": 0.8692066993564367, |
| "num_tokens": 39350819.0, |
| "step": 5530 |
| }, |
| { |
| "entropy": 0.4051651638466865, |
| "epoch": 2.216, |
| "grad_norm": 42.5, |
| "learning_rate": 9.913333333333334e-06, |
| "loss": 6.3632, |
| "mean_token_accuracy": 0.8777267079800367, |
| "num_tokens": 39422640.0, |
| "step": 5540 |
| }, |
| { |
| "entropy": 0.4295547602698207, |
| "epoch": 2.22, |
| "grad_norm": 54.5, |
| "learning_rate": 9.891111111111113e-06, |
| "loss": 6.8093, |
| "mean_token_accuracy": 0.8717206422239542, |
| "num_tokens": 39499029.0, |
| "step": 5550 |
| }, |
| { |
| "entropy": 0.4212003557942808, |
| "epoch": 2.224, |
| "grad_norm": 48.5, |
| "learning_rate": 9.86888888888889e-06, |
| "loss": 6.8865, |
| "mean_token_accuracy": 0.871536310017109, |
| "num_tokens": 39569030.0, |
| "step": 5560 |
| }, |
| { |
| "entropy": 0.4268992354162037, |
| "epoch": 2.228, |
| "grad_norm": 50.25, |
| "learning_rate": 9.846666666666668e-06, |
| "loss": 6.6546, |
| "mean_token_accuracy": 0.8737176164984704, |
| "num_tokens": 39641377.0, |
| "step": 5570 |
| }, |
| { |
| "entropy": 0.42734280228614807, |
| "epoch": 2.232, |
| "grad_norm": 55.75, |
| "learning_rate": 9.824444444444446e-06, |
| "loss": 6.927, |
| "mean_token_accuracy": 0.8698129173368215, |
| "num_tokens": 39711909.0, |
| "step": 5580 |
| }, |
| { |
| "entropy": 0.4118302929215133, |
| "epoch": 2.2359999999999998, |
| "grad_norm": 46.0, |
| "learning_rate": 9.802222222222224e-06, |
| "loss": 6.6016, |
| "mean_token_accuracy": 0.8747963376343251, |
| "num_tokens": 39778802.0, |
| "step": 5590 |
| }, |
| { |
| "entropy": 0.44969167895615103, |
| "epoch": 2.24, |
| "grad_norm": 46.25, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 7.2577, |
| "mean_token_accuracy": 0.8630360651761293, |
| "num_tokens": 39854873.0, |
| "step": 5600 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 0.6057550595998764, |
| "eval_biology_loss": 1.96756911277771, |
| "eval_biology_mean_token_accuracy": 0.6547689053416252, |
| "eval_biology_num_tokens": 39854873.0, |
| "eval_biology_runtime": 40.343, |
| "eval_biology_samples_per_second": 12.394, |
| "eval_biology_steps_per_second": 12.394, |
| "step": 5600 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.4735337852537632, |
| "eval_math_loss": 0.7460785508155823, |
| "eval_math_mean_token_accuracy": 0.8141546627283096, |
| "eval_math_num_tokens": 39854873.0, |
| "eval_math_runtime": 48.0671, |
| "eval_math_samples_per_second": 10.402, |
| "eval_math_steps_per_second": 10.402, |
| "step": 5600 |
| }, |
| { |
| "entropy": 0.42828079536557195, |
| "epoch": 2.2439999999999998, |
| "grad_norm": 55.75, |
| "learning_rate": 9.757777777777779e-06, |
| "loss": 6.864, |
| "mean_token_accuracy": 0.8715633232146501, |
| "num_tokens": 39930381.0, |
| "step": 5610 |
| }, |
| { |
| "entropy": 0.4341730220243335, |
| "epoch": 2.248, |
| "grad_norm": 49.0, |
| "learning_rate": 9.735555555555556e-06, |
| "loss": 6.9127, |
| "mean_token_accuracy": 0.8678554892539978, |
| "num_tokens": 40001461.0, |
| "step": 5620 |
| }, |
| { |
| "entropy": 0.42248312048614023, |
| "epoch": 2.252, |
| "grad_norm": 43.5, |
| "learning_rate": 9.713333333333334e-06, |
| "loss": 6.7504, |
| "mean_token_accuracy": 0.8722781769931316, |
| "num_tokens": 40073826.0, |
| "step": 5630 |
| }, |
| { |
| "entropy": 0.42506562154740096, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 49.5, |
| "learning_rate": 9.691111111111112e-06, |
| "loss": 6.7171, |
| "mean_token_accuracy": 0.8711165145039559, |
| "num_tokens": 40141190.0, |
| "step": 5640 |
| }, |
| { |
| "entropy": 0.42791012856177985, |
| "epoch": 2.26, |
| "grad_norm": 52.0, |
| "learning_rate": 9.66888888888889e-06, |
| "loss": 6.7408, |
| "mean_token_accuracy": 0.8737585876137018, |
| "num_tokens": 40206181.0, |
| "step": 5650 |
| }, |
| { |
| "entropy": 0.43890425460413096, |
| "epoch": 2.2640000000000002, |
| "grad_norm": 49.25, |
| "learning_rate": 9.646666666666667e-06, |
| "loss": 7.1119, |
| "mean_token_accuracy": 0.8661639731377363, |
| "num_tokens": 40278495.0, |
| "step": 5660 |
| }, |
| { |
| "entropy": 0.39845298742875457, |
| "epoch": 2.268, |
| "grad_norm": 57.0, |
| "learning_rate": 9.624444444444445e-06, |
| "loss": 6.4125, |
| "mean_token_accuracy": 0.8783936399966479, |
| "num_tokens": 40349980.0, |
| "step": 5670 |
| }, |
| { |
| "entropy": 0.41560715418308974, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 60.0, |
| "learning_rate": 9.602222222222222e-06, |
| "loss": 6.5269, |
| "mean_token_accuracy": 0.8741395603865385, |
| "num_tokens": 40415203.0, |
| "step": 5680 |
| }, |
| { |
| "entropy": 0.4151542292907834, |
| "epoch": 2.276, |
| "grad_norm": 49.5, |
| "learning_rate": 9.58e-06, |
| "loss": 6.8056, |
| "mean_token_accuracy": 0.8695466585457325, |
| "num_tokens": 40486789.0, |
| "step": 5690 |
| }, |
| { |
| "entropy": 0.422638895874843, |
| "epoch": 2.2800000000000002, |
| "grad_norm": 45.75, |
| "learning_rate": 9.557777777777777e-06, |
| "loss": 6.768, |
| "mean_token_accuracy": 0.871444109082222, |
| "num_tokens": 40564273.0, |
| "step": 5700 |
| }, |
| { |
| "epoch": 2.2800000000000002, |
| "eval_biology_entropy": 0.5819249742627144, |
| "eval_biology_loss": 2.0149455070495605, |
| "eval_biology_mean_token_accuracy": 0.6538055400848388, |
| "eval_biology_num_tokens": 40564273.0, |
| "eval_biology_runtime": 39.9098, |
| "eval_biology_samples_per_second": 12.528, |
| "eval_biology_steps_per_second": 12.528, |
| "step": 5700 |
| }, |
| { |
| "epoch": 2.2800000000000002, |
| "eval_math_entropy": 0.45950557124614716, |
| "eval_math_loss": 0.7581589221954346, |
| "eval_math_mean_token_accuracy": 0.8132617511749267, |
| "eval_math_num_tokens": 40564273.0, |
| "eval_math_runtime": 47.8602, |
| "eval_math_samples_per_second": 10.447, |
| "eval_math_steps_per_second": 10.447, |
| "step": 5700 |
| }, |
| { |
| "entropy": 0.4103171478956938, |
| "epoch": 2.284, |
| "grad_norm": 47.75, |
| "learning_rate": 9.535555555555557e-06, |
| "loss": 6.4813, |
| "mean_token_accuracy": 0.8742847621440888, |
| "num_tokens": 40633728.0, |
| "step": 5710 |
| }, |
| { |
| "entropy": 0.44584370190277695, |
| "epoch": 2.288, |
| "grad_norm": 52.5, |
| "learning_rate": 9.513333333333334e-06, |
| "loss": 7.1733, |
| "mean_token_accuracy": 0.8652041930705309, |
| "num_tokens": 40702393.0, |
| "step": 5720 |
| }, |
| { |
| "entropy": 0.4124108956195414, |
| "epoch": 2.292, |
| "grad_norm": 46.75, |
| "learning_rate": 9.491111111111112e-06, |
| "loss": 6.5962, |
| "mean_token_accuracy": 0.8741286195814609, |
| "num_tokens": 40776862.0, |
| "step": 5730 |
| }, |
| { |
| "entropy": 0.39886171496473255, |
| "epoch": 2.296, |
| "grad_norm": 45.25, |
| "learning_rate": 9.46888888888889e-06, |
| "loss": 6.3687, |
| "mean_token_accuracy": 0.8789331752806901, |
| "num_tokens": 40848688.0, |
| "step": 5740 |
| }, |
| { |
| "entropy": 0.40309340143576267, |
| "epoch": 2.3, |
| "grad_norm": 44.25, |
| "learning_rate": 9.446666666666667e-06, |
| "loss": 6.4136, |
| "mean_token_accuracy": 0.8775001037865877, |
| "num_tokens": 40920604.0, |
| "step": 5750 |
| }, |
| { |
| "entropy": 0.4116678783670068, |
| "epoch": 2.304, |
| "grad_norm": 51.25, |
| "learning_rate": 9.424444444444445e-06, |
| "loss": 6.6682, |
| "mean_token_accuracy": 0.8723521053791046, |
| "num_tokens": 40982775.0, |
| "step": 5760 |
| }, |
| { |
| "entropy": 0.45275690061971546, |
| "epoch": 2.308, |
| "grad_norm": 50.25, |
| "learning_rate": 9.402222222222222e-06, |
| "loss": 7.247, |
| "mean_token_accuracy": 0.8651459068059921, |
| "num_tokens": 41053395.0, |
| "step": 5770 |
| }, |
| { |
| "entropy": 0.4326243043411523, |
| "epoch": 2.312, |
| "grad_norm": 57.25, |
| "learning_rate": 9.38e-06, |
| "loss": 6.8593, |
| "mean_token_accuracy": 0.8701514005661011, |
| "num_tokens": 41121934.0, |
| "step": 5780 |
| }, |
| { |
| "entropy": 0.419553127232939, |
| "epoch": 2.316, |
| "grad_norm": 45.0, |
| "learning_rate": 9.35777777777778e-06, |
| "loss": 6.6708, |
| "mean_token_accuracy": 0.8724352307617664, |
| "num_tokens": 41194023.0, |
| "step": 5790 |
| }, |
| { |
| "entropy": 0.4197757411748171, |
| "epoch": 2.32, |
| "grad_norm": 50.25, |
| "learning_rate": 9.335555555555557e-06, |
| "loss": 6.7908, |
| "mean_token_accuracy": 0.8714965227991343, |
| "num_tokens": 41263356.0, |
| "step": 5800 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_biology_entropy": 0.5869233113527298, |
| "eval_biology_loss": 2.0093045234680176, |
| "eval_biology_mean_token_accuracy": 0.6542842233181, |
| "eval_biology_num_tokens": 41263356.0, |
| "eval_biology_runtime": 39.8661, |
| "eval_biology_samples_per_second": 12.542, |
| "eval_biology_steps_per_second": 12.542, |
| "step": 5800 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_math_entropy": 0.46763592533767223, |
| "eval_math_loss": 0.7556903958320618, |
| "eval_math_mean_token_accuracy": 0.8134210889339447, |
| "eval_math_num_tokens": 41263356.0, |
| "eval_math_runtime": 47.7848, |
| "eval_math_samples_per_second": 10.464, |
| "eval_math_steps_per_second": 10.464, |
| "step": 5800 |
| }, |
| { |
| "entropy": 0.3964392309077084, |
| "epoch": 2.324, |
| "grad_norm": 50.5, |
| "learning_rate": 9.313333333333335e-06, |
| "loss": 6.4021, |
| "mean_token_accuracy": 0.878456924110651, |
| "num_tokens": 41331119.0, |
| "step": 5810 |
| }, |
| { |
| "entropy": 0.43976585315540434, |
| "epoch": 2.328, |
| "grad_norm": 45.0, |
| "learning_rate": 9.291111111111112e-06, |
| "loss": 6.8545, |
| "mean_token_accuracy": 0.8717822723090649, |
| "num_tokens": 41401200.0, |
| "step": 5820 |
| }, |
| { |
| "entropy": 0.4158212966285646, |
| "epoch": 2.332, |
| "grad_norm": 48.0, |
| "learning_rate": 9.26888888888889e-06, |
| "loss": 6.6812, |
| "mean_token_accuracy": 0.8741118922829628, |
| "num_tokens": 41473604.0, |
| "step": 5830 |
| }, |
| { |
| "entropy": 0.4344927713274956, |
| "epoch": 2.336, |
| "grad_norm": 55.5, |
| "learning_rate": 9.246666666666667e-06, |
| "loss": 6.9947, |
| "mean_token_accuracy": 0.8661442644894123, |
| "num_tokens": 41545235.0, |
| "step": 5840 |
| }, |
| { |
| "entropy": 0.41850304855033754, |
| "epoch": 2.34, |
| "grad_norm": 48.5, |
| "learning_rate": 9.224444444444445e-06, |
| "loss": 6.6379, |
| "mean_token_accuracy": 0.8751402664929628, |
| "num_tokens": 41615505.0, |
| "step": 5850 |
| }, |
| { |
| "entropy": 0.4300383645109832, |
| "epoch": 2.344, |
| "grad_norm": 45.0, |
| "learning_rate": 9.202222222222224e-06, |
| "loss": 6.9447, |
| "mean_token_accuracy": 0.8679604087024927, |
| "num_tokens": 41689311.0, |
| "step": 5860 |
| }, |
| { |
| "entropy": 0.4523826284334064, |
| "epoch": 2.348, |
| "grad_norm": 56.5, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 7.2144, |
| "mean_token_accuracy": 0.8654012832790613, |
| "num_tokens": 41759609.0, |
| "step": 5870 |
| }, |
| { |
| "entropy": 0.44451016243547203, |
| "epoch": 2.352, |
| "grad_norm": 55.25, |
| "learning_rate": 9.15777777777778e-06, |
| "loss": 7.0684, |
| "mean_token_accuracy": 0.8669733438640833, |
| "num_tokens": 41833417.0, |
| "step": 5880 |
| }, |
| { |
| "entropy": 0.39929521968588233, |
| "epoch": 2.356, |
| "grad_norm": 52.75, |
| "learning_rate": 9.135555555555557e-06, |
| "loss": 6.427, |
| "mean_token_accuracy": 0.8776704370975494, |
| "num_tokens": 41899020.0, |
| "step": 5890 |
| }, |
| { |
| "entropy": 0.4467870525550097, |
| "epoch": 2.36, |
| "grad_norm": 53.0, |
| "learning_rate": 9.113333333333335e-06, |
| "loss": 7.2367, |
| "mean_token_accuracy": 0.863744517788291, |
| "num_tokens": 41974532.0, |
| "step": 5900 |
| }, |
| { |
| "epoch": 2.36, |
| "eval_biology_entropy": 0.6006575649976731, |
| "eval_biology_loss": 1.9817614555358887, |
| "eval_biology_mean_token_accuracy": 0.6542308796048164, |
| "eval_biology_num_tokens": 41974532.0, |
| "eval_biology_runtime": 40.0403, |
| "eval_biology_samples_per_second": 12.487, |
| "eval_biology_steps_per_second": 12.487, |
| "step": 5900 |
| }, |
| { |
| "epoch": 2.36, |
| "eval_math_entropy": 0.4754800496697426, |
| "eval_math_loss": 0.7414502501487732, |
| "eval_math_mean_token_accuracy": 0.814188836812973, |
| "eval_math_num_tokens": 41974532.0, |
| "eval_math_runtime": 47.8885, |
| "eval_math_samples_per_second": 10.441, |
| "eval_math_steps_per_second": 10.441, |
| "step": 5900 |
| }, |
| { |
| "entropy": 0.4196365540847182, |
| "epoch": 2.364, |
| "grad_norm": 68.5, |
| "learning_rate": 9.091111111111112e-06, |
| "loss": 6.583, |
| "mean_token_accuracy": 0.8739133566617966, |
| "num_tokens": 42044951.0, |
| "step": 5910 |
| }, |
| { |
| "entropy": 0.4233891054056585, |
| "epoch": 2.368, |
| "grad_norm": 55.5, |
| "learning_rate": 9.06888888888889e-06, |
| "loss": 7.0078, |
| "mean_token_accuracy": 0.8674500647932291, |
| "num_tokens": 42117030.0, |
| "step": 5920 |
| }, |
| { |
| "entropy": 0.446973239723593, |
| "epoch": 2.372, |
| "grad_norm": 45.0, |
| "learning_rate": 9.046666666666668e-06, |
| "loss": 6.9911, |
| "mean_token_accuracy": 0.8701009515672922, |
| "num_tokens": 42192688.0, |
| "step": 5930 |
| }, |
| { |
| "entropy": 0.43473182059824467, |
| "epoch": 2.376, |
| "grad_norm": 49.25, |
| "learning_rate": 9.024444444444445e-06, |
| "loss": 7.0587, |
| "mean_token_accuracy": 0.8688194397836924, |
| "num_tokens": 42267341.0, |
| "step": 5940 |
| }, |
| { |
| "entropy": 0.4359048457816243, |
| "epoch": 2.38, |
| "grad_norm": 51.25, |
| "learning_rate": 9.002222222222223e-06, |
| "loss": 7.08, |
| "mean_token_accuracy": 0.8660250119864941, |
| "num_tokens": 42340623.0, |
| "step": 5950 |
| }, |
| { |
| "entropy": 0.42556518195196985, |
| "epoch": 2.384, |
| "grad_norm": 50.0, |
| "learning_rate": 8.98e-06, |
| "loss": 6.7953, |
| "mean_token_accuracy": 0.8733339115977288, |
| "num_tokens": 42410990.0, |
| "step": 5960 |
| }, |
| { |
| "entropy": 0.4303522454574704, |
| "epoch": 2.388, |
| "grad_norm": 58.25, |
| "learning_rate": 8.957777777777778e-06, |
| "loss": 6.8711, |
| "mean_token_accuracy": 0.8718335065990687, |
| "num_tokens": 42480569.0, |
| "step": 5970 |
| }, |
| { |
| "entropy": 0.4231502369046211, |
