| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.335784598439932, |
| "epoch": 0.016, |
| "grad_norm": 157.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 34.5026, |
| "mean_token_accuracy": 0.6119794614613057, |
| "num_tokens": 338352.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.3430390540510415, |
| "epoch": 0.032, |
| "grad_norm": 137.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 34.2867, |
| "mean_token_accuracy": 0.6166948016732932, |
| "num_tokens": 671193.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.3716558743268252, |
| "epoch": 0.048, |
| "grad_norm": 121.5, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 33.3472, |
| "mean_token_accuracy": 0.6195640843361616, |
| "num_tokens": 1001008.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.492756663635373, |
| "epoch": 0.064, |
| "grad_norm": 91.5, |
| "learning_rate": 2.6e-06, |
| "loss": 32.9244, |
| "mean_token_accuracy": 0.6165778396651149, |
| "num_tokens": 1331208.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.5643325500190257, |
| "epoch": 0.08, |
| "grad_norm": 64.0, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 30.4382, |
| "mean_token_accuracy": 0.6357192637398839, |
| "num_tokens": 1667532.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.5284064427018165, |
| "epoch": 0.096, |
| "grad_norm": 43.5, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 27.3833, |
| "mean_token_accuracy": 0.6618047762662173, |
| "num_tokens": 2007176.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.5410372719168663, |
| "epoch": 0.112, |
| "grad_norm": 36.25, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 26.009, |
| "mean_token_accuracy": 0.6689645217731595, |
| "num_tokens": 2342259.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.5293128810822965, |
| "epoch": 0.128, |
| "grad_norm": 30.5, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 24.7913, |
| "mean_token_accuracy": 0.6720394570380449, |
| "num_tokens": 2663987.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.4024027574807405, |
| "epoch": 0.144, |
| "grad_norm": 22.0, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 22.4089, |
| "mean_token_accuracy": 0.6952127493917942, |
| "num_tokens": 2997824.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.3540743235498667, |
| "epoch": 0.16, |
| "grad_norm": 19.875, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 21.2896, |
| "mean_token_accuracy": 0.7063459653407336, |
| "num_tokens": 3330597.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 2.8890565853118897, |
| "eval_biology_loss": 3.7317216396331787, |
| "eval_biology_mean_token_accuracy": 0.4269867980480194, |
| "eval_biology_num_tokens": 3330597.0, |
| "eval_biology_runtime": 49.6093, |
| "eval_biology_samples_per_second": 10.079, |
| "eval_biology_steps_per_second": 2.52, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 1.682255081653595, |
| "eval_chemistry_loss": 1.6963560581207275, |
| "eval_chemistry_mean_token_accuracy": 0.6472484331130981, |
| "eval_chemistry_num_tokens": 3330597.0, |
| "eval_chemistry_runtime": 61.4961, |
| "eval_chemistry_samples_per_second": 8.131, |
| "eval_chemistry_steps_per_second": 2.033, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.142211413383484, |
| "eval_math_loss": 1.4016218185424805, |
| "eval_math_mean_token_accuracy": 0.7029124236106873, |
| "eval_math_num_tokens": 3330597.0, |
| "eval_math_runtime": 63.0568, |
| "eval_math_samples_per_second": 7.929, |
| "eval_math_steps_per_second": 1.982, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.3429004917144776, |
| "eval_physics_loss": 1.3169747591018677, |
| "eval_physics_mean_token_accuracy": 0.7064582467079162, |
| "eval_physics_num_tokens": 3330597.0, |
| "eval_physics_runtime": 71.749, |
| "eval_physics_samples_per_second": 6.969, |
| "eval_physics_steps_per_second": 1.742, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.3304455887526274, |
| "epoch": 0.176, |
| "grad_norm": 22.375, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 20.7106, |
| "mean_token_accuracy": 0.7095530580729246, |
| "num_tokens": 3658264.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.2620876263827085, |
| "epoch": 0.192, |
| "grad_norm": 16.375, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 19.8944, |
| "mean_token_accuracy": 0.7141206510365009, |
| "num_tokens": 3995568.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.201955909654498, |
| "epoch": 0.208, |
| "grad_norm": 17.375, |
| "learning_rate": 8.6e-06, |
| "loss": 18.9243, |
| "mean_token_accuracy": 0.7253232792019844, |
| "num_tokens": 4321436.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.178970755264163, |
| "epoch": 0.224, |
| "grad_norm": 15.6875, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 18.583, |
| "mean_token_accuracy": 0.7276350263506174, |
| "num_tokens": 4648491.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.155185490846634, |
| "epoch": 0.24, |
| "grad_norm": 16.625, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 18.4686, |
| "mean_token_accuracy": 0.7289414200931787, |
| "num_tokens": 4986175.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.1447543263435365, |
