| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.6, |
| "eval_steps": 100, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.9036113740876317, |
| "epoch": 0.016, |
| "grad_norm": 128.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 34.907, |
| "mean_token_accuracy": 0.6652832090854645, |
| "num_tokens": 338352.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.9087308555841446, |
| "epoch": 0.032, |
| "grad_norm": 119.5, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 34.745, |
| "mean_token_accuracy": 0.6691140927374363, |
| "num_tokens": 671193.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.9059555266052485, |
| "epoch": 0.048, |
| "grad_norm": 109.5, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 34.1928, |
| "mean_token_accuracy": 0.6685802672058344, |
| "num_tokens": 1001008.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.9450886320322752, |
| "epoch": 0.064, |
| "grad_norm": 102.0, |
| "learning_rate": 2.6e-06, |
| "loss": 34.8515, |
| "mean_token_accuracy": 0.6614982625469565, |
| "num_tokens": 1331208.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.9593638770282269, |
| "epoch": 0.08, |
| "grad_norm": 94.5, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 33.1046, |
| "mean_token_accuracy": 0.6680185409262777, |
| "num_tokens": 1667532.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9606023650616408, |
| "epoch": 0.096, |
| "grad_norm": 80.0, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 31.2876, |
| "mean_token_accuracy": 0.678196857497096, |
| "num_tokens": 2007176.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.069316839799285, |
| "epoch": 0.112, |
| "grad_norm": 63.25, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 30.7512, |
| "mean_token_accuracy": 0.6743977405130863, |
| "num_tokens": 2342259.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.2628379821777345, |
| "epoch": 0.128, |
| "grad_norm": 48.0, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 30.3935, |
| "mean_token_accuracy": 0.6708028431981802, |
| "num_tokens": 2663987.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.3055987607687711, |
| "epoch": 0.144, |
| "grad_norm": 32.25, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 27.3167, |
| "mean_token_accuracy": 0.6920903529971838, |
| "num_tokens": 2997824.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.3579398691654205, |
| "epoch": 0.16, |
| "grad_norm": 24.625, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 25.5362, |
| "mean_token_accuracy": 0.7021430663764476, |
| "num_tokens": 3330597.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.549320913314819, |
| "eval_biology_loss": 4.242035388946533, |
| "eval_biology_mean_token_accuracy": 0.4143077299594879, |
| "eval_biology_num_tokens": 3330597.0, |
| "eval_biology_runtime": 55.9019, |
| "eval_biology_samples_per_second": 8.944, |
| "eval_biology_steps_per_second": 2.236, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 1.9644288024902343, |
| "eval_chemistry_loss": 2.1523141860961914, |
| "eval_chemistry_mean_token_accuracy": 0.635781112909317, |
| "eval_chemistry_num_tokens": 3330597.0, |
| "eval_chemistry_runtime": 69.6126, |
| "eval_chemistry_samples_per_second": 7.183, |
| "eval_chemistry_steps_per_second": 1.796, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.1708643288612366, |
| "eval_math_loss": 1.6400102376937866, |
| "eval_math_mean_token_accuracy": 0.7028316712379455, |
| "eval_math_num_tokens": 3330597.0, |
| "eval_math_runtime": 71.6855, |
| "eval_math_samples_per_second": 6.975, |
| "eval_math_steps_per_second": 1.744, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.388730420589447, |
| "eval_physics_loss": 1.5701876878738403, |
| "eval_physics_mean_token_accuracy": 0.7025446329116821, |
| "eval_physics_num_tokens": 3330597.0, |
| "eval_physics_runtime": 83.1014, |
| "eval_physics_samples_per_second": 6.017, |
| "eval_physics_steps_per_second": 1.504, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.3951868526637554, |
| "epoch": 0.176, |
| "grad_norm": 23.0, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 24.797, |
| "mean_token_accuracy": 0.7061628632247448, |
| "num_tokens": 3658264.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.3387951213866471, |
| "epoch": 0.192, |
| "grad_norm": 15.375, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 23.7842, |
| "mean_token_accuracy": 0.713041927665472, |
| "num_tokens": 3995568.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.3034498687833547, |
| "epoch": 0.208, |
| "grad_norm": 17.5, |
| "learning_rate": 8.6e-06, |
| "loss": 22.4839, |
| "mean_token_accuracy": 0.7252091735601425, |
| "num_tokens": 4321436.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.2866257604211568, |
| "epoch": 0.224, |
| "grad_norm": 14.6875, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 22.0837, |