| "epoch": 2.392, |
| "grad_norm": 46.5, |
| "learning_rate": 8.935555555555556e-06, |
| "loss": 6.7364, |
| "mean_token_accuracy": 0.8731079772114754, |
| "num_tokens": 42552892.0, |
| "step": 5980 |
| }, |
| { |
| "entropy": 0.4250686950981617, |
| "epoch": 2.396, |
| "grad_norm": 47.0, |
| "learning_rate": 8.913333333333333e-06, |
| "loss": 6.7386, |
| "mean_token_accuracy": 0.8727667387574911, |
| "num_tokens": 42624560.0, |
| "step": 5990 |
| }, |
| { |
| "entropy": 0.4225637829862535, |
| "epoch": 2.4, |
| "grad_norm": 65.0, |
| "learning_rate": 8.891111111111111e-06, |
| "loss": 6.6958, |
| "mean_token_accuracy": 0.8730304487049579, |
| "num_tokens": 42691890.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 0.575371405005455, |
| "eval_biology_loss": 2.044929027557373, |
| "eval_biology_mean_token_accuracy": 0.6542533279061318, |
| "eval_biology_num_tokens": 42691890.0, |
| "eval_biology_runtime": 40.0232, |
| "eval_biology_samples_per_second": 12.493, |
| "eval_biology_steps_per_second": 12.493, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.4633637537062168, |
| "eval_math_loss": 0.7558966875076294, |
| "eval_math_mean_token_accuracy": 0.8143890473842621, |
| "eval_math_num_tokens": 42691890.0, |
| "eval_math_runtime": 49.341, |
| "eval_math_samples_per_second": 10.134, |
| "eval_math_steps_per_second": 10.134, |
| "step": 6000 |
| }, |
| { |
| "entropy": 0.42499869139865043, |
| "epoch": 2.404, |
| "grad_norm": 51.75, |
| "learning_rate": 8.868888888888888e-06, |
| "loss": 6.7462, |
| "mean_token_accuracy": 0.8716801207512617, |
| "num_tokens": 42761561.0, |
| "step": 6010 |
| }, |
| { |
| "entropy": 0.45049548922106625, |
| "epoch": 2.408, |
| "grad_norm": 54.5, |
| "learning_rate": 8.846666666666668e-06, |
| "loss": 7.208, |
| "mean_token_accuracy": 0.8635994680225849, |
| "num_tokens": 42830910.0, |
| "step": 6020 |
| }, |
| { |
| "entropy": 0.4525366496760398, |
| "epoch": 2.412, |
| "grad_norm": 50.0, |
| "learning_rate": 8.824444444444445e-06, |
| "loss": 7.3573, |
| "mean_token_accuracy": 0.8630804970860482, |
| "num_tokens": 42904236.0, |
| "step": 6030 |
| }, |
| { |
| "entropy": 0.42418601056560873, |
| "epoch": 2.416, |
| "grad_norm": 55.75, |
| "learning_rate": 8.802222222222223e-06, |
| "loss": 6.7437, |
| "mean_token_accuracy": 0.8715368177741766, |
| "num_tokens": 42971588.0, |
| "step": 6040 |
| }, |
| { |
| "entropy": 0.4163145710714161, |
| "epoch": 2.42, |
| "grad_norm": 49.75, |
| "learning_rate": 8.78e-06, |
| "loss": 6.6546, |
| "mean_token_accuracy": 0.8735284950584173, |
| "num_tokens": 43040919.0, |
| "step": 6050 |
| }, |
| { |
| "entropy": 0.4353508242405951, |
| "epoch": 2.424, |
| "grad_norm": 42.0, |
| "learning_rate": 8.757777777777778e-06, |
| "loss": 7.0011, |
| "mean_token_accuracy": 0.8691820878535509, |
| "num_tokens": 43111006.0, |
| "step": 6060 |
| }, |
| { |
| "entropy": 0.4030357056297362, |
| "epoch": 2.428, |
| "grad_norm": 51.25, |
| "learning_rate": 8.735555555555556e-06, |
| "loss": 6.4415, |
| "mean_token_accuracy": 0.8770109038800001, |
| "num_tokens": 43182076.0, |
| "step": 6070 |
| }, |
| { |
| "entropy": 0.4154713830910623, |
| "epoch": 2.432, |
| "grad_norm": 47.0, |
| "learning_rate": 8.713333333333333e-06, |
| "loss": 6.6295, |
| "mean_token_accuracy": 0.8738619331270456, |
| "num_tokens": 43253821.0, |
| "step": 6080 |
| }, |
| { |
| "entropy": 0.40787449646741153, |
| "epoch": 2.436, |
| "grad_norm": 48.75, |
| "learning_rate": 8.691111111111111e-06, |
| "loss": 6.411, |
| "mean_token_accuracy": 0.8778625573962927, |
| "num_tokens": 43330875.0, |
| "step": 6090 |
| }, |
| { |
| "entropy": 0.40236510494723915, |
| "epoch": 2.44, |
| "grad_norm": 49.5, |
| "learning_rate": 8.66888888888889e-06, |
| "loss": 6.5016, |
| "mean_token_accuracy": 0.8759409468621016, |
| "num_tokens": 43404210.0, |
| "step": 6100 |
| }, |
| { |
| "epoch": 2.44, |
| "eval_biology_entropy": 0.5770821847915649, |
| "eval_biology_loss": 2.047761917114258, |
| "eval_biology_mean_token_accuracy": 0.6530932180285454, |
| "eval_biology_num_tokens": 43404210.0, |
| "eval_biology_runtime": 40.318, |
| "eval_biology_samples_per_second": 12.401, |
| "eval_biology_steps_per_second": 12.401, |
| "step": 6100 |
| }, |
| { |
| "epoch": 2.44, |
| "eval_math_entropy": 0.46366557210683823, |
| "eval_math_loss": 0.7578248381614685, |
| "eval_math_mean_token_accuracy": 0.8139072825908661, |
| "eval_math_num_tokens": 43404210.0, |
| "eval_math_runtime": 47.9825, |
| "eval_math_samples_per_second": 10.42, |
| "eval_math_steps_per_second": 10.42, |
| "step": 6100 |
| }, |
| { |
| "entropy": 0.4569242998957634, |
| "epoch": 2.444, |
| "grad_norm": 43.0, |
| "learning_rate": 8.646666666666668e-06, |
| "loss": 7.3247, |
| "mean_token_accuracy": 0.8623149659484625, |
| "num_tokens": 43476262.0, |
| "step": 6110 |
| }, |
| { |
| "entropy": 0.43789711138233545, |
| "epoch": 2.448, |
| "grad_norm": 69.0, |
| "learning_rate": 8.624444444444446e-06, |
| "loss": 7.0216, |
| "mean_token_accuracy": 0.8687020387500525, |
| "num_tokens": 43550902.0, |
| "step": 6120 |
| }, |
| { |
| "entropy": 0.4508271205238998, |
| "epoch": 2.452, |
| "grad_norm": 44.0, |
| "learning_rate": 8.602222222222223e-06, |
| "loss": 7.1292, |
| "mean_token_accuracy": 0.86444644741714, |
| "num_tokens": 43627440.0, |
| "step": 6130 |
| }, |
| { |
| "entropy": 0.4288553704507649, |
| "epoch": 2.456, |
| "grad_norm": 40.0, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 6.8878, |
| "mean_token_accuracy": 0.871123394370079, |
| "num_tokens": 43697363.0, |
| "step": 6140 |
| }, |
| { |
| "entropy": 0.4387895753607154, |
| "epoch": 2.46, |
| "grad_norm": 42.25, |
| "learning_rate": 8.557777777777778e-06, |
| "loss": 6.9059, |
| "mean_token_accuracy": 0.869594781473279, |
| "num_tokens": 43772857.0, |
| "step": 6150 |
| }, |
| { |
| "entropy": 0.42130946470424535, |
| "epoch": 2.464, |
| "grad_norm": 55.0, |
| "learning_rate": 8.535555555555556e-06, |
| "loss": 6.9002, |
| "mean_token_accuracy": 0.870204196497798, |
| "num_tokens": 43844259.0, |
| "step": 6160 |
| }, |
| { |
| "entropy": 0.4506405842490494, |
| "epoch": 2.468, |
| "grad_norm": 53.75, |
| "learning_rate": 8.513333333333335e-06, |
| "loss": 6.9993, |
| "mean_token_accuracy": 0.8688790615648031, |
| "num_tokens": 43909390.0, |
| "step": 6170 |
| }, |
| { |
| "entropy": 0.4409319592639804, |
| "epoch": 2.472, |
| "grad_norm": 54.75, |
| "learning_rate": 8.491111111111113e-06, |
| "loss": 7.1326, |
| "mean_token_accuracy": 0.867237013950944, |
| "num_tokens": 43980336.0, |
| "step": 6180 |
| }, |
| { |
| "entropy": 0.42059532226994634, |
| "epoch": 2.476, |
| "grad_norm": 43.75, |
| "learning_rate": 8.46888888888889e-06, |
| "loss": 6.6837, |
| "mean_token_accuracy": 0.8718780767172575, |
| "num_tokens": 44046412.0, |
| "step": 6190 |
| }, |
| { |
| "entropy": 0.41430794666521253, |
| "epoch": 2.48, |
| "grad_norm": 58.0, |
| "learning_rate": 8.446666666666668e-06, |
| "loss": 6.7082, |
| "mean_token_accuracy": 0.872522134706378, |
| "num_tokens": 44115701.0, |
| "step": 6200 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_biology_entropy": 0.5884713095426559, |
| "eval_biology_loss": 2.013845443725586, |
| "eval_biology_mean_token_accuracy": 0.653822160243988, |
| "eval_biology_num_tokens": 44115701.0, |
| "eval_biology_runtime": 40.4714, |
| "eval_biology_samples_per_second": 12.354, |
| "eval_biology_steps_per_second": 12.354, |
| "step": 6200 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_math_entropy": 0.4692519321739674, |
| "eval_math_loss": 0.7483494877815247, |
| "eval_math_mean_token_accuracy": 0.8143787887096405, |
| "eval_math_num_tokens": 44115701.0, |
| "eval_math_runtime": 47.9056, |
| "eval_math_samples_per_second": 10.437, |
| "eval_math_steps_per_second": 10.437, |
| "step": 6200 |
| }, |
| { |
| "entropy": 0.4241214304696769, |
| "epoch": 2.484, |
| "grad_norm": 57.0, |
| "learning_rate": 8.424444444444446e-06, |
| "loss": 6.6723, |
| "mean_token_accuracy": 0.8724103335291147, |
| "num_tokens": 44186838.0, |
| "step": 6210 |
| }, |
| { |
| "entropy": 0.4078469776548445, |
| "epoch": 2.488, |
| "grad_norm": 54.25, |
| "learning_rate": 8.402222222222223e-06, |
| "loss": 6.6715, |
| "mean_token_accuracy": 0.873164689168334, |
| "num_tokens": 44259716.0, |
| "step": 6220 |
| }, |
| { |
| "entropy": 0.45215398278087376, |
| "epoch": 2.492, |
| "grad_norm": 54.75, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 7.1615, |
| "mean_token_accuracy": 0.8672402109950781, |
| "num_tokens": 44332420.0, |
| "step": 6230 |
| }, |
| { |
| "entropy": 0.4097982753068209, |
| "epoch": 2.496, |
| "grad_norm": 59.5, |
| "learning_rate": 8.357777777777779e-06, |
| "loss": 6.7823, |
| "mean_token_accuracy": 0.8722094774246216, |
| "num_tokens": 44405520.0, |
| "step": 6240 |
| }, |
| { |
| "entropy": 0.42705487716011703, |
| "epoch": 2.5, |
| "grad_norm": 46.0, |
| "learning_rate": 8.335555555555556e-06, |
| "loss": 6.739, |
| "mean_token_accuracy": 0.8730598337948322, |
| "num_tokens": 44474272.0, |
| "step": 6250 |
| }, |
| { |
| "entropy": 0.449821054097265, |
| "epoch": 2.504, |
| "grad_norm": 59.25, |
| "learning_rate": 8.313333333333334e-06, |
| "loss": 7.1814, |
| "mean_token_accuracy": 0.8647784728556871, |
| "num_tokens": 44546843.0, |
| "step": 6260 |
| }, |
| { |
| "entropy": 0.41240001767873763, |
| "epoch": 2.508, |
| "grad_norm": 52.25, |
| "learning_rate": 8.291111111111112e-06, |
| "loss": 6.6114, |
| "mean_token_accuracy": 0.8742719057947397, |
| "num_tokens": 44617788.0, |
| "step": 6270 |
| }, |
| { |
| "entropy": 0.42788790753111244, |
| "epoch": 2.512, |
| "grad_norm": 49.0, |
| "learning_rate": 8.268888888888889e-06, |
| "loss": 6.8232, |
| "mean_token_accuracy": 0.8698410071432591, |
| "num_tokens": 44686477.0, |
| "step": 6280 |
| }, |
| { |
| "entropy": 0.3856860482133925, |
| "epoch": 2.516, |
| "grad_norm": 48.0, |
| "learning_rate": 8.246666666666667e-06, |
| "loss": 6.1045, |
| "mean_token_accuracy": 0.8834345221519471, |
| "num_tokens": 44755652.0, |
| "step": 6290 |
| }, |
| { |
| "entropy": 0.4138589259237051, |
| "epoch": 2.52, |
| "grad_norm": 42.5, |
| "learning_rate": 8.224444444444444e-06, |
| "loss": 6.6538, |
| "mean_token_accuracy": 0.8740550264716148, |
| "num_tokens": 44825888.0, |
| "step": 6300 |
| }, |
| { |
| "epoch": 2.52, |
| "eval_biology_entropy": 0.6034090686440468, |
| "eval_biology_loss": 1.9779011011123657, |
| "eval_biology_mean_token_accuracy": 0.6545934562683106, |
| "eval_biology_num_tokens": 44825888.0, |
| "eval_biology_runtime": 41.4535, |
| "eval_biology_samples_per_second": 12.062, |
| "eval_biology_steps_per_second": 12.062, |
| "step": 6300 |
| }, |
| { |
| "epoch": 2.52, |
| "eval_math_entropy": 0.47658546568453314, |
| "eval_math_loss": 0.7405884861946106, |
| "eval_math_mean_token_accuracy": 0.8148692543506623, |
| "eval_math_num_tokens": 44825888.0, |
| "eval_math_runtime": 48.363, |
| "eval_math_samples_per_second": 10.338, |
| "eval_math_steps_per_second": 10.338, |
| "step": 6300 |
| }, |
| { |
| "entropy": 0.4453656492754817, |
| "epoch": 2.524, |
| "grad_norm": 65.5, |
| "learning_rate": 8.202222222222222e-06, |
| "loss": 7.1146, |
| "mean_token_accuracy": 0.8656911112368106, |
| "num_tokens": 44896046.0, |
| "step": 6310 |
| }, |
| { |
| "entropy": 0.44410878997296094, |
| "epoch": 2.528, |
| "grad_norm": 43.75, |
| "learning_rate": 8.18e-06, |
| "loss": 7.0397, |
| "mean_token_accuracy": 0.8671455517411232, |
| "num_tokens": 44969760.0, |
| "step": 6320 |
| }, |
| { |
| "entropy": 0.425704933796078, |
| "epoch": 2.532, |
| "grad_norm": 49.0, |
| "learning_rate": 8.157777777777779e-06, |
| "loss": 6.741, |
| "mean_token_accuracy": 0.87345957942307, |
| "num_tokens": 45036721.0, |
| "step": 6330 |
| }, |
| { |
| "entropy": 0.43970570946112275, |
| "epoch": 2.536, |
| "grad_norm": 52.5, |
| "learning_rate": 8.135555555555557e-06, |
| "loss": 7.1269, |
| "mean_token_accuracy": 0.8662078116089106, |
| "num_tokens": 45114115.0, |
| "step": 6340 |
| }, |
| { |
| "entropy": 0.4180485351011157, |
| "epoch": 2.54, |
| "grad_norm": 52.75, |
| "learning_rate": 8.113333333333334e-06, |
| "loss": 6.6542, |
| "mean_token_accuracy": 0.8732343025505542, |
| "num_tokens": 45186485.0, |
| "step": 6350 |
| }, |
| { |
| "entropy": 0.3982725726440549, |
| "epoch": 2.544, |
| "grad_norm": 62.5, |
| "learning_rate": 8.091111111111112e-06, |
| "loss": 6.3922, |
| "mean_token_accuracy": 0.8768733851611614, |
| "num_tokens": 45255326.0, |
| "step": 6360 |
| }, |
| { |
| "entropy": 0.412839618511498, |
| "epoch": 2.548, |
| "grad_norm": 51.75, |
| "learning_rate": 8.06888888888889e-06, |
| "loss": 6.5843, |
| "mean_token_accuracy": 0.874532961845398, |
| "num_tokens": 45319961.0, |
| "step": 6370 |
| }, |
| { |
| "entropy": 0.42857724605128167, |
| "epoch": 2.552, |
| "grad_norm": 44.0, |
| "learning_rate": 8.046666666666667e-06, |
| "loss": 6.8479, |
| "mean_token_accuracy": 0.8690481420606375, |
| "num_tokens": 45388515.0, |
| "step": 6380 |
| }, |
| { |
| "entropy": 0.4207867604680359, |
| "epoch": 2.556, |
| "grad_norm": 60.25, |
| "learning_rate": 8.024444444444445e-06, |
| "loss": 6.6898, |
| "mean_token_accuracy": 0.8720209904015064, |
| "num_tokens": 45458722.0, |
| "step": 6390 |
| }, |
| { |
| "entropy": 0.4003525773063302, |
| "epoch": 2.56, |
| "grad_norm": 46.25, |
| "learning_rate": 8.002222222222222e-06, |
| "loss": 6.5451, |
| "mean_token_accuracy": 0.8745481837540865, |
| "num_tokens": 45532525.0, |
| "step": 6400 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 0.5866417605280876, |
| "eval_biology_loss": 2.0179808139801025, |
| "eval_biology_mean_token_accuracy": 0.6536050434112549, |
| "eval_biology_num_tokens": 45532525.0, |
| "eval_biology_runtime": 43.0353, |
| "eval_biology_samples_per_second": 11.618, |
| "eval_biology_steps_per_second": 11.618, |
| "step": 6400 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.46750668485462665, |
| "eval_math_loss": 0.7480150461196899, |
| "eval_math_mean_token_accuracy": 0.8147590626478195, |
