| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.1780160423368216, |
| "epoch": 0.016, |
| "grad_norm": 13.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 1.8406, |
| "mean_token_accuracy": 0.6489301804453135, |
| "num_tokens": 340696.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.1818634796887637, |
| "epoch": 0.032, |
| "grad_norm": 11.5625, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 1.8191, |
| "mean_token_accuracy": 0.6528046734631061, |
| "num_tokens": 675930.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.195955842360854, |
| "epoch": 0.048, |
| "grad_norm": 9.3125, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 1.7817, |
| "mean_token_accuracy": 0.6578715395182371, |
| "num_tokens": 1007956.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.2672166559845208, |
| "epoch": 0.064, |
| "grad_norm": 6.5625, |
| "learning_rate": 2.6e-06, |
| "loss": 1.7272, |
| "mean_token_accuracy": 0.6625342659652234, |
| "num_tokens": 1340560.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.2964693307876587, |
| "epoch": 0.08, |
| "grad_norm": 5.0, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 1.5867, |
| "mean_token_accuracy": 0.6811951555311679, |
| "num_tokens": 1679432.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.2539724007248878, |
| "epoch": 0.096, |
| "grad_norm": 3.375, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 1.3962, |
| "mean_token_accuracy": 0.7081016473472118, |
| "num_tokens": 2021570.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.2302619956433773, |
| "epoch": 0.112, |
| "grad_norm": 2.28125, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 1.2974, |
| "mean_token_accuracy": 0.7239668637514114, |
| "num_tokens": 2359297.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.169797332212329, |
| "epoch": 0.128, |
| "grad_norm": 1.875, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 1.2306, |
| "mean_token_accuracy": 0.7328833002597094, |
| "num_tokens": 2683168.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.0165224198251963, |
| "epoch": 0.144, |
| "grad_norm": 1.59375, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 1.1129, |
| "mean_token_accuracy": 0.7561763934791088, |
| "num_tokens": 3019426.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.9457759071141482, |
| "epoch": 0.16, |
| "grad_norm": 1.546875, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 1.0448, |
| "mean_token_accuracy": 0.7662712432444095, |
| "num_tokens": 3354390.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 2.4009983863830566, |
| "eval_biology_loss": 3.090766429901123, |
| "eval_biology_mean_token_accuracy": 0.5075433547496796, |
| "eval_biology_num_tokens": 3354390.0, |
| "eval_biology_runtime": 38.801, |
| "eval_biology_samples_per_second": 12.886, |
| "eval_biology_steps_per_second": 3.222, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 1.1818295245170594, |
| "eval_chemistry_loss": 1.4003204107284546, |
| "eval_chemistry_mean_token_accuracy": 0.7158680257797241, |
| "eval_chemistry_num_tokens": 3354390.0, |
| "eval_chemistry_runtime": 48.2819, |
| "eval_chemistry_samples_per_second": 10.356, |
| "eval_chemistry_steps_per_second": 2.589, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 0.8902422695159912, |
| "eval_math_loss": 1.2323389053344727, |
| "eval_math_mean_token_accuracy": 0.750218500137329, |
| "eval_math_num_tokens": 3354390.0, |
| "eval_math_runtime": 49.6484, |
| "eval_math_samples_per_second": 10.071, |
| "eval_math_steps_per_second": 2.518, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 0.9318590335845948, |
| "eval_physics_loss": 1.0569193363189697, |
| "eval_physics_mean_token_accuracy": 0.7675474114418029, |
| "eval_physics_num_tokens": 3354390.0, |
| "eval_physics_runtime": 57.1057, |
| "eval_physics_samples_per_second": 8.756, |
| "eval_physics_steps_per_second": 2.189, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.9224597703665495, |
| "epoch": 0.176, |
| "grad_norm": 1.4921875, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 1.0192, |
| "mean_token_accuracy": 0.7696165222674608, |
| "num_tokens": 3684374.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8823430232703686, |
| "epoch": 0.192, |
| "grad_norm": 1.125, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 0.9854, |
| "mean_token_accuracy": 0.7739298477768898, |
| "num_tokens": 4023915.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8393984897062182, |
| "epoch": 0.208, |
| "grad_norm": 1.4375, |
| "learning_rate": 8.6e-06, |
| "loss": 0.9379, |
| "mean_token_accuracy": 0.7841779347509146, |
| "num_tokens": 4352045.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.8179828137159347, |
| "epoch": 0.224, |
| "grad_norm": 1.234375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 0.9191, |
| "mean_token_accuracy": 0.7869349300861359, |
| "num_tokens": 4681371.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.805096386373043, |
| "epoch": 0.24, |
| "grad_norm": 1.09375, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 0.9012, |
| "mean_token_accuracy": 0.7888233289122581, |
| "num_tokens": 5021784.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.798224457167089, |
| "epoch": 0.256, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 0.8922, |
| "mean_token_accuracy": 0.7903898701071739, |
| "num_tokens": 5367308.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.805234762467444, |
| "epoch": 0.272, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 0.8993, |
| "mean_token_accuracy": 0.7877147275954485, |
| "num_tokens": 5699101.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7946224914863705, |
| "epoch": 0.288, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 0.8976, |
| "mean_token_accuracy": 0.7911442808806897, |
| "num_tokens": 6022837.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7837886592373252, |
| "epoch": 0.304, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 0.8805, |
| "mean_token_accuracy": 0.7934082143008709, |
| "num_tokens": 6343826.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.778077344968915, |
| "epoch": 0.32, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 0.8781, |
| "mean_token_accuracy": 0.7940127164125442, |
| "num_tokens": 6670164.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 2.1669651889801025, |
| "eval_biology_loss": 2.686201810836792, |
| "eval_biology_mean_token_accuracy": 0.5460970797538758, |
| "eval_biology_num_tokens": 6670164.0, |
| "eval_biology_runtime": 38.7901, |
| "eval_biology_samples_per_second": 12.89, |
| "eval_biology_steps_per_second": 3.222, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.9981456413269043, |
| "eval_chemistry_loss": 1.1842883825302124, |
| "eval_chemistry_mean_token_accuracy": 0.7467401041984558, |
| "eval_chemistry_num_tokens": 6670164.0, |
| "eval_chemistry_runtime": 48.3154, |
| "eval_chemistry_samples_per_second": 10.349, |
| "eval_chemistry_steps_per_second": 2.587, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 0.7862319254875183, |
| "eval_math_loss": 1.0977362394332886, |
| "eval_math_mean_token_accuracy": 0.7683374147415161, |
| "eval_math_num_tokens": 6670164.0, |
| "eval_math_runtime": 49.687, |
| "eval_math_samples_per_second": 10.063, |
| "eval_math_steps_per_second": 2.516, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 0.7679825134277344, |
| "eval_physics_loss": 0.886142373085022, |
| "eval_physics_mean_token_accuracy": 0.7957743234634399, |
| "eval_physics_num_tokens": 6670164.0, |
| "eval_physics_runtime": 57.1202, |
| "eval_physics_samples_per_second": 8.753, |
| "eval_physics_steps_per_second": 2.188, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7617403082549572, |
| "epoch": 0.336, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 0.8531, |
| "mean_token_accuracy": 0.7977134332060813, |
| "num_tokens": 7003404.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.737130863033235, |
| "epoch": 0.352, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.46e-05, |
| "loss": 0.8339, |
| "mean_token_accuracy": 0.8024804938584567, |
| "num_tokens": 7342593.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7299130430445075, |
| "epoch": 0.368, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 0.8252, |
| "mean_token_accuracy": 0.805871631577611, |
| "num_tokens": 7675569.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7597114410251379, |
| "epoch": 0.384, |
| "grad_norm": 1.171875, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 0.8558, |
| "mean_token_accuracy": 0.7943590730428696, |
| "num_tokens": 8012218.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.7247350050136447, |
| "epoch": 0.4, |
| "grad_norm": 1.125, |
| "learning_rate": 1.66e-05, |
| "loss": 0.8193, |
| "mean_token_accuracy": 0.8040182612836361, |
| "num_tokens": 8350120.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7144488081336021, |
| "epoch": 0.416, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 0.7998, |
| "mean_token_accuracy": 0.8067968346178531, |
| "num_tokens": 8688773.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7226355630904436, |
| "epoch": 0.432, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 0.8125, |
| "mean_token_accuracy": 0.8042349684983492, |
| "num_tokens": 9022094.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.6943683221936225, |
| "epoch": 0.448, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.86e-05, |
| "loss": 0.7924, |
| "mean_token_accuracy": 0.8109067149460316, |