| "mean_token_accuracy": 0.7270880017429591, |
| "num_tokens": 4648491.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.268964756652713, |
| "epoch": 0.24, |
| "grad_norm": 14.25, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 21.6129, |
| "mean_token_accuracy": 0.7301765695214272, |
| "num_tokens": 4986175.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.2642716094851494, |
| "epoch": 0.256, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 21.2618, |
| "mean_token_accuracy": 0.734415826946497, |
| "num_tokens": 5329320.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.2717196103185415, |
| "epoch": 0.272, |
| "grad_norm": 17.5, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 21.4982, |
| "mean_token_accuracy": 0.7297716859728098, |
| "num_tokens": 5658796.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.2510493908077478, |
| "epoch": 0.288, |
| "grad_norm": 14.5625, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 21.0954, |
| "mean_token_accuracy": 0.7358391582965851, |
| "num_tokens": 5980201.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.2463122855871915, |
| "epoch": 0.304, |
| "grad_norm": 19.5, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 20.8303, |
| "mean_token_accuracy": 0.7361553307622671, |
| "num_tokens": 6298858.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.215079977735877, |
| "epoch": 0.32, |
| "grad_norm": 15.875, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 20.4326, |
| "mean_token_accuracy": 0.7394854992628097, |
| "num_tokens": 6622798.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.1598935775756836, |
| "eval_biology_loss": 3.6490557193756104, |
| "eval_biology_mean_token_accuracy": 0.45875698828697203, |
| "eval_biology_num_tokens": 6622798.0, |
| "eval_biology_runtime": 56.3027, |
| "eval_biology_samples_per_second": 8.881, |
| "eval_biology_steps_per_second": 2.22, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 1.6644244351387023, |
| "eval_chemistry_loss": 1.768973469734192, |
| "eval_chemistry_mean_token_accuracy": 0.6738141541481018, |
| "eval_chemistry_num_tokens": 6622798.0, |
| "eval_chemistry_runtime": 69.6913, |
| "eval_chemistry_samples_per_second": 7.174, |
| "eval_chemistry_steps_per_second": 1.794, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.124363118648529, |
| "eval_math_loss": 1.4134961366653442, |
| "eval_math_mean_token_accuracy": 0.7205373592376709, |
| "eval_math_num_tokens": 6622798.0, |
| "eval_math_runtime": 71.6804, |
| "eval_math_samples_per_second": 6.975, |
| "eval_math_steps_per_second": 1.744, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.2113610072135925, |
| "eval_physics_loss": 1.2715808153152466, |
| "eval_physics_mean_token_accuracy": 0.7393231744766235, |
| "eval_physics_num_tokens": 6622798.0, |
| "eval_physics_runtime": 83.0035, |
| "eval_physics_samples_per_second": 6.024, |
| "eval_physics_steps_per_second": 1.506, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.2057966101914643, |
| "epoch": 0.336, |
| "grad_norm": 13.875, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 20.1637, |
| "mean_token_accuracy": 0.740947175398469, |
| "num_tokens": 6953611.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.1682135988026858, |
| "epoch": 0.352, |
| "grad_norm": 14.625, |
| "learning_rate": 1.46e-05, |
| "loss": 19.5323, |
| "mean_token_accuracy": 0.748641949146986, |
| "num_tokens": 7290378.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.1533767595887183, |
| "epoch": 0.368, |
| "grad_norm": 14.75, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 19.3393, |
| "mean_token_accuracy": 0.7496714886277914, |
| "num_tokens": 7621082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.2154120814055205, |
| "epoch": 0.384, |
| "grad_norm": 13.875, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 20.1519, |
| "mean_token_accuracy": 0.7383838955312967, |
| "num_tokens": 7955570.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.162219974398613, |
| "epoch": 0.4, |
| "grad_norm": 17.0, |
| "learning_rate": 1.66e-05, |
| "loss": 19.3544, |
| "mean_token_accuracy": 0.7475786007940769, |
| "num_tokens": 8291049.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.1398306384682655, |
| "epoch": 0.416, |
| "grad_norm": 17.0, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 18.9335, |
| "mean_token_accuracy": 0.7522856764495373, |
| "num_tokens": 8627310.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.160587527230382, |
| "epoch": 0.432, |
| "grad_norm": 14.5, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 19.2248, |
| "mean_token_accuracy": 0.7491097755730152, |
| "num_tokens": 8958371.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.1003841817378999, |
| "epoch": 0.448, |
| "grad_norm": 14.0, |
| "learning_rate": 1.86e-05, |
