| { |
| "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": 0.7980934290215373, |
| "epoch": 0.016, |
| "grad_norm": 49.0, |
| "learning_rate": 1.8e-07, |
| "loss": 13.873, |
| "mean_token_accuracy": 0.77677470035851, |
| "num_tokens": 338352.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.7983348269015551, |
| "epoch": 0.032, |
| "grad_norm": 39.75, |
| "learning_rate": 3.8e-07, |
| "loss": 13.832, |
| "mean_token_accuracy": 0.7767552722245454, |
| "num_tokens": 671193.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.7940070297569036, |
| "epoch": 0.048, |
| "grad_norm": 36.5, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 13.6955, |
| "mean_token_accuracy": 0.7781575798988343, |
| "num_tokens": 1001008.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8190817344933748, |
| "epoch": 0.064, |
| "grad_norm": 34.0, |
| "learning_rate": 7.8e-07, |
| "loss": 13.9863, |
| "mean_token_accuracy": 0.7736780602484942, |
| "num_tokens": 1331208.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.8228552235290408, |
| "epoch": 0.08, |
| "grad_norm": 33.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 13.7981, |
| "mean_token_accuracy": 0.7759026233106852, |
| "num_tokens": 1667532.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8120608307421208, |
| "epoch": 0.096, |
| "grad_norm": 34.0, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 13.4686, |
| "mean_token_accuracy": 0.7815129280090332, |
| "num_tokens": 2007176.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8287483796477317, |
| "epoch": 0.112, |
| "grad_norm": 36.5, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 13.7526, |
| "mean_token_accuracy": 0.776063147187233, |
| "num_tokens": 2342259.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8556615479290486, |
| "epoch": 0.128, |
| "grad_norm": 27.125, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 13.9267, |
| "mean_token_accuracy": 0.773414358869195, |
| "num_tokens": 2663987.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8306544724851846, |
| "epoch": 0.144, |
| "grad_norm": 25.75, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 13.6531, |
| "mean_token_accuracy": 0.7779791738837958, |
| "num_tokens": 2997824.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8303449299186468, |
| "epoch": 0.16, |
| "grad_norm": 26.5, |
| "learning_rate": 1.98e-06, |
| "loss": 13.4855, |
| "mean_token_accuracy": 0.7814826928079128, |
| "num_tokens": 3330597.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.125799958705902, |
| "eval_biology_loss": 1.1626057624816895, |
| "eval_biology_mean_token_accuracy": 0.7041610617637635, |
| "eval_biology_num_tokens": 3330597.0, |
| "eval_biology_runtime": 29.5726, |
| "eval_biology_samples_per_second": 16.908, |
| "eval_biology_steps_per_second": 4.227, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 0.8412163820266724, |
| "eval_physics_loss": 0.8517085909843445, |
| "eval_physics_mean_token_accuracy": 0.7784765930175781, |
| "eval_physics_num_tokens": 3330597.0, |
| "eval_physics_runtime": 42.3208, |
| "eval_physics_samples_per_second": 11.815, |
| "eval_physics_steps_per_second": 2.954, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8291829921305179, |
| "epoch": 0.176, |
| "grad_norm": 32.5, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 13.4483, |
| "mean_token_accuracy": 0.7799590565264225, |
| "num_tokens": 3658264.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8301662255078555, |
| "epoch": 0.192, |
| "grad_norm": 25.5, |
| "learning_rate": 2.38e-06, |
| "loss": 13.4143, |
| "mean_token_accuracy": 0.7796540033072233, |
| "num_tokens": 3995568.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.821339176222682, |
| "epoch": 0.208, |
| "grad_norm": 26.125, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 13.2097, |
| "mean_token_accuracy": 0.7838085852563381, |
| "num_tokens": 4321436.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.8153277274221182, |
| "epoch": 0.224, |
| "grad_norm": 24.0, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 13.0671, |
| "mean_token_accuracy": 0.7830340649932623, |
| "num_tokens": 4648491.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.8157381068915128, |
| "epoch": 0.24, |
| "grad_norm": 23.375, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 13.1148, |
| "mean_token_accuracy": 0.7832087766379118, |
| "num_tokens": 4986175.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8061813484877348, |
| "epoch": 0.256, |
| "grad_norm": 26.625, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 13.0481, |
| "mean_token_accuracy": 0.7853319209069014, |
| "num_tokens": 5329320.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.8233318269252777, |
| "epoch": 0.272, |
| "grad_norm": 26.0, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 13.1699, |
| "mean_token_accuracy": 0.7812638122588396, |
| "num_tokens": 5658796.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.8212542906403542, |
| "epoch": 0.288, |
| "grad_norm": 22.0, |
| "learning_rate": 3.58e-06, |
| "loss": 13.1426, |
| "mean_token_accuracy": 0.783899562805891, |
| "num_tokens": 5980201.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8011876806616783, |
| "epoch": 0.304, |
| "grad_norm": 24.375, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 12.8823, |
| "mean_token_accuracy": 0.7862063981592655, |
| "num_tokens": 6298858.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.8119899664074183, |
| "epoch": 0.32, |
| "grad_norm": 22.375, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 13.017, |
| "mean_token_accuracy": 0.7848218575119972, |
| "num_tokens": 6622798.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.137313835144043, |
| "eval_biology_loss": 1.1703258752822876, |
| "eval_biology_mean_token_accuracy": 0.7031810960769653, |
| "eval_biology_num_tokens": 6622798.0, |
| "eval_biology_runtime": 29.6247, |
| "eval_biology_samples_per_second": 16.878, |
| "eval_biology_steps_per_second": 4.219, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 0.8116096572875977, |
| "eval_physics_loss": 0.8106478452682495, |
| "eval_physics_mean_token_accuracy": 0.7854588747024536, |
| "eval_physics_num_tokens": 6622798.0, |
| "eval_physics_runtime": 42.332, |
| "eval_physics_samples_per_second": 11.811, |
| "eval_physics_steps_per_second": 2.953, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8008256774395704, |
| "epoch": 0.336, |
| "grad_norm": 21.25, |
| "learning_rate": 4.18e-06, |
| "loss": 12.8337, |
| "mean_token_accuracy": 0.7872962314635515, |
| "num_tokens": 6953611.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.7870257034897804, |
| "epoch": 0.352, |
| "grad_norm": 23.75, |
| "learning_rate": 4.38e-06, |
| "loss": 12.629, |
| "mean_token_accuracy": 0.7910902902483941, |
| "num_tokens": 7290378.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7753207640722394, |
| "epoch": 0.368, |
| "grad_norm": 23.625, |
| "learning_rate": 4.58e-06, |
| "loss": 12.3974, |
| "mean_token_accuracy": 0.7931954108178616, |
| "num_tokens": 7621082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.8035390684381127, |
| "epoch": 0.384, |
| "grad_norm": 23.375, |
| "learning_rate": 4.78e-06, |
| "loss": 12.8833, |
| "mean_token_accuracy": 0.7841386552900076, |
| "num_tokens": 7955570.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.7771880365908146, |
| "epoch": 0.4, |
| "grad_norm": 23.5, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 12.462, |
| "mean_token_accuracy": 0.7913563769310713, |
| "num_tokens": 8291049.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7649452563375234, |
| "epoch": 0.416, |
| "grad_norm": 22.25, |
| "learning_rate": 5.18e-06, |
| "loss": 12.2876, |
| "mean_token_accuracy": 0.7948728159070015, |
| "num_tokens": 8627310.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7790930971503258, |
| "epoch": 0.432, |
| "grad_norm": 23.0, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 12.4896, |