| "eval_math_num_tokens": 45532525.0, |
| "eval_math_runtime": 49.1095, |
| "eval_math_samples_per_second": 10.181, |
| "eval_math_steps_per_second": 10.181, |
| "step": 6400 |
| }, |
| { |
| "entropy": 0.3977276818826795, |
| "epoch": 2.564, |
| "grad_norm": 50.75, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 6.2796, |
| "mean_token_accuracy": 0.8802063751965761, |
| "num_tokens": 45604892.0, |
| "step": 6410 |
| }, |
| { |
| "entropy": 0.4499839426018298, |
| "epoch": 2.568, |
| "grad_norm": 40.0, |
| "learning_rate": 7.957777777777779e-06, |
| "loss": 7.2953, |
| "mean_token_accuracy": 0.8636733632534742, |
| "num_tokens": 45677949.0, |
| "step": 6420 |
| }, |
| { |
| "entropy": 0.430719330534339, |
| "epoch": 2.572, |
| "grad_norm": 60.0, |
| "learning_rate": 7.935555555555557e-06, |
| "loss": 6.7024, |
| "mean_token_accuracy": 0.8731723293662071, |
| "num_tokens": 45746010.0, |
| "step": 6430 |
| }, |
| { |
| "entropy": 0.4180364768952131, |
| "epoch": 2.576, |
| "grad_norm": 54.75, |
| "learning_rate": 7.913333333333334e-06, |
| "loss": 6.8925, |
| "mean_token_accuracy": 0.8709745384752751, |
| "num_tokens": 45817478.0, |
| "step": 6440 |
| }, |
| { |
| "entropy": 0.4299122693017125, |
| "epoch": 2.58, |
| "grad_norm": 47.25, |
| "learning_rate": 7.891111111111112e-06, |
| "loss": 6.7983, |
| "mean_token_accuracy": 0.8713275451213122, |
| "num_tokens": 45888559.0, |
| "step": 6450 |
| }, |
| { |
| "entropy": 0.413304222561419, |
| "epoch": 2.584, |
| "grad_norm": 40.75, |
| "learning_rate": 7.86888888888889e-06, |
| "loss": 6.7049, |
| "mean_token_accuracy": 0.8721549216657877, |
| "num_tokens": 45963170.0, |
| "step": 6460 |
| }, |
| { |
| "entropy": 0.41771556427702305, |
| "epoch": 2.588, |
| "grad_norm": 49.0, |
| "learning_rate": 7.846666666666667e-06, |
| "loss": 6.5152, |
| "mean_token_accuracy": 0.8753284107893705, |
| "num_tokens": 46035817.0, |
| "step": 6470 |
| }, |
| { |
| "entropy": 0.42811128138564525, |
| "epoch": 2.592, |
| "grad_norm": 37.75, |
| "learning_rate": 7.824444444444445e-06, |
| "loss": 7.056, |
| "mean_token_accuracy": 0.867966702952981, |
| "num_tokens": 46109575.0, |
| "step": 6480 |
| }, |
| { |
| "entropy": 0.42363819666206837, |
| "epoch": 2.596, |
| "grad_norm": 57.5, |
| "learning_rate": 7.802222222222222e-06, |
| "loss": 6.6463, |
| "mean_token_accuracy": 0.8737729638814926, |
| "num_tokens": 46182182.0, |
| "step": 6490 |
| }, |
| { |
| "entropy": 0.4127486293204129, |
| "epoch": 2.6, |
| "grad_norm": 60.75, |
| "learning_rate": 7.78e-06, |
| "loss": 6.6972, |
| "mean_token_accuracy": 0.8734816189855337, |
| "num_tokens": 46252455.0, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.6, |
| "eval_biology_entropy": 0.586870098054409, |
| "eval_biology_loss": 2.011416435241699, |
| "eval_biology_mean_token_accuracy": 0.6541973667144776, |
| "eval_biology_num_tokens": 46252455.0, |
| "eval_biology_runtime": 42.2284, |
| "eval_biology_samples_per_second": 11.84, |
| "eval_biology_steps_per_second": 11.84, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.6, |
| "eval_math_entropy": 0.4718260209560394, |
| "eval_math_loss": 0.7444885969161987, |
| "eval_math_mean_token_accuracy": 0.8148928849697114, |
| "eval_math_num_tokens": 46252455.0, |
| "eval_math_runtime": 49.4452, |
| "eval_math_samples_per_second": 10.112, |
| "eval_math_steps_per_second": 10.112, |
| "step": 6500 |
| }, |
| { |
| "entropy": 0.4371944203041494, |
| "epoch": 2.604, |
| "grad_norm": 49.0, |
| "learning_rate": 7.75777777777778e-06, |
| "loss": 6.8996, |
| "mean_token_accuracy": 0.871408598124981, |
| "num_tokens": 46324106.0, |
| "step": 6510 |
| }, |
| { |
| "entropy": 0.463454591948539, |
| "epoch": 2.608, |
| "grad_norm": 56.25, |
| "learning_rate": 7.735555555555557e-06, |
| "loss": 7.3462, |
| "mean_token_accuracy": 0.861482036113739, |
| "num_tokens": 46391461.0, |
| "step": 6520 |
| }, |
| { |
| "entropy": 0.45144426431506873, |
| "epoch": 2.612, |
| "grad_norm": 59.0, |
| "learning_rate": 7.713333333333335e-06, |
| "loss": 7.2085, |
| "mean_token_accuracy": 0.8657496828585863, |
| "num_tokens": 46468589.0, |
| "step": 6530 |
| }, |
| { |
| "entropy": 0.3921378825325519, |
| "epoch": 2.616, |
| "grad_norm": 47.0, |
| "learning_rate": 7.691111111111112e-06, |
| "loss": 6.2026, |
| "mean_token_accuracy": 0.8815145649015903, |
| "num_tokens": 46540541.0, |
| "step": 6540 |
| }, |
| { |
| "entropy": 0.38311725817620756, |
| "epoch": 2.62, |
| "grad_norm": 58.75, |
| "learning_rate": 7.66888888888889e-06, |
| "loss": 6.2397, |
| "mean_token_accuracy": 0.8799462229013443, |
| "num_tokens": 46611027.0, |
| "step": 6550 |
| }, |
| { |
| "entropy": 0.40485583515837786, |
| "epoch": 2.624, |
| "grad_norm": 46.5, |
| "learning_rate": 7.646666666666667e-06, |
| "loss": 6.4306, |
| "mean_token_accuracy": 0.8775864738970995, |
| "num_tokens": 46683117.0, |
| "step": 6560 |
| }, |
| { |
| "entropy": 0.4200917361304164, |
| "epoch": 2.628, |
| "grad_norm": 55.5, |
| "learning_rate": 7.624444444444445e-06, |
| "loss": 6.7371, |
| "mean_token_accuracy": 0.8720926944166422, |
| "num_tokens": 46753287.0, |
| "step": 6570 |
| }, |
| { |
| "entropy": 0.42967795189470054, |
| "epoch": 2.632, |
| "grad_norm": 52.25, |
| "learning_rate": 7.602222222222223e-06, |
| "loss": 6.8535, |
| "mean_token_accuracy": 0.871143440157175, |
| "num_tokens": 46822076.0, |
| "step": 6580 |
| }, |
| { |
| "entropy": 0.419647815823555, |
| "epoch": 2.636, |
| "grad_norm": 56.75, |
| "learning_rate": 7.58e-06, |
| "loss": 6.7994, |
| "mean_token_accuracy": 0.871767944842577, |
| "num_tokens": 46894811.0, |
| "step": 6590 |
| }, |
| { |
| "entropy": 0.41958459550514815, |
| "epoch": 2.64, |
| "grad_norm": 53.75, |
| "learning_rate": 7.557777777777779e-06, |
| "loss": 6.6745, |
| "mean_token_accuracy": 0.8726473856717348, |
| "num_tokens": 46965534.0, |
| "step": 6600 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_biology_entropy": 0.5868029145598411, |
| "eval_biology_loss": 2.013819932937622, |
| "eval_biology_mean_token_accuracy": 0.653265580356121, |
| "eval_biology_num_tokens": 46965534.0, |
| "eval_biology_runtime": 41.9509, |
| "eval_biology_samples_per_second": 11.919, |
| "eval_biology_steps_per_second": 11.919, |
| "step": 6600 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_math_entropy": 0.46807982882857324, |
| "eval_math_loss": 0.7483139634132385, |
| "eval_math_mean_token_accuracy": 0.8148642436265946, |
| "eval_math_num_tokens": 46965534.0, |
| "eval_math_runtime": 48.0919, |
| "eval_math_samples_per_second": 10.397, |
| "eval_math_steps_per_second": 10.397, |
| "step": 6600 |
| }, |
| { |
| "entropy": 0.4037216482684016, |
| "epoch": 2.644, |
| "grad_norm": 61.0, |
| "learning_rate": 7.535555555555556e-06, |
| "loss": 6.3633, |
| "mean_token_accuracy": 0.8756832357496023, |
| "num_tokens": 47031057.0, |
| "step": 6610 |
| }, |
| { |
| "entropy": 0.4063232786953449, |
| "epoch": 2.648, |
| "grad_norm": 50.25, |
| "learning_rate": 7.513333333333334e-06, |
| "loss": 6.4956, |
| "mean_token_accuracy": 0.876214187592268, |
| "num_tokens": 47102766.0, |
| "step": 6620 |
| }, |
| { |
| "entropy": 0.41445388714782894, |
| "epoch": 2.652, |
| "grad_norm": 52.5, |
| "learning_rate": 7.4911111111111116e-06, |
| "loss": 6.6369, |
| "mean_token_accuracy": 0.8721906170248985, |
| "num_tokens": 47176476.0, |
| "step": 6630 |
| }, |
| { |
| "entropy": 0.4423921186942607, |
| "epoch": 2.656, |
| "grad_norm": 52.75, |
| "learning_rate": 7.46888888888889e-06, |
| "loss": 7.114, |
| "mean_token_accuracy": 0.8636913981288672, |
| "num_tokens": 47245852.0, |
| "step": 6640 |
| }, |
| { |
| "entropy": 0.4251969835720956, |
| "epoch": 2.66, |
| "grad_norm": 43.75, |
| "learning_rate": 7.446666666666668e-06, |
| "loss": 6.7523, |
| "mean_token_accuracy": 0.8728545539081096, |
| "num_tokens": 47317663.0, |
| "step": 6650 |
| }, |
| { |
| "entropy": 0.41862577814608815, |
| "epoch": 2.664, |
| "grad_norm": 45.5, |
| "learning_rate": 7.424444444444445e-06, |
| "loss": 6.6372, |
| "mean_token_accuracy": 0.8741007048636675, |
| "num_tokens": 47389744.0, |
| "step": 6660 |
| }, |
| { |
| "entropy": 0.3969283942133188, |
| "epoch": 2.668, |
| "grad_norm": 55.75, |
| "learning_rate": 7.402222222222223e-06, |
| "loss": 6.4573, |
| "mean_token_accuracy": 0.8766754295676946, |
| "num_tokens": 47456518.0, |
| "step": 6670 |
| }, |
| { |
| "entropy": 0.45196999963372947, |
| "epoch": 2.672, |
| "grad_norm": 67.0, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 7.1528, |
| "mean_token_accuracy": 0.8635958023369312, |
| "num_tokens": 47524916.0, |
| "step": 6680 |
| }, |
| { |
| "entropy": 0.41666053170338274, |
| "epoch": 2.676, |
| "grad_norm": 72.5, |
| "learning_rate": 7.357777777777778e-06, |
| "loss": 6.7333, |
| "mean_token_accuracy": 0.8736396338790655, |
| "num_tokens": 47596080.0, |
| "step": 6690 |
| }, |
| { |
| "entropy": 0.4345975358970463, |
| "epoch": 2.68, |
| "grad_norm": 52.0, |
| "learning_rate": 7.335555555555556e-06, |
| "loss": 6.8677, |
| "mean_token_accuracy": 0.8691735249012709, |
| "num_tokens": 47665004.0, |
| "step": 6700 |
| }, |
| { |
| "epoch": 2.68, |
| "eval_biology_entropy": 0.5962486489415169, |
| "eval_biology_loss": 1.9968345165252686, |
| "eval_biology_mean_token_accuracy": 0.6534018060564994, |
| "eval_biology_num_tokens": 47665004.0, |
| "eval_biology_runtime": 42.21, |
| "eval_biology_samples_per_second": 11.846, |
| "eval_biology_steps_per_second": 11.846, |
| "step": 6700 |
| }, |
| { |
| "epoch": 2.68, |
| "eval_math_entropy": 0.4732825907766819, |
| "eval_math_loss": 0.7430461645126343, |
| "eval_math_mean_token_accuracy": 0.8152086834907531, |
| "eval_math_num_tokens": 47665004.0, |
| "eval_math_runtime": 48.4756, |
| "eval_math_samples_per_second": 10.314, |
| "eval_math_steps_per_second": 10.314, |
| "step": 6700 |
| }, |
| { |
| "entropy": 0.42849911991506817, |
| "epoch": 2.684, |
| "grad_norm": 51.25, |
| "learning_rate": 7.313333333333333e-06, |
| "loss": 6.9437, |
| "mean_token_accuracy": 0.869233100861311, |
| "num_tokens": 47735417.0, |
| "step": 6710 |
| }, |
| { |
| "entropy": 0.4111584144644439, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 48.25, |
| "learning_rate": 7.291111111111112e-06, |
| "loss": 6.6967, |
| "mean_token_accuracy": 0.872507344558835, |
| "num_tokens": 47807131.0, |
| "step": 6720 |
| }, |
| { |
| "entropy": 0.4163995720911771, |
| "epoch": 2.692, |
| "grad_norm": 48.5, |
| "learning_rate": 7.268888888888889e-06, |
| "loss": 6.4657, |
| "mean_token_accuracy": 0.8772596966475248, |
| "num_tokens": 47878388.0, |
| "step": 6730 |
| }, |
| { |
| "entropy": 0.38970125149935486, |
| "epoch": 2.6959999999999997, |
| "grad_norm": 59.0, |
| "learning_rate": 7.246666666666667e-06, |
| "loss": 6.2677, |
| "mean_token_accuracy": 0.8803575787693262, |
| "num_tokens": 47950106.0, |
| "step": 6740 |
| }, |
| { |
| "entropy": 0.4708751678932458, |
| "epoch": 2.7, |
| "grad_norm": 56.5, |
| "learning_rate": 7.224444444444445e-06, |
| "loss": 7.5372, |
| "mean_token_accuracy": 0.8596292629837989, |
| "num_tokens": 48027671.0, |
| "step": 6750 |
| }, |
| { |
| "entropy": 0.4660518255084753, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 51.25, |
| "learning_rate": 7.202222222222222e-06, |
| "loss": 7.5225, |
| "mean_token_accuracy": 0.8585685279220343, |
| "num_tokens": 48099654.0, |
| "step": 6760 |
| }, |
| { |
| "entropy": 0.4261979955248535, |
| "epoch": 2.708, |
| "grad_norm": 43.0, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 6.7521, |
| "mean_token_accuracy": 0.8706818737089634, |
| "num_tokens": 48167542.0, |
| "step": 6770 |
| }, |
| { |
| "entropy": 0.43399141170084476, |
| "epoch": 2.7119999999999997, |
| "grad_norm": 60.0, |
| "learning_rate": 7.157777777777778e-06, |
| "loss": 7.0452, |
| "mean_token_accuracy": 0.8658308994024992, |
| "num_tokens": 48238121.0, |
| "step": 6780 |
| }, |
| { |
| "entropy": 0.40664876801893113, |
| "epoch": 2.716, |
| "grad_norm": 44.75, |
| "learning_rate": 7.135555555555557e-06, |
| "loss": 6.4772, |
| "mean_token_accuracy": 0.8766309142112731, |
| "num_tokens": 48309383.0, |
| "step": 6790 |
| }, |
| { |
| "entropy": 0.37334191938862205, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 49.75, |
| "learning_rate": 7.113333333333334e-06, |
| "loss": 5.9748, |
| "mean_token_accuracy": 0.8858205169439316, |
| "num_tokens": 48375019.0, |
| "step": 6800 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 0.5746917700767518, |
| "eval_biology_loss": 2.0361363887786865, |
| "eval_biology_mean_token_accuracy": 0.6539033913612365, |
| "eval_biology_num_tokens": 48375019.0, |
| "eval_biology_runtime": 40.4639, |
| "eval_biology_samples_per_second": 12.357, |
| "eval_biology_steps_per_second": 12.357, |
| "step": 6800 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.4615035522431135, |
| "eval_math_loss": 0.7516216039657593, |
| "eval_math_mean_token_accuracy": 0.8153433845043182, |
| "eval_math_num_tokens": 48375019.0, |
| "eval_math_runtime": 48.0038, |
| "eval_math_samples_per_second": 10.416, |
| "eval_math_steps_per_second": 10.416, |
| "step": 6800 |
| }, |
| { |
| "entropy": 0.42069728681817653, |
| "epoch": 2.724, |
| "grad_norm": 53.75, |
| "learning_rate": 7.091111111111112e-06, |
| "loss": 6.6423, |
| "mean_token_accuracy": 0.8729877796024084, |
| "num_tokens": 48445937.0, |
| "step": 6810 |
| }, |
| { |
| "entropy": 0.4233226933516562, |
| "epoch": 2.7279999999999998, |
| "grad_norm": 51.0, |
| "learning_rate": 7.06888888888889e-06, |
| "loss": 6.7264, |
| "mean_token_accuracy": 0.8710582949221134, |
| "num_tokens": 48517409.0, |
| "step": 6820 |
| }, |
| { |
| "entropy": 0.41732067372649906, |
| "epoch": 2.732, |
| "grad_norm": 49.25, |
| "learning_rate": 7.046666666666667e-06, |
| "loss": 6.7369, |
| "mean_token_accuracy": 0.8743715040385723, |
| "num_tokens": 48588525.0, |
| "step": 6830 |
| }, |
| { |
| "entropy": 0.41687058927491305, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 55.25, |
| "learning_rate": 7.024444444444445e-06, |