| "loss": 18.3143, |
| "mean_token_accuracy": 0.7592806451022625, |
| "num_tokens": 9294388.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.1066032487899065, |
| "epoch": 0.464, |
| "grad_norm": 15.375, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 18.3092, |
| "mean_token_accuracy": 0.7574895836412907, |
| "num_tokens": 9620366.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.1134729530662297, |
| "epoch": 0.48, |
| "grad_norm": 14.875, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 18.5931, |
| "mean_token_accuracy": 0.7540634006261826, |
| "num_tokens": 9955431.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.058082123756409, |
| "eval_biology_loss": 3.4944381713867188, |
| "eval_biology_mean_token_accuracy": 0.4677197203636169, |
| "eval_biology_num_tokens": 9955431.0, |
| "eval_biology_runtime": 56.129, |
| "eval_biology_samples_per_second": 8.908, |
| "eval_biology_steps_per_second": 2.227, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 1.584096737384796, |
| "eval_chemistry_loss": 1.6560814380645752, |
| "eval_chemistry_mean_token_accuracy": 0.6847201838493348, |
| "eval_chemistry_num_tokens": 9955431.0, |
| "eval_chemistry_runtime": 69.5947, |
| "eval_chemistry_samples_per_second": 7.184, |
| "eval_chemistry_steps_per_second": 1.796, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.1037719421386718, |
| "eval_math_loss": 1.3507343530654907, |
| "eval_math_mean_token_accuracy": 0.7274009571075439, |
| "eval_math_num_tokens": 9955431.0, |
| "eval_math_runtime": 71.8212, |
| "eval_math_samples_per_second": 6.962, |
| "eval_math_steps_per_second": 1.74, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.1398421788215638, |
| "eval_physics_loss": 1.1733136177062988, |
| "eval_physics_mean_token_accuracy": 0.7520353999137879, |
| "eval_physics_num_tokens": 9955431.0, |
| "eval_physics_runtime": 83.0575, |
| "eval_physics_samples_per_second": 6.02, |
| "eval_physics_steps_per_second": 1.505, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.1404459465295076, |
| "epoch": 0.496, |
| "grad_norm": 14.25, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 18.8326, |
| "mean_token_accuracy": 0.7513993989676238, |
| "num_tokens": 10289883.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.114009590819478, |
| "epoch": 0.512, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 18.526, |
| "mean_token_accuracy": 0.7558605384081603, |
| "num_tokens": 10619468.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.126872155070305, |
| "epoch": 0.528, |
| "grad_norm": 13.5625, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 18.6944, |
| "mean_token_accuracy": 0.75301427654922, |
| "num_tokens": 10944904.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.11179842017591, |
| "epoch": 0.544, |
| "grad_norm": 14.625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 18.3887, |
| "mean_token_accuracy": 0.7559159055352211, |
| "num_tokens": 11266409.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.0923905447125435, |
| "epoch": 0.56, |
| "grad_norm": 13.5, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 18.0431, |
| "mean_token_accuracy": 0.7601441144943237, |
| "num_tokens": 11605544.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.075535924360156, |
| "epoch": 0.576, |
| "grad_norm": 15.375, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 17.9134, |
| "mean_token_accuracy": 0.762933773174882, |
| "num_tokens": 11931715.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.0857708267867565, |
| "epoch": 0.592, |
| "grad_norm": 16.5, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 17.9806, |
| "mean_token_accuracy": 0.7590868677943945, |
| "num_tokens": 12257243.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.077372272312641, |
| "epoch": 0.608, |
| "grad_norm": 14.25, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 17.8473, |
| "mean_token_accuracy": 0.7623418286442757, |
| "num_tokens": 12583154.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.101408040151, |
| "epoch": 0.624, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 18.1987, |
| "mean_token_accuracy": 0.7578636299818754, |
| "num_tokens": 12905392.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.0936459716409446, |
| "epoch": 0.64, |
| "grad_norm": 15.0, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 18.2164, |
| "mean_token_accuracy": 0.7581360317766667, |
| "num_tokens": 13232198.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 2.985183084487915, |
| "eval_biology_loss": 3.3934555053710938, |
| "eval_biology_mean_token_accuracy": 0.47690209436416625, |
| "eval_biology_num_tokens": 13232198.0, |
| "eval_biology_runtime": 56.2992, |
| "eval_biology_samples_per_second": 8.881, |
| "eval_biology_steps_per_second": 2.22, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 1.5230218110084535, |