| "mean_token_accuracy": 0.790573175624013, |
| "num_tokens": 8958371.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7571947271004319, |
| "epoch": 0.448, |
| "grad_norm": 22.75, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 12.1797, |
| "mean_token_accuracy": 0.795942148193717, |
| "num_tokens": 9294388.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.7509586235508323, |
| "epoch": 0.464, |
| "grad_norm": 22.875, |
| "learning_rate": 5.78e-06, |
| "loss": 11.9835, |
| "mean_token_accuracy": 0.7969875477254391, |
| "num_tokens": 9620366.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7548794999718667, |
| "epoch": 0.48, |
| "grad_norm": 20.875, |
| "learning_rate": 5.98e-06, |
| "loss": 12.0853, |
| "mean_token_accuracy": 0.7961488746106624, |
| "num_tokens": 9955431.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.1340459384918213, |
| "eval_biology_loss": 1.1766083240509033, |
| "eval_biology_mean_token_accuracy": 0.7020830793380737, |
| "eval_biology_num_tokens": 9955431.0, |
| "eval_biology_runtime": 29.8733, |
| "eval_biology_samples_per_second": 16.737, |
| "eval_biology_steps_per_second": 4.184, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 0.7825248126983643, |
| "eval_physics_loss": 0.7770141363143921, |
| "eval_physics_mean_token_accuracy": 0.7918553309440612, |
| "eval_physics_num_tokens": 9955431.0, |
| "eval_physics_runtime": 42.3494, |
| "eval_physics_samples_per_second": 11.807, |
| "eval_physics_steps_per_second": 2.952, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7715075328946114, |
| "epoch": 0.496, |
| "grad_norm": 20.375, |
| "learning_rate": 6.18e-06, |
| "loss": 12.3217, |
| "mean_token_accuracy": 0.7934962294995784, |
| "num_tokens": 10289883.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.7606870967894792, |
| "epoch": 0.512, |
| "grad_norm": 24.625, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 12.2158, |
| "mean_token_accuracy": 0.795098090544343, |
| "num_tokens": 10619468.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.7731524888426066, |
| "epoch": 0.528, |
| "grad_norm": 20.5, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 12.3882, |
| "mean_token_accuracy": 0.7916440792381764, |
| "num_tokens": 10944904.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.755775578506291, |
| "epoch": 0.544, |
| "grad_norm": 22.875, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 12.1334, |
| "mean_token_accuracy": 0.7953360237181186, |
| "num_tokens": 11266409.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.7537049597129226, |
| "epoch": 0.56, |
| "grad_norm": 18.875, |
| "learning_rate": 6.98e-06, |
| "loss": 12.0364, |
| "mean_token_accuracy": 0.7975195907056332, |
| "num_tokens": 11605544.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.7410756940022111, |
| "epoch": 0.576, |
| "grad_norm": 21.375, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 11.8459, |
| "mean_token_accuracy": 0.8007426861673593, |
| "num_tokens": 11931715.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.7542769065126776, |
| "epoch": 0.592, |
| "grad_norm": 21.375, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 12.0135, |
| "mean_token_accuracy": 0.7966224331408739, |
| "num_tokens": 12257243.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.7349719753488898, |
| "epoch": 0.608, |
| "grad_norm": 20.375, |
| "learning_rate": 7.58e-06, |
| "loss": 11.7827, |
| "mean_token_accuracy": 0.7996757850050926, |
| "num_tokens": 12583154.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.7510534537956118, |
| "epoch": 0.624, |
| "grad_norm": 20.875, |
| "learning_rate": 7.78e-06, |
| "loss": 12.0433, |
| "mean_token_accuracy": 0.7961709950119257, |
| "num_tokens": 12905392.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.746537908166647, |
| "epoch": 0.64, |
| "grad_norm": 18.75, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 11.9585, |
| "mean_token_accuracy": 0.798334264755249, |
| "num_tokens": 13232198.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.100157816886902, |
| "eval_biology_loss": 1.1817364692687988, |
| "eval_biology_mean_token_accuracy": 0.7015948586463928, |
| "eval_biology_num_tokens": 13232198.0, |
| "eval_biology_runtime": 29.5996, |
| "eval_biology_samples_per_second": 16.892, |
| "eval_biology_steps_per_second": 4.223, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.7428050627708435, |
| "eval_physics_loss": 0.7491776943206787, |
| "eval_physics_mean_token_accuracy": 0.7974028449058532, |
| "eval_physics_num_tokens": 13232198.0, |
| "eval_physics_runtime": 42.3622, |
| "eval_physics_samples_per_second": 11.803, |
| "eval_physics_steps_per_second": 2.951, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.7335962850600481, |
| "epoch": 0.656, |
| "grad_norm": 18.125, |
| "learning_rate": 8.18e-06, |
| "loss": 11.7731, |
| "mean_token_accuracy": 0.8001615725457668, |
| "num_tokens": 13575902.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.7217804849147796, |
| "epoch": 0.672, |
| "grad_norm": 21.25, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 11.5198, |
| "mean_token_accuracy": 0.8043914053589105, |
| "num_tokens": 13895997.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.7267325738444924, |
| "epoch": 0.688, |
| "grad_norm": 20.75, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 11.6352, |
| "mean_token_accuracy": 0.8020107671618462, |
| "num_tokens": 14234888.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.741226227581501, |
| "epoch": 0.704, |
| "grad_norm": 21.5, |
| "learning_rate": 8.78e-06, |
| "loss": 11.9022, |
| "mean_token_accuracy": 0.7984758585691452, |
| "num_tokens": 14567908.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.722562575340271, |
| "epoch": 0.72, |
| "grad_norm": 22.875, |
| "learning_rate": 8.98e-06, |
| "loss": 11.5666, |
| "mean_token_accuracy": 0.8033338606357574, |
| "num_tokens": 14882927.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.7157323375344277, |
| "epoch": 0.736, |
| "grad_norm": 22.25, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 11.464, |
| "mean_token_accuracy": 0.8032864931970835, |
| "num_tokens": 15217342.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.710782004147768, |
| "epoch": 0.752, |
| "grad_norm": 18.5, |
| "learning_rate": 9.38e-06, |
| "loss": 11.3824, |
| "mean_token_accuracy": 0.8050779778510332, |
| "num_tokens": 15547710.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.7326425770297647, |
| "epoch": 0.768, |
| "grad_norm": 20.375, |
| "learning_rate": 9.58e-06, |
| "loss": 11.7364, |
| "mean_token_accuracy": 0.7999915853142738, |
| "num_tokens": 15877177.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.7081343689933419, |
| "epoch": 0.784, |
| "grad_norm": 18.375, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 11.3161, |
| "mean_token_accuracy": 0.8068799041211605, |
| "num_tokens": 16219640.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.7145562319085002, |
| "epoch": 0.8, |
| "grad_norm": 19.375, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 11.4904, |
| "mean_token_accuracy": 0.8040819689631462, |
| "num_tokens": 16548261.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.1381474986076354, |
| "eval_biology_loss": 1.1847245693206787, |
| "eval_biology_mean_token_accuracy": 0.6994160051345826, |
| "eval_biology_num_tokens": 16548261.0, |
| "eval_biology_runtime": 29.6161, |
| "eval_biology_samples_per_second": 16.883, |
| "eval_biology_steps_per_second": 4.221, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.7353026342391967, |
| "eval_physics_loss": 0.7241687178611755, |
| "eval_physics_mean_token_accuracy": 0.8026187562942505, |
| "eval_physics_num_tokens": 16548261.0, |
| "eval_physics_runtime": 42.1602, |
| "eval_physics_samples_per_second": 11.86, |
| "eval_physics_steps_per_second": 2.965, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.715489343367517, |
| "epoch": 0.816, |
| "grad_norm": 18.75, |