| "num_tokens": 9360608.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.6898288525640964, |
| "epoch": 0.464, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 0.7814, |
| "mean_token_accuracy": 0.8110810052603483, |
| "num_tokens": 9688896.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.6964065950363875, |
| "epoch": 0.48, |
| "grad_norm": 1.125, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 0.7896, |
| "mean_token_accuracy": 0.8094950247555971, |
| "num_tokens": 10025864.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 2.1021561794281007, |
| "eval_biology_loss": 2.546415090560913, |
| "eval_biology_mean_token_accuracy": 0.5616441056728363, |
| "eval_biology_num_tokens": 10025864.0, |
| "eval_biology_runtime": 38.8255, |
| "eval_biology_samples_per_second": 12.878, |
| "eval_biology_steps_per_second": 3.22, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.9578249802589417, |
| "eval_chemistry_loss": 1.1104036569595337, |
| "eval_chemistry_mean_token_accuracy": 0.7582236580848694, |
| "eval_chemistry_num_tokens": 10025864.0, |
| "eval_chemistry_runtime": 48.5149, |
| "eval_chemistry_samples_per_second": 10.306, |
| "eval_chemistry_steps_per_second": 2.577, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.7676350421905518, |
| "eval_math_loss": 1.0549193620681763, |
| "eval_math_mean_token_accuracy": 0.7745008988380432, |
| "eval_math_num_tokens": 10025864.0, |
| "eval_math_runtime": 49.8707, |
| "eval_math_samples_per_second": 10.026, |
| "eval_math_steps_per_second": 2.506, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 0.7307665984630585, |
| "eval_physics_loss": 0.8242064118385315, |
| "eval_physics_mean_token_accuracy": 0.8064617681503295, |
| "eval_physics_num_tokens": 10025864.0, |
| "eval_physics_runtime": 57.1693, |
| "eval_physics_samples_per_second": 8.746, |
| "eval_physics_steps_per_second": 2.186, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7168749757111073, |
| "epoch": 0.496, |
| "grad_norm": 0.9921875, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 0.8032, |
| "mean_token_accuracy": 0.8051175128668546, |
| "num_tokens": 10362573.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.6993480321019888, |
| "epoch": 0.512, |
| "grad_norm": 1.15625, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 0.7903, |
| "mean_token_accuracy": 0.8099954195320607, |
| "num_tokens": 10694440.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.7138712629675865, |
| "epoch": 0.528, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 0.8101, |
| "mean_token_accuracy": 0.805506169050932, |
| "num_tokens": 11022357.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.6928766580298543, |
| "epoch": 0.544, |
| "grad_norm": 1.0546875, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 0.7916, |
| "mean_token_accuracy": 0.8087165944278241, |
| "num_tokens": 11345911.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.682861409150064, |
| "epoch": 0.56, |
| "grad_norm": 0.8984375, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 0.7735, |
| "mean_token_accuracy": 0.8126827124506235, |
| "num_tokens": 11687225.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.6824749782681465, |
| "epoch": 0.576, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 0.7723, |
| "mean_token_accuracy": 0.8146394658833742, |
| "num_tokens": 12015720.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.6805058639496565, |
| "epoch": 0.592, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 0.776, |
| "mean_token_accuracy": 0.811661035567522, |
| "num_tokens": 12343393.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.6700849516317249, |
| "epoch": 0.608, |
| "grad_norm": 0.9921875, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 0.7658, |
| "mean_token_accuracy": 0.8141583666205406, |
| "num_tokens": 12671474.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.6893410481512546, |
| "epoch": 0.624, |
| "grad_norm": 0.98046875, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 0.7782, |
| "mean_token_accuracy": 0.8111887093633413, |
| "num_tokens": 12996077.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.6852772971615195, |
| "epoch": 0.64, |
| "grad_norm": 0.9609375, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 0.7799, |
| "mean_token_accuracy": 0.8112590182572603, |
| "num_tokens": 13325291.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 2.0124985208511355, |
| "eval_biology_loss": 2.4155702590942383, |
| "eval_biology_mean_token_accuracy": 0.5735694291591644, |
| "eval_biology_num_tokens": 13325291.0, |
| "eval_biology_runtime": 38.6667, |
| "eval_biology_samples_per_second": 12.931, |
| "eval_biology_steps_per_second": 3.233, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.8911373369693756, |
| "eval_chemistry_loss": 1.0692609548568726, |
| "eval_chemistry_mean_token_accuracy": 0.7647108516693115, |
| "eval_chemistry_num_tokens": 13325291.0, |
| "eval_chemistry_runtime": 48.2089, |
| "eval_chemistry_samples_per_second": 10.372, |
| "eval_chemistry_steps_per_second": 2.593, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.7330298454761505, |
| "eval_math_loss": 1.0336663722991943, |
| "eval_math_mean_token_accuracy": 0.7786638278961182, |
| "eval_math_num_tokens": 13325291.0, |
| "eval_math_runtime": 49.7905, |
| "eval_math_samples_per_second": 10.042, |
| "eval_math_steps_per_second": 2.511, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.681670261144638, |
| "eval_physics_loss": 0.7883204817771912, |
| "eval_physics_mean_token_accuracy": 0.8129381031990052, |
| "eval_physics_num_tokens": 13325291.0, |
| "eval_physics_runtime": 57.2563, |
| "eval_physics_samples_per_second": 8.733, |
| "eval_physics_steps_per_second": 2.183, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.6723099524155259, |
| "epoch": 0.656, |
| "grad_norm": 0.953125, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 0.757, |
| "mean_token_accuracy": 0.8151150114834309, |
| "num_tokens": 13671434.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.6621538577601314, |
| "epoch": 0.672, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 0.7563, |
| "mean_token_accuracy": 0.8172798678278923, |
| "num_tokens": 13993668.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.6673029117286206, |
| "epoch": 0.688, |
| "grad_norm": 0.921875, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 0.7529, |
| "mean_token_accuracy": 0.8164214458316564, |
| "num_tokens": 14334907.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.6810694945976138, |
| "epoch": 0.704, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 0.7695, |
| "mean_token_accuracy": 0.8127802673727273, |
| "num_tokens": 14670639.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.6582699194550514, |
| "epoch": 0.72, |
| "grad_norm": 1.0, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 0.7544, |
| "mean_token_accuracy": 0.8171826928853989, |
| "num_tokens": 14987995.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.6564427128061652, |
| "epoch": 0.736, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 0.7446, |
| "mean_token_accuracy": 0.8173684533685446, |
| "num_tokens": 15324659.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.6495672106742859, |
| "epoch": 0.752, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 0.7416, |
| "mean_token_accuracy": 0.8195757914334536, |
| "num_tokens": 15657342.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.6738376861438156, |
| "epoch": 0.768, |
| "grad_norm": 0.95703125, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 0.7642, |
| "mean_token_accuracy": 0.8124835971742869, |
| "num_tokens": 15988993.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.6495399951934815, |
| "epoch": 0.784, |
| "grad_norm": 0.9609375, |
| "learning_rate": 1.86e-05, |
| "loss": 0.7324, |
| "mean_token_accuracy": 0.8208612345159054, |
| "num_tokens": 16333857.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.6567428983747959, |
| "epoch": 0.8, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 0.7488, |
| "mean_token_accuracy": 0.817277068644762, |
| "num_tokens": 16664690.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 2.039887357711792, |
| "eval_biology_loss": 2.4157471656799316, |
| "eval_biology_mean_token_accuracy": 0.5770127189159393, |
| "eval_biology_num_tokens": 16664690.0, |
| "eval_biology_runtime": 38.6815, |
| "eval_biology_samples_per_second": 12.926, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.8955893061161041, |
| "eval_chemistry_loss": 1.048737645149231, |
| "eval_chemistry_mean_token_accuracy": 0.768093888759613, |
| "eval_chemistry_num_tokens": 16664690.0, |
| "eval_chemistry_runtime": 48.42, |
| "eval_chemistry_samples_per_second": 10.326, |
| "eval_chemistry_steps_per_second": 2.582, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.7318526220321655, |
| "eval_math_loss": 1.0199034214019775, |
| "eval_math_mean_token_accuracy": 0.7799897599220276, |
| "eval_math_num_tokens": 16664690.0, |
| "eval_math_runtime": 49.8882, |
| "eval_math_samples_per_second": 10.022, |
| "eval_math_steps_per_second": 2.506, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.6712099099159241, |
| "eval_physics_loss": 0.7689476609230042, |
| "eval_physics_mean_token_accuracy": 0.8163677668571472, |
| "eval_physics_num_tokens": 16664690.0, |
| "eval_physics_runtime": 57.4517, |
| "eval_physics_samples_per_second": 8.703, |
| "eval_physics_steps_per_second": 2.176, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.6573688389733434, |
| "epoch": 0.816, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 0.7477, |
