| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8, |
| "eval_steps": 100, |
| "global_step": 500, |
| "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 |
| } |
| ], |
| "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": 1.652179086470234e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