| "loss": 6.6707, |
| "mean_token_accuracy": 0.8740807961672544, |
| "num_tokens": 48659284.0, |
| "step": 6840 |
| }, |
| { |
| "entropy": 0.4094483610242605, |
| "epoch": 2.74, |
| "grad_norm": 44.5, |
| "learning_rate": 7.0022222222222225e-06, |
| "loss": 6.6007, |
| "mean_token_accuracy": 0.874849434196949, |
| "num_tokens": 48731232.0, |
| "step": 6850 |
| }, |
| { |
| "entropy": 0.4151701061055064, |
| "epoch": 2.7439999999999998, |
| "grad_norm": 47.25, |
| "learning_rate": 6.98e-06, |
| "loss": 6.6054, |
| "mean_token_accuracy": 0.8755607772618532, |
| "num_tokens": 48804539.0, |
| "step": 6860 |
| }, |
| { |
| "entropy": 0.42676802910864353, |
| "epoch": 2.748, |
| "grad_norm": 47.0, |
| "learning_rate": 6.9577777777777785e-06, |
| "loss": 6.7767, |
| "mean_token_accuracy": 0.8701058030128479, |
| "num_tokens": 48872276.0, |
| "step": 6870 |
| }, |
| { |
| "entropy": 0.4327113514766097, |
| "epoch": 2.752, |
| "grad_norm": 51.5, |
| "learning_rate": 6.935555555555556e-06, |
| "loss": 6.9283, |
| "mean_token_accuracy": 0.8684029880911112, |
| "num_tokens": 48943804.0, |
| "step": 6880 |
| }, |
| { |
| "entropy": 0.45012712278403344, |
| "epoch": 2.7560000000000002, |
| "grad_norm": 56.0, |
| "learning_rate": 6.913333333333334e-06, |
| "loss": 7.2201, |
| "mean_token_accuracy": 0.8644526150077582, |
| "num_tokens": 49013828.0, |
| "step": 6890 |
| }, |
| { |
| "entropy": 0.4122659021988511, |
| "epoch": 2.76, |
| "grad_norm": 53.75, |
| "learning_rate": 6.891111111111111e-06, |
| "loss": 6.5767, |
| "mean_token_accuracy": 0.8738088518381119, |
| "num_tokens": 49085791.0, |
| "step": 6900 |
| }, |
| { |
| "epoch": 2.76, |
| "eval_biology_entropy": 0.5880849555730819, |
| "eval_biology_loss": 2.0232467651367188, |
| "eval_biology_mean_token_accuracy": 0.6533576596975327, |
| "eval_biology_num_tokens": 49085791.0, |
| "eval_biology_runtime": 42.1104, |
| "eval_biology_samples_per_second": 11.874, |
| "eval_biology_steps_per_second": 11.874, |
| "step": 6900 |
| }, |
| { |
| "epoch": 2.76, |
| "eval_math_entropy": 0.4741788964420557, |
| "eval_math_loss": 0.7446678280830383, |
| "eval_math_mean_token_accuracy": 0.8148188325166702, |
| "eval_math_num_tokens": 49085791.0, |
| "eval_math_runtime": 48.2138, |
| "eval_math_samples_per_second": 10.37, |
| "eval_math_steps_per_second": 10.37, |
| "step": 6900 |
| }, |
| { |
| "entropy": 0.4254678047262132, |
| "epoch": 2.7640000000000002, |
| "grad_norm": 56.25, |
| "learning_rate": 6.868888888888889e-06, |
| "loss": 6.7103, |
| "mean_token_accuracy": 0.8719491966068744, |
| "num_tokens": 49154103.0, |
| "step": 6910 |
| }, |
| { |
| "entropy": 0.4228336410596967, |
| "epoch": 2.768, |
| "grad_norm": 49.0, |
| "learning_rate": 6.846666666666667e-06, |
| "loss": 6.8817, |
| "mean_token_accuracy": 0.871215769276023, |
| "num_tokens": 49230939.0, |
| "step": 6920 |
| }, |
| { |
| "entropy": 0.4168972961604595, |
| "epoch": 2.7720000000000002, |
| "grad_norm": 53.0, |
| "learning_rate": 6.824444444444444e-06, |
| "loss": 6.5716, |
| "mean_token_accuracy": 0.8755698557943106, |
| "num_tokens": 49297612.0, |
| "step": 6930 |
| }, |
| { |
| "entropy": 0.4234836363233626, |
| "epoch": 2.776, |
| "grad_norm": 44.5, |
| "learning_rate": 6.802222222222223e-06, |
| "loss": 6.7706, |
| "mean_token_accuracy": 0.8723371788859368, |
| "num_tokens": 49364615.0, |
| "step": 6940 |
| }, |
| { |
| "entropy": 0.4253668738529086, |
| "epoch": 2.7800000000000002, |
| "grad_norm": 52.25, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 6.8657, |
| "mean_token_accuracy": 0.8708317507058382, |
| "num_tokens": 49435350.0, |
| "step": 6950 |
| }, |
| { |
| "entropy": 0.41709458697587254, |
| "epoch": 2.784, |
| "grad_norm": 62.25, |
| "learning_rate": 6.757777777777779e-06, |
| "loss": 6.7797, |
| "mean_token_accuracy": 0.8713479198515415, |
| "num_tokens": 49504425.0, |
| "step": 6960 |
| }, |
| { |
| "entropy": 0.41700680470094087, |
| "epoch": 2.7880000000000003, |
| "grad_norm": 61.25, |
| "learning_rate": 6.735555555555556e-06, |
| "loss": 6.653, |
| "mean_token_accuracy": 0.8729936812072993, |
| "num_tokens": 49572597.0, |
| "step": 6970 |
| }, |
| { |
| "entropy": 0.41320684365928173, |
| "epoch": 2.792, |
| "grad_norm": 60.75, |
| "learning_rate": 6.713333333333334e-06, |
| "loss": 6.616, |
| "mean_token_accuracy": 0.8752215579152107, |
| "num_tokens": 49641963.0, |
| "step": 6980 |
| }, |
| { |
| "entropy": 0.4137293916195631, |
| "epoch": 2.7960000000000003, |
| "grad_norm": 53.25, |
| "learning_rate": 6.691111111111112e-06, |
| "loss": 6.618, |
| "mean_token_accuracy": 0.8730958495289087, |
| "num_tokens": 49713314.0, |
| "step": 6990 |
| }, |
| { |
| "entropy": 0.4147604891099036, |
| "epoch": 2.8, |
| "grad_norm": 48.75, |
| "learning_rate": 6.668888888888889e-06, |
| "loss": 6.5247, |
| "mean_token_accuracy": 0.8765051685273647, |
| "num_tokens": 49782661.0, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_biology_entropy": 0.5701675927639007, |
| "eval_biology_loss": 2.059124231338501, |
| "eval_biology_mean_token_accuracy": 0.6532024182081223, |
| "eval_biology_num_tokens": 49782661.0, |
| "eval_biology_runtime": 41.9715, |
| "eval_biology_samples_per_second": 11.913, |
| "eval_biology_steps_per_second": 11.913, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_math_entropy": 0.4600864284336567, |
| "eval_math_loss": 0.753042995929718, |
| "eval_math_mean_token_accuracy": 0.8148794600963593, |
| "eval_math_num_tokens": 49782661.0, |
| "eval_math_runtime": 48.0598, |
| "eval_math_samples_per_second": 10.404, |
| "eval_math_steps_per_second": 10.404, |
| "step": 7000 |
| }, |
| { |
| "entropy": 0.420876379776746, |
| "epoch": 2.8040000000000003, |
| "grad_norm": 47.25, |
| "learning_rate": 6.646666666666667e-06, |
| "loss": 6.8484, |
| "mean_token_accuracy": 0.8690746333450079, |
| "num_tokens": 49855053.0, |
| "step": 7010 |
| }, |
| { |
| "entropy": 0.439685994759202, |
| "epoch": 2.808, |
| "grad_norm": 50.0, |
| "learning_rate": 6.6244444444444445e-06, |
| "loss": 6.9914, |
| "mean_token_accuracy": 0.8675274383276701, |
| "num_tokens": 49928796.0, |
| "step": 7020 |
| }, |
| { |
| "entropy": 0.42355858338996766, |
| "epoch": 2.8120000000000003, |
| "grad_norm": 42.75, |
| "learning_rate": 6.602222222222223e-06, |
| "loss": 6.8479, |
| "mean_token_accuracy": 0.868702232092619, |
| "num_tokens": 50004063.0, |
| "step": 7030 |
| }, |
| { |
| "entropy": 0.4072733450680971, |
| "epoch": 2.816, |
| "grad_norm": 52.0, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 6.4578, |
| "mean_token_accuracy": 0.8754336308687926, |
| "num_tokens": 50073632.0, |
| "step": 7040 |
| }, |
| { |
| "entropy": 0.4220940926577896, |
| "epoch": 2.82, |
| "grad_norm": 46.0, |
| "learning_rate": 6.557777777777778e-06, |
| "loss": 6.6736, |
| "mean_token_accuracy": 0.8728714480996131, |
| "num_tokens": 50142158.0, |
| "step": 7050 |
| }, |
| { |
| "entropy": 0.4003599639981985, |
| "epoch": 2.824, |
| "grad_norm": 59.25, |
| "learning_rate": 6.535555555555556e-06, |
| "loss": 6.471, |
| "mean_token_accuracy": 0.8761493194848299, |
| "num_tokens": 50214300.0, |
| "step": 7060 |
| }, |
| { |
| "entropy": 0.4396217320114374, |
| "epoch": 2.828, |
| "grad_norm": 49.25, |
| "learning_rate": 6.513333333333333e-06, |
| "loss": 7.0062, |
| "mean_token_accuracy": 0.8678940020501613, |
| "num_tokens": 50288549.0, |
| "step": 7070 |
| }, |
| { |
| "entropy": 0.4146323238965124, |
| "epoch": 2.832, |
| "grad_norm": 61.5, |
| "learning_rate": 6.491111111111111e-06, |
| "loss": 6.6069, |
| "mean_token_accuracy": 0.8730486676096916, |
| "num_tokens": 50356964.0, |
| "step": 7080 |
| }, |
| { |
| "entropy": 0.42287809927947817, |
| "epoch": 2.836, |
| "grad_norm": 53.25, |
| "learning_rate": 6.468888888888889e-06, |
| "loss": 6.7031, |
| "mean_token_accuracy": 0.8723801571875811, |
| "num_tokens": 50427808.0, |
| "step": 7090 |
| }, |
| { |
| "entropy": 0.40453361459076403, |
| "epoch": 2.84, |
| "grad_norm": 54.0, |
| "learning_rate": 6.446666666666668e-06, |
| "loss": 6.6241, |
| "mean_token_accuracy": 0.8743327666074038, |
| "num_tokens": 50499105.0, |
| "step": 7100 |
| }, |
| { |
| "epoch": 2.84, |
| "eval_biology_entropy": 0.5755265356898308, |
| "eval_biology_loss": 2.04902982711792, |
| "eval_biology_mean_token_accuracy": 0.6534116899967194, |
| "eval_biology_num_tokens": 50499105.0, |
| "eval_biology_runtime": 40.2791, |
| "eval_biology_samples_per_second": 12.413, |
| "eval_biology_steps_per_second": 12.413, |
| "step": 7100 |
| }, |
| { |
| "epoch": 2.84, |
| "eval_math_entropy": 0.4622981072962284, |
| "eval_math_loss": 0.749148964881897, |
| "eval_math_mean_token_accuracy": 0.8149455053806305, |
| "eval_math_num_tokens": 50499105.0, |
| "eval_math_runtime": 48.1208, |
| "eval_math_samples_per_second": 10.391, |
| "eval_math_steps_per_second": 10.391, |
| "step": 7100 |
| }, |
| { |
| "entropy": 0.430855705216527, |
| "epoch": 2.844, |
| "grad_norm": 50.25, |
| "learning_rate": 6.4244444444444455e-06, |
| "loss": 6.9517, |
| "mean_token_accuracy": 0.8701685316860676, |
| "num_tokens": 50569307.0, |
| "step": 7110 |
| }, |
| { |
| "entropy": 0.44624668206088247, |
| "epoch": 2.848, |
| "grad_norm": 52.0, |
| "learning_rate": 6.402222222222223e-06, |
| "loss": 7.0886, |
| "mean_token_accuracy": 0.8676021881401539, |
| "num_tokens": 50644535.0, |
| "step": 7120 |
| }, |
| { |
| "entropy": 0.41043754341080785, |
| "epoch": 2.852, |
| "grad_norm": 49.5, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 6.5749, |
| "mean_token_accuracy": 0.8746382817626, |
| "num_tokens": 50714709.0, |
| "step": 7130 |
| }, |
| { |
| "entropy": 0.43770591551437976, |
| "epoch": 2.856, |
| "grad_norm": 53.0, |
| "learning_rate": 6.357777777777778e-06, |
| "loss": 7.0007, |
| "mean_token_accuracy": 0.8664165291935205, |
| "num_tokens": 50784952.0, |
| "step": 7140 |
| }, |
| { |
| "entropy": 0.43440137477591634, |
| "epoch": 2.86, |
| "grad_norm": 50.75, |
| "learning_rate": 6.335555555555556e-06, |
| "loss": 6.8187, |
| "mean_token_accuracy": 0.8716139234602451, |
| "num_tokens": 50855392.0, |
| "step": 7150 |
| }, |
| { |
| "entropy": 0.40293501038104296, |
| "epoch": 2.864, |
| "grad_norm": 47.75, |
| "learning_rate": 6.313333333333334e-06, |
| "loss": 6.555, |
| "mean_token_accuracy": 0.8768540095537901, |
| "num_tokens": 50925653.0, |
| "step": 7160 |
| }, |
| { |
| "entropy": 0.4266944283619523, |
| "epoch": 2.868, |
| "grad_norm": 64.5, |
| "learning_rate": 6.291111111111111e-06, |
| "loss": 6.8825, |
| "mean_token_accuracy": 0.8687186989933252, |
| "num_tokens": 50996698.0, |
| "step": 7170 |
| }, |
| { |
| "entropy": 0.39816700732335447, |
| "epoch": 2.872, |
| "grad_norm": 61.25, |
| "learning_rate": 6.26888888888889e-06, |
| "loss": 6.4226, |
| "mean_token_accuracy": 0.8773866448551416, |
| "num_tokens": 51065482.0, |
| "step": 7180 |
| }, |
| { |
| "entropy": 0.44183889096602796, |
| "epoch": 2.876, |
| "grad_norm": 60.75, |
| "learning_rate": 6.246666666666667e-06, |
| "loss": 6.9231, |
| "mean_token_accuracy": 0.8688192542642355, |
| "num_tokens": 51137259.0, |
| "step": 7190 |
| }, |
| { |
| "entropy": 0.4061698867939413, |
| "epoch": 2.88, |
| "grad_norm": 51.25, |
| "learning_rate": 6.224444444444445e-06, |
| "loss": 6.3989, |
| "mean_token_accuracy": 0.8756192050874233, |
| "num_tokens": 51204374.0, |
| "step": 7200 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 0.5688342335820198, |
| "eval_biology_loss": 2.06632137298584, |
| "eval_biology_mean_token_accuracy": 0.6529575547575951, |
| "eval_biology_num_tokens": 51204374.0, |
| "eval_biology_runtime": 40.153, |
| "eval_biology_samples_per_second": 12.452, |
| "eval_biology_steps_per_second": 12.452, |
| "step": 7200 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.4588739917576313, |
| "eval_math_loss": 0.7518824338912964, |
| "eval_math_mean_token_accuracy": 0.8151850702762604, |
| "eval_math_num_tokens": 51204374.0, |
| "eval_math_runtime": 47.8677, |
| "eval_math_samples_per_second": 10.445, |
| "eval_math_steps_per_second": 10.445, |
| "step": 7200 |
| }, |
| { |
| "entropy": 0.41158931888639927, |
| "epoch": 2.884, |
| "grad_norm": 49.25, |
| "learning_rate": 6.2022222222222225e-06, |
| "loss": 6.709, |
| "mean_token_accuracy": 0.8745049592107534, |
| "num_tokens": 51272969.0, |
| "step": 7210 |
| }, |
| { |
| "entropy": 0.39727600691840054, |
| "epoch": 2.888, |
| "grad_norm": 64.5, |
| "learning_rate": 6.18e-06, |
| "loss": 6.384, |
| "mean_token_accuracy": 0.879120760411024, |
| "num_tokens": 51341393.0, |
| "step": 7220 |
| }, |
| { |
| "entropy": 0.41550225345417857, |
| "epoch": 2.892, |
| "grad_norm": 64.5, |
| "learning_rate": 6.157777777777778e-06, |
| "loss": 6.6891, |
| "mean_token_accuracy": 0.8737437456846238, |
| "num_tokens": 51415270.0, |
| "step": 7230 |
| }, |
| { |
| "entropy": 0.42290679141879084, |
| "epoch": 2.896, |
| "grad_norm": 47.0, |
| "learning_rate": 6.135555555555555e-06, |
| "loss": 6.6741, |
| "mean_token_accuracy": 0.8721366498619318, |
| "num_tokens": 51483944.0, |
| "step": 7240 |
| }, |
| { |
| "entropy": 0.41524764224886895, |
| "epoch": 2.9, |
| "grad_norm": 52.75, |
| "learning_rate": 6.113333333333333e-06, |
| "loss": 6.6982, |
| "mean_token_accuracy": 0.8737081781029701, |
| "num_tokens": 51555569.0, |
| "step": 7250 |
| }, |
| { |
| "entropy": 0.4218080253340304, |
| "epoch": 2.904, |
| "grad_norm": 53.25, |
| "learning_rate": 6.091111111111112e-06, |
| "loss": 6.7799, |
| "mean_token_accuracy": 0.8702561281621456, |
| "num_tokens": 51626595.0, |
| "step": 7260 |
| }, |
| { |
| "entropy": 0.40329353911802174, |
| "epoch": 2.908, |
| "grad_norm": 85.5, |
| "learning_rate": 6.06888888888889e-06, |
| "loss": 6.4685, |
| "mean_token_accuracy": 0.8770487900823355, |
| "num_tokens": 51695049.0, |
| "step": 7270 |
| }, |
| { |
| "entropy": 0.42409236086532476, |
| "epoch": 2.912, |
| "grad_norm": 56.75, |
| "learning_rate": 6.0466666666666675e-06, |
| "loss": 6.836, |
| "mean_token_accuracy": 0.8699745990335941, |
| "num_tokens": 51765755.0, |
| "step": 7280 |
| }, |