| "mean_token_accuracy": 0.8164514761418105, |
| "num_tokens": 16989307.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.6648645078763366, |
| "epoch": 0.832, |
| "grad_norm": 0.9609375, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 0.749, |
| "mean_token_accuracy": 0.8170726090669632, |
| "num_tokens": 17317524.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.6526606786996126, |
| "epoch": 0.848, |
| "grad_norm": 0.8515625, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 0.7388, |
| "mean_token_accuracy": 0.8195879191160202, |
| "num_tokens": 17650424.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.6565553491935134, |
| "epoch": 0.864, |
| "grad_norm": 0.8984375, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 0.7464, |
| "mean_token_accuracy": 0.8177060768008232, |
| "num_tokens": 17985547.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.6623865978792309, |
| "epoch": 0.88, |
| "grad_norm": 0.9453125, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 0.7495, |
| "mean_token_accuracy": 0.8162953305989504, |
| "num_tokens": 18315392.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.6179373754188419, |
| "epoch": 0.896, |
| "grad_norm": 0.8359375, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 0.7023, |
| "mean_token_accuracy": 0.8264330130070447, |
| "num_tokens": 18667124.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.6555242039263248, |
| "epoch": 0.912, |
| "grad_norm": 0.8984375, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 0.7393, |
| "mean_token_accuracy": 0.8174499638378621, |
| "num_tokens": 19002186.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.6485247412696481, |
| "epoch": 0.928, |
| "grad_norm": 0.98046875, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 0.7363, |
| "mean_token_accuracy": 0.8195535041391849, |
| "num_tokens": 19335013.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.63772834520787, |
| "epoch": 0.944, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 0.7269, |
| "mean_token_accuracy": 0.821834321692586, |
| "num_tokens": 19662733.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6372630735859275, |
| "epoch": 0.96, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 0.7318, |
| "mean_token_accuracy": 0.8204116970300674, |
| "num_tokens": 19996789.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 2.0030080833435058, |
| "eval_biology_loss": 2.361283302307129, |
| "eval_biology_mean_token_accuracy": 0.5793048655986786, |
| "eval_biology_num_tokens": 19996789.0, |
| "eval_biology_runtime": 38.6981, |
| "eval_biology_samples_per_second": 12.921, |
| "eval_biology_steps_per_second": 3.23, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.8840513801574708, |
| "eval_chemistry_loss": 1.0323643684387207, |
| "eval_chemistry_mean_token_accuracy": 0.7703397722244263, |
| "eval_chemistry_num_tokens": 19996789.0, |
| "eval_chemistry_runtime": 48.2039, |
| "eval_chemistry_samples_per_second": 10.373, |
| "eval_chemistry_steps_per_second": 2.593, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.7270034260749817, |
| "eval_math_loss": 1.009481430053711, |
| "eval_math_mean_token_accuracy": 0.7820751585960388, |
| "eval_math_num_tokens": 19996789.0, |
| "eval_math_runtime": 49.571, |
| "eval_math_samples_per_second": 10.087, |
| "eval_math_steps_per_second": 2.522, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.6698693735599518, |
| "eval_physics_loss": 0.7564050555229187, |
| "eval_physics_mean_token_accuracy": 0.8185040464401245, |
| "eval_physics_num_tokens": 19996789.0, |
| "eval_physics_runtime": 56.9517, |
| "eval_physics_samples_per_second": 8.779, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6384279150515795, |
| "epoch": 0.976, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 0.7283, |
| "mean_token_accuracy": 0.82187210470438, |
| "num_tokens": 20332636.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.6450063675642014, |
| "epoch": 0.992, |
| "grad_norm": 0.8515625, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 0.7272, |
| "mean_token_accuracy": 0.8209770727902651, |
| "num_tokens": 20662209.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.6167608626186848, |
| "epoch": 1.008, |
| "grad_norm": 0.8984375, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 0.7008, |
| "mean_token_accuracy": 0.825862829759717, |
| "num_tokens": 21005731.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.6117491278797388, |
| "epoch": 1.024, |
| "grad_norm": 0.93359375, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 0.7058, |
| "mean_token_accuracy": 0.8266724064946175, |
| "num_tokens": 21335407.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6068670526146889, |
| "epoch": 1.04, |
| "grad_norm": 0.97265625, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 0.6989, |
| "mean_token_accuracy": 0.8283324401825667, |
| "num_tokens": 21668490.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.601451874896884, |
| "epoch": 1.056, |
| "grad_norm": 0.95703125, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 0.6864, |
| "mean_token_accuracy": 0.829675118252635, |
| "num_tokens": 22006039.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.6137524953112006, |
| "epoch": 1.072, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 0.7008, |
| "mean_token_accuracy": 0.8262542523443699, |
| "num_tokens": 22348206.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6093455260619521, |
| "epoch": 1.088, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 0.6956, |
| "mean_token_accuracy": 0.8260304640978575, |
| "num_tokens": 22676163.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6026462253183127, |
| "epoch": 1.104, |
| "grad_norm": 0.87109375, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 0.6967, |
| "mean_token_accuracy": 0.8273924000561237, |
| "num_tokens": 23009279.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6122452523559332, |
| "epoch": 1.12, |
| "grad_norm": 0.9453125, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 0.7, |
| "mean_token_accuracy": 0.8257350366562605, |
| "num_tokens": 23346828.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.8616097135543823, |
| "eval_biology_loss": 2.4014246463775635, |
| "eval_biology_mean_token_accuracy": 0.578062358379364, |
| "eval_biology_num_tokens": 23346828.0, |
| "eval_biology_runtime": 38.9662, |
| "eval_biology_samples_per_second": 12.832, |
| "eval_biology_steps_per_second": 3.208, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.8116880433559418, |
| "eval_chemistry_loss": 1.0317949056625366, |
| "eval_chemistry_mean_token_accuracy": 0.7715164208412171, |
| "eval_chemistry_num_tokens": 23346828.0, |
| "eval_chemistry_runtime": 48.496, |
| "eval_chemistry_samples_per_second": 10.31, |
| "eval_chemistry_steps_per_second": 2.578, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.6895896532535553, |
| "eval_math_loss": 1.0153776407241821, |
| "eval_math_mean_token_accuracy": 0.7820956745147705, |
| "eval_math_num_tokens": 23346828.0, |
| "eval_math_runtime": 49.7815, |
| "eval_math_samples_per_second": 10.044, |
| "eval_math_steps_per_second": 2.511, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.6192426791191101, |
| "eval_physics_loss": 0.7497905492782593, |
| "eval_physics_mean_token_accuracy": 0.8200045080184937, |
| "eval_physics_num_tokens": 23346828.0, |
| "eval_physics_runtime": 57.1287, |
| "eval_physics_samples_per_second": 8.752, |
| "eval_physics_steps_per_second": 2.188, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.588855667039752, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 0.9375, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 0.6772, |
| "mean_token_accuracy": 0.8326364874839782, |
| "num_tokens": 23689161.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6099351227283478, |
| "epoch": 1.152, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 0.7039, |
| "mean_token_accuracy": 0.8282962709665298, |
| "num_tokens": 24016439.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6140455640852451, |
| "epoch": 1.168, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 0.7027, |
| "mean_token_accuracy": 0.8253309100866317, |
| "num_tokens": 24344431.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6041248327121138, |
| "epoch": 1.184, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 0.6955, |
| "mean_token_accuracy": 0.8286070462316275, |
| "num_tokens": 24668087.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.5895567566156388, |
| "epoch": 1.2, |
| "grad_norm": 0.9140625, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 0.6822, |
| "mean_token_accuracy": 0.8305550657212735, |
| "num_tokens": 25000629.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6073450578376651, |
| "epoch": 1.216, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.66e-05, |
| "loss": 0.7023, |
| "mean_token_accuracy": 0.8261869914829731, |
| "num_tokens": 25324324.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6059388216584921, |
| "epoch": 1.232, |
| "grad_norm": 0.9453125, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 0.6863, |
| "mean_token_accuracy": 0.8298216536641121, |
| "num_tokens": 25658795.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.5957553267478943, |
| "epoch": 1.248, |
| "grad_norm": 0.92578125, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 0.6864, |
| "mean_token_accuracy": 0.830686765909195, |
| "num_tokens": 25993177.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.5959657493978738, |