| "learning_rate": 1.018e-05, |
| "loss": 11.4335, |
| "mean_token_accuracy": 0.8032866880297661, |
| "num_tokens": 16870533.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.7197281535714865, |
| "epoch": 0.832, |
| "grad_norm": 22.625, |
| "learning_rate": 1.038e-05, |
| "loss": 11.5253, |
| "mean_token_accuracy": 0.803961057588458, |
| "num_tokens": 17196371.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.7022356629371643, |
| "epoch": 0.848, |
| "grad_norm": 17.375, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 11.2777, |
| "mean_token_accuracy": 0.8062990486621857, |
| "num_tokens": 17527066.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.7121493600308895, |
| "epoch": 0.864, |
| "grad_norm": 17.625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 11.3616, |
| "mean_token_accuracy": 0.8058860942721366, |
| "num_tokens": 17859605.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.7143916061148048, |
| "epoch": 0.88, |
| "grad_norm": 19.875, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 11.4128, |
| "mean_token_accuracy": 0.8041808195412159, |
| "num_tokens": 18187315.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.6755495458841324, |
| "epoch": 0.896, |
| "grad_norm": 17.375, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 10.8244, |
| "mean_token_accuracy": 0.8135501727461815, |
| "num_tokens": 18536790.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.7111412728205323, |
| "epoch": 0.912, |
| "grad_norm": 18.5, |
| "learning_rate": 1.138e-05, |
| "loss": 11.3738, |
| "mean_token_accuracy": 0.8046688307076693, |
| "num_tokens": 18869344.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.7016011888161302, |
| "epoch": 0.928, |
| "grad_norm": 20.75, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 11.2569, |
| "mean_token_accuracy": 0.8068617224693299, |
| "num_tokens": 19199834.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.6901180900633335, |
| "epoch": 0.944, |
| "grad_norm": 19.375, |
| "learning_rate": 1.178e-05, |
| "loss": 11.0197, |
| "mean_token_accuracy": 0.8105497155338526, |
| "num_tokens": 19525385.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6980544172227383, |
| "epoch": 0.96, |
| "grad_norm": 18.0, |
| "learning_rate": 1.198e-05, |
| "loss": 11.1812, |
| "mean_token_accuracy": 0.8075801994651556, |
| "num_tokens": 19856973.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.10544912815094, |
| "eval_biology_loss": 1.1910302639007568, |
| "eval_biology_mean_token_accuracy": 0.7007143182754516, |
| "eval_biology_num_tokens": 19856973.0, |
| "eval_biology_runtime": 29.6701, |
| "eval_biology_samples_per_second": 16.852, |
| "eval_biology_steps_per_second": 4.213, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.7146178777217865, |
| "eval_physics_loss": 0.7040261030197144, |
| "eval_physics_mean_token_accuracy": 0.8063621273040772, |
| "eval_physics_num_tokens": 19856973.0, |
| "eval_physics_runtime": 42.3789, |
| "eval_physics_samples_per_second": 11.798, |
| "eval_physics_steps_per_second": 2.95, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6889748936519027, |
| "epoch": 0.976, |
| "grad_norm": 18.375, |
| "learning_rate": 1.218e-05, |
| "loss": 10.9979, |
| "mean_token_accuracy": 0.8105814348906278, |
| "num_tokens": 20190600.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.6845765860751272, |
| "epoch": 0.992, |
| "grad_norm": 17.875, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 10.9961, |
| "mean_token_accuracy": 0.8099663183093071, |
| "num_tokens": 20518014.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.6710353532806039, |
| "epoch": 1.008, |
| "grad_norm": 17.375, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 10.6734, |
| "mean_token_accuracy": 0.8143552124500275, |
| "num_tokens": 20859215.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.671150723285973, |
| "epoch": 1.024, |
| "grad_norm": 18.875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 10.7577, |
| "mean_token_accuracy": 0.8129799973219634, |
| "num_tokens": 21186658.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.670030782930553, |
| "epoch": 1.04, |
| "grad_norm": 17.75, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 10.7736, |
| "mean_token_accuracy": 0.8133355241268874, |
| "num_tokens": 21517237.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.6583073100075125, |
| "epoch": 1.056, |
| "grad_norm": 16.75, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 10.521, |
| "mean_token_accuracy": 0.8164310045540333, |
| "num_tokens": 21852528.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.67260313462466, |
| "epoch": 1.072, |
| "grad_norm": 17.875, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 10.711, |
| "mean_token_accuracy": 0.8140104580670595, |
| "num_tokens": 22192380.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6687374956905842, |
| "epoch": 1.088, |
| "grad_norm": 20.625, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 10.7304, |
| "mean_token_accuracy": 0.8136457096785307, |
| "num_tokens": 22517971.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6568242866545916, |
| "epoch": 1.104, |
| "grad_norm": 16.5, |
| "learning_rate": 1.378e-05, |
| "loss": 10.5146, |
| "mean_token_accuracy": 0.8156068984419107, |
| "num_tokens": 22848694.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6624812703579664, |
| "epoch": 1.12, |
| "grad_norm": 18.25, |
| "learning_rate": 1.398e-05, |
| "loss": 10.6908, |
| "mean_token_accuracy": 0.8136210646480322, |
| "num_tokens": 23183889.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.096513487815857, |
| "eval_biology_loss": 1.1950488090515137, |
| "eval_biology_mean_token_accuracy": 0.6985604496002197, |
| "eval_biology_num_tokens": 23183889.0, |
| "eval_biology_runtime": 29.5973, |
| "eval_biology_samples_per_second": 16.893, |
| "eval_biology_steps_per_second": 4.223, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.6853570246696472, |
| "eval_physics_loss": 0.6883610486984253, |
| "eval_physics_mean_token_accuracy": 0.8095321311950684, |
| "eval_physics_num_tokens": 23183889.0, |
| "eval_physics_runtime": 42.3537, |
| "eval_physics_samples_per_second": 11.805, |
| "eval_physics_steps_per_second": 2.951, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.6393926108255983, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 17.125, |
| "learning_rate": 1.418e-05, |
| "loss": 10.2691, |
| "mean_token_accuracy": 0.8204710930585861, |
| "num_tokens": 23523903.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6592076497152448, |
| "epoch": 1.152, |
| "grad_norm": 19.5, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 10.525, |
| "mean_token_accuracy": 0.8170250646770001, |
| "num_tokens": 23848848.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6672398639842868, |
| "epoch": 1.168, |
| "grad_norm": 19.25, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 10.7446, |
| "mean_token_accuracy": 0.8126429941505193, |
| "num_tokens": 24174589.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6579326229169965, |
| "epoch": 1.184, |
| "grad_norm": 20.0, |
| "learning_rate": 1.478e-05, |
| "loss": 10.5484, |
| "mean_token_accuracy": 0.8169561930000782, |
| "num_tokens": 24495981.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6486070465296507, |
| "epoch": 1.2, |
| "grad_norm": 17.875, |
| "learning_rate": 1.498e-05, |
| "loss": 10.4176, |
| "mean_token_accuracy": 0.818020261451602, |
| "num_tokens": 24826245.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6617947239428759, |
| "epoch": 1.216, |
| "grad_norm": 19.25, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 10.5833, |
| "mean_token_accuracy": 0.8140788938850164, |
| "num_tokens": 25147775.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6527341572567821, |
| "epoch": 1.232, |
| "grad_norm": 17.875, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 10.4663, |
| "mean_token_accuracy": 0.8178805992007255, |