| { |
| "entropy": 0.42378859128803015, |
| "epoch": 2.916, |
| "grad_norm": 57.0, |
| "learning_rate": 6.024444444444445e-06, |
| "loss": 6.6576, |
| "mean_token_accuracy": 0.875879754498601, |
| "num_tokens": 51837472.0, |
| "step": 7290 |
| }, |
| { |
| "entropy": 0.40245743859559296, |
| "epoch": 2.92, |
| "grad_norm": 52.0, |
| "learning_rate": 6.002222222222223e-06, |
| "loss": 6.6049, |
| "mean_token_accuracy": 0.8747074913233519, |
| "num_tokens": 51909054.0, |
| "step": 7300 |
| }, |
| { |
| "epoch": 2.92, |
| "eval_biology_entropy": 0.5713666259646416, |
| "eval_biology_loss": 2.059365749359131, |
| "eval_biology_mean_token_accuracy": 0.6528208155035973, |
| "eval_biology_num_tokens": 51909054.0, |
| "eval_biology_runtime": 39.9977, |
| "eval_biology_samples_per_second": 12.501, |
| "eval_biology_steps_per_second": 12.501, |
| "step": 7300 |
| }, |
| { |
| "epoch": 2.92, |
| "eval_math_entropy": 0.4593735449016094, |
| "eval_math_loss": 0.7520932555198669, |
| "eval_math_mean_token_accuracy": 0.814995592713356, |
| "eval_math_num_tokens": 51909054.0, |
| "eval_math_runtime": 47.7765, |
| "eval_math_samples_per_second": 10.465, |
| "eval_math_steps_per_second": 10.465, |
| "step": 7300 |
| }, |
| { |
| "entropy": 0.4271112416870892, |
| "epoch": 2.924, |
| "grad_norm": 47.0, |
| "learning_rate": 5.98e-06, |
| "loss": 6.8087, |
| "mean_token_accuracy": 0.8703517261892557, |
| "num_tokens": 51982314.0, |
| "step": 7310 |
| }, |
| { |
| "entropy": 0.4414162654429674, |
| "epoch": 2.928, |
| "grad_norm": 49.0, |
| "learning_rate": 5.957777777777778e-06, |
| "loss": 6.9483, |
| "mean_token_accuracy": 0.8697590786963701, |
| "num_tokens": 52056379.0, |
| "step": 7320 |
| }, |
| { |
| "entropy": 0.43352372804656625, |
| "epoch": 2.932, |
| "grad_norm": 50.25, |
| "learning_rate": 5.9355555555555556e-06, |
| "loss": 7.0335, |
| "mean_token_accuracy": 0.867513294517994, |
| "num_tokens": 52125747.0, |
| "step": 7330 |
| }, |
| { |
| "entropy": 0.41296316143125295, |
| "epoch": 2.936, |
| "grad_norm": 62.25, |
| "learning_rate": 5.913333333333334e-06, |
| "loss": 6.6491, |
| "mean_token_accuracy": 0.8731504786759615, |
| "num_tokens": 52195099.0, |
| "step": 7340 |
| }, |
| { |
| "entropy": 0.4088913880288601, |
| "epoch": 2.94, |
| "grad_norm": 74.0, |
| "learning_rate": 5.891111111111112e-06, |
| "loss": 6.4919, |
| "mean_token_accuracy": 0.8755192022770644, |
| "num_tokens": 52268583.0, |
| "step": 7350 |
| }, |
| { |
| "entropy": 0.418209442216903, |
| "epoch": 2.944, |
| "grad_norm": 57.75, |
| "learning_rate": 5.868888888888889e-06, |
| "loss": 6.7707, |
| "mean_token_accuracy": 0.8708242069929838, |
| "num_tokens": 52346232.0, |
| "step": 7360 |
| }, |
| { |
| "entropy": 0.4261803382076323, |
| "epoch": 2.948, |
| "grad_norm": 61.5, |
| "learning_rate": 5.846666666666667e-06, |
| "loss": 6.8978, |
| "mean_token_accuracy": 0.867920583486557, |
| "num_tokens": 52417353.0, |
| "step": 7370 |
| }, |
| { |
| "entropy": 0.4419269938953221, |
| "epoch": 2.952, |
| "grad_norm": 55.25, |
| "learning_rate": 5.8244444444444445e-06, |
| "loss": 7.1038, |
| "mean_token_accuracy": 0.8650728393346071, |
| "num_tokens": 52489803.0, |
| "step": 7380 |
| }, |
| { |
| "entropy": 0.42536340802907946, |
| "epoch": 2.956, |
| "grad_norm": 47.0, |
| "learning_rate": 5.802222222222222e-06, |
| "loss": 6.7066, |
| "mean_token_accuracy": 0.8735882114619017, |
| "num_tokens": 52559419.0, |
| "step": 7390 |
| }, |
| { |
| "entropy": 0.4226967167109251, |
| "epoch": 2.96, |
| "grad_norm": 54.75, |
| "learning_rate": 5.78e-06, |
| "loss": 6.8842, |
| "mean_token_accuracy": 0.8703772567212582, |
| "num_tokens": 52633789.0, |
| "step": 7400 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_biology_entropy": 0.567707498550415, |
| "eval_biology_loss": 2.0689804553985596, |
| "eval_biology_mean_token_accuracy": 0.6533778265118599, |
| "eval_biology_num_tokens": 52633789.0, |
| "eval_biology_runtime": 40.0504, |
| "eval_biology_samples_per_second": 12.484, |
| "eval_biology_steps_per_second": 12.484, |
| "step": 7400 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_math_entropy": 0.4593480526953936, |
| "eval_math_loss": 0.7509452104568481, |
| "eval_math_mean_token_accuracy": 0.8156206517219543, |
| "eval_math_num_tokens": 52633789.0, |
| "eval_math_runtime": 47.8306, |
| "eval_math_samples_per_second": 10.454, |
| "eval_math_steps_per_second": 10.454, |
| "step": 7400 |
| }, |
| { |
| "entropy": 0.41576854828745125, |
| "epoch": 2.964, |
| "grad_norm": 55.75, |
| "learning_rate": 5.757777777777779e-06, |
| "loss": 6.5634, |
| "mean_token_accuracy": 0.8743478409945965, |
| "num_tokens": 52700925.0, |
| "step": 7410 |
| }, |
| { |
| "entropy": 0.4251088222488761, |
| "epoch": 2.968, |
| "grad_norm": 56.5, |
| "learning_rate": 5.735555555555557e-06, |
| "loss": 6.7435, |
| "mean_token_accuracy": 0.8705783415585756, |
| "num_tokens": 52765602.0, |
| "step": 7420 |
| }, |
| { |
| "entropy": 0.4234215301461518, |
| "epoch": 2.972, |
| "grad_norm": 47.25, |
| "learning_rate": 5.713333333333334e-06, |
| "loss": 6.8971, |
| "mean_token_accuracy": 0.8695938050746918, |
| "num_tokens": 52838638.0, |
| "step": 7430 |
| }, |
| { |
| "entropy": 0.4232695480808616, |
| "epoch": 2.976, |
| "grad_norm": 49.0, |
| "learning_rate": 5.691111111111112e-06, |
| "loss": 6.6237, |
| "mean_token_accuracy": 0.8742138776928187, |
| "num_tokens": 52911755.0, |
| "step": 7440 |
| }, |
| { |
| "entropy": 0.4332481697201729, |
| "epoch": 2.98, |
| "grad_norm": 53.75, |
| "learning_rate": 5.6688888888888895e-06, |
| "loss": 6.8348, |
| "mean_token_accuracy": 0.8695711780339479, |
| "num_tokens": 52982200.0, |
| "step": 7450 |
| }, |
| { |
| "entropy": 0.42631542691960933, |
| "epoch": 2.984, |
| "grad_norm": 52.0, |
| "learning_rate": 5.646666666666667e-06, |
| "loss": 6.8524, |
| "mean_token_accuracy": 0.8712625961750746, |
| "num_tokens": 53054434.0, |
| "step": 7460 |
| }, |
| { |
| "entropy": 0.4202953160740435, |
| "epoch": 2.988, |
| "grad_norm": 44.5, |
| "learning_rate": 5.624444444444445e-06, |
| "loss": 6.8501, |
| "mean_token_accuracy": 0.8697643611580134, |
| "num_tokens": 53129935.0, |
| "step": 7470 |
| }, |
| { |
| "entropy": 0.3870232336688787, |
| "epoch": 2.992, |
| "grad_norm": 53.25, |
| "learning_rate": 5.602222222222222e-06, |
| "loss": 6.1791, |
| "mean_token_accuracy": 0.8826418906450272, |
| "num_tokens": 53198176.0, |
| "step": 7480 |
| }, |
| { |
| "entropy": 0.41608267249539493, |
| "epoch": 2.996, |
| "grad_norm": 58.25, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 6.6058, |
| "mean_token_accuracy": 0.8742454893887043, |
| "num_tokens": 53272486.0, |
| "step": 7490 |
| }, |
| { |
| "entropy": 0.45272362693212925, |
| "epoch": 3.0, |
| "grad_norm": 54.5, |
| "learning_rate": 5.557777777777778e-06, |
| "loss": 7.2983, |
| "mean_token_accuracy": 0.8642814248800278, |
| "num_tokens": 53341296.0, |
| "step": 7500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_biology_entropy": 0.5785715200901032, |
| "eval_biology_loss": 2.0478339195251465, |
| "eval_biology_mean_token_accuracy": 0.6534382239580154, |
| "eval_biology_num_tokens": 53341296.0, |
| "eval_biology_runtime": 39.9529, |
| "eval_biology_samples_per_second": 12.515, |
| "eval_biology_steps_per_second": 12.515, |
| "step": 7500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_math_entropy": 0.4669304715096951, |
| "eval_math_loss": 0.7433039546012878, |
| "eval_math_mean_token_accuracy": 0.8156477793455124, |
| "eval_math_num_tokens": 53341296.0, |
| "eval_math_runtime": 47.8579, |
| "eval_math_samples_per_second": 10.448, |
| "eval_math_steps_per_second": 10.448, |
| "step": 7500 |
| }, |
| { |
| "entropy": 0.35361110465601087, |
| "epoch": 3.004, |
| "grad_norm": 50.5, |
| "learning_rate": 5.535555555555556e-06, |
| "loss": 5.2063, |
| "mean_token_accuracy": 0.9028309009969234, |
| "num_tokens": 53408027.0, |
| "step": 7510 |
| }, |
| { |
| "entropy": 0.317386430921033, |
| "epoch": 3.008, |
| "grad_norm": 58.5, |
| "learning_rate": 5.513333333333334e-06, |
| "loss": 5.2665, |
| "mean_token_accuracy": 0.9008146651089192, |
| "num_tokens": 53481893.0, |
| "step": 7520 |
| }, |
| { |
| "entropy": 0.33771364130079745, |
| "epoch": 3.012, |
| "grad_norm": 54.5, |
| "learning_rate": 5.491111111111111e-06, |
| "loss": 5.1724, |
| "mean_token_accuracy": 0.9037490069866181, |
| "num_tokens": 53553569.0, |
| "step": 7530 |
| }, |
| { |
| "entropy": 0.33333308547735213, |
| "epoch": 3.016, |
| "grad_norm": 61.5, |
| "learning_rate": 5.468888888888889e-06, |
| "loss": 5.1333, |
| "mean_token_accuracy": 0.9025349728763103, |
| "num_tokens": 53625259.0, |
| "step": 7540 |
| }, |
| { |
| "entropy": 0.33144948063418267, |
| "epoch": 3.02, |
| "grad_norm": 54.75, |
| "learning_rate": 5.4466666666666665e-06, |
| "loss": 5.2293, |
| "mean_token_accuracy": 0.8993874792009592, |
| "num_tokens": 53697511.0, |
| "step": 7550 |
| }, |
| { |
| "entropy": 0.3690175848547369, |
| "epoch": 3.024, |
| "grad_norm": 59.0, |
| "learning_rate": 5.424444444444444e-06, |
| "loss": 5.8487, |
| "mean_token_accuracy": 0.8902920853346586, |
| "num_tokens": 53771717.0, |
| "step": 7560 |
| }, |
| { |
| "entropy": 0.32172863064333795, |
| "epoch": 3.028, |
| "grad_norm": 57.25, |
| "learning_rate": 5.402222222222223e-06, |
| "loss": 5.0283, |
| "mean_token_accuracy": 0.9051333077251911, |
| "num_tokens": 53840569.0, |
| "step": 7570 |
| }, |
| { |
| "entropy": 0.31612554416060445, |
| "epoch": 3.032, |
| "grad_norm": 65.0, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 4.8943, |
| "mean_token_accuracy": 0.9070401955395937, |
| "num_tokens": 53912836.0, |
| "step": 7580 |
| }, |
| { |
| "entropy": 0.3095882005058229, |
| "epoch": 3.036, |
| "grad_norm": 57.75, |
| "learning_rate": 5.357777777777779e-06, |
| "loss": 4.8251, |
| "mean_token_accuracy": 0.9079424347728491, |
| "num_tokens": 53982242.0, |
| "step": 7590 |
| }, |
| { |
| "entropy": 0.36460896488279104, |
| "epoch": 3.04, |
| "grad_norm": 59.0, |
| "learning_rate": 5.335555555555556e-06, |
| "loss": 5.7738, |
| "mean_token_accuracy": 0.8920512955635787, |
| "num_tokens": 54058045.0, |
| "step": 7600 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 0.4659742852151394, |
| "eval_biology_loss": 2.464916467666626, |
| "eval_biology_mean_token_accuracy": 0.6449353910684585, |
| "eval_biology_num_tokens": 54058045.0, |
| "eval_biology_runtime": 40.4398, |
| "eval_biology_samples_per_second": 12.364, |
| "eval_biology_steps_per_second": 12.364, |
| "step": 7600 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.37724404206871986, |
| "eval_math_loss": 0.8946568965911865, |
| "eval_math_mean_token_accuracy": 0.809667916893959, |
| "eval_math_num_tokens": 54058045.0, |
| "eval_math_runtime": 47.9431, |
| "eval_math_samples_per_second": 10.429, |
| "eval_math_steps_per_second": 10.429, |
| "step": 7600 |
| }, |
| { |
| "entropy": 0.3237521236296743, |
| "epoch": 3.044, |
| "grad_norm": 57.75, |
| "learning_rate": 5.313333333333334e-06, |
| "loss": 5.0842, |
| "mean_token_accuracy": 0.9027916241437197, |
| "num_tokens": 54126366.0, |
| "step": 7610 |
| }, |
| { |
| "entropy": 0.3265122418757528, |
| "epoch": 3.048, |
| "grad_norm": 55.75, |
| "learning_rate": 5.2911111111111115e-06, |
| "loss": 5.0782, |
| "mean_token_accuracy": 0.9034842975437641, |
| "num_tokens": 54194627.0, |
| "step": 7620 |
| }, |
| { |
| "entropy": 0.3139278646558523, |
| "epoch": 3.052, |
| "grad_norm": 58.0, |
| "learning_rate": 5.268888888888889e-06, |
| "loss": 4.8781, |
| "mean_token_accuracy": 0.9051724404096604, |
| "num_tokens": 54265092.0, |
| "step": 7630 |
| }, |
| { |
| "entropy": 0.30669141178950665, |
| "epoch": 3.056, |
| "grad_norm": 52.0, |
| "learning_rate": 5.246666666666667e-06, |
| "loss": 4.7466, |
| "mean_token_accuracy": 0.9087039653211832, |
| "num_tokens": 54334332.0, |
| "step": 7640 |
| }, |
| { |
| "entropy": 0.31843363838270305, |
| "epoch": 3.06, |
| "grad_norm": 67.0, |
| "learning_rate": 5.224444444444445e-06, |
| "loss": 5.0549, |
| "mean_token_accuracy": 0.9038249392062425, |
| "num_tokens": 54405624.0, |
| "step": 7650 |
| }, |
| { |
| "entropy": 0.34820014336146415, |
| "epoch": 3.064, |
| "grad_norm": 60.75, |
| "learning_rate": 5.202222222222223e-06, |
| "loss": 5.4398, |
| "mean_token_accuracy": 0.8965415615588427, |
| "num_tokens": 54480429.0, |
| "step": 7660 |
| }, |
| { |
| "entropy": 0.3183653435204178, |
| "epoch": 3.068, |
| "grad_norm": 58.5, |
| "learning_rate": 5.18e-06, |
| "loss": 4.9982, |
| "mean_token_accuracy": 0.9055410612374544, |
| "num_tokens": 54552485.0, |
| "step": 7670 |
| }, |
| { |
| "entropy": 0.33736218037083743, |
| "epoch": 3.072, |
| "grad_norm": 54.5, |
| "learning_rate": 5.157777777777778e-06, |
| "loss": 5.2106, |
| "mean_token_accuracy": 0.9005741581320763, |
| "num_tokens": 54624543.0, |
| "step": 7680 |
| }, |
| { |
| "entropy": 0.2914720828644931, |
| "epoch": 3.076, |
| "grad_norm": 68.0, |
| "learning_rate": 5.135555555555556e-06, |
| "loss": 4.6177, |
| "mean_token_accuracy": 0.9120103523135186, |
| "num_tokens": 54692074.0, |
| "step": 7690 |
| }, |
| { |
| "entropy": 0.31254780581220987, |
| "epoch": 3.08, |
| "grad_norm": 74.0, |
| "learning_rate": 5.113333333333333e-06, |
| "loss": 4.8767, |
| "mean_token_accuracy": 0.9058825578540564, |
| "num_tokens": 54761124.0, |
| "step": 7700 |
| }, |
| { |
| "epoch": 3.08, |
| "eval_biology_entropy": 0.4661154082417488, |
| "eval_biology_loss": 2.4701883792877197, |
| "eval_biology_mean_token_accuracy": 0.6445987159013749, |
| "eval_biology_num_tokens": 54761124.0, |
| "eval_biology_runtime": 39.9976, |
| "eval_biology_samples_per_second": 12.501, |
| "eval_biology_steps_per_second": 12.501, |
| "step": 7700 |
| }, |
| { |
| "epoch": 3.08, |
| "eval_math_entropy": 0.382908758610487, |
| "eval_math_loss": 0.8919873237609863, |
| "eval_math_mean_token_accuracy": 0.8095510637760163, |
| "eval_math_num_tokens": 54761124.0, |
| "eval_math_runtime": 47.8554, |
| "eval_math_samples_per_second": 10.448, |