| "eval_chemistry_loss": 1.5974879264831543, |
| "eval_chemistry_mean_token_accuracy": 0.6924252939224244, |
| "eval_chemistry_num_tokens": 13232198.0, |
| "eval_chemistry_runtime": 69.6914, |
| "eval_chemistry_samples_per_second": 7.174, |
| "eval_chemistry_steps_per_second": 1.794, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 1.0690260066986084, |
| "eval_math_loss": 1.3214365243911743, |
| "eval_math_mean_token_accuracy": 0.7313346996307373, |
| "eval_math_num_tokens": 13232198.0, |
| "eval_math_runtime": 71.8154, |
| "eval_math_samples_per_second": 6.962, |
| "eval_math_steps_per_second": 1.741, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.079647916316986, |
| "eval_physics_loss": 1.122850775718689, |
| "eval_physics_mean_token_accuracy": 0.7587474570274353, |
| "eval_physics_num_tokens": 13232198.0, |
| "eval_physics_runtime": 83.3676, |
| "eval_physics_samples_per_second": 5.998, |
| "eval_physics_steps_per_second": 1.499, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.0861666969954968, |
| "epoch": 0.656, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 17.9599, |
| "mean_token_accuracy": 0.7603602163493633, |
| "num_tokens": 13575902.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.0595566183328629, |
| "epoch": 0.672, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 17.6133, |
| "mean_token_accuracy": 0.7644100178033113, |
| "num_tokens": 13895997.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.0779153060168027, |
| "epoch": 0.688, |
| "grad_norm": 13.625, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 17.708, |
| "mean_token_accuracy": 0.7641124926507473, |
| "num_tokens": 14234888.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.0761012937873602, |
| "epoch": 0.704, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 17.9412, |
| "mean_token_accuracy": 0.7593970015645027, |
| "num_tokens": 14567908.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.0635488711297512, |
| "epoch": 0.72, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 17.5409, |
| "mean_token_accuracy": 0.7655271884053946, |
| "num_tokens": 14882927.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.0563195772469043, |
| "epoch": 0.736, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 17.5759, |
| "mean_token_accuracy": 0.7641396790742874, |
| "num_tokens": 15217342.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.037678936868906, |
| "epoch": 0.752, |
| "grad_norm": 13.125, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 17.1107, |
| "mean_token_accuracy": 0.7690521724522114, |
| "num_tokens": 15547710.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.0838343646377324, |
| "epoch": 0.768, |
| "grad_norm": 16.75, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 18.0446, |
| "mean_token_accuracy": 0.759132880717516, |
| "num_tokens": 15877177.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.0296639803797007, |
| "epoch": 0.784, |
| "grad_norm": 12.5, |
| "learning_rate": 1.86e-05, |
| "loss": 17.1015, |
| "mean_token_accuracy": 0.7697366833686828, |
| "num_tokens": 16219640.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.0563473679125308, |
| "epoch": 0.8, |
| "grad_norm": 15.5, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 17.5054, |
| "mean_token_accuracy": 0.7657136749476194, |
| "num_tokens": 16548261.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 2.9737762069702147, |
| "eval_biology_loss": 3.3795254230499268, |
| "eval_biology_mean_token_accuracy": 0.48070894145965576, |
| "eval_biology_num_tokens": 16548261.0, |
| "eval_biology_runtime": 56.18, |
| "eval_biology_samples_per_second": 8.9, |
| "eval_biology_steps_per_second": 2.225, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 1.5115246725082399, |
| "eval_chemistry_loss": 1.573546051979065, |
| "eval_chemistry_mean_token_accuracy": 0.6963776540756226, |
| "eval_chemistry_num_tokens": 16548261.0, |
| "eval_chemistry_runtime": 69.0701, |
| "eval_chemistry_samples_per_second": 7.239, |
| "eval_chemistry_steps_per_second": 1.81, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 1.0560337038040162, |
| "eval_math_loss": 1.306509017944336, |
| "eval_math_mean_token_accuracy": 0.7335917987823486, |
| "eval_math_num_tokens": 16548261.0, |
| "eval_math_runtime": 71.2835, |
| "eval_math_samples_per_second": 7.014, |
| "eval_math_steps_per_second": 1.754, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.0595522990226747, |
| "eval_physics_loss": 1.0949721336364746, |
| "eval_physics_mean_token_accuracy": 0.7634064397811889, |
| "eval_physics_num_tokens": 16548261.0, |
| "eval_physics_runtime": 83.0981, |
| "eval_physics_samples_per_second": 6.017, |
| "eval_physics_steps_per_second": 1.504, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.042676868662238, |