| "epoch": 1.264, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 0.6826, |
| "mean_token_accuracy": 0.8305899318307638, |
| "num_tokens": 26329726.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6204709148034453, |
| "epoch": 1.28, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 0.7071, |
| "mean_token_accuracy": 0.8238335218280554, |
| "num_tokens": 26664093.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 1.7922346639633178, |
| "eval_biology_loss": 2.4233291149139404, |
| "eval_biology_mean_token_accuracy": 0.5778127768039704, |
| "eval_biology_num_tokens": 26664093.0, |
| "eval_biology_runtime": 39.1057, |
| "eval_biology_samples_per_second": 12.786, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.7962674243450165, |
| "eval_chemistry_loss": 1.0286256074905396, |
| "eval_chemistry_mean_token_accuracy": 0.7727626013755798, |
| "eval_chemistry_num_tokens": 26664093.0, |
| "eval_chemistry_runtime": 48.3131, |
| "eval_chemistry_samples_per_second": 10.349, |
| "eval_chemistry_steps_per_second": 2.587, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.6702238636016846, |
| "eval_math_loss": 1.0157709121704102, |
| "eval_math_mean_token_accuracy": 0.7829344019889831, |
| "eval_math_num_tokens": 26664093.0, |
| "eval_math_runtime": 49.7647, |
| "eval_math_samples_per_second": 10.047, |
| "eval_math_steps_per_second": 2.512, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.6072953283786774, |
| "eval_physics_loss": 0.7434503436088562, |
| "eval_physics_mean_token_accuracy": 0.8212288799285888, |
| "eval_physics_num_tokens": 26664093.0, |
| "eval_physics_runtime": 57.1238, |
| "eval_physics_samples_per_second": 8.753, |
| "eval_physics_steps_per_second": 2.188, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.593550406768918, |
| "epoch": 1.296, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 0.6825, |
| "mean_token_accuracy": 0.8320997886359691, |
| "num_tokens": 26994877.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6156477816402912, |
| "epoch": 1.312, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 0.703, |
| "mean_token_accuracy": 0.8259445391595364, |
| "num_tokens": 27328198.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.5969830378890038, |
| "epoch": 1.328, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 0.6901, |
| "mean_token_accuracy": 0.8290002550929785, |
| "num_tokens": 27658818.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.5997996777296066, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 1.1015625, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 0.6904, |
| "mean_token_accuracy": 0.8284455709159374, |
| "num_tokens": 27991108.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6052390130236744, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 0.6852, |
| "mean_token_accuracy": 0.8274706263095141, |
| "num_tokens": 28327775.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6018602728843689, |
| "epoch": 1.376, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 0.6827, |
| "mean_token_accuracy": 0.8304568257182836, |
| "num_tokens": 28666116.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.5851238770410419, |
| "epoch": 1.392, |
| "grad_norm": 0.98828125, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 0.6746, |
| "mean_token_accuracy": 0.8333170894533396, |
| "num_tokens": 29004293.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.58795285820961, |
| "epoch": 1.408, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 0.6857, |
| "mean_token_accuracy": 0.8305332105606794, |
| "num_tokens": 29330131.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6189510561525822, |
| "epoch": 1.424, |
| "grad_norm": 0.90234375, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 0.7072, |
| "mean_token_accuracy": 0.8251793116331101, |
| "num_tokens": 29661850.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.5798567572608591, |
| "epoch": 1.44, |
| "grad_norm": 0.90625, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 0.6695, |
| "mean_token_accuracy": 0.8342564977705479, |
| "num_tokens": 29999392.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 1.8147089042663573, |
| "eval_biology_loss": 2.417410135269165, |
| "eval_biology_mean_token_accuracy": 0.5787473826408386, |
| "eval_biology_num_tokens": 29999392.0, |
| "eval_biology_runtime": 38.9072, |
| "eval_biology_samples_per_second": 12.851, |
| "eval_biology_steps_per_second": 3.213, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.7944381227493286, |
| "eval_chemistry_loss": 1.0228049755096436, |
| "eval_chemistry_mean_token_accuracy": 0.7732948322296143, |
| "eval_chemistry_num_tokens": 29999392.0, |
| "eval_chemistry_runtime": 48.4429, |
| "eval_chemistry_samples_per_second": 10.321, |
| "eval_chemistry_steps_per_second": 2.58, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.6757729785442352, |
| "eval_math_loss": 1.0138108730316162, |
| "eval_math_mean_token_accuracy": 0.7829539861679077, |
| "eval_math_num_tokens": 29999392.0, |
| "eval_math_runtime": 49.813, |
| "eval_math_samples_per_second": 10.038, |
| "eval_math_steps_per_second": 2.509, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.607404308795929, |
| "eval_physics_loss": 0.737089991569519, |
| "eval_physics_mean_token_accuracy": 0.8226114134788514, |
| "eval_physics_num_tokens": 29999392.0, |
| "eval_physics_runtime": 57.3115, |
| "eval_physics_samples_per_second": 8.724, |
| "eval_physics_steps_per_second": 2.181, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.5827234297990799, |
| "epoch": 1.456, |
| "grad_norm": 0.8828125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 0.6728, |
| "mean_token_accuracy": 0.8342432040721178, |
| "num_tokens": 30341004.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.613773494027555, |
| "epoch": 1.472, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 0.7035, |
| "mean_token_accuracy": 0.8268327709287405, |
| "num_tokens": 30667689.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.5988023646175862, |
| "epoch": 1.488, |
| "grad_norm": 0.93359375, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 0.6897, |
| "mean_token_accuracy": 0.8293631616979837, |
| "num_tokens": 30999697.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.5896453000605106, |
| "epoch": 1.504, |
| "grad_norm": 0.8828125, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 0.6728, |
| "mean_token_accuracy": 0.8324251122772693, |
| "num_tokens": 31332775.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.605005569756031, |
| "epoch": 1.52, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 0.6942, |
| "mean_token_accuracy": 0.8275358382612467, |
| "num_tokens": 31666521.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.5999037871137262, |
| "epoch": 1.536, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 0.6904, |
| "mean_token_accuracy": 0.8276842717081309, |
| "num_tokens": 31998232.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.5867987772449851, |
| "epoch": 1.552, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 0.6754, |
| "mean_token_accuracy": 0.8337812848389149, |
| "num_tokens": 32328085.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.5942975046113134, |
| "epoch": 1.568, |
| "grad_norm": 1.0, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 0.6807, |
| "mean_token_accuracy": 0.831771444156766, |
| "num_tokens": 32656603.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.6016290852800011, |
| "epoch": 1.584, |
| "grad_norm": 0.98828125, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 0.6947, |
| "mean_token_accuracy": 0.8264373868703843, |
| "num_tokens": 32983769.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.5999371835961937, |
| "epoch": 1.6, |
| "grad_norm": 0.92578125, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 0.6882, |
| "mean_token_accuracy": 0.8293469067662954, |
| "num_tokens": 33309282.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 1.788435049057007, |
| "eval_biology_loss": 2.369239330291748, |
| "eval_biology_mean_token_accuracy": 0.582524644613266, |
| "eval_biology_num_tokens": 33309282.0, |
| "eval_biology_runtime": 38.8968, |
| "eval_biology_samples_per_second": 12.855, |
| "eval_biology_steps_per_second": 3.214, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.7873289968967437, |
| "eval_chemistry_loss": 1.0177081823349, |
| "eval_chemistry_mean_token_accuracy": 0.774118812084198, |
| "eval_chemistry_num_tokens": 33309282.0, |
| "eval_chemistry_runtime": 48.3649, |
| "eval_chemistry_samples_per_second": 10.338, |
| "eval_chemistry_steps_per_second": 2.585, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.6668027784824372, |
| "eval_math_loss": 1.0126802921295166, |
| "eval_math_mean_token_accuracy": 0.7838437123298645, |
| "eval_math_num_tokens": 33309282.0, |
| "eval_math_runtime": 49.8108, |
| "eval_math_samples_per_second": 10.038, |
| "eval_math_steps_per_second": 2.509, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.5990195829868317, |
| "eval_physics_loss": 0.7323749661445618, |
| "eval_physics_mean_token_accuracy": 0.8238483490943909, |
| "eval_physics_num_tokens": 33309282.0, |
| "eval_physics_runtime": 57.3705, |
| "eval_physics_samples_per_second": 8.715, |
| "eval_physics_steps_per_second": 2.179, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.5766303434967994, |
| "epoch": 1.616, |
| "grad_norm": 1.0, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 0.6687, |
| "mean_token_accuracy": 0.8350894570350647, |