| "num_tokens": 25480033.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.6440409064292908, |
| "epoch": 1.248, |
| "grad_norm": 18.25, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 10.3188, |
| "mean_token_accuracy": 0.819524721801281, |
| "num_tokens": 25811978.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.6459297401830554, |
| "epoch": 1.264, |
| "grad_norm": 17.5, |
| "learning_rate": 1.578e-05, |
| "loss": 10.3328, |
| "mean_token_accuracy": 0.8194583039730787, |
| "num_tokens": 26146299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.6661811180412769, |
| "epoch": 1.28, |
| "grad_norm": 18.5, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 10.6612, |
| "mean_token_accuracy": 0.813135676458478, |
| "num_tokens": 26478207.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 1.0779736218452454, |
| "eval_biology_loss": 1.1969928741455078, |
| "eval_biology_mean_token_accuracy": 0.6996237397193908, |
| "eval_biology_num_tokens": 26478207.0, |
| "eval_biology_runtime": 29.6158, |
| "eval_biology_samples_per_second": 16.883, |
| "eval_biology_steps_per_second": 4.221, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.6634284896850586, |
| "eval_physics_loss": 0.6750168800354004, |
| "eval_physics_mean_token_accuracy": 0.8123881487846375, |
| "eval_physics_num_tokens": 26478207.0, |
| "eval_physics_runtime": 42.3789, |
| "eval_physics_samples_per_second": 11.798, |
| "eval_physics_steps_per_second": 2.95, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.6371305795386434, |
| "epoch": 1.296, |
| "grad_norm": 18.375, |
| "learning_rate": 1.618e-05, |
| "loss": 10.2353, |
| "mean_token_accuracy": 0.8211836166679859, |
| "num_tokens": 26806477.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6579740369692445, |
| "epoch": 1.312, |
| "grad_norm": 18.5, |
| "learning_rate": 1.638e-05, |
| "loss": 10.5754, |
| "mean_token_accuracy": 0.8152522802352905, |
| "num_tokens": 27137528.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.6427882853895426, |
| "epoch": 1.328, |
| "grad_norm": 18.5, |
| "learning_rate": 1.658e-05, |
| "loss": 10.3247, |
| "mean_token_accuracy": 0.8178475011140108, |
| "num_tokens": 27465859.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.6479509010910988, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 19.0, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 10.4341, |
| "mean_token_accuracy": 0.8170580971986056, |
| "num_tokens": 27795568.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.6500799937173725, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.698e-05, |
| "loss": 10.383, |
| "mean_token_accuracy": 0.8165912210941315, |
| "num_tokens": 28129988.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6481674930080772, |
| "epoch": 1.376, |
| "grad_norm": 20.125, |
| "learning_rate": 1.718e-05, |
| "loss": 10.3763, |
| "mean_token_accuracy": 0.8186531469225884, |
| "num_tokens": 28465903.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.6292730543762446, |
| "epoch": 1.392, |
| "grad_norm": 17.0, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 10.0886, |
| "mean_token_accuracy": 0.8228901036083698, |
| "num_tokens": 28801554.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6366163183003664, |
| "epoch": 1.408, |
| "grad_norm": 18.375, |
| "learning_rate": 1.758e-05, |
| "loss": 10.2299, |
| "mean_token_accuracy": 0.8191725321114063, |
| "num_tokens": 29125114.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6577673699706793, |
| "epoch": 1.424, |
| "grad_norm": 18.75, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 10.5993, |
| "mean_token_accuracy": 0.8148027937859297, |
| "num_tokens": 29454334.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.629906609095633, |
| "epoch": 1.44, |
| "grad_norm": 17.25, |
| "learning_rate": 1.798e-05, |
| "loss": 10.0722, |
| "mean_token_accuracy": 0.8225096724927425, |
| "num_tokens": 29789253.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 1.0743142724037171, |
| "eval_biology_loss": 1.2018756866455078, |
| "eval_biology_mean_token_accuracy": 0.6978029899597168, |
| "eval_biology_num_tokens": 29789253.0, |
| "eval_biology_runtime": 29.941, |
| "eval_biology_samples_per_second": 16.7, |
| "eval_biology_steps_per_second": 4.175, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.6614522356986999, |
| "eval_physics_loss": 0.6613803505897522, |
| "eval_physics_mean_token_accuracy": 0.8159068145751953, |
| "eval_physics_num_tokens": 29789253.0, |
| "eval_physics_runtime": 42.4043, |
| "eval_physics_samples_per_second": 11.791, |
| "eval_physics_steps_per_second": 2.948, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.6264998497441411, |
| "epoch": 1.456, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 10.0407, |
| "mean_token_accuracy": 0.8242419868707657, |
| "num_tokens": 30128576.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.6577600754797459, |
| "epoch": 1.472, |
| "grad_norm": 18.875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 10.5867, |
| "mean_token_accuracy": 0.8155373357236385, |
| "num_tokens": 30452719.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.6394334685057401, |
| "epoch": 1.488, |
| "grad_norm": 16.125, |
| "learning_rate": 1.858e-05, |
| "loss": 10.2302, |
| "mean_token_accuracy": 0.8199584804475307, |
| "num_tokens": 30782343.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.6304329918697477, |
| "epoch": 1.504, |
| "grad_norm": 15.625, |
| "learning_rate": 1.878e-05, |
| "loss": 10.1144, |
| "mean_token_accuracy": 0.8226039990782738, |
| "num_tokens": 31113238.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.649897889047861, |
| "epoch": 1.52, |
| "grad_norm": 17.25, |
| "learning_rate": 1.898e-05, |
| "loss": 10.4456, |
| "mean_token_accuracy": 0.8167380917817354, |
| "num_tokens": 31444589.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.6425180744379759, |
| "epoch": 1.536, |
| "grad_norm": 18.375, |
| "learning_rate": 1.918e-05, |
| "loss": 10.3223, |
| "mean_token_accuracy": 0.8181851245462894, |
| "num_tokens": 31774137.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.6240503186360001, |
| "epoch": 1.552, |
| "grad_norm": 17.375, |
| "learning_rate": 1.938e-05, |
| "loss": 9.9901, |
| "mean_token_accuracy": 0.8241735894232989, |
| "num_tokens": 32101776.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.6293262420222163, |
| "epoch": 1.568, |
| "grad_norm": 16.625, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 10.0922, |
| "mean_token_accuracy": 0.8220919147133827, |
| "num_tokens": 32428021.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.6412162970751524, |
| "epoch": 1.584, |
| "grad_norm": 17.5, |
| "learning_rate": 1.978e-05, |
| "loss": 10.3053, |
| "mean_token_accuracy": 0.8172495547682047, |
| "num_tokens": 32752735.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.6345738202333451, |
| "epoch": 1.6, |
| "grad_norm": 16.375, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 10.2193, |
| "mean_token_accuracy": 0.8202366653829813, |
| "num_tokens": 33075822.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 1.0772727332115173, |
| "eval_biology_loss": 1.1990994215011597, |
| "eval_biology_mean_token_accuracy": 0.7000869359970093, |
| "eval_biology_num_tokens": 33075822.0, |
| "eval_biology_runtime": 29.6208, |
| "eval_biology_samples_per_second": 16.88, |
| "eval_biology_steps_per_second": 4.22, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.6503664858341217, |
| "eval_physics_loss": 0.651718020439148, |
| "eval_physics_mean_token_accuracy": 0.8178416604995727, |
| "eval_physics_num_tokens": 33075822.0, |
| "eval_physics_runtime": 42.1601, |
| "eval_physics_samples_per_second": 11.86, |
| "eval_physics_steps_per_second": 2.965, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.6182714030146599, |
| "epoch": 1.616, |
| "grad_norm": 17.375, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 9.8964, |
| "mean_token_accuracy": 0.8249843519181013, |