| "epoch": 0.816, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 17.3757, |
| "mean_token_accuracy": 0.7653235342353583, |
| "num_tokens": 16870533.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.060330757498741, |
| "epoch": 0.832, |
| "grad_norm": 17.875, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 17.5577, |
| "mean_token_accuracy": 0.7640024948865175, |
| "num_tokens": 17196371.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.045187360420823, |
| "epoch": 0.848, |
| "grad_norm": 12.75, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 17.2392, |
| "mean_token_accuracy": 0.7683587469160557, |
| "num_tokens": 17527066.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.0516619730740786, |
| "epoch": 0.864, |
| "grad_norm": 13.9375, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 17.4228, |
| "mean_token_accuracy": 0.7653108265250921, |
| "num_tokens": 17859605.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.0745812084525823, |
| "epoch": 0.88, |
| "grad_norm": 16.0, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 17.7266, |
| "mean_token_accuracy": 0.7631923343986273, |
| "num_tokens": 18187315.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.9957344826310873, |
| "epoch": 0.896, |
| "grad_norm": 12.625, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 16.4857, |
| "mean_token_accuracy": 0.7752151470631361, |
| "num_tokens": 18536790.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.0575346298515798, |
| "epoch": 0.912, |
| "grad_norm": 14.9375, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 17.4739, |
| "mean_token_accuracy": 0.7637861587107182, |
| "num_tokens": 18869344.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.0460248701274395, |
| "epoch": 0.928, |
| "grad_norm": 16.125, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 17.3717, |
| "mean_token_accuracy": 0.7649744093418122, |
| "num_tokens": 19199834.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.0241859197616576, |
| "epoch": 0.944, |
| "grad_norm": 15.5, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 16.9257, |
| "mean_token_accuracy": 0.7700011480599642, |
| "num_tokens": 19525385.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.0272901140153408, |
| "epoch": 0.96, |
| "grad_norm": 17.125, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 17.0788, |
| "mean_token_accuracy": 0.7674089819192886, |
| "num_tokens": 19856973.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 2.8927044954299928, |
| "eval_biology_loss": 3.3549082279205322, |
| "eval_biology_mean_token_accuracy": 0.48264122939109805, |
| "eval_biology_num_tokens": 19856973.0, |
| "eval_biology_runtime": 55.8568, |
| "eval_biology_samples_per_second": 8.951, |
| "eval_biology_steps_per_second": 2.238, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 1.464129506111145, |
| "eval_chemistry_loss": 1.5512925386428833, |
| "eval_chemistry_mean_token_accuracy": 0.6986791820526123, |
| "eval_chemistry_num_tokens": 19856973.0, |
| "eval_chemistry_runtime": 69.6627, |
| "eval_chemistry_samples_per_second": 7.177, |
| "eval_chemistry_steps_per_second": 1.794, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 1.0420140266418456, |
| "eval_math_loss": 1.297760248184204, |
| "eval_math_mean_token_accuracy": 0.7350094408988953, |
| "eval_math_num_tokens": 19856973.0, |
| "eval_math_runtime": 71.5855, |
| "eval_math_samples_per_second": 6.985, |
| "eval_math_steps_per_second": 1.746, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.0300544242858887, |
| "eval_physics_loss": 1.076904058456421, |
| "eval_physics_mean_token_accuracy": 0.7661643705368042, |
| "eval_physics_num_tokens": 19856973.0, |
| "eval_physics_runtime": 83.7599, |
| "eval_physics_samples_per_second": 5.969, |
| "eval_physics_steps_per_second": 1.492, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.029942188039422, |
| "epoch": 0.976, |
| "grad_norm": 14.5625, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 17.0833, |
| "mean_token_accuracy": 0.7702321965247393, |
| "num_tokens": 20190600.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.043160342797637, |
| "epoch": 0.992, |
| "grad_norm": 15.0, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 17.2354, |
| "mean_token_accuracy": 0.7673168186098337, |
| "num_tokens": 20518014.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.0153347484767437, |
| "epoch": 1.008, |
| "grad_norm": 14.8125, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 16.634, |
| "mean_token_accuracy": 0.7728456899523735, |
| "num_tokens": 20859215.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.010590048506856, |
| "epoch": 1.024, |
| "grad_norm": 15.25, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 16.8635, |
| "mean_token_accuracy": 0.770306408405304, |
| "num_tokens": 21186658.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.0216664642095565, |