| "num_tokens": 33647616.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.5906545480713248, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 0.6798, |
| "mean_token_accuracy": 0.8322980519384146, |
| "num_tokens": 33977586.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.5677616313099861, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 0.98828125, |
| "learning_rate": 1.46e-05, |
| "loss": 0.658, |
| "mean_token_accuracy": 0.8373314294964075, |
| "num_tokens": 34312436.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.5948906594887375, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 0.682, |
| "mean_token_accuracy": 0.8317620534449816, |
| "num_tokens": 34641549.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.5764713797718286, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 0.94921875, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 0.6715, |
| "mean_token_accuracy": 0.8332184217870235, |
| "num_tokens": 34977031.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.5952808676287532, |
| "epoch": 1.696, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 0.6834, |
| "mean_token_accuracy": 0.8309958126395941, |
| "num_tokens": 35300962.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.5965396504849195, |
| "epoch": 1.712, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 0.6786, |
| "mean_token_accuracy": 0.829968997463584, |
| "num_tokens": 35642662.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.5757137715816498, |
| "epoch": 1.728, |
| "grad_norm": 0.9765625, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 0.6695, |
| "mean_token_accuracy": 0.833732133358717, |
| "num_tokens": 35980993.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.5878799825906753, |
| "epoch": 1.744, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 0.6717, |
| "mean_token_accuracy": 0.8321238547563553, |
| "num_tokens": 36326797.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.577763288281858, |
| "epoch": 1.76, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 0.6698, |
| "mean_token_accuracy": 0.8338598430156707, |
| "num_tokens": 36654797.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 1.76575110912323, |
| "eval_biology_loss": 2.379521369934082, |
| "eval_biology_mean_token_accuracy": 0.5822040309906006, |
| "eval_biology_num_tokens": 36654797.0, |
| "eval_biology_runtime": 38.7201, |
| "eval_biology_samples_per_second": 12.913, |
| "eval_biology_steps_per_second": 3.228, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.7854319989681244, |
| "eval_chemistry_loss": 1.01557457447052, |
| "eval_chemistry_mean_token_accuracy": 0.7744541010856628, |
| "eval_chemistry_num_tokens": 36654797.0, |
| "eval_chemistry_runtime": 48.2464, |
| "eval_chemistry_samples_per_second": 10.363, |
| "eval_chemistry_steps_per_second": 2.591, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.6641156001091003, |
| "eval_math_loss": 1.0110862255096436, |
| "eval_math_mean_token_accuracy": 0.783713164806366, |
| "eval_math_num_tokens": 36654797.0, |
| "eval_math_runtime": 49.5907, |
| "eval_math_samples_per_second": 10.083, |
| "eval_math_steps_per_second": 2.521, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.5953411047458649, |
| "eval_physics_loss": 0.7290456295013428, |
| "eval_physics_mean_token_accuracy": 0.8248090887069702, |
| "eval_physics_num_tokens": 36654797.0, |
| "eval_physics_runtime": 57.3959, |
| "eval_physics_samples_per_second": 8.711, |
| "eval_physics_steps_per_second": 2.178, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.5980557221919298, |
| "epoch": 1.776, |
| "grad_norm": 0.95703125, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 0.6911, |
| "mean_token_accuracy": 0.8282759781926871, |
| "num_tokens": 36989310.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.5839062621816993, |
| "epoch": 1.792, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 0.6708, |
| "mean_token_accuracy": 0.8331681247800589, |
| "num_tokens": 37318535.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.5731059337034822, |
| "epoch": 1.808, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 0.6585, |
| "mean_token_accuracy": 0.8363862674683332, |
| "num_tokens": 37658889.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.5865196855738759, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 0.984375, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 0.6698, |
| "mean_token_accuracy": 0.8333104524761439, |
| "num_tokens": 37995198.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.5773211907595396, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 1.0, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 0.6726, |
| "mean_token_accuracy": 0.8339518435299397, |
| "num_tokens": 38325542.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.5834932073950767, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 0.97265625, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 0.6632, |
| "mean_token_accuracy": 0.8345976937562227, |
| "num_tokens": 38668148.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.5739704865962267, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 0.95703125, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 0.6687, |
| "mean_token_accuracy": 0.8333972290158271, |
| "num_tokens": 39001652.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.5951925914734602, |
| "epoch": 1.888, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 0.682, |
| "mean_token_accuracy": 0.8303560864180326, |
| "num_tokens": 39333474.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.6042962603271007, |
| "epoch": 1.904, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 0.6915, |
| "mean_token_accuracy": 0.8288168527185917, |
| "num_tokens": 39663055.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.601089458540082, |
| "epoch": 1.92, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 0.6888, |
| "mean_token_accuracy": 0.8287999380379916, |
| "num_tokens": 39989824.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 1.759890214920044, |
| "eval_biology_loss": 2.368649959564209, |
| "eval_biology_mean_token_accuracy": 0.5835101284980774, |
| "eval_biology_num_tokens": 39989824.0, |
| "eval_biology_runtime": 38.6774, |
| "eval_biology_samples_per_second": 12.927, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.776658641576767, |
| "eval_chemistry_loss": 1.0122178792953491, |
| "eval_chemistry_mean_token_accuracy": 0.7752061448097229, |
| "eval_chemistry_num_tokens": 39989824.0, |
| "eval_chemistry_runtime": 48.1689, |
| "eval_chemistry_samples_per_second": 10.38, |
| "eval_chemistry_steps_per_second": 2.595, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.6634243364334107, |
| "eval_math_loss": 1.0101300477981567, |
| "eval_math_mean_token_accuracy": 0.7840519022941589, |
| "eval_math_num_tokens": 39989824.0, |
| "eval_math_runtime": 49.5286, |
| "eval_math_samples_per_second": 10.095, |
| "eval_math_steps_per_second": 2.524, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.5912484722137451, |
| "eval_physics_loss": 0.7268282175064087, |
| "eval_physics_mean_token_accuracy": 0.825360511302948, |
| "eval_physics_num_tokens": 39989824.0, |
| "eval_physics_runtime": 56.9412, |
| "eval_physics_samples_per_second": 8.781, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.5859691947698593, |
| "epoch": 1.936, |
| "grad_norm": 0.96875, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 0.676, |
| "mean_token_accuracy": 0.8319578696042299, |
| "num_tokens": 40325070.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5632861316204071, |
| "epoch": 1.952, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 0.6508, |
| "mean_token_accuracy": 0.83807716332376, |
| "num_tokens": 40659614.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.5868391951546073, |
| "epoch": 1.968, |
| "grad_norm": 0.91796875, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 0.6756, |
| "mean_token_accuracy": 0.8323294088244438, |
| "num_tokens": 40991508.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.5867437845095992, |
| "epoch": 1.984, |
| "grad_norm": 1.078125, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 0.676, |
| "mean_token_accuracy": 0.8312037277966737, |
| "num_tokens": 41328779.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.593030778504908, |
| "epoch": 2.0, |
| "grad_norm": 0.9765625, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 0.6752, |
| "mean_token_accuracy": 0.8317596733570098, |
| "num_tokens": 41664296.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.5428176861256361, |
| "epoch": 2.016, |
| "grad_norm": 0.91796875, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 0.6304, |
| "mean_token_accuracy": 0.8421510916203261, |
| "num_tokens": 42001193.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.5524626910686493, |
| "epoch": 2.032, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 0.633, |
| "mean_token_accuracy": 0.8409431543201208, |
| "num_tokens": 42340757.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.5630420710891485, |
| "epoch": 2.048, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 0.6561, |
| "mean_token_accuracy": 0.8348792966455221, |
| "num_tokens": 42670324.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.5547426689416171, |
| "epoch": 2.064, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 0.6368, |
| "mean_token_accuracy": 0.8413027279078961, |
| "num_tokens": 43010231.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.5610322959721088, |
| "epoch": 2.08, |