| "num_tokens": 33411850.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.6229340618476271, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 16.5, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 10.0248, |
| "mean_token_accuracy": 0.8223889589309692, |
| "num_tokens": 33739456.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.6091084901243449, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 17.5, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 9.7378, |
| "mean_token_accuracy": 0.8279588807374239, |
| "num_tokens": 34072022.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.628602746501565, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 18.25, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 10.1041, |
| "mean_token_accuracy": 0.8225820947438478, |
| "num_tokens": 34398959.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.6163907073438167, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 9.8895, |
| "mean_token_accuracy": 0.824999624863267, |
| "num_tokens": 34732284.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.6272942896932363, |
| "epoch": 1.696, |
| "grad_norm": 16.875, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 10.0732, |
| "mean_token_accuracy": 0.822333462536335, |
| "num_tokens": 35053712.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.6298937538638711, |
| "epoch": 1.712, |
| "grad_norm": 16.5, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 10.1089, |
| "mean_token_accuracy": 0.8207412119954824, |
| "num_tokens": 35393088.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.6155203450471163, |
| "epoch": 1.728, |
| "grad_norm": 16.125, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 9.9054, |
| "mean_token_accuracy": 0.8255299925804138, |
| "num_tokens": 35729007.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.6240347031503916, |
| "epoch": 1.744, |
| "grad_norm": 16.75, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 9.9979, |
| "mean_token_accuracy": 0.8231176160275936, |
| "num_tokens": 36072489.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.6143232045695186, |
| "epoch": 1.76, |
| "grad_norm": 17.375, |
| "learning_rate": 1.978e-05, |
| "loss": 9.8637, |
| "mean_token_accuracy": 0.8249252289533615, |
| "num_tokens": 36398285.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 1.0265369310379029, |
| "eval_biology_loss": 1.2094172239303589, |
| "eval_biology_mean_token_accuracy": 0.6984671940803527, |
| "eval_biology_num_tokens": 36398285.0, |
| "eval_biology_runtime": 29.6897, |
| "eval_biology_samples_per_second": 16.841, |
| "eval_biology_steps_per_second": 4.21, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.6072226724624634, |
| "eval_physics_loss": 0.6429793238639832, |
| "eval_physics_mean_token_accuracy": 0.8200512890815734, |
| "eval_physics_num_tokens": 36398285.0, |
| "eval_physics_runtime": 42.4662, |
| "eval_physics_samples_per_second": 11.774, |
| "eval_physics_steps_per_second": 2.944, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.6318713866174221, |
| "epoch": 1.776, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 10.1889, |
| "mean_token_accuracy": 0.8199843384325505, |
| "num_tokens": 36730479.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.6133379619568586, |
| "epoch": 1.792, |
| "grad_norm": 18.125, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 9.9002, |
| "mean_token_accuracy": 0.8245061755180358, |
| "num_tokens": 37057425.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.6075753191486001, |
| "epoch": 1.808, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 9.7216, |
| "mean_token_accuracy": 0.827298716455698, |
| "num_tokens": 37395449.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.6157500127330422, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 9.869, |
| "mean_token_accuracy": 0.8250419706106186, |
| "num_tokens": 37729535.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.6038926662877202, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 9.7492, |
| "mean_token_accuracy": 0.8271189860999584, |
| "num_tokens": 38057839.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.6130170978605747, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 16.0, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 9.7944, |
| "mean_token_accuracy": 0.8268033184111119, |
| "num_tokens": 38398033.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.6028442984446883, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 16.5, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 9.6695, |
| "mean_token_accuracy": 0.8263881340622902, |
| "num_tokens": 38729203.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.623160850442946, |
| "epoch": 1.888, |
| "grad_norm": 16.5, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 9.9773, |
| "mean_token_accuracy": 0.8230186630040407, |
| "num_tokens": 39058791.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.6327232969924808, |
| "epoch": 1.904, |
| "grad_norm": 16.0, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 10.1963, |
| "mean_token_accuracy": 0.8210305809974671, |
| "num_tokens": 39386170.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.6258664343506097, |
| "epoch": 1.92, |
| "grad_norm": 15.875, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 10.0632, |
| "mean_token_accuracy": 0.8224393729120493, |
| "num_tokens": 39710381.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 1.0642137598991395, |
| "eval_biology_loss": 1.2021751403808594, |
| "eval_biology_mean_token_accuracy": 0.6998101119995117, |
| "eval_biology_num_tokens": 39710381.0, |
| "eval_biology_runtime": 29.6797, |
| "eval_biology_samples_per_second": 16.847, |
| "eval_biology_steps_per_second": 4.212, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.6103473696708679, |
| "eval_physics_loss": 0.6331072449684143, |
| "eval_physics_mean_token_accuracy": 0.8220849981307984, |
| "eval_physics_num_tokens": 39710381.0, |
| "eval_physics_runtime": 42.4391, |
| "eval_physics_samples_per_second": 11.782, |
| "eval_physics_steps_per_second": 2.945, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.6147853201255202, |
| "epoch": 1.936, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 9.8494, |
| "mean_token_accuracy": 0.8241737693548202, |
| "num_tokens": 40043354.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5860691763460636, |
| "epoch": 1.952, |
| "grad_norm": 16.625, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 9.4581, |
| "mean_token_accuracy": 0.8311044085770846, |
| "num_tokens": 40375790.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.6076713547110557, |
| "epoch": 1.968, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 9.7817, |
| "mean_token_accuracy": 0.8264643102884293, |
| "num_tokens": 40705451.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.6071135563775897, |
| "epoch": 1.984, |
| "grad_norm": 16.625, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 9.7942, |
| "mean_token_accuracy": 0.8252186521887779, |
| "num_tokens": 41040313.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.6113535448908806, |
| "epoch": 2.0, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 9.7519, |
| "mean_token_accuracy": 0.8272297196090221, |
| "num_tokens": 41373608.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.530447144433856, |
| "epoch": 2.016, |
| "grad_norm": 17.0, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 8.3959, |
| "mean_token_accuracy": 0.8450614836066961, |
| "num_tokens": 41708226.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.5259812783449889, |
| "epoch": 2.032, |
| "grad_norm": 17.5, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 8.4414, |
| "mean_token_accuracy": 0.8447725884616375, |
| "num_tokens": 42045433.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.5413103114813567, |
| "epoch": 2.048, |
| "grad_norm": 19.75, |
| "learning_rate": 1.938e-05, |
| "loss": 8.6582, |