| "eval_math_steps_per_second": 10.448, |
| "step": 7700 |
| }, |
| { |
| "entropy": 0.30812596646137536, |
| "epoch": 3.084, |
| "grad_norm": 62.25, |
| "learning_rate": 5.091111111111111e-06, |
| "loss": 4.8079, |
| "mean_token_accuracy": 0.9083172097802162, |
| "num_tokens": 54835373.0, |
| "step": 7710 |
| }, |
| { |
| "entropy": 0.33229771852493284, |
| "epoch": 3.088, |
| "grad_norm": 62.0, |
| "learning_rate": 5.06888888888889e-06, |
| "loss": 5.24, |
| "mean_token_accuracy": 0.9021586898714304, |
| "num_tokens": 54907550.0, |
| "step": 7720 |
| }, |
| { |
| "entropy": 0.3008203926961869, |
| "epoch": 3.092, |
| "grad_norm": 57.25, |
| "learning_rate": 5.046666666666668e-06, |
| "loss": 4.6546, |
| "mean_token_accuracy": 0.9120187398046256, |
| "num_tokens": 54976238.0, |
| "step": 7730 |
| }, |
| { |
| "entropy": 0.3413645211607218, |
| "epoch": 3.096, |
| "grad_norm": 52.5, |
| "learning_rate": 5.024444444444445e-06, |
| "loss": 5.3974, |
| "mean_token_accuracy": 0.8978584710508585, |
| "num_tokens": 55048134.0, |
| "step": 7740 |
| }, |
| { |
| "entropy": 0.3375399123877287, |
| "epoch": 3.1, |
| "grad_norm": 65.5, |
| "learning_rate": 5.002222222222223e-06, |
| "loss": 5.2498, |
| "mean_token_accuracy": 0.9009373385459185, |
| "num_tokens": 55122955.0, |
| "step": 7750 |
| }, |
| { |
| "entropy": 0.310310556832701, |
| "epoch": 3.104, |
| "grad_norm": 63.75, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 4.8122, |
| "mean_token_accuracy": 0.9075421027839183, |
| "num_tokens": 55190959.0, |
| "step": 7760 |
| }, |
| { |
| "entropy": 0.3247022640425712, |
| "epoch": 3.108, |
| "grad_norm": 53.5, |
| "learning_rate": 4.957777777777778e-06, |
| "loss": 5.0894, |
| "mean_token_accuracy": 0.9036967132240534, |
| "num_tokens": 55264918.0, |
| "step": 7770 |
| }, |
| { |
| "entropy": 0.32933325627818705, |
| "epoch": 3.112, |
| "grad_norm": 60.75, |
| "learning_rate": 4.935555555555556e-06, |
| "loss": 5.1041, |
| "mean_token_accuracy": 0.9041826728731394, |
| "num_tokens": 55335549.0, |
| "step": 7780 |
| }, |
| { |
| "entropy": 0.3293325945734978, |
| "epoch": 3.116, |
| "grad_norm": 68.5, |
| "learning_rate": 4.9133333333333334e-06, |
| "loss": 5.0896, |
| "mean_token_accuracy": 0.9034241706132888, |
| "num_tokens": 55406698.0, |
| "step": 7790 |
| }, |
| { |
| "entropy": 0.33265041671693324, |
| "epoch": 3.12, |
| "grad_norm": 52.5, |
| "learning_rate": 4.891111111111111e-06, |
| "loss": 5.2903, |
| "mean_token_accuracy": 0.8996705986559391, |
| "num_tokens": 55481635.0, |
| "step": 7800 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_biology_entropy": 0.46094007897377015, |
| "eval_biology_loss": 2.486146926879883, |
| "eval_biology_mean_token_accuracy": 0.6444572919011116, |
| "eval_biology_num_tokens": 55481635.0, |
| "eval_biology_runtime": 40.109, |
| "eval_biology_samples_per_second": 12.466, |
| "eval_biology_steps_per_second": 12.466, |
| "step": 7800 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_math_entropy": 0.3829066038131714, |
| "eval_math_loss": 0.8934834003448486, |
| "eval_math_mean_token_accuracy": 0.8094851752519607, |
| "eval_math_num_tokens": 55481635.0, |
| "eval_math_runtime": 47.8316, |
| "eval_math_samples_per_second": 10.453, |
| "eval_math_steps_per_second": 10.453, |
| "step": 7800 |
| }, |
| { |
| "entropy": 0.3011790055781603, |
| "epoch": 3.124, |
| "grad_norm": 64.5, |
| "learning_rate": 4.8688888888888895e-06, |
| "loss": 4.7284, |
| "mean_token_accuracy": 0.9100992254912853, |
| "num_tokens": 55551147.0, |
| "step": 7810 |
| }, |
| { |
| "entropy": 0.3434626078233123, |
| "epoch": 3.128, |
| "grad_norm": 58.5, |
| "learning_rate": 4.846666666666667e-06, |
| "loss": 5.3955, |
| "mean_token_accuracy": 0.898734774067998, |
| "num_tokens": 55621324.0, |
| "step": 7820 |
| }, |
| { |
| "entropy": 0.3306236209347844, |
| "epoch": 3.132, |
| "grad_norm": 53.0, |
| "learning_rate": 4.824444444444445e-06, |
| "loss": 5.2207, |
| "mean_token_accuracy": 0.9004638835787773, |
| "num_tokens": 55692210.0, |
| "step": 7830 |
| }, |
| { |
| "entropy": 0.3404700209852308, |
| "epoch": 3.136, |
| "grad_norm": 56.75, |
| "learning_rate": 4.802222222222222e-06, |
| "loss": 5.3601, |
| "mean_token_accuracy": 0.8985728345811367, |
| "num_tokens": 55769010.0, |
| "step": 7840 |
| }, |
| { |
| "entropy": 0.3126563511323184, |
| "epoch": 3.14, |
| "grad_norm": 71.0, |
| "learning_rate": 4.78e-06, |
| "loss": 4.7756, |
| "mean_token_accuracy": 0.9084874775260687, |
| "num_tokens": 55835370.0, |
| "step": 7850 |
| }, |
| { |
| "entropy": 0.3216227440163493, |
| "epoch": 3.144, |
| "grad_norm": 53.25, |
| "learning_rate": 4.7577777777777784e-06, |
| "loss": 5.0378, |
| "mean_token_accuracy": 0.9044648971408605, |
| "num_tokens": 55906927.0, |
| "step": 7860 |
| }, |
| { |
| "entropy": 0.3090505673550069, |
| "epoch": 3.148, |
| "grad_norm": 59.75, |
| "learning_rate": 4.735555555555556e-06, |
| "loss": 4.9203, |
| "mean_token_accuracy": 0.9066134061664343, |
| "num_tokens": 55981416.0, |
| "step": 7870 |
| }, |
| { |
| "entropy": 0.33745630346238614, |
| "epoch": 3.152, |
| "grad_norm": 54.75, |
| "learning_rate": 4.713333333333334e-06, |
| "loss": 5.2338, |
| "mean_token_accuracy": 0.901577778160572, |
| "num_tokens": 56053160.0, |
| "step": 7880 |
| }, |
| { |
| "entropy": 0.3518794354517013, |
| "epoch": 3.156, |
| "grad_norm": 56.25, |
| "learning_rate": 4.691111111111111e-06, |
| "loss": 5.4518, |
| "mean_token_accuracy": 0.8971395801752806, |
| "num_tokens": 56120945.0, |
| "step": 7890 |
| }, |
| { |
| "entropy": 0.32917964705266056, |
| "epoch": 3.16, |
| "grad_norm": 58.0, |
| "learning_rate": 4.66888888888889e-06, |
| "loss": 5.2869, |
| "mean_token_accuracy": 0.8997520461678505, |
| "num_tokens": 56197082.0, |
| "step": 7900 |
| }, |
| { |
| "epoch": 3.16, |
| "eval_biology_entropy": 0.4709624526500702, |
| "eval_biology_loss": 2.4603536128997803, |
| "eval_biology_mean_token_accuracy": 0.6438030276298523, |
| "eval_biology_num_tokens": 56197082.0, |
| "eval_biology_runtime": 44.2502, |
| "eval_biology_samples_per_second": 11.299, |
| "eval_biology_steps_per_second": 11.299, |
| "step": 7900 |
| }, |
| { |
| "epoch": 3.16, |
| "eval_math_entropy": 0.38973184932768346, |
| "eval_math_loss": 0.8849496841430664, |
| "eval_math_mean_token_accuracy": 0.8093812301158905, |
| "eval_math_num_tokens": 56197082.0, |
| "eval_math_runtime": 51.0245, |
| "eval_math_samples_per_second": 9.799, |
| "eval_math_steps_per_second": 9.799, |
| "step": 7900 |
| }, |
| { |
| "entropy": 0.33233238277025523, |
| "epoch": 3.164, |
| "grad_norm": 57.0, |
| "learning_rate": 4.646666666666667e-06, |
| "loss": 5.2171, |
| "mean_token_accuracy": 0.9030471380800009, |
| "num_tokens": 56265132.0, |
| "step": 7910 |
| }, |
| { |
| "entropy": 0.32855810816399755, |
| "epoch": 3.168, |
| "grad_norm": 60.0, |
| "learning_rate": 4.624444444444445e-06, |
| "loss": 5.2152, |
| "mean_token_accuracy": 0.9011328600347042, |
| "num_tokens": 56337066.0, |
| "step": 7920 |
| }, |
| { |
| "entropy": 0.3474338168743998, |
| "epoch": 3.172, |
| "grad_norm": 50.75, |
| "learning_rate": 4.602222222222223e-06, |
| "loss": 5.2758, |
| "mean_token_accuracy": 0.8982305862009525, |
| "num_tokens": 56409953.0, |
| "step": 7930 |
| }, |
| { |
| "entropy": 0.30498403031378984, |
| "epoch": 3.176, |
| "grad_norm": 72.5, |
| "learning_rate": 4.58e-06, |
| "loss": 4.7825, |
| "mean_token_accuracy": 0.9080491527915001, |
| "num_tokens": 56478421.0, |
| "step": 7940 |
| }, |
| { |
| "entropy": 0.3291911333799362, |
| "epoch": 3.18, |
| "grad_norm": 65.5, |
| "learning_rate": 4.557777777777778e-06, |
| "loss": 5.2038, |
| "mean_token_accuracy": 0.9005887977778911, |
| "num_tokens": 56548259.0, |
| "step": 7950 |
| }, |
| { |
| "entropy": 0.3255664438940585, |
| "epoch": 3.184, |
| "grad_norm": 76.5, |
| "learning_rate": 4.535555555555555e-06, |
| "loss": 5.132, |
| "mean_token_accuracy": 0.9030448354780674, |
| "num_tokens": 56618899.0, |
| "step": 7960 |
| }, |
| { |
| "entropy": 0.32358582559973004, |
| "epoch": 3.188, |
| "grad_norm": 58.0, |
| "learning_rate": 4.513333333333333e-06, |
| "loss": 5.0937, |
| "mean_token_accuracy": 0.9020735800266266, |
| "num_tokens": 56693376.0, |
| "step": 7970 |
| }, |
| { |
| "entropy": 0.3105836093891412, |
| "epoch": 3.192, |
| "grad_norm": 55.0, |
| "learning_rate": 4.4911111111111115e-06, |
| "loss": 4.7987, |
| "mean_token_accuracy": 0.9090862285345793, |
| "num_tokens": 56764351.0, |
| "step": 7980 |
| }, |
| { |
| "entropy": 0.3293394946027547, |
| "epoch": 3.196, |
| "grad_norm": 60.75, |
| "learning_rate": 4.468888888888889e-06, |
| "loss": 5.2327, |
| "mean_token_accuracy": 0.9007721368223429, |
| "num_tokens": 56832454.0, |
| "step": 7990 |
| }, |
| { |
| "entropy": 0.3379898440092802, |
| "epoch": 3.2, |
| "grad_norm": 58.0, |
| "learning_rate": 4.446666666666667e-06, |
| "loss": 5.2898, |
| "mean_token_accuracy": 0.8989723846316338, |
| "num_tokens": 56910071.0, |
| "step": 8000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 0.4689203085899353, |
| "eval_biology_loss": 2.463636636734009, |
| "eval_biology_mean_token_accuracy": 0.6452909532785416, |
| "eval_biology_num_tokens": 56910071.0, |
| "eval_biology_runtime": 40.2927, |
| "eval_biology_samples_per_second": 12.409, |
| "eval_biology_steps_per_second": 12.409, |
| "step": 8000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.38982494167983534, |
| "eval_math_loss": 0.8843812346458435, |
| "eval_math_mean_token_accuracy": 0.8096272931098938, |
| "eval_math_num_tokens": 56910071.0, |
| "eval_math_runtime": 53.1425, |
| "eval_math_samples_per_second": 9.409, |
| "eval_math_steps_per_second": 9.409, |
| "step": 8000 |
| }, |
| { |
| "entropy": 0.33605060377158225, |
| "epoch": 3.204, |
| "grad_norm": 55.5, |
| "learning_rate": 4.424444444444444e-06, |
| "loss": 5.2136, |
| "mean_token_accuracy": 0.9022354893386364, |
| "num_tokens": 56980995.0, |
| "step": 8010 |
| }, |
| { |
| "entropy": 0.3417088720947504, |
| "epoch": 3.208, |
| "grad_norm": 59.0, |
| "learning_rate": 4.402222222222223e-06, |
| "loss": 5.3576, |
| "mean_token_accuracy": 0.898954252526164, |
| "num_tokens": 57050321.0, |
| "step": 8020 |
| }, |
| { |
| "entropy": 0.3144971864297986, |
| "epoch": 3.212, |
| "grad_norm": 52.5, |
| "learning_rate": 4.38e-06, |
| "loss": 4.9416, |
| "mean_token_accuracy": 0.9072338283061981, |
| "num_tokens": 57118485.0, |
| "step": 8030 |
| }, |
| { |
| "entropy": 0.3332448072731495, |
| "epoch": 3.216, |
| "grad_norm": 63.25, |
| "learning_rate": 4.357777777777778e-06, |
| "loss": 5.1862, |
| "mean_token_accuracy": 0.9009304217994213, |
| "num_tokens": 57190265.0, |
| "step": 8040 |
| }, |
| { |
| "entropy": 0.29999592429958283, |
| "epoch": 3.22, |
| "grad_norm": 56.5, |
| "learning_rate": 4.3355555555555565e-06, |
| "loss": 4.622, |
| "mean_token_accuracy": 0.9102712884545326, |
| "num_tokens": 57258187.0, |
| "step": 8050 |
| }, |
| { |
| "entropy": 0.3377910345327109, |
| "epoch": 3.224, |
| "grad_norm": 53.0, |
| "learning_rate": 4.313333333333334e-06, |
| "loss": 5.4402, |
| "mean_token_accuracy": 0.8972452398389578, |
| "num_tokens": 57335361.0, |
| "step": 8060 |
| }, |
| { |
| "entropy": 0.3233450698433444, |
| "epoch": 3.228, |
| "grad_norm": 59.25, |
| "learning_rate": 4.291111111111112e-06, |
| "loss": 4.9922, |
| "mean_token_accuracy": 0.9053470551967621, |
| "num_tokens": 57405287.0, |
| "step": 8070 |
| }, |
| { |
| "entropy": 0.3319830815307796, |
| "epoch": 3.232, |
| "grad_norm": 60.0, |
| "learning_rate": 4.268888888888889e-06, |
| "loss": 5.2539, |
| "mean_token_accuracy": 0.9016107145696879, |
| "num_tokens": 57471355.0, |
| "step": 8080 |
| }, |
| { |
| "entropy": 0.3109321851748973, |
| "epoch": 3.2359999999999998, |
| "grad_norm": 66.0, |
| "learning_rate": 4.246666666666667e-06, |
| "loss": 4.7775, |
| "mean_token_accuracy": 0.9089783221483231, |
| "num_tokens": 57541601.0, |
| "step": 8090 |
| }, |
| { |
| "entropy": 0.3036802196409553, |
| "epoch": 3.24, |
| "grad_norm": 72.5, |
| "learning_rate": 4.2244444444444446e-06, |
| "loss": 4.7453, |
| "mean_token_accuracy": 0.9081243328750134, |
| "num_tokens": 57611361.0, |
| "step": 8100 |
| }, |
| { |
| "epoch": 3.24, |
| "eval_biology_entropy": 0.4556665977239609, |
| "eval_biology_loss": 2.5050101280212402, |
| "eval_biology_mean_token_accuracy": 0.6445098651647567, |
| "eval_biology_num_tokens": 57611361.0, |
| "eval_biology_runtime": 42.2787, |
| "eval_biology_samples_per_second": 11.826, |
| "eval_biology_steps_per_second": 11.826, |
| "step": 8100 |
| }, |
| { |
| "epoch": 3.24, |
| "eval_math_entropy": 0.3757398373633623, |
| "eval_math_loss": 0.9050271511077881, |
| "eval_math_mean_token_accuracy": 0.809360929608345, |
| "eval_math_num_tokens": 57611361.0, |
| "eval_math_runtime": 50.9419, |
| "eval_math_samples_per_second": 9.815, |
| "eval_math_steps_per_second": 9.815, |
| "step": 8100 |
| }, |
| { |
| "entropy": 0.3032497449312359, |
| "epoch": 3.2439999999999998, |
| "grad_norm": 56.0, |
| "learning_rate": 4.202222222222222e-06, |
| "loss": 4.7054, |
| "mean_token_accuracy": 0.9095405578613281, |
| "num_tokens": 57680423.0, |
| "step": 8110 |
| }, |
| { |
| "entropy": 0.3054811102338135, |
| "epoch": 3.248, |
| "grad_norm": 59.0, |
| "learning_rate": 4.18e-06, |
| "loss": 4.6826, |
| "mean_token_accuracy": 0.9101293407380581, |
| "num_tokens": 57749030.0, |
| "step": 8120 |
| }, |
| { |
| "entropy": 0.31911276560276747, |
| "epoch": 3.252, |
| "grad_norm": 67.0, |
| "learning_rate": 4.157777777777778e-06, |
| "loss": 5.1334, |
| "mean_token_accuracy": 0.9037488639354706, |
| "num_tokens": 57821437.0, |
| "step": 8130 |
| }, |
| { |
| "entropy": 0.30937101258896293, |
| "epoch": 3.2560000000000002, |
| "grad_norm": 58.25, |
| "learning_rate": 4.135555555555556e-06, |
| "loss": 4.8639, |
| "mean_token_accuracy": 0.9091051619499921, |
| "num_tokens": 57888017.0, |
| "step": 8140 |
| }, |
| { |
| "entropy": 0.31460040775127707, |
| "epoch": 3.26, |
| "grad_norm": 49.5, |
| "learning_rate": 4.1133333333333335e-06, |