| "grad_norm": 1.0, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 0.6516, |
| "mean_token_accuracy": 0.8378075629472732, |
| "num_tokens": 43340227.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 1.6757129163742066, |
| "eval_biology_loss": 2.434731960296631, |
| "eval_biology_mean_token_accuracy": 0.5802675273418426, |
| "eval_biology_num_tokens": 43340227.0, |
| "eval_biology_runtime": 38.7759, |
| "eval_biology_samples_per_second": 12.895, |
| "eval_biology_steps_per_second": 3.224, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.7416200067996979, |
| "eval_chemistry_loss": 1.027290940284729, |
| "eval_chemistry_mean_token_accuracy": 0.774878448009491, |
| "eval_chemistry_num_tokens": 43340227.0, |
| "eval_chemistry_runtime": 48.262, |
| "eval_chemistry_samples_per_second": 10.36, |
| "eval_chemistry_steps_per_second": 2.59, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.6465173182487488, |
| "eval_math_loss": 1.023693561553955, |
| "eval_math_mean_token_accuracy": 0.7827271037101745, |
| "eval_math_num_tokens": 43340227.0, |
| "eval_math_runtime": 49.631, |
| "eval_math_samples_per_second": 10.074, |
| "eval_math_steps_per_second": 2.519, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.5699445073604583, |
| "eval_physics_loss": 0.7285439372062683, |
| "eval_physics_mean_token_accuracy": 0.8254043416976928, |
| "eval_physics_num_tokens": 43340227.0, |
| "eval_physics_runtime": 57.0739, |
| "eval_physics_samples_per_second": 8.761, |
| "eval_physics_steps_per_second": 2.19, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.5561112010851502, |
| "epoch": 2.096, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 0.6449, |
| "mean_token_accuracy": 0.838719493150711, |
| "num_tokens": 43673099.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5452175224199891, |
| "epoch": 2.112, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 0.6321, |
| "mean_token_accuracy": 0.8417959384620189, |
| "num_tokens": 44005986.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.5562603289261461, |
| "epoch": 2.128, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 0.653, |
| "mean_token_accuracy": 0.8379485800862312, |
| "num_tokens": 44321465.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5499181509017944, |
| "epoch": 2.144, |
| "grad_norm": 0.9921875, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 0.638, |
| "mean_token_accuracy": 0.8397632710635662, |
| "num_tokens": 44661027.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5740855507552624, |
| "epoch": 2.16, |
| "grad_norm": 1.0703125, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 0.6641, |
| "mean_token_accuracy": 0.8336062435060739, |
| "num_tokens": 44994236.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.5468110611662269, |
| "epoch": 2.176, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 0.6341, |
| "mean_token_accuracy": 0.8411031287163496, |
| "num_tokens": 45334280.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.5445626365020871, |
| "epoch": 2.192, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 0.6379, |
| "mean_token_accuracy": 0.8422058593481779, |
| "num_tokens": 45666954.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.5418444711714983, |
| "epoch": 2.208, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 0.6314, |
| "mean_token_accuracy": 0.8422281835228205, |
| "num_tokens": 45999546.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.54025251083076, |
| "epoch": 2.224, |
| "grad_norm": 1.0, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 0.6343, |
| "mean_token_accuracy": 0.8418932400643826, |
| "num_tokens": 46333809.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.5499276254326105, |
| "epoch": 2.24, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 0.6369, |
| "mean_token_accuracy": 0.8407698534429073, |
| "num_tokens": 46673328.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 1.6533626160621644, |
| "eval_biology_loss": 2.441138744354248, |
| "eval_biology_mean_token_accuracy": 0.5818718819618225, |
| "eval_biology_num_tokens": 46673328.0, |
| "eval_biology_runtime": 38.7889, |
| "eval_biology_samples_per_second": 12.89, |
| "eval_biology_steps_per_second": 3.223, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.7358436925411225, |
| "eval_chemistry_loss": 1.0279189348220825, |
| "eval_chemistry_mean_token_accuracy": 0.7754086136817933, |
| "eval_chemistry_num_tokens": 46673328.0, |
| "eval_chemistry_runtime": 48.2677, |
| "eval_chemistry_samples_per_second": 10.359, |
| "eval_chemistry_steps_per_second": 2.59, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.6436889851093293, |
| "eval_math_loss": 1.0210638046264648, |
| "eval_math_mean_token_accuracy": 0.7834169192314148, |
| "eval_math_num_tokens": 46673328.0, |
| "eval_math_runtime": 49.6493, |
| "eval_math_samples_per_second": 10.071, |
| "eval_math_steps_per_second": 2.518, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.5687800683975219, |
| "eval_physics_loss": 0.7271425127983093, |
| "eval_physics_mean_token_accuracy": 0.8259177951812744, |
| "eval_physics_num_tokens": 46673328.0, |
| "eval_physics_runtime": 57.064, |
| "eval_physics_samples_per_second": 8.762, |
| "eval_physics_steps_per_second": 2.191, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.5672985427081585, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 0.6585, |
| "mean_token_accuracy": 0.8361219819635153, |
| "num_tokens": 47001509.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.5440221125259995, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 0.6308, |
| "mean_token_accuracy": 0.842396317794919, |
| "num_tokens": 47340142.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.5601490139961243, |
| "epoch": 2.288, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 0.6515, |
| "mean_token_accuracy": 0.8371975239366293, |
| "num_tokens": 47677550.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.5608395885676145, |
| "epoch": 2.304, |
| "grad_norm": 1.15625, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 0.65, |
| "mean_token_accuracy": 0.8390717066824436, |
| "num_tokens": 47993851.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.5507894741371274, |
| "epoch": 2.32, |
| "grad_norm": 0.96484375, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 0.6392, |
| "mean_token_accuracy": 0.8389977443963289, |
| "num_tokens": 48332577.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.5349721314385534, |
| "epoch": 2.336, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 0.6212, |
| "mean_token_accuracy": 0.8431226223707199, |
| "num_tokens": 48676623.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5349597102031112, |
| "epoch": 2.352, |
| "grad_norm": 1.015625, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 0.6199, |
| "mean_token_accuracy": 0.8435559894889593, |
| "num_tokens": 49007770.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.5474015891551971, |
| "epoch": 2.368, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 0.6384, |
| "mean_token_accuracy": 0.8399505577981472, |
| "num_tokens": 49350016.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.5547215724363923, |
| "epoch": 2.384, |
| "grad_norm": 1.171875, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 0.6461, |
| "mean_token_accuracy": 0.8389234948903322, |
| "num_tokens": 49679074.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.5506776092574001, |
| "epoch": 2.4, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 0.6381, |
| "mean_token_accuracy": 0.8396286979317665, |
| "num_tokens": 50020648.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 1.6101444239616394, |
| "eval_biology_loss": 2.4393153190612793, |
| "eval_biology_mean_token_accuracy": 0.5827285711765289, |
| "eval_biology_num_tokens": 50020648.0, |
| "eval_biology_runtime": 38.6357, |
| "eval_biology_samples_per_second": 12.941, |
| "eval_biology_steps_per_second": 3.235, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.7312388877868652, |
| "eval_chemistry_loss": 1.029062271118164, |
| "eval_chemistry_mean_token_accuracy": 0.7757972526550293, |
| "eval_chemistry_num_tokens": 50020648.0, |
| "eval_chemistry_runtime": 48.0523, |
| "eval_chemistry_samples_per_second": 10.405, |
| "eval_chemistry_steps_per_second": 2.601, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.6399386277198792, |
| "eval_math_loss": 1.0237832069396973, |
| "eval_math_mean_token_accuracy": 0.7829466652870178, |
| "eval_math_num_tokens": 50020648.0, |
| "eval_math_runtime": 49.5118, |
| "eval_math_samples_per_second": 10.099, |
| "eval_math_steps_per_second": 2.525, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.5676034677028656, |
| "eval_physics_loss": 0.7261826395988464, |
| "eval_physics_mean_token_accuracy": 0.825977876663208, |
| "eval_physics_num_tokens": 50020648.0, |
| "eval_physics_runtime": 56.9466, |
| "eval_physics_samples_per_second": 8.78, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.546076669357717, |
| "epoch": 2.416, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 0.6307, |
| "mean_token_accuracy": 0.8414915602654218, |
| "num_tokens": 50370135.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.5511986341327428, |
| "epoch": 2.432, |
| "grad_norm": 1.0859375, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 0.6434, |
| "mean_token_accuracy": 0.8396999359130859, |
| "num_tokens": 50695363.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.568221763893962, |
| "epoch": 2.448, |