| "epoch": 0.256, |
| "grad_norm": 17.125, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 18.0999, |
| "mean_token_accuracy": 0.7330310437828302, |
| "num_tokens": 5329320.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.142161527276039, |
| "epoch": 0.272, |
| "grad_norm": 17.625, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 18.0884, |
| "mean_token_accuracy": 0.7317263871431351, |
| "num_tokens": 5658796.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.1180811140686273, |
| "epoch": 0.288, |
| "grad_norm": 15.75, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 17.8484, |
| "mean_token_accuracy": 0.7360638868063688, |
| "num_tokens": 5980201.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.0995653536170722, |
| "epoch": 0.304, |
| "grad_norm": 19.75, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 17.5458, |
| "mean_token_accuracy": 0.7386808756738901, |
| "num_tokens": 6298858.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.0919909674674273, |
| "epoch": 0.32, |
| "grad_norm": 17.75, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 17.376, |
| "mean_token_accuracy": 0.7401029217988253, |
| "num_tokens": 6622798.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 2.781595293998718, |
| "eval_biology_loss": 3.2204973697662354, |
| "eval_biology_mean_token_accuracy": 0.46290496277809146, |
| "eval_biology_num_tokens": 6622798.0, |
| "eval_biology_runtime": 49.6202, |
| "eval_biology_samples_per_second": 10.077, |
| "eval_biology_steps_per_second": 2.519, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 1.4062482738494873, |
| "eval_chemistry_loss": 1.4335906505584717, |
| "eval_chemistry_mean_token_accuracy": 0.6814104890823365, |
| "eval_chemistry_num_tokens": 6622798.0, |
| "eval_chemistry_runtime": 61.4965, |
| "eval_chemistry_samples_per_second": 8.131, |
| "eval_chemistry_steps_per_second": 2.033, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.0162367067337037, |
| "eval_math_loss": 1.2369706630706787, |
| "eval_math_mean_token_accuracy": 0.721817741394043, |
| "eval_math_num_tokens": 6622798.0, |
| "eval_math_runtime": 63.0765, |
| "eval_math_samples_per_second": 7.927, |
| "eval_math_steps_per_second": 1.982, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.067228192806244, |
| "eval_physics_loss": 1.0861860513687134, |
| "eval_physics_mean_token_accuracy": 0.7410436091423035, |
| "eval_physics_num_tokens": 6622798.0, |
| "eval_physics_runtime": 71.8547, |
| "eval_physics_samples_per_second": 6.958, |
| "eval_physics_steps_per_second": 1.74, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0740001201629639, |
| "epoch": 0.336, |
| "grad_norm": 13.8125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 17.1669, |
| "mean_token_accuracy": 0.74343366548419, |
| "num_tokens": 6953611.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.042375946417451, |
| "epoch": 0.352, |
| "grad_norm": 16.5, |
| "learning_rate": 1.46e-05, |
| "loss": 16.6192, |
| "mean_token_accuracy": 0.7505168370902539, |
| "num_tokens": 7290378.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.0228992454707622, |
| "epoch": 0.368, |
| "grad_norm": 16.375, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 16.3576, |
| "mean_token_accuracy": 0.7527059823274612, |
| "num_tokens": 7621082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.0688911959528924, |
| "epoch": 0.384, |
| "grad_norm": 17.0, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 17.0574, |
| "mean_token_accuracy": 0.7418320391327142, |
| "num_tokens": 7955570.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.016674182936549, |
| "epoch": 0.4, |
| "grad_norm": 17.375, |
| "learning_rate": 1.66e-05, |
| "loss": 16.2747, |
| "mean_token_accuracy": 0.7516404036432505, |
| "num_tokens": 8291049.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.9974442631006241, |
| "epoch": 0.416, |
| "grad_norm": 16.0, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 15.9357, |
| "mean_token_accuracy": 0.7562779419124126, |
| "num_tokens": 8627310.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.0080594643950462, |
| "epoch": 0.432, |
| "grad_norm": 16.625, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 16.1472, |
| "mean_token_accuracy": 0.7539955742657185, |
| "num_tokens": 8958371.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9669299107044935, |
| "epoch": 0.448, |
| "grad_norm": 15.625, |
| "learning_rate": 1.86e-05, |
| "loss": 15.5349, |
| "mean_token_accuracy": 0.7622108798474073, |
| "num_tokens": 9294388.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.9649454571306706, |
| "epoch": 0.464, |
| "grad_norm": 16.625, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 15.3696, |
| "mean_token_accuracy": 0.7622643817216158, |
| "num_tokens": 9620366.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.9703133786097169, |
| "epoch": 0.48, |
| "grad_norm": 17.125, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.5958, |
| "mean_token_accuracy": 0.7595039043575526, |
| "num_tokens": 9955431.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 2.6708189754486082, |
| "eval_biology_loss": 3.0823237895965576, |
| "eval_biology_mean_token_accuracy": 0.4763511681556702, |