| "loss": 4.8762, |
| "mean_token_accuracy": 0.9060662046074868, |
| "num_tokens": 57959696.0, |
| "step": 8150 |
| }, |
| { |
| "entropy": 0.3173192950431257, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 83.0, |
| "learning_rate": 4.091111111111111e-06, |
| "loss": 4.9344, |
| "mean_token_accuracy": 0.9058732099831104, |
| "num_tokens": 58027567.0, |
| "step": 8160 |
| }, |
| { |
| "entropy": 0.3330814838409424, |
| "epoch": 3.268, |
| "grad_norm": 51.0, |
| "learning_rate": 4.0688888888888896e-06, |
| "loss": 5.1492, |
| "mean_token_accuracy": 0.9008448019623756, |
| "num_tokens": 58099947.0, |
| "step": 8170 |
| }, |
| { |
| "entropy": 0.3291762791574001, |
| "epoch": 3.2720000000000002, |
| "grad_norm": 88.0, |
| "learning_rate": 4.046666666666667e-06, |
| "loss": 5.2467, |
| "mean_token_accuracy": 0.8991876218467951, |
| "num_tokens": 58168938.0, |
| "step": 8180 |
| }, |
| { |
| "entropy": 0.32676368644461035, |
| "epoch": 3.276, |
| "grad_norm": 64.0, |
| "learning_rate": 4.024444444444445e-06, |
| "loss": 5.1019, |
| "mean_token_accuracy": 0.9029897533357143, |
| "num_tokens": 58235885.0, |
| "step": 8190 |
| }, |
| { |
| "entropy": 0.3010506847407669, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 67.0, |
| "learning_rate": 4.002222222222222e-06, |
| "loss": 4.7349, |
| "mean_token_accuracy": 0.9098265107721091, |
| "num_tokens": 58304346.0, |
| "step": 8200 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_biology_entropy": 0.4539010265171528, |
| "eval_biology_loss": 2.519207000732422, |
| "eval_biology_mean_token_accuracy": 0.643939290523529, |
| "eval_biology_num_tokens": 58304346.0, |
| "eval_biology_runtime": 43.097, |
| "eval_biology_samples_per_second": 11.602, |
| "eval_biology_steps_per_second": 11.602, |
| "step": 8200 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_math_entropy": 0.37609435790777207, |
| "eval_math_loss": 0.9081472158432007, |
| "eval_math_mean_token_accuracy": 0.8089789019823075, |
| "eval_math_num_tokens": 58304346.0, |
| "eval_math_runtime": 51.1045, |
| "eval_math_samples_per_second": 9.784, |
| "eval_math_steps_per_second": 9.784, |
| "step": 8200 |
| }, |
| { |
| "entropy": 0.31467316136695445, |
| "epoch": 3.284, |
| "grad_norm": 53.5, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 4.8935, |
| "mean_token_accuracy": 0.9085107143968344, |
| "num_tokens": 58368912.0, |
| "step": 8210 |
| }, |
| { |
| "entropy": 0.3181184314657003, |
| "epoch": 3.288, |
| "grad_norm": 54.5, |
| "learning_rate": 3.9577777777777785e-06, |
| "loss": 5.0762, |
| "mean_token_accuracy": 0.905138299614191, |
| "num_tokens": 58438385.0, |
| "step": 8220 |
| }, |
| { |
| "entropy": 0.32442491985857486, |
| "epoch": 3.292, |
| "grad_norm": 60.25, |
| "learning_rate": 3.935555555555556e-06, |
| "loss": 5.0986, |
| "mean_token_accuracy": 0.9026278845965863, |
| "num_tokens": 58509521.0, |
| "step": 8230 |
| }, |
| { |
| "entropy": 0.33693485665135087, |
| "epoch": 3.296, |
| "grad_norm": 71.5, |
| "learning_rate": 3.913333333333334e-06, |
| "loss": 5.2706, |
| "mean_token_accuracy": 0.9006870277225971, |
| "num_tokens": 58579831.0, |
| "step": 8240 |
| }, |
| { |
| "entropy": 0.31584364133886994, |
| "epoch": 3.3, |
| "grad_norm": 64.5, |
| "learning_rate": 3.891111111111111e-06, |
| "loss": 4.9536, |
| "mean_token_accuracy": 0.9037735447287559, |
| "num_tokens": 58651304.0, |
| "step": 8250 |
| }, |
| { |
| "entropy": 0.3204303595703095, |
| "epoch": 3.304, |
| "grad_norm": 57.0, |
| "learning_rate": 3.868888888888889e-06, |
| "loss": 5.052, |
| "mean_token_accuracy": 0.9031656358391047, |
| "num_tokens": 58718635.0, |
| "step": 8260 |
| }, |
| { |
| "entropy": 0.34388326364569366, |
| "epoch": 3.308, |
| "grad_norm": 61.5, |
| "learning_rate": 3.8466666666666665e-06, |
| "loss": 5.3232, |
| "mean_token_accuracy": 0.9000855289399624, |
| "num_tokens": 58789377.0, |
| "step": 8270 |
| }, |
| { |
| "entropy": 0.31120314756408335, |
| "epoch": 3.312, |
| "grad_norm": 55.75, |
| "learning_rate": 3.824444444444444e-06, |
| "loss": 4.842, |
| "mean_token_accuracy": 0.9061641365289688, |
| "num_tokens": 58858426.0, |
| "step": 8280 |
| }, |
| { |
| "entropy": 0.311283694813028, |
| "epoch": 3.316, |
| "grad_norm": 65.0, |
| "learning_rate": 3.8022222222222226e-06, |
| "loss": 4.9103, |
| "mean_token_accuracy": 0.9049538910388947, |
| "num_tokens": 58925787.0, |
| "step": 8290 |
| }, |
| { |
| "entropy": 0.32050827383063735, |
| "epoch": 3.32, |
| "grad_norm": 61.25, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 5.0813, |
| "mean_token_accuracy": 0.9042157370597124, |
| "num_tokens": 59000124.0, |
| "step": 8300 |
| }, |
| { |
| "epoch": 3.32, |
| "eval_biology_entropy": 0.46080008935928346, |
| "eval_biology_loss": 2.493774175643921, |
| "eval_biology_mean_token_accuracy": 0.6444614965319634, |
| "eval_biology_num_tokens": 59000124.0, |
| "eval_biology_runtime": 47.1044, |
| "eval_biology_samples_per_second": 10.615, |
| "eval_biology_steps_per_second": 10.615, |
| "step": 8300 |
| }, |
| { |
| "epoch": 3.32, |
| "eval_math_entropy": 0.3854194475859404, |
| "eval_math_loss": 0.8927767872810364, |
| "eval_math_mean_token_accuracy": 0.8096117595434189, |
| "eval_math_num_tokens": 59000124.0, |
| "eval_math_runtime": 50.1695, |
| "eval_math_samples_per_second": 9.966, |
| "eval_math_steps_per_second": 9.966, |
| "step": 8300 |
| }, |
| { |
| "entropy": 0.32121019610203805, |
| "epoch": 3.324, |
| "grad_norm": 57.0, |
| "learning_rate": 3.757777777777778e-06, |
| "loss": 4.9488, |
| "mean_token_accuracy": 0.9059676744043828, |
| "num_tokens": 59071669.0, |
| "step": 8310 |
| }, |
| { |
| "entropy": 0.322776048257947, |
| "epoch": 3.328, |
| "grad_norm": 63.25, |
| "learning_rate": 3.7355555555555555e-06, |
| "loss": 5.1272, |
| "mean_token_accuracy": 0.9031867858022451, |
| "num_tokens": 59142949.0, |
| "step": 8320 |
| }, |
| { |
| "entropy": 0.3197089948691428, |
| "epoch": 3.332, |
| "grad_norm": 74.0, |
| "learning_rate": 3.713333333333334e-06, |
| "loss": 5.0116, |
| "mean_token_accuracy": 0.9048744354397058, |
| "num_tokens": 59211050.0, |
| "step": 8330 |
| }, |
| { |
| "entropy": 0.3080874865874648, |
| "epoch": 3.336, |
| "grad_norm": 64.0, |
| "learning_rate": 3.6911111111111115e-06, |
| "loss": 4.8536, |
| "mean_token_accuracy": 0.9081154361367225, |
| "num_tokens": 59281180.0, |
| "step": 8340 |
| }, |
| { |
| "entropy": 0.325222243135795, |
| "epoch": 3.34, |
| "grad_norm": 61.5, |
| "learning_rate": 3.668888888888889e-06, |
| "loss": 5.1005, |
| "mean_token_accuracy": 0.9023119129240513, |
| "num_tokens": 59353288.0, |
| "step": 8350 |
| }, |
| { |
| "entropy": 0.32920175404287877, |
| "epoch": 3.344, |
| "grad_norm": 70.5, |
| "learning_rate": 3.6466666666666668e-06, |
| "loss": 5.1863, |
| "mean_token_accuracy": 0.902720507606864, |
| "num_tokens": 59424654.0, |
| "step": 8360 |
| }, |
| { |
| "entropy": 0.3661952294409275, |
| "epoch": 3.348, |
| "grad_norm": 59.25, |
| "learning_rate": 3.624444444444445e-06, |
| "loss": 5.7423, |
| "mean_token_accuracy": 0.8904729921370744, |
| "num_tokens": 59500378.0, |
| "step": 8370 |
| }, |
| { |
| "entropy": 0.32444571228697894, |
| "epoch": 3.352, |
| "grad_norm": 54.5, |
| "learning_rate": 3.6022222222222224e-06, |
| "loss": 5.0703, |
| "mean_token_accuracy": 0.9044077355414629, |
| "num_tokens": 59568201.0, |
| "step": 8380 |
| }, |
| { |
| "entropy": 0.3351575843989849, |
| "epoch": 3.356, |
| "grad_norm": 58.5, |
| "learning_rate": 3.58e-06, |
| "loss": 5.2568, |
| "mean_token_accuracy": 0.9016330175101757, |
| "num_tokens": 59639761.0, |
| "step": 8390 |
| }, |
| { |
| "entropy": 0.32153961188159885, |
| "epoch": 3.36, |
| "grad_norm": 70.0, |
| "learning_rate": 3.5577777777777785e-06, |
| "loss": 5.0975, |
| "mean_token_accuracy": 0.9034799814224244, |
| "num_tokens": 59704876.0, |
| "step": 8400 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_biology_entropy": 0.45955329650640486, |
| "eval_biology_loss": 2.5006103515625, |
| "eval_biology_mean_token_accuracy": 0.6444163115620614, |
| "eval_biology_num_tokens": 59704876.0, |
| "eval_biology_runtime": 41.042, |
| "eval_biology_samples_per_second": 12.183, |
| "eval_biology_steps_per_second": 12.183, |
| "step": 8400 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_math_entropy": 0.3818207141757011, |
| "eval_math_loss": 0.9000546336174011, |
| "eval_math_mean_token_accuracy": 0.8093910000324249, |
| "eval_math_num_tokens": 59704876.0, |
| "eval_math_runtime": 56.9062, |
| "eval_math_samples_per_second": 8.786, |
| "eval_math_steps_per_second": 8.786, |
| "step": 8400 |
| }, |
| { |
| "entropy": 0.3185452503152192, |
| "epoch": 3.364, |
| "grad_norm": 71.0, |
| "learning_rate": 3.535555555555556e-06, |
| "loss": 5.0278, |
| "mean_token_accuracy": 0.9045491706579923, |
| "num_tokens": 59774481.0, |
| "step": 8410 |
| }, |
| { |
| "entropy": 0.32128109140321615, |
| "epoch": 3.368, |
| "grad_norm": 62.75, |
| "learning_rate": 3.5133333333333337e-06, |
| "loss": 4.9508, |
| "mean_token_accuracy": 0.9060647714883089, |
| "num_tokens": 59845620.0, |
| "step": 8420 |
| }, |
| { |
| "entropy": 0.326228050282225, |
| "epoch": 3.372, |
| "grad_norm": 61.75, |
| "learning_rate": 3.4911111111111113e-06, |
| "loss": 5.2598, |
| "mean_token_accuracy": 0.901841563358903, |
| "num_tokens": 59911603.0, |
| "step": 8430 |
| }, |
| { |
| "entropy": 0.3499437707941979, |
| "epoch": 3.376, |
| "grad_norm": 56.25, |
| "learning_rate": 3.4688888888888894e-06, |
| "loss": 5.5278, |
| "mean_token_accuracy": 0.8960554044693708, |
| "num_tokens": 59988211.0, |
| "step": 8440 |
| }, |
| { |
| "entropy": 0.3221739827189595, |
| "epoch": 3.38, |
| "grad_norm": 68.5, |
| "learning_rate": 3.446666666666667e-06, |
| "loss": 4.9963, |
| "mean_token_accuracy": 0.9054488070309162, |
| "num_tokens": 60058387.0, |
| "step": 8450 |
| }, |
| { |
| "entropy": 0.3187526231165975, |
| "epoch": 3.384, |
| "grad_norm": 83.0, |
| "learning_rate": 3.4244444444444446e-06, |
| "loss": 5.0363, |
| "mean_token_accuracy": 0.9045701380819082, |
| "num_tokens": 60133127.0, |
| "step": 8460 |
| }, |
| { |
| "entropy": 0.30613085981458427, |
| "epoch": 3.388, |
| "grad_norm": 55.0, |
| "learning_rate": 3.4022222222222222e-06, |
| "loss": 4.7594, |
| "mean_token_accuracy": 0.9095892880111933, |
| "num_tokens": 60206568.0, |
| "step": 8470 |
| }, |
| { |
| "entropy": 0.31459893845021725, |
| "epoch": 3.392, |
| "grad_norm": 71.5, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 4.9252, |
| "mean_token_accuracy": 0.9050246458500624, |
| "num_tokens": 60276606.0, |
| "step": 8480 |
| }, |
| { |
| "entropy": 0.32208725847303865, |
| "epoch": 3.396, |
| "grad_norm": 59.75, |
| "learning_rate": 3.3577777777777783e-06, |
| "loss": 5.1952, |
| "mean_token_accuracy": 0.9004988580942154, |
| "num_tokens": 60351006.0, |
| "step": 8490 |
| }, |
| { |
| "entropy": 0.30095921964384614, |
| "epoch": 3.4, |
| "grad_norm": 60.5, |
| "learning_rate": 3.335555555555556e-06, |
| "loss": 4.6436, |
| "mean_token_accuracy": 0.9135570351034403, |
| "num_tokens": 60419419.0, |
| "step": 8500 |
| }, |
| { |
| "epoch": 3.4, |
| "eval_biology_entropy": 0.4618914158642292, |
| "eval_biology_loss": 2.4888687133789062, |
| "eval_biology_mean_token_accuracy": 0.6445204463601112, |
| "eval_biology_num_tokens": 60419419.0, |
| "eval_biology_runtime": 40.8341, |
| "eval_biology_samples_per_second": 12.245, |
| "eval_biology_steps_per_second": 12.245, |
| "step": 8500 |
| }, |
| { |
| "epoch": 3.4, |
| "eval_math_entropy": 0.38344445544481276, |
| "eval_math_loss": 0.8954219818115234, |
| "eval_math_mean_token_accuracy": 0.8096882755756378, |
| "eval_math_num_tokens": 60419419.0, |
| "eval_math_runtime": 48.2854, |
| "eval_math_samples_per_second": 10.355, |
| "eval_math_steps_per_second": 10.355, |
| "step": 8500 |
| }, |
| { |
| "entropy": 0.3422982081305236, |
| "epoch": 3.404, |
| "grad_norm": 77.0, |
| "learning_rate": 3.3133333333333335e-06, |
| "loss": 5.3932, |
| "mean_token_accuracy": 0.8967350706458092, |
| "num_tokens": 60494360.0, |
| "step": 8510 |
| }, |
| { |
| "entropy": 0.2991619905922562, |
| "epoch": 3.408, |
| "grad_norm": 55.0, |
| "learning_rate": 3.2911111111111116e-06, |
| "loss": 4.7409, |
| "mean_token_accuracy": 0.909938033297658, |
| "num_tokens": 60560228.0, |
| "step": 8520 |
| }, |
| { |
| "entropy": 0.3384838871657848, |
| "epoch": 3.412, |
| "grad_norm": 55.25, |
| "learning_rate": 3.268888888888889e-06, |
| "loss": 5.3135, |
| "mean_token_accuracy": 0.8985142655670643, |
| "num_tokens": 60633677.0, |
| "step": 8530 |
| }, |
| { |
| "entropy": 0.33267364245839415, |
| "epoch": 3.416, |
| "grad_norm": 56.5, |
| "learning_rate": 3.2466666666666668e-06, |
| "loss": 5.2464, |
| "mean_token_accuracy": 0.8996383219957351, |
| "num_tokens": 60704176.0, |
| "step": 8540 |
| }, |
| { |
| "entropy": 0.3180673863273114, |
| "epoch": 3.42, |
| "grad_norm": 60.0, |
| "learning_rate": 3.2244444444444444e-06, |
| "loss": 4.942, |
| "mean_token_accuracy": 0.906225797906518, |
| "num_tokens": 60774355.0, |
| "step": 8550 |
| }, |
| { |
| "entropy": 0.3137379629071802, |
| "epoch": 3.424, |
| "grad_norm": 56.75, |
| "learning_rate": 3.202222222222223e-06, |
| "loss": 4.9836, |
| "mean_token_accuracy": 0.9062975227832795, |
| "num_tokens": 60845893.0, |
| "step": 8560 |
| }, |
| { |
| "entropy": 0.3455097742844373, |
| "epoch": 3.428, |
| "grad_norm": 67.5, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 5.4394, |
| "mean_token_accuracy": 0.8962368164211512, |
| "num_tokens": 60919804.0, |
| "step": 8570 |
| }, |
| { |
| "entropy": 0.2858659246005118, |
| "epoch": 3.432, |
| "grad_norm": 60.75, |
| "learning_rate": 3.157777777777778e-06, |
| "loss": 4.4308, |
| "mean_token_accuracy": 0.916533924266696, |
| "num_tokens": 60986244.0, |
| "step": 8580 |
| }, |
| { |
| "entropy": 0.31317990412935615, |
| "epoch": 3.436, |
| "grad_norm": 53.75, |
| "learning_rate": 3.1355555555555557e-06, |
| "loss": 4.903, |
| "mean_token_accuracy": 0.9058194417506457, |
| "num_tokens": 61053779.0, |