| "grad_norm": 1.0234375, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 0.6577, |
| "mean_token_accuracy": 0.8357432372868061, |
| "num_tokens": 51023498.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.5589647406712175, |
| "epoch": 2.464, |
| "grad_norm": 1.1328125, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 0.6484, |
| "mean_token_accuracy": 0.8391309097409249, |
| "num_tokens": 51349816.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.5509988136589528, |
| "epoch": 2.48, |
| "grad_norm": 1.109375, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 0.6417, |
| "mean_token_accuracy": 0.8385909989476203, |
| "num_tokens": 51680507.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.5563702458515764, |
| "epoch": 2.496, |
| "grad_norm": 1.046875, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 0.6462, |
| "mean_token_accuracy": 0.8388838239014149, |
| "num_tokens": 52017752.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5431887688115239, |
| "epoch": 2.512, |
| "grad_norm": 1.140625, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 0.6342, |
| "mean_token_accuracy": 0.8411614701151848, |
| "num_tokens": 52345439.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.543870740942657, |
| "epoch": 2.528, |
| "grad_norm": 0.99609375, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 0.6315, |
| "mean_token_accuracy": 0.8416834581643343, |
| "num_tokens": 52683886.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.5500559687614441, |
| "epoch": 2.544, |
| "grad_norm": 1.125, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 0.6438, |
| "mean_token_accuracy": 0.8382218111306429, |
| "num_tokens": 53004529.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.5530376594513655, |
| "epoch": 2.56, |
| "grad_norm": 1.1171875, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 0.6366, |
| "mean_token_accuracy": 0.8400206513702869, |
| "num_tokens": 53336280.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 1.6161543126106261, |
| "eval_biology_loss": 2.4465079307556152, |
| "eval_biology_mean_token_accuracy": 0.5826826608180999, |
| "eval_biology_num_tokens": 53336280.0, |
| "eval_biology_runtime": 38.7006, |
| "eval_biology_samples_per_second": 12.92, |
| "eval_biology_steps_per_second": 3.23, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 0.7235212452411651, |
| "eval_chemistry_loss": 1.0302128791809082, |
| "eval_chemistry_mean_token_accuracy": 0.776062783241272, |
| "eval_chemistry_num_tokens": 53336280.0, |
| "eval_chemistry_runtime": 48.1934, |
| "eval_chemistry_samples_per_second": 10.375, |
| "eval_chemistry_steps_per_second": 2.594, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.6339258260726929, |
| "eval_math_loss": 1.025081753730774, |
| "eval_math_mean_token_accuracy": 0.7834318013191223, |
| "eval_math_num_tokens": 53336280.0, |
| "eval_math_runtime": 49.912, |
| "eval_math_samples_per_second": 10.018, |
| "eval_math_steps_per_second": 2.504, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.5578962779045105, |
| "eval_physics_loss": 0.7250717878341675, |
| "eval_physics_mean_token_accuracy": 0.8264809098243714, |
| "eval_physics_num_tokens": 53336280.0, |
| "eval_physics_runtime": 56.9593, |
| "eval_physics_samples_per_second": 8.778, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.537515789270401, |
| "epoch": 2.576, |
| "grad_norm": 1.0625, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 0.6322, |
| "mean_token_accuracy": 0.8432505313307047, |
| "num_tokens": 53667647.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.5506160443648696, |
| "epoch": 2.592, |
| "grad_norm": 0.97265625, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 0.6402, |
| "mean_token_accuracy": 0.8399668127298355, |
| "num_tokens": 53997614.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.5386643601581454, |
| "epoch": 2.608, |
| "grad_norm": 1.0078125, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 0.6243, |
| "mean_token_accuracy": 0.8430452451109887, |
| "num_tokens": 54330195.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.5495816670358181, |
| "epoch": 2.624, |
| "grad_norm": 1.03125, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 0.6364, |
| "mean_token_accuracy": 0.8401540901511908, |
| "num_tokens": 54664342.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.5469144558534026, |
| "epoch": 2.64, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 0.636, |
| "mean_token_accuracy": 0.8406256098300219, |
| "num_tokens": 54993626.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.5694095639511942, |
| "epoch": 2.656, |
| "grad_norm": 1.1953125, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 0.656, |
| "mean_token_accuracy": 0.8346509717404842, |
| "num_tokens": 55339962.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5597302883863449, |
| "epoch": 2.672, |
| "grad_norm": 1.140625, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 0.6462, |
| "mean_token_accuracy": 0.838615670055151, |
| "num_tokens": 55670567.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.5369519403204321, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 1.1015625, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 0.6323, |
| "mean_token_accuracy": 0.843082357198, |
| "num_tokens": 56003156.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.5516389394178987, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 1.046875, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 0.6369, |
| "mean_token_accuracy": 0.8410698171705008, |
| "num_tokens": 56342926.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.5484106032177806, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 1.203125, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 0.6328, |
| "mean_token_accuracy": 0.8402947820723057, |
| "num_tokens": 56677521.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 1.5972739911079408, |
| "eval_biology_loss": 2.458798408508301, |
| "eval_biology_mean_token_accuracy": 0.5806125638484955, |
| "eval_biology_num_tokens": 56677521.0, |
| "eval_biology_runtime": 38.6662, |
| "eval_biology_samples_per_second": 12.931, |
| "eval_biology_steps_per_second": 3.233, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 0.7181811017990112, |
| "eval_chemistry_loss": 1.0299792289733887, |
| "eval_chemistry_mean_token_accuracy": 0.7751972675323486, |
| "eval_chemistry_num_tokens": 56677521.0, |
| "eval_chemistry_runtime": 48.1483, |
| "eval_chemistry_samples_per_second": 10.385, |
| "eval_chemistry_steps_per_second": 2.596, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.6283527569770813, |
| "eval_math_loss": 1.0243655443191528, |
| "eval_math_mean_token_accuracy": 0.7836487565040589, |
| "eval_math_num_tokens": 56677521.0, |
| "eval_math_runtime": 49.508, |
| "eval_math_samples_per_second": 10.099, |
| "eval_math_steps_per_second": 2.525, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.5540005767345428, |
| "eval_physics_loss": 0.7235716581344604, |
| "eval_physics_mean_token_accuracy": 0.8267813692092896, |
| "eval_physics_num_tokens": 56677521.0, |
| "eval_physics_runtime": 56.9575, |
| "eval_physics_samples_per_second": 8.778, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.5626651704311371, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 1.0859375, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 0.6551, |
| "mean_token_accuracy": 0.8361094355583191, |
| "num_tokens": 57003161.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.5559584245085716, |
| "epoch": 2.752, |
| "grad_norm": 1.0859375, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 0.6447, |
| "mean_token_accuracy": 0.8389039475470781, |
| "num_tokens": 57335850.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.5382809387519956, |
| "epoch": 2.768, |
| "grad_norm": 1.1171875, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 0.6267, |
| "mean_token_accuracy": 0.8432002298533916, |
| "num_tokens": 57672649.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.5509585844352841, |
| "epoch": 2.784, |
| "grad_norm": 1.09375, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 0.6395, |
| "mean_token_accuracy": 0.8397889394313097, |
| "num_tokens": 58007431.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5838387541472911, |
| "epoch": 2.8, |
| "grad_norm": 1.0859375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 0.6711, |
| "mean_token_accuracy": 0.8319451794028282, |
| "num_tokens": 58332730.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.5306204471737146, |
| "epoch": 2.816, |
| "grad_norm": 1.078125, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 0.6171, |
| "mean_token_accuracy": 0.8446537777781487, |
| "num_tokens": 58672106.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.553738858550787, |
| "epoch": 2.832, |
| "grad_norm": 1.078125, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 0.6486, |
| "mean_token_accuracy": 0.8376123756170273, |
| "num_tokens": 58997592.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.5568923223763704, |
| "epoch": 2.848, |
| "grad_norm": 1.0625, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 0.6446, |
| "mean_token_accuracy": 0.8393427152186632, |
| "num_tokens": 59326336.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.5415080957114696, |
| "epoch": 2.864, |
| "grad_norm": 0.94921875, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 0.6308, |
| "mean_token_accuracy": 0.8419267870485783, |
| "num_tokens": 59668586.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.5516748385503888, |