| "epoch": 1.04, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 16.8228, |
| "mean_token_accuracy": 0.771704015135765, |
| "num_tokens": 21517237.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.9916718799620867, |
| "epoch": 1.056, |
| "grad_norm": 13.9375, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 16.4977, |
| "mean_token_accuracy": 0.7741733457893133, |
| "num_tokens": 21852528.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.0193660650402308, |
| "epoch": 1.072, |
| "grad_norm": 12.6875, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 16.7942, |
| "mean_token_accuracy": 0.7709830049425364, |
| "num_tokens": 22192380.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.017938245087862, |
| "epoch": 1.088, |
| "grad_norm": 17.5, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 17.004, |
| "mean_token_accuracy": 0.7690784070640803, |
| "num_tokens": 22517971.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.009276295825839, |
| "epoch": 1.104, |
| "grad_norm": 13.5625, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 16.5871, |
| "mean_token_accuracy": 0.773436525464058, |
| "num_tokens": 22848694.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.0195886611938476, |
| "epoch": 1.12, |
| "grad_norm": 14.125, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 17.0053, |
| "mean_token_accuracy": 0.7680860497057438, |
| "num_tokens": 23183889.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 2.837118970870972, |
| "eval_biology_loss": 3.3420016765594482, |
| "eval_biology_mean_token_accuracy": 0.48326786851882936, |
| "eval_biology_num_tokens": 23183889.0, |
| "eval_biology_runtime": 56.3022, |
| "eval_biology_samples_per_second": 8.881, |
| "eval_biology_steps_per_second": 2.22, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 1.433508418083191, |
| "eval_chemistry_loss": 1.5409785509109497, |
| "eval_chemistry_mean_token_accuracy": 0.6995256505012513, |
| "eval_chemistry_num_tokens": 23183889.0, |
| "eval_chemistry_runtime": 69.4682, |
| "eval_chemistry_samples_per_second": 7.198, |
| "eval_chemistry_steps_per_second": 1.799, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 1.0230936131477355, |
| "eval_math_loss": 1.2898101806640625, |
| "eval_math_mean_token_accuracy": 0.736490508556366, |
| "eval_math_num_tokens": 23183889.0, |
| "eval_math_runtime": 71.817, |
| "eval_math_samples_per_second": 6.962, |
| "eval_math_steps_per_second": 1.741, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 1.0090678911209106, |
| "eval_physics_loss": 1.062973976135254, |
| "eval_physics_mean_token_accuracy": 0.7677445983886719, |
| "eval_physics_num_tokens": 23183889.0, |
| "eval_physics_runtime": 82.8728, |
| "eval_physics_samples_per_second": 6.033, |
| "eval_physics_steps_per_second": 1.508, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.9851093973964453, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 16.2335, |
| "mean_token_accuracy": 0.7770712476223707, |
| "num_tokens": 23523903.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.0010987296700478, |
| "epoch": 1.152, |
| "grad_norm": 16.375, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 16.6749, |
| "mean_token_accuracy": 0.7745790995657444, |
| "num_tokens": 23848848.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.0347776237875224, |
| "epoch": 1.168, |
| "grad_norm": 14.375, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 16.9856, |
| "mean_token_accuracy": 0.7672174524515867, |
| "num_tokens": 24174589.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.0073780696839094, |
| "epoch": 1.184, |
| "grad_norm": 16.875, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 16.7297, |
| "mean_token_accuracy": 0.7723128654062748, |
| "num_tokens": 24495981.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.9867725741118193, |
| "epoch": 1.2, |
| "grad_norm": 14.25, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 16.3695, |
| "mean_token_accuracy": 0.7768875520676375, |
| "num_tokens": 24826245.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.018533943966031, |
| "epoch": 1.216, |
| "grad_norm": 17.875, |
| "learning_rate": 1.66e-05, |
| "loss": 16.7892, |
| "mean_token_accuracy": 0.7702058039605617, |
| "num_tokens": 25147775.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.9930577903985978, |
| "epoch": 1.232, |
| "grad_norm": 16.125, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 16.5313, |
| "mean_token_accuracy": 0.7744447194039822, |
| "num_tokens": 25480033.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.0064157444983721, |
| "epoch": 1.248, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 16.6421, |
| "mean_token_accuracy": 0.773406020924449, |
| "num_tokens": 25811978.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.9985887616872787, |