| "mean_token_accuracy": 0.8407626215368509, |
| "num_tokens": 42372694.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.5283081056550145, |
| "epoch": 2.064, |
| "grad_norm": 17.75, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 8.4357, |
| "mean_token_accuracy": 0.8448646664619446, |
| "num_tokens": 42710240.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.5373278863728046, |
| "epoch": 2.08, |
| "grad_norm": 17.375, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 8.6189, |
| "mean_token_accuracy": 0.8424648057669402, |
| "num_tokens": 43037844.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 0.9063808312416076, |
| "eval_biology_loss": 1.2580033540725708, |
| "eval_biology_mean_token_accuracy": 0.6951368255615235, |
| "eval_biology_num_tokens": 43037844.0, |
| "eval_biology_runtime": 29.6286, |
| "eval_biology_samples_per_second": 16.876, |
| "eval_biology_steps_per_second": 4.219, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.552986116886139, |
| "eval_physics_loss": 0.6381449103355408, |
| "eval_physics_mean_token_accuracy": 0.8220356435775756, |
| "eval_physics_num_tokens": 43037844.0, |
| "eval_physics_runtime": 42.3885, |
| "eval_physics_samples_per_second": 11.796, |
| "eval_physics_steps_per_second": 2.949, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.5337282193824648, |
| "epoch": 2.096, |
| "grad_norm": 16.75, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 8.5123, |
| "mean_token_accuracy": 0.8428315419703722, |
| "num_tokens": 43368675.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5185356924310327, |
| "epoch": 2.112, |
| "grad_norm": 17.0, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 8.3554, |
| "mean_token_accuracy": 0.8454546939581633, |
| "num_tokens": 43699544.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.5311926949769259, |
| "epoch": 2.128, |
| "grad_norm": 17.5, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 8.4936, |
| "mean_token_accuracy": 0.8429730042815209, |
| "num_tokens": 44012632.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5302754450589419, |
| "epoch": 2.144, |
| "grad_norm": 16.875, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 8.4927, |
| "mean_token_accuracy": 0.8428822424262762, |
| "num_tokens": 44349732.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5503825809806585, |
| "epoch": 2.16, |
| "grad_norm": 18.0, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 8.8253, |
| "mean_token_accuracy": 0.8380614928901196, |
| "num_tokens": 44680410.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.5267998551949858, |
| "epoch": 2.176, |
| "grad_norm": 16.75, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 8.4749, |
| "mean_token_accuracy": 0.8448099918663502, |
| "num_tokens": 45017922.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.5206218842417002, |
| "epoch": 2.192, |
| "grad_norm": 17.0, |
| "learning_rate": 1.918e-05, |
| "loss": 8.3612, |
| "mean_token_accuracy": 0.8453428644686938, |
| "num_tokens": 45348179.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.5240026475861669, |
| "epoch": 2.208, |
| "grad_norm": 18.875, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 8.369, |
| "mean_token_accuracy": 0.8455433510243893, |
| "num_tokens": 45678397.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.5279045928269624, |
| "epoch": 2.224, |
| "grad_norm": 17.0, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 8.434, |
| "mean_token_accuracy": 0.8444005087018013, |
| "num_tokens": 46010052.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.5289217280223966, |
| "epoch": 2.24, |
| "grad_norm": 17.375, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 8.44, |
| "mean_token_accuracy": 0.8444702833890915, |
| "num_tokens": 46347116.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 0.9073377561569214, |
| "eval_biology_loss": 1.2652989625930786, |
| "eval_biology_mean_token_accuracy": 0.6944872965812683, |
| "eval_biology_num_tokens": 46347116.0, |
| "eval_biology_runtime": 29.9507, |
| "eval_biology_samples_per_second": 16.694, |
| "eval_biology_steps_per_second": 4.174, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.5540535972118378, |
| "eval_physics_loss": 0.6368283629417419, |
| "eval_physics_mean_token_accuracy": 0.8228535957336426, |
| "eval_physics_num_tokens": 46347116.0, |
| "eval_physics_runtime": 42.3816, |
| "eval_physics_samples_per_second": 11.798, |
| "eval_physics_steps_per_second": 2.949, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.5402740124613047, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 18.5, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 8.7016, |
| "mean_token_accuracy": 0.8409985613077879, |
| "num_tokens": 46673074.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.5238024350255728, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 18.75, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 8.3724, |
| "mean_token_accuracy": 0.8452999964356422, |
| "num_tokens": 47009386.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.5359452277421951, |
| "epoch": 2.288, |
| "grad_norm": 17.625, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 8.6322, |
| "mean_token_accuracy": 0.8418019615113735, |
| "num_tokens": 47344231.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.5372845523059369, |
| "epoch": 2.304, |
| "grad_norm": 19.625, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 8.5597, |
| "mean_token_accuracy": 0.8422956768423319, |
| "num_tokens": 47658392.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.5315724445506931, |
| "epoch": 2.32, |
| "grad_norm": 16.125, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 8.5227, |
| "mean_token_accuracy": 0.8427653182297945, |
| "num_tokens": 47995015.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.5130708247423172, |
| "epoch": 2.336, |
| "grad_norm": 18.125, |
| "learning_rate": 1.898e-05, |
| "loss": 8.2514, |
| "mean_token_accuracy": 0.8464119978249073, |
| "num_tokens": 48336708.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5171284712851048, |
| "epoch": 2.352, |
| "grad_norm": 17.625, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 8.2526, |
| "mean_token_accuracy": 0.8467188984155655, |
| "num_tokens": 48665311.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.5282838650047779, |
| "epoch": 2.368, |
| "grad_norm": 17.25, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 8.4841, |
| "mean_token_accuracy": 0.8436643645167351, |
| "num_tokens": 49005334.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.534802301414311, |
| "epoch": 2.384, |
| "grad_norm": 19.375, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 8.5751, |
| "mean_token_accuracy": 0.8415927398949862, |
| "num_tokens": 49331936.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.5311837416142225, |
| "epoch": 2.4, |
| "grad_norm": 18.625, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 8.5063, |
| "mean_token_accuracy": 0.8436322450637818, |
| "num_tokens": 49671048.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 0.8830818734169006, |
| "eval_biology_loss": 1.2704880237579346, |
| "eval_biology_mean_token_accuracy": 0.6929694819450378, |
| "eval_biology_num_tokens": 49671048.0, |
| "eval_biology_runtime": 29.626, |
| "eval_biology_samples_per_second": 16.877, |
| "eval_biology_steps_per_second": 4.219, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.540393363237381, |
| "eval_physics_loss": 0.6307452321052551, |
| "eval_physics_mean_token_accuracy": 0.8239317517280579, |
| "eval_physics_num_tokens": 49671048.0, |
| "eval_physics_runtime": 42.2546, |
| "eval_physics_samples_per_second": 11.833, |
| "eval_physics_steps_per_second": 2.958, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.5313838511705399, |
| "epoch": 2.416, |
| "grad_norm": 17.25, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 8.5314, |
| "mean_token_accuracy": 0.8434251178056001, |
| "num_tokens": 50018182.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.5314315013587475, |