| "step": 8590 |
| }, |
| { |
| "entropy": 0.3295704510528594, |
| "epoch": 3.44, |
| "grad_norm": 62.25, |
| "learning_rate": 3.1133333333333337e-06, |
| "loss": 5.214, |
| "mean_token_accuracy": 0.90303308442235, |
| "num_tokens": 61126410.0, |
| "step": 8600 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_biology_entropy": 0.45893249300122263, |
| "eval_biology_loss": 2.4983081817626953, |
| "eval_biology_mean_token_accuracy": 0.6446919196248054, |
| "eval_biology_num_tokens": 61126410.0, |
| "eval_biology_runtime": 41.5094, |
| "eval_biology_samples_per_second": 12.045, |
| "eval_biology_steps_per_second": 12.045, |
| "step": 8600 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_math_entropy": 0.3818259998857975, |
| "eval_math_loss": 0.8984816074371338, |
| "eval_math_mean_token_accuracy": 0.8096093845367431, |
| "eval_math_num_tokens": 61126410.0, |
| "eval_math_runtime": 48.517, |
| "eval_math_samples_per_second": 10.306, |
| "eval_math_steps_per_second": 10.306, |
| "step": 8600 |
| }, |
| { |
| "entropy": 0.3354396115988493, |
| "epoch": 3.444, |
| "grad_norm": 56.75, |
| "learning_rate": 3.0911111111111114e-06, |
| "loss": 5.2871, |
| "mean_token_accuracy": 0.9007753081619739, |
| "num_tokens": 61197846.0, |
| "step": 8610 |
| }, |
| { |
| "entropy": 0.3385889260098338, |
| "epoch": 3.448, |
| "grad_norm": 66.5, |
| "learning_rate": 3.068888888888889e-06, |
| "loss": 5.2063, |
| "mean_token_accuracy": 0.901242345944047, |
| "num_tokens": 61271920.0, |
| "step": 8620 |
| }, |
| { |
| "entropy": 0.3089803828392178, |
| "epoch": 3.452, |
| "grad_norm": 83.5, |
| "learning_rate": 3.0466666666666666e-06, |
| "loss": 4.8301, |
| "mean_token_accuracy": 0.906913922727108, |
| "num_tokens": 61338272.0, |
| "step": 8630 |
| }, |
| { |
| "entropy": 0.3152537756599486, |
| "epoch": 3.456, |
| "grad_norm": 57.25, |
| "learning_rate": 3.024444444444445e-06, |
| "loss": 4.8629, |
| "mean_token_accuracy": 0.9073093626648188, |
| "num_tokens": 61406501.0, |
| "step": 8640 |
| }, |
| { |
| "entropy": 0.33658872786909344, |
| "epoch": 3.46, |
| "grad_norm": 52.75, |
| "learning_rate": 3.0022222222222227e-06, |
| "loss": 5.3616, |
| "mean_token_accuracy": 0.8991640008985996, |
| "num_tokens": 61480130.0, |
| "step": 8650 |
| }, |
| { |
| "entropy": 0.3485852759797126, |
| "epoch": 3.464, |
| "grad_norm": 56.75, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 5.4684, |
| "mean_token_accuracy": 0.8964336782693862, |
| "num_tokens": 61552208.0, |
| "step": 8660 |
| }, |
| { |
| "entropy": 0.30440401141531764, |
| "epoch": 3.468, |
| "grad_norm": 60.25, |
| "learning_rate": 2.957777777777778e-06, |
| "loss": 4.8814, |
| "mean_token_accuracy": 0.9077559147030115, |
| "num_tokens": 61621502.0, |
| "step": 8670 |
| }, |
| { |
| "entropy": 0.35066906227730216, |
| "epoch": 3.472, |
| "grad_norm": 53.5, |
| "learning_rate": 2.935555555555556e-06, |
| "loss": 5.613, |
| "mean_token_accuracy": 0.8939222235232591, |
| "num_tokens": 61693412.0, |
| "step": 8680 |
| }, |
| { |
| "entropy": 0.34795715771615504, |
| "epoch": 3.476, |
| "grad_norm": 61.5, |
| "learning_rate": 2.9133333333333335e-06, |
| "loss": 5.4236, |
| "mean_token_accuracy": 0.8969675276428462, |
| "num_tokens": 61768453.0, |
| "step": 8690 |
| }, |
| { |
| "entropy": 0.3176584153901786, |
| "epoch": 3.48, |
| "grad_norm": 55.5, |
| "learning_rate": 2.891111111111111e-06, |
| "loss": 4.932, |
| "mean_token_accuracy": 0.9064708258956671, |
| "num_tokens": 61837105.0, |
| "step": 8700 |
| }, |
| { |
| "epoch": 3.48, |
| "eval_biology_entropy": 0.4575452301502228, |
| "eval_biology_loss": 2.5073909759521484, |
| "eval_biology_mean_token_accuracy": 0.6444264451265335, |
| "eval_biology_num_tokens": 61837105.0, |
| "eval_biology_runtime": 40.1178, |
| "eval_biology_samples_per_second": 12.463, |
| "eval_biology_steps_per_second": 12.463, |
| "step": 8700 |
| }, |
| { |
| "epoch": 3.48, |
| "eval_math_entropy": 0.38166464218497276, |
| "eval_math_loss": 0.9021702408790588, |
| "eval_math_mean_token_accuracy": 0.8093274117708206, |
| "eval_math_num_tokens": 61837105.0, |
| "eval_math_runtime": 47.8425, |
| "eval_math_samples_per_second": 10.451, |
| "eval_math_steps_per_second": 10.451, |
| "step": 8700 |
| }, |
| { |
| "entropy": 0.309402941679582, |
| "epoch": 3.484, |
| "grad_norm": 58.75, |
| "learning_rate": 2.868888888888889e-06, |
| "loss": 4.8677, |
| "mean_token_accuracy": 0.9072772365063428, |
| "num_tokens": 61906773.0, |
| "step": 8710 |
| }, |
| { |
| "entropy": 0.323706172965467, |
| "epoch": 3.488, |
| "grad_norm": 65.0, |
| "learning_rate": 2.8466666666666672e-06, |
| "loss": 5.0908, |
| "mean_token_accuracy": 0.9036669682711362, |
| "num_tokens": 61977303.0, |
| "step": 8720 |
| }, |
| { |
| "entropy": 0.33917504535056653, |
| "epoch": 3.492, |
| "grad_norm": 50.25, |
| "learning_rate": 2.824444444444445e-06, |
| "loss": 5.3339, |
| "mean_token_accuracy": 0.8993751354515552, |
| "num_tokens": 62048872.0, |
| "step": 8730 |
| }, |
| { |
| "entropy": 0.32198751070536674, |
| "epoch": 3.496, |
| "grad_norm": 58.5, |
| "learning_rate": 2.8022222222222225e-06, |
| "loss": 5.1126, |
| "mean_token_accuracy": 0.9029149748384953, |
| "num_tokens": 62118423.0, |
| "step": 8740 |
| }, |
| { |
| "entropy": 0.3269650877919048, |
| "epoch": 3.5, |
| "grad_norm": 63.0, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 5.1376, |
| "mean_token_accuracy": 0.9015126034617424, |
| "num_tokens": 62192083.0, |
| "step": 8750 |
| }, |
| { |
| "entropy": 0.31776649891398845, |
| "epoch": 3.504, |
| "grad_norm": 60.75, |
| "learning_rate": 2.757777777777778e-06, |
| "loss": 4.9552, |
| "mean_token_accuracy": 0.9048555921763182, |
| "num_tokens": 62265256.0, |
| "step": 8760 |
| }, |
| { |
| "entropy": 0.3078855854459107, |
| "epoch": 3.508, |
| "grad_norm": 68.0, |
| "learning_rate": 2.7355555555555557e-06, |
| "loss": 4.8388, |
| "mean_token_accuracy": 0.9083955936133862, |
| "num_tokens": 62332700.0, |
| "step": 8770 |
| }, |
| { |
| "entropy": 0.329991118516773, |
| "epoch": 3.512, |
| "grad_norm": 61.5, |
| "learning_rate": 2.7133333333333333e-06, |
| "loss": 5.161, |
| "mean_token_accuracy": 0.9019669067114592, |
| "num_tokens": 62406574.0, |
| "step": 8780 |
| }, |
| { |
| "entropy": 0.3270020863041282, |
| "epoch": 3.516, |
| "grad_norm": 64.0, |
| "learning_rate": 2.6911111111111114e-06, |
| "loss": 5.1009, |
| "mean_token_accuracy": 0.9015716716647149, |
| "num_tokens": 62478198.0, |
| "step": 8790 |
| }, |
| { |
| "entropy": 0.3344034475274384, |
| "epoch": 3.52, |
| "grad_norm": 59.0, |
| "learning_rate": 2.6688888888888894e-06, |
| "loss": 5.3657, |
| "mean_token_accuracy": 0.8987553268671036, |
| "num_tokens": 62554970.0, |
| "step": 8800 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_biology_entropy": 0.4630870677828789, |
| "eval_biology_loss": 2.489217758178711, |
| "eval_biology_mean_token_accuracy": 0.6445771997570992, |
| "eval_biology_num_tokens": 62554970.0, |
| "eval_biology_runtime": 41.3949, |
| "eval_biology_samples_per_second": 12.079, |
| "eval_biology_steps_per_second": 12.079, |
| "step": 8800 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_math_entropy": 0.38753711888194087, |
| "eval_math_loss": 0.8926445841789246, |
| "eval_math_mean_token_accuracy": 0.809493475317955, |
| "eval_math_num_tokens": 62554970.0, |
| "eval_math_runtime": 48.2564, |
| "eval_math_samples_per_second": 10.361, |
| "eval_math_steps_per_second": 10.361, |
| "step": 8800 |
| }, |
| { |
| "entropy": 0.33188236416317524, |
| "epoch": 3.524, |
| "grad_norm": 56.75, |
| "learning_rate": 2.646666666666667e-06, |
| "loss": 5.1661, |
| "mean_token_accuracy": 0.9023394897580147, |
| "num_tokens": 62625368.0, |
| "step": 8810 |
| }, |
| { |
| "entropy": 0.3277266817865893, |
| "epoch": 3.528, |
| "grad_norm": 59.5, |
| "learning_rate": 2.6244444444444446e-06, |
| "loss": 5.004, |
| "mean_token_accuracy": 0.9045140277594328, |
| "num_tokens": 62694627.0, |
| "step": 8820 |
| }, |
| { |
| "entropy": 0.32899440452456474, |
| "epoch": 3.532, |
| "grad_norm": 64.5, |
| "learning_rate": 2.6022222222222227e-06, |
| "loss": 5.0702, |
| "mean_token_accuracy": 0.9026375714689493, |
| "num_tokens": 62768033.0, |
| "step": 8830 |
| }, |
| { |
| "entropy": 0.32320210337638855, |
| "epoch": 3.536, |
| "grad_norm": 73.5, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 5.1967, |
| "mean_token_accuracy": 0.9000289402902126, |
| "num_tokens": 62837055.0, |
| "step": 8840 |
| }, |
| { |
| "entropy": 0.3478081680368632, |
| "epoch": 3.54, |
| "grad_norm": 58.0, |
| "learning_rate": 2.557777777777778e-06, |
| "loss": 5.3842, |
| "mean_token_accuracy": 0.8974248945713044, |
| "num_tokens": 62910139.0, |
| "step": 8850 |
| }, |
| { |
| "entropy": 0.3250414053443819, |
| "epoch": 3.544, |
| "grad_norm": 54.25, |
| "learning_rate": 2.5355555555555555e-06, |
| "loss": 5.0988, |
| "mean_token_accuracy": 0.9036026947200299, |
| "num_tokens": 62986593.0, |
| "step": 8860 |
| }, |
| { |
| "entropy": 0.3140929968096316, |
| "epoch": 3.548, |
| "grad_norm": 53.25, |
| "learning_rate": 2.5133333333333336e-06, |
| "loss": 5.0216, |
| "mean_token_accuracy": 0.9068751383572817, |
| "num_tokens": 63056975.0, |
| "step": 8870 |
| }, |
| { |
| "entropy": 0.3189876807387918, |
| "epoch": 3.552, |
| "grad_norm": 48.75, |
| "learning_rate": 2.491111111111111e-06, |
| "loss": 5.0587, |
| "mean_token_accuracy": 0.9045934800058604, |
| "num_tokens": 63129767.0, |
| "step": 8880 |
| }, |
| { |
| "entropy": 0.34246243285015227, |
| "epoch": 3.556, |
| "grad_norm": 69.5, |
| "learning_rate": 2.468888888888889e-06, |
| "loss": 5.3604, |
| "mean_token_accuracy": 0.8986260533332825, |
| "num_tokens": 63204642.0, |
| "step": 8890 |
| }, |
| { |
| "entropy": 0.3256260816939175, |
| "epoch": 3.56, |
| "grad_norm": 69.0, |
| "learning_rate": 2.446666666666667e-06, |
| "loss": 5.0916, |
| "mean_token_accuracy": 0.9034204427152872, |
| "num_tokens": 63278003.0, |
| "step": 8900 |
| }, |
| { |
| "epoch": 3.56, |
| "eval_biology_entropy": 0.45685322165489195, |
| "eval_biology_loss": 2.510265588760376, |
| "eval_biology_mean_token_accuracy": 0.6443586536645889, |
| "eval_biology_num_tokens": 63278003.0, |
| "eval_biology_runtime": 40.3075, |
| "eval_biology_samples_per_second": 12.405, |
| "eval_biology_steps_per_second": 12.405, |
| "step": 8900 |
| }, |
| { |
| "epoch": 3.56, |
| "eval_math_entropy": 0.38097130963206294, |
| "eval_math_loss": 0.9018134474754333, |
| "eval_math_mean_token_accuracy": 0.8094553341865539, |
| "eval_math_num_tokens": 63278003.0, |
| "eval_math_runtime": 47.948, |
| "eval_math_samples_per_second": 10.428, |
| "eval_math_steps_per_second": 10.428, |
| "step": 8900 |
| }, |
| { |
| "entropy": 0.3001694985665381, |
| "epoch": 3.564, |
| "grad_norm": 64.5, |
| "learning_rate": 2.4244444444444444e-06, |
| "loss": 4.7241, |
| "mean_token_accuracy": 0.909381739422679, |
| "num_tokens": 63348595.0, |
| "step": 8910 |
| }, |
| { |
| "entropy": 0.32415162757970395, |
| "epoch": 3.568, |
| "grad_norm": 64.5, |
| "learning_rate": 2.4022222222222225e-06, |
| "loss": 5.1221, |
| "mean_token_accuracy": 0.9029840309172869, |
| "num_tokens": 63424239.0, |
| "step": 8920 |
| }, |
| { |
| "entropy": 0.3208468010649085, |
| "epoch": 3.572, |
| "grad_norm": 64.5, |
| "learning_rate": 2.38e-06, |
| "loss": 5.0152, |
| "mean_token_accuracy": 0.9038519725203514, |
| "num_tokens": 63492765.0, |
| "step": 8930 |
| }, |
| { |
| "entropy": 0.3628962025977671, |
| "epoch": 3.576, |
| "grad_norm": 62.5, |
| "learning_rate": 2.357777777777778e-06, |
| "loss": 5.7444, |
| "mean_token_accuracy": 0.8923406567424536, |
| "num_tokens": 63563563.0, |
| "step": 8940 |
| }, |
| { |
| "entropy": 0.3093442699406296, |
| "epoch": 3.58, |
| "grad_norm": 65.5, |
| "learning_rate": 2.3355555555555557e-06, |
| "loss": 4.797, |
| "mean_token_accuracy": 0.9081469882279635, |
| "num_tokens": 63635002.0, |
| "step": 8950 |
| }, |
| { |
| "entropy": 0.3075421143323183, |
| "epoch": 3.584, |
| "grad_norm": 59.75, |
| "learning_rate": 2.3133333333333333e-06, |
| "loss": 4.7813, |
| "mean_token_accuracy": 0.9090614575892687, |
| "num_tokens": 63704243.0, |
| "step": 8960 |
| }, |
| { |
| "entropy": 0.32340707452967765, |
| "epoch": 3.588, |
| "grad_norm": 63.5, |
| "learning_rate": 2.2911111111111114e-06, |
| "loss": 5.121, |
| "mean_token_accuracy": 0.9032158050686121, |
| "num_tokens": 63777682.0, |
| "step": 8970 |
| }, |
| { |
| "entropy": 0.3268561732955277, |
| "epoch": 3.592, |
| "grad_norm": 54.5, |
| "learning_rate": 2.268888888888889e-06, |
| "loss": 5.1293, |
| "mean_token_accuracy": 0.903079216554761, |
| "num_tokens": 63851832.0, |
| "step": 8980 |
| }, |
| { |
| "entropy": 0.3031598216388375, |
| "epoch": 3.596, |
| "grad_norm": 59.75, |
| "learning_rate": 2.2466666666666666e-06, |
| "loss": 4.756, |
| "mean_token_accuracy": 0.909818933904171, |
| "num_tokens": 63923695.0, |
| "step": 8990 |
| }, |
| { |
| "entropy": 0.3280855935998261, |
| "epoch": 3.6, |
| "grad_norm": 54.5, |
| "learning_rate": 2.2244444444444447e-06, |
| "loss": 5.1088, |
| "mean_token_accuracy": 0.9040178053081036, |
| "num_tokens": 63994604.0, |
| "step": 9000 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_biology_entropy": 0.4595108395814896, |
| "eval_biology_loss": 2.499375343322754, |
| "eval_biology_mean_token_accuracy": 0.6447158735394478, |
| "eval_biology_num_tokens": 63994604.0, |
| "eval_biology_runtime": 40.3519, |
| "eval_biology_samples_per_second": 12.391, |
| "eval_biology_steps_per_second": 12.391, |
| "step": 9000 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_math_entropy": 0.3830903619080782, |
| "eval_math_loss": 0.8974357843399048, |
| "eval_math_mean_token_accuracy": 0.809648564338684, |
| "eval_math_num_tokens": 63994604.0, |
| "eval_math_runtime": 48.1676, |
| "eval_math_samples_per_second": 10.38, |
| "eval_math_steps_per_second": 10.38, |
| "step": 9000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.2911176221910774e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