| "epoch": 1.264, |
| "grad_norm": 16.5, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 16.5153, |
| "mean_token_accuracy": 0.774219336360693, |
| "num_tokens": 26146299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.0295063827186823, |
| "epoch": 1.28, |
| "grad_norm": 16.125, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 17.0668, |
| "mean_token_accuracy": 0.7661171369254589, |
| "num_tokens": 26478207.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 2.898805745124817, |
| "eval_biology_loss": 3.341296911239624, |
| "eval_biology_mean_token_accuracy": 0.48686809062957764, |
| "eval_biology_num_tokens": 26478207.0, |
| "eval_biology_runtime": 56.3495, |
| "eval_biology_samples_per_second": 8.873, |
| "eval_biology_steps_per_second": 2.218, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.4466705160140991, |
| "eval_chemistry_loss": 1.5262336730957031, |
| "eval_chemistry_mean_token_accuracy": 0.7025133318901062, |
| "eval_chemistry_num_tokens": 26478207.0, |
| "eval_chemistry_runtime": 69.3655, |
| "eval_chemistry_samples_per_second": 7.208, |
| "eval_chemistry_steps_per_second": 1.802, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 1.024560465335846, |
| "eval_math_loss": 1.2816721200942993, |
| "eval_math_mean_token_accuracy": 0.737667652130127, |
| "eval_math_num_tokens": 26478207.0, |
| "eval_math_runtime": 71.7104, |
| "eval_math_samples_per_second": 6.972, |
| "eval_math_steps_per_second": 1.743, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 1.0090858311653137, |
| "eval_physics_loss": 1.0530790090560913, |
| "eval_physics_mean_token_accuracy": 0.7696638231277466, |
| "eval_physics_num_tokens": 26478207.0, |
| "eval_physics_runtime": 83.3567, |
| "eval_physics_samples_per_second": 5.998, |
| "eval_physics_steps_per_second": 1.5, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.9791497606784105, |
| "epoch": 1.296, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 16.2738, |
| "mean_token_accuracy": 0.777652832493186, |
| "num_tokens": 26806477.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.0318988490849734, |
| "epoch": 1.312, |
| "grad_norm": 17.125, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 16.9474, |
| "mean_token_accuracy": 0.7686588883399963, |
| "num_tokens": 27137528.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.007080302760005, |
| "epoch": 1.328, |
| "grad_norm": 17.875, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 16.6844, |
| "mean_token_accuracy": 0.771523043885827, |
| "num_tokens": 27465859.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.0098473474383354, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 16.5, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 16.8087, |
| "mean_token_accuracy": 0.7707158491015434, |
| "num_tokens": 27795568.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.0220563992857934, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 16.768, |
| "mean_token_accuracy": 0.770363200828433, |
| "num_tokens": 28129988.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.9859919909387826, |
| "epoch": 1.376, |
| "grad_norm": 17.875, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 16.4135, |
| "mean_token_accuracy": 0.7744484882801771, |
| "num_tokens": 28465903.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.988647548854351, |
| "epoch": 1.392, |
| "grad_norm": 17.125, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 16.2043, |
| "mean_token_accuracy": 0.7775332607328892, |
| "num_tokens": 28801554.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.9736344590783119, |
| "epoch": 1.408, |
| "grad_norm": 15.25, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 16.332, |
| "mean_token_accuracy": 0.7759838547557593, |
| "num_tokens": 29125114.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.024445422179997, |
| "epoch": 1.424, |
| "grad_norm": 14.125, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 16.9175, |
| "mean_token_accuracy": 0.7679786011576653, |
| "num_tokens": 29454334.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.9635057615116238, |
| "epoch": 1.44, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 15.9192, |
| "mean_token_accuracy": 0.7802809584885836, |
| "num_tokens": 29789253.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 2.8437174711227415, |
| "eval_biology_loss": 3.318575143814087, |
| "eval_biology_mean_token_accuracy": 0.4861623268127441, |
| "eval_biology_num_tokens": 29789253.0, |
| "eval_biology_runtime": 56.4644, |
| "eval_biology_samples_per_second": 8.855, |
| "eval_biology_steps_per_second": 2.214, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.420595220565796, |
| "eval_chemistry_loss": 1.5160387754440308, |
| "eval_chemistry_mean_token_accuracy": 0.7028642320632934, |
| "eval_chemistry_num_tokens": 29789253.0, |