| "epoch": 2.432, |
| "grad_norm": 18.875, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 8.5551, |
| "mean_token_accuracy": 0.8438528429716825, |
| "num_tokens": 50341187.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.5497623635455966, |
| "epoch": 2.448, |
| "grad_norm": 17.875, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 8.769, |
| "mean_token_accuracy": 0.839348716661334, |
| "num_tokens": 50667019.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.5337263094261289, |
| "epoch": 2.464, |
| "grad_norm": 18.125, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 8.5788, |
| "mean_token_accuracy": 0.8423588499426842, |
| "num_tokens": 50991113.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.533509024977684, |
| "epoch": 2.48, |
| "grad_norm": 18.25, |
| "learning_rate": 1.878e-05, |
| "loss": 8.5669, |
| "mean_token_accuracy": 0.8423088423907756, |
| "num_tokens": 51319624.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.5355414936318994, |
| "epoch": 2.496, |
| "grad_norm": 18.0, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 8.5578, |
| "mean_token_accuracy": 0.8425650514662266, |
| "num_tokens": 51654633.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5151204424910247, |
| "epoch": 2.512, |
| "grad_norm": 18.75, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 8.3018, |
| "mean_token_accuracy": 0.8466352391988039, |
| "num_tokens": 51980051.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.5199752731248737, |
| "epoch": 2.528, |
| "grad_norm": 16.375, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 8.3413, |
| "mean_token_accuracy": 0.8458983488380909, |
| "num_tokens": 52316109.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.5363344758749008, |
| "epoch": 2.544, |
| "grad_norm": 19.0, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 8.5941, |
| "mean_token_accuracy": 0.8419183246791363, |
| "num_tokens": 52634528.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.5285956017673016, |
| "epoch": 2.56, |
| "grad_norm": 17.625, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 8.433, |
| "mean_token_accuracy": 0.8443803623318672, |
| "num_tokens": 52964035.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 0.9072599239349365, |
| "eval_biology_loss": 1.264472246170044, |
| "eval_biology_mean_token_accuracy": 0.6935372724533081, |
| "eval_biology_num_tokens": 52964035.0, |
| "eval_biology_runtime": 29.6445, |
| "eval_biology_samples_per_second": 16.867, |
| "eval_biology_steps_per_second": 4.217, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.5502645847797394, |
| "eval_physics_loss": 0.6250398755073547, |
| "eval_physics_mean_token_accuracy": 0.8245352702140808, |
| "eval_physics_num_tokens": 52964035.0, |
| "eval_physics_runtime": 42.3903, |
| "eval_physics_samples_per_second": 11.795, |
| "eval_physics_steps_per_second": 2.949, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.5171479491516948, |
| "epoch": 2.576, |
| "grad_norm": 18.125, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 8.3356, |
| "mean_token_accuracy": 0.8456523548811674, |
| "num_tokens": 53293345.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.5298814970999957, |
| "epoch": 2.592, |
| "grad_norm": 16.125, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 8.4943, |
| "mean_token_accuracy": 0.8433435622602701, |
| "num_tokens": 53620896.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.5211699862033129, |
| "epoch": 2.608, |
| "grad_norm": 16.25, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 8.3802, |
| "mean_token_accuracy": 0.8462613090872765, |
| "num_tokens": 53950989.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.5339049361646175, |
| "epoch": 2.624, |
| "grad_norm": 17.0, |
| "learning_rate": 1.858e-05, |
| "loss": 8.5302, |
| "mean_token_accuracy": 0.8435598284006118, |
| "num_tokens": 54282732.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.5256838278844953, |
| "epoch": 2.64, |
| "grad_norm": 17.75, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 8.473, |
| "mean_token_accuracy": 0.8441216733306647, |
| "num_tokens": 54609748.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.5526315724477172, |
| "epoch": 2.656, |
| "grad_norm": 19.5, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 8.8463, |
| "mean_token_accuracy": 0.8381063390523196, |
| "num_tokens": 54953583.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5346001913771034, |
| "epoch": 2.672, |
| "grad_norm": 17.875, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 8.5872, |
| "mean_token_accuracy": 0.8416836660355329, |
| "num_tokens": 55281945.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.507072365656495, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 17.0, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 8.1723, |
| "mean_token_accuracy": 0.848843240365386, |
| "num_tokens": 55612391.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.539692627824843, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 17.25, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 8.6013, |
| "mean_token_accuracy": 0.8423799473792315, |
| "num_tokens": 55949791.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.5213707761839033, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 20.875, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 8.3781, |
| "mean_token_accuracy": 0.8448753044009208, |
| "num_tokens": 56282204.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 0.8998830423355103, |
| "eval_biology_loss": 1.269989013671875, |
| "eval_biology_mean_token_accuracy": 0.6936762638092041, |
| "eval_biology_num_tokens": 56282204.0, |
| "eval_biology_runtime": 29.629, |
| "eval_biology_samples_per_second": 16.875, |
| "eval_biology_steps_per_second": 4.219, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.5432275445461273, |
| "eval_physics_loss": 0.6207777261734009, |
| "eval_physics_mean_token_accuracy": 0.8257771005630493, |
| "eval_physics_num_tokens": 56282204.0, |
| "eval_physics_runtime": 42.3855, |
| "eval_physics_samples_per_second": 11.796, |
| "eval_physics_steps_per_second": 2.949, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.5400118172168732, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 18.625, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 8.6625, |
| "mean_token_accuracy": 0.8404322098940611, |
| "num_tokens": 56605478.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.5394736461341381, |
| "epoch": 2.752, |
| "grad_norm": 17.625, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 8.6358, |
| "mean_token_accuracy": 0.8410456858575344, |
| "num_tokens": 56935874.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.5178206816315651, |
| "epoch": 2.768, |
| "grad_norm": 17.875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 8.337, |
| "mean_token_accuracy": 0.8459576744586229, |
| "num_tokens": 57270552.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.5335245341062546, |
| "epoch": 2.784, |
| "grad_norm": 18.125, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 8.5249, |
| "mean_token_accuracy": 0.8427888829261064, |
| "num_tokens": 57602968.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5528621880337596, |
| "epoch": 2.8, |
| "grad_norm": 18.25, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 8.8497, |
| "mean_token_accuracy": 0.8369761358946561, |
| "num_tokens": 57925886.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.5134829036891461, |
| "epoch": 2.816, |
| "grad_norm": 16.75, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 8.1668, |
| "mean_token_accuracy": 0.848399018868804, |
| "num_tokens": 58262960.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.5281499922275543, |
| "epoch": 2.832, |
| "grad_norm": 17.75, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 8.5747, |
| "mean_token_accuracy": 0.8418189071118831, |
| "num_tokens": 58586324.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.5403068274259567, |