| "epoch": 2.88, |
| "grad_norm": 1.2578125, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 0.6441, |
| "mean_token_accuracy": 0.840485867485404, |
| "num_tokens": 59993572.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 1.6067570729255676, |
| "eval_biology_loss": 2.442582130432129, |
| "eval_biology_mean_token_accuracy": 0.5815401375293732, |
| "eval_biology_num_tokens": 59993572.0, |
| "eval_biology_runtime": 38.7204, |
| "eval_biology_samples_per_second": 12.913, |
| "eval_biology_steps_per_second": 3.228, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 0.7207075932025909, |
| "eval_chemistry_loss": 1.0278831720352173, |
| "eval_chemistry_mean_token_accuracy": 0.7756774797439575, |
| "eval_chemistry_num_tokens": 59993572.0, |
| "eval_chemistry_runtime": 48.2116, |
| "eval_chemistry_samples_per_second": 10.371, |
| "eval_chemistry_steps_per_second": 2.593, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.629777349948883, |
| "eval_math_loss": 1.0234665870666504, |
| "eval_math_mean_token_accuracy": 0.7837573509216309, |
| "eval_math_num_tokens": 59993572.0, |
| "eval_math_runtime": 49.5638, |
| "eval_math_samples_per_second": 10.088, |
| "eval_math_steps_per_second": 2.522, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.5562911832332611, |
| "eval_physics_loss": 0.7223864793777466, |
| "eval_physics_mean_token_accuracy": 0.8269883937835694, |
| "eval_physics_num_tokens": 59993572.0, |
| "eval_physics_runtime": 56.9994, |
| "eval_physics_samples_per_second": 8.772, |
| "eval_physics_steps_per_second": 2.193, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.5429861357435584, |
| "epoch": 2.896, |
| "grad_norm": 1.1171875, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 0.6347, |
| "mean_token_accuracy": 0.8413365628570318, |
| "num_tokens": 60324128.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.5474321844056249, |
| "epoch": 2.912, |
| "grad_norm": 1.1015625, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 0.635, |
| "mean_token_accuracy": 0.8426228888332844, |
| "num_tokens": 60657399.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.5416806817054749, |
| "epoch": 2.928, |
| "grad_norm": 1.0703125, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 0.6306, |
| "mean_token_accuracy": 0.8423144549131394, |
| "num_tokens": 60984711.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.5390040006488561, |
| "epoch": 2.944, |
| "grad_norm": 1.1875, |
| "learning_rate": 8.6e-06, |
| "loss": 0.6304, |
| "mean_token_accuracy": 0.8424391083419323, |
| "num_tokens": 61321359.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.5532678855583072, |
| "epoch": 2.96, |
| "grad_norm": 0.984375, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 0.6378, |
| "mean_token_accuracy": 0.8402687277644872, |
| "num_tokens": 61659042.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.5464650699868798, |
| "epoch": 2.976, |
| "grad_norm": 1.0546875, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 0.6345, |
| "mean_token_accuracy": 0.8401576526463032, |
| "num_tokens": 61993595.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.5306900983676315, |
| "epoch": 2.992, |
| "grad_norm": 1.0390625, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 0.6196, |
| "mean_token_accuracy": 0.8441225662827492, |
| "num_tokens": 62334012.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.5364003209397197, |
| "epoch": 3.008, |
| "grad_norm": 1.0703125, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 0.6242, |
| "mean_token_accuracy": 0.8451750382781029, |
| "num_tokens": 62660928.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.5261571481823921, |
| "epoch": 3.024, |
| "grad_norm": 1.078125, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 0.614, |
| "mean_token_accuracy": 0.8461304292082786, |
| "num_tokens": 62992670.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.5176527475938201, |
| "epoch": 3.04, |
| "grad_norm": 1.1328125, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 0.6074, |
| "mean_token_accuracy": 0.8469121795147657, |
| "num_tokens": 63335157.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 1.5792802815437317, |
| "eval_biology_loss": 2.4808876514434814, |
| "eval_biology_mean_token_accuracy": 0.5793148455619812, |
| "eval_biology_num_tokens": 63335157.0, |
| "eval_biology_runtime": 38.7055, |
| "eval_biology_samples_per_second": 12.918, |
| "eval_biology_steps_per_second": 3.23, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 0.7061369748115539, |
| "eval_chemistry_loss": 1.0390231609344482, |
| "eval_chemistry_mean_token_accuracy": 0.7745346717834473, |
| "eval_chemistry_num_tokens": 63335157.0, |
| "eval_chemistry_runtime": 48.1872, |
| "eval_chemistry_samples_per_second": 10.376, |
| "eval_chemistry_steps_per_second": 2.594, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.6224793126583099, |
| "eval_math_loss": 1.034122347831726, |
| "eval_math_mean_token_accuracy": 0.7828424015045166, |
| "eval_math_num_tokens": 63335157.0, |
| "eval_math_runtime": 49.5497, |
| "eval_math_samples_per_second": 10.091, |
| "eval_math_steps_per_second": 2.523, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.5448903846740722, |
| "eval_physics_loss": 0.7253366708755493, |
| "eval_physics_mean_token_accuracy": 0.8264335384368896, |
| "eval_physics_num_tokens": 63335157.0, |
| "eval_physics_runtime": 56.9581, |
| "eval_physics_samples_per_second": 8.778, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.5329983660951256, |
| "epoch": 3.056, |
| "grad_norm": 1.140625, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 0.6183, |
| "mean_token_accuracy": 0.8438076838850975, |
| "num_tokens": 63662314.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.5316095747053623, |
| "epoch": 3.072, |
| "grad_norm": 1.171875, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 0.6223, |
| "mean_token_accuracy": 0.8441695164889097, |
| "num_tokens": 63997780.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.541679815761745, |
| "epoch": 3.088, |
| "grad_norm": 1.15625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 0.6328, |
| "mean_token_accuracy": 0.8417537044733763, |
| "num_tokens": 64325274.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.5170316396281123, |
| "epoch": 3.104, |
| "grad_norm": 1.140625, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 0.6062, |
| "mean_token_accuracy": 0.8478735946118832, |
| "num_tokens": 64659683.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.5163595724850893, |
| "epoch": 3.12, |
| "grad_norm": 1.1328125, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 0.603, |
| "mean_token_accuracy": 0.8484522052109241, |
| "num_tokens": 64998212.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.5422403154894709, |
| "epoch": 3.136, |
| "grad_norm": 1.1015625, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 0.6357, |
| "mean_token_accuracy": 0.8405985131859779, |
| "num_tokens": 65328436.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.5195852382108569, |
| "epoch": 3.152, |
| "grad_norm": 1.125, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 0.602, |
| "mean_token_accuracy": 0.8487723391503096, |
| "num_tokens": 65659346.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.5302856534719467, |
| "epoch": 3.168, |
| "grad_norm": 1.03125, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 0.6216, |
| "mean_token_accuracy": 0.8445643980056048, |
| "num_tokens": 65986382.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.5176609115675092, |
| "epoch": 3.184, |
| "grad_norm": 1.1953125, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 0.6056, |
| "mean_token_accuracy": 0.8478365700691939, |
| "num_tokens": 66324308.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.5267078908160329, |
| "epoch": 3.2, |
| "grad_norm": 1.1015625, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 0.6149, |
| "mean_token_accuracy": 0.8465773615986109, |
| "num_tokens": 66658712.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 1.5625373516082763, |
| "eval_biology_loss": 2.4981629848480225, |
| "eval_biology_mean_token_accuracy": 0.57876149559021, |
| "eval_biology_num_tokens": 66658712.0, |
| "eval_biology_runtime": 38.9285, |
| "eval_biology_samples_per_second": 12.844, |
| "eval_biology_steps_per_second": 3.211, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 0.70146466588974, |
| "eval_chemistry_loss": 1.0409005880355835, |
| "eval_chemistry_mean_token_accuracy": 0.7748316297531128, |
| "eval_chemistry_num_tokens": 66658712.0, |
| "eval_chemistry_runtime": 48.3893, |
| "eval_chemistry_samples_per_second": 10.333, |
| "eval_chemistry_steps_per_second": 2.583, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.6218927059173583, |
| "eval_math_loss": 1.0327448844909668, |
| "eval_math_mean_token_accuracy": 0.7830894327163697, |
| "eval_math_num_tokens": 66658712.0, |
| "eval_math_runtime": 49.899, |
| "eval_math_samples_per_second": 10.02, |
| "eval_math_steps_per_second": 2.505, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.5445813267230988, |
| "eval_physics_loss": 0.725453794002533, |
| "eval_physics_mean_token_accuracy": 0.8266869735717773, |
| "eval_physics_num_tokens": 66658712.0, |
| "eval_physics_runtime": 57.3527, |
| "eval_physics_samples_per_second": 8.718, |
| "eval_physics_steps_per_second": 2.179, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "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": 4.956042200464073e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