| "eval_chemistry_runtime": 69.7177, |
| "eval_chemistry_samples_per_second": 7.172, |
| "eval_chemistry_steps_per_second": 1.793, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 1.0116905975341797, |
| "eval_math_loss": 1.2779794931411743, |
| "eval_math_mean_token_accuracy": 0.7384266419410705, |
| "eval_math_num_tokens": 29789253.0, |
| "eval_math_runtime": 71.6683, |
| "eval_math_samples_per_second": 6.977, |
| "eval_math_steps_per_second": 1.744, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.9905482964515686, |
| "eval_physics_loss": 1.0435165166854858, |
| "eval_physics_mean_token_accuracy": 0.770782280921936, |
| "eval_physics_num_tokens": 29789253.0, |
| "eval_physics_runtime": 83.2228, |
| "eval_physics_samples_per_second": 6.008, |
| "eval_physics_steps_per_second": 1.502, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.9581953696906567, |
| "epoch": 1.456, |
| "grad_norm": 13.1875, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 15.9515, |
| "mean_token_accuracy": 0.7805543716996908, |
| "num_tokens": 30128576.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.0345965854823589, |
| "epoch": 1.472, |
| "grad_norm": 17.125, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 17.1026, |
| "mean_token_accuracy": 0.7668049655854702, |
| "num_tokens": 30452719.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.0023833878338337, |
| "epoch": 1.488, |
| "grad_norm": 13.8125, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 16.5595, |
| "mean_token_accuracy": 0.7728984329849482, |
| "num_tokens": 30782343.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.9856885317713022, |
| "epoch": 1.504, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 16.267, |
| "mean_token_accuracy": 0.7769489102065563, |
| "num_tokens": 31113238.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.0002378012984992, |
| "epoch": 1.52, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 16.5956, |
| "mean_token_accuracy": 0.7721422206610441, |
| "num_tokens": 31444589.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.0214108165353537, |
| "epoch": 1.536, |
| "grad_norm": 17.625, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 16.8896, |
| "mean_token_accuracy": 0.7694170981645584, |
| "num_tokens": 31774137.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.9722279075533151, |
| "epoch": 1.552, |
| "grad_norm": 16.375, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 16.0224, |
| "mean_token_accuracy": 0.7783945810049773, |
| "num_tokens": 32101776.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.9741225987672806, |
| "epoch": 1.568, |
| "grad_norm": 15.25, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 16.145, |
| "mean_token_accuracy": 0.778332832083106, |
| "num_tokens": 32428021.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.0168811423704027, |
| "epoch": 1.584, |
| "grad_norm": 16.5, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 16.8197, |
| "mean_token_accuracy": 0.7688806146383286, |
| "num_tokens": 32752735.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.0012462731450795, |
| "epoch": 1.6, |
| "grad_norm": 14.625, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 16.6785, |
| "mean_token_accuracy": 0.7718310851603747, |
| "num_tokens": 33075822.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 2.871693316459656, |
| "eval_biology_loss": 3.2952377796173096, |
| "eval_biology_mean_token_accuracy": 0.48806494069099426, |
| "eval_biology_num_tokens": 33075822.0, |
| "eval_biology_runtime": 56.2343, |
| "eval_biology_samples_per_second": 8.891, |
| "eval_biology_steps_per_second": 2.223, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.435364501476288, |
| "eval_chemistry_loss": 1.5101022720336914, |
| "eval_chemistry_mean_token_accuracy": 0.7037839899063111, |
| "eval_chemistry_num_tokens": 33075822.0, |
| "eval_chemistry_runtime": 69.2005, |
| "eval_chemistry_samples_per_second": 7.225, |
| "eval_chemistry_steps_per_second": 1.806, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 1.0202706332206726, |
| "eval_math_loss": 1.2738072872161865, |
| "eval_math_mean_token_accuracy": 0.7389369897842407, |
| "eval_math_num_tokens": 33075822.0, |
| "eval_math_runtime": 71.9142, |
| "eval_math_samples_per_second": 6.953, |
| "eval_math_steps_per_second": 1.738, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.9962421131134033, |
| "eval_physics_loss": 1.0371240377426147, |
| "eval_physics_mean_token_accuracy": 0.7717792973518371, |
| "eval_physics_num_tokens": 33075822.0, |
| "eval_physics_runtime": 83.5174, |
| "eval_physics_samples_per_second": 5.987, |
| "eval_physics_steps_per_second": 1.497, |
| "step": 1000 |
| } |
| ], |
| "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": 3.3005113980506685e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