| "epoch": 2.848, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 8.6062, |
| "mean_token_accuracy": 0.8411106329411269, |
| "num_tokens": 58912949.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.5275976810604334, |
| "epoch": 2.864, |
| "grad_norm": 16.125, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 8.4806, |
| "mean_token_accuracy": 0.844698841124773, |
| "num_tokens": 59252494.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.5257823524996639, |
| "epoch": 2.88, |
| "grad_norm": 18.875, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 8.4506, |
| "mean_token_accuracy": 0.8446649070829153, |
| "num_tokens": 59574951.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 0.91778861618042, |
| "eval_biology_loss": 1.2650281190872192, |
| "eval_biology_mean_token_accuracy": 0.69282026720047, |
| "eval_biology_num_tokens": 59574951.0, |
| "eval_biology_runtime": 29.6226, |
| "eval_biology_samples_per_second": 16.879, |
| "eval_biology_steps_per_second": 4.22, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.5498799192905426, |
| "eval_physics_loss": 0.6160721182823181, |
| "eval_physics_mean_token_accuracy": 0.8270136556625366, |
| "eval_physics_num_tokens": 59574951.0, |
| "eval_physics_runtime": 42.3762, |
| "eval_physics_samples_per_second": 11.799, |
| "eval_physics_steps_per_second": 2.95, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.5232849149033427, |
| "epoch": 2.896, |
| "grad_norm": 17.625, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 8.4238, |
| "mean_token_accuracy": 0.8446525242179632, |
| "num_tokens": 59903274.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.5211154764518142, |
| "epoch": 2.912, |
| "grad_norm": 17.375, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 8.3389, |
| "mean_token_accuracy": 0.8464620877057314, |
| "num_tokens": 60234203.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.5203663635998964, |
| "epoch": 2.928, |
| "grad_norm": 17.75, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 8.3211, |
| "mean_token_accuracy": 0.8467351488769055, |
| "num_tokens": 60559123.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.5172538913786411, |
| "epoch": 2.944, |
| "grad_norm": 19.375, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 8.3406, |
| "mean_token_accuracy": 0.8462818019092083, |
| "num_tokens": 60893423.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.5285945516079664, |
| "epoch": 2.96, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 8.4533, |
| "mean_token_accuracy": 0.8438907958567142, |
| "num_tokens": 61228692.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.5289302395656705, |
| "epoch": 2.976, |
| "grad_norm": 17.5, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 8.4515, |
| "mean_token_accuracy": 0.8438733883202076, |
| "num_tokens": 61560852.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.512842470407486, |
| "epoch": 2.992, |
| "grad_norm": 17.625, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 8.2149, |
| "mean_token_accuracy": 0.8474430676549674, |
| "num_tokens": 61898951.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.4715210357680917, |
| "epoch": 3.008, |
| "grad_norm": 23.625, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 7.2166, |
| "mean_token_accuracy": 0.8635042041540146, |
| "num_tokens": 62223875.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.37948312778025867, |
| "epoch": 3.024, |
| "grad_norm": 23.5, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 6.161, |
| "mean_token_accuracy": 0.880422555282712, |
| "num_tokens": 62553166.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.3847277104854584, |
| "epoch": 3.04, |
| "grad_norm": 24.0, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 6.1092, |
| "mean_token_accuracy": 0.8813776560127735, |
| "num_tokens": 62893113.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 0.6869136800765991, |
| "eval_biology_loss": 1.5112156867980957, |
| "eval_biology_mean_token_accuracy": 0.6802826480865478, |
| "eval_biology_num_tokens": 62893113.0, |
| "eval_biology_runtime": 29.5983, |
| "eval_biology_samples_per_second": 16.893, |
| "eval_biology_steps_per_second": 4.223, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.42201741313934327, |
| "eval_physics_loss": 0.6906136870384216, |
| "eval_physics_mean_token_accuracy": 0.8213669605255127, |
| "eval_physics_num_tokens": 62893113.0, |
| "eval_physics_runtime": 42.6547, |
| "eval_physics_samples_per_second": 11.722, |
| "eval_physics_steps_per_second": 2.931, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.37959762578830125, |
| "epoch": 3.056, |
| "grad_norm": 23.25, |
| "learning_rate": 1.798e-05, |
| "loss": 6.0285, |
| "mean_token_accuracy": 0.8816115573048592, |
| "num_tokens": 63218088.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.3821444820612669, |
| "epoch": 3.072, |
| "grad_norm": 24.875, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 6.0732, |
| "mean_token_accuracy": 0.8815719917416572, |
| "num_tokens": 63551201.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.39367833686992526, |
| "epoch": 3.088, |
| "grad_norm": 24.875, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 6.2704, |
| "mean_token_accuracy": 0.8780518177896738, |
| "num_tokens": 63876362.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.3746034217998385, |
| "epoch": 3.104, |
| "grad_norm": 23.75, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 5.9319, |
| "mean_token_accuracy": 0.8839064195752144, |
| "num_tokens": 64208171.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.37151786470785736, |
| "epoch": 3.12, |
| "grad_norm": 23.875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 5.9241, |
| "mean_token_accuracy": 0.884624307230115, |
| "num_tokens": 64544493.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.39232265017926693, |
| "epoch": 3.136, |
| "grad_norm": 23.625, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 6.2659, |
| "mean_token_accuracy": 0.87808175124228, |
| "num_tokens": 64872488.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.37443468971177934, |
| "epoch": 3.152, |
| "grad_norm": 24.0, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 5.9417, |
| "mean_token_accuracy": 0.8836083229631185, |
| "num_tokens": 65201124.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.385872579459101, |
| "epoch": 3.168, |
| "grad_norm": 23.0, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 6.1646, |
| "mean_token_accuracy": 0.8805806953459978, |
| "num_tokens": 65525868.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.37636062121018765, |
| "epoch": 3.184, |
| "grad_norm": 24.0, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 5.9906, |
| "mean_token_accuracy": 0.883328914642334, |
| "num_tokens": 65861603.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.3804218841716647, |
| "epoch": 3.2, |
| "grad_norm": 23.75, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 6.1031, |
| "mean_token_accuracy": 0.8820050243288279, |
| "num_tokens": 66193675.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 0.6348137440681457, |
| "eval_biology_loss": 1.5773308277130127, |
| "eval_biology_mean_token_accuracy": 0.6818766250610352, |
| "eval_biology_num_tokens": 66193675.0, |
| "eval_biology_runtime": 29.6404, |
| "eval_biology_samples_per_second": 16.869, |
| "eval_biology_steps_per_second": 4.217, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.4191882491111755, |
| "eval_physics_loss": 0.690969705581665, |
| "eval_physics_mean_token_accuracy": 0.8224700026512146, |
| "eval_physics_num_tokens": 66193675.0, |
| "eval_physics_runtime": 42.228, |
| "eval_physics_samples_per_second": 11.84, |
| "eval_physics_steps_per_second": 2.96, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 16, |
| "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": 6.614072648666411e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