| "eval_biology_num_tokens": 9955431.0, |
| "eval_biology_runtime": 49.4876, |
| "eval_biology_samples_per_second": 10.104, |
| "eval_biology_steps_per_second": 2.526, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 1.320955008983612, |
| "eval_chemistry_loss": 1.3409372568130493, |
| "eval_chemistry_mean_token_accuracy": 0.6957354559898377, |
| "eval_chemistry_num_tokens": 9955431.0, |
| "eval_chemistry_runtime": 61.3511, |
| "eval_chemistry_samples_per_second": 8.15, |
| "eval_chemistry_steps_per_second": 2.037, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.9890776557922363, |
| "eval_math_loss": 1.1766003370285034, |
| "eval_math_mean_token_accuracy": 0.732050290107727, |
| "eval_math_num_tokens": 9955431.0, |
| "eval_math_runtime": 62.9199, |
| "eval_math_samples_per_second": 7.947, |
| "eval_math_steps_per_second": 1.987, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 0.9889620594978332, |
| "eval_physics_loss": 0.9930320978164673, |
| "eval_physics_mean_token_accuracy": 0.7561589913368225, |
| "eval_physics_num_tokens": 9955431.0, |
| "eval_physics_runtime": 71.6691, |
| "eval_physics_samples_per_second": 6.977, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9819621413946151, |
| "epoch": 0.496, |
| "grad_norm": 14.5625, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.7494, |
| "mean_token_accuracy": 0.7583937518298626, |
| "num_tokens": 10289883.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.9672155775129795, |
| "epoch": 0.512, |
| "grad_norm": 18.625, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 15.5391, |
| "mean_token_accuracy": 0.7612126469612122, |
| "num_tokens": 10619468.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.978820975869894, |
| "epoch": 0.528, |
| "grad_norm": 16.0, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 15.6782, |
| "mean_token_accuracy": 0.7582353662699461, |
| "num_tokens": 10944904.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.963908814266324, |
| "epoch": 0.544, |
| "grad_norm": 16.625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 15.4658, |
| "mean_token_accuracy": 0.7615648847073316, |
| "num_tokens": 11266409.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.9442941181361675, |
| "epoch": 0.56, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 15.0885, |
| "mean_token_accuracy": 0.7654793575406075, |
| "num_tokens": 11605544.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.9328314460813999, |
| "epoch": 0.576, |
| "grad_norm": 16.25, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 15.0173, |
| "mean_token_accuracy": 0.768263740092516, |
| "num_tokens": 11931715.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.9495550885796546, |
| "epoch": 0.592, |
| "grad_norm": 16.5, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 15.1932, |
| "mean_token_accuracy": 0.7637792613357306, |
| "num_tokens": 12257243.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9208816282451153, |
| "epoch": 0.608, |
| "grad_norm": 14.75, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 14.8406, |
| "mean_token_accuracy": 0.7683777522295714, |
| "num_tokens": 12583154.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.9426417943090201, |
| "epoch": 0.624, |
| "grad_norm": 15.5, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 15.1649, |
| "mean_token_accuracy": 0.7648395039141178, |
| "num_tokens": 12905392.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.9430808376520872, |
| "epoch": 0.64, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 15.1432, |
| "mean_token_accuracy": 0.7654214475303889, |
| "num_tokens": 13232198.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 2.6180445404052732, |
| "eval_biology_loss": 2.958848237991333, |
| "eval_biology_mean_token_accuracy": 0.48831122827529905, |
| "eval_biology_num_tokens": 13232198.0, |
| "eval_biology_runtime": 49.452, |
| "eval_biology_samples_per_second": 10.111, |
| "eval_biology_steps_per_second": 2.528, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 1.2611312751770019, |
| "eval_chemistry_loss": 1.2864106893539429, |
| "eval_chemistry_mean_token_accuracy": 0.7043857793807984, |
| "eval_chemistry_num_tokens": 13232198.0, |
| "eval_chemistry_runtime": 61.4019, |
| "eval_chemistry_samples_per_second": 8.143, |
| "eval_chemistry_steps_per_second": 2.036, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.9471726126670837, |
| "eval_math_loss": 1.148703694343567, |
| "eval_math_mean_token_accuracy": 0.7371885905265808, |
| "eval_math_num_tokens": 13232198.0, |
| "eval_math_runtime": 62.9339, |
| "eval_math_samples_per_second": 7.945, |
| "eval_math_steps_per_second": 1.986, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.9362427434921264, |
| "eval_physics_loss": 0.9423627853393555, |
| "eval_physics_mean_token_accuracy": 0.7653819880485535, |
| "eval_physics_num_tokens": 13232198.0, |
| "eval_physics_runtime": 71.631, |
| "eval_physics_samples_per_second": 6.98, |
| "eval_physics_steps_per_second": 1.745, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.9263238042593003, |
| "epoch": 0.656, |
| "grad_norm": 16.125, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 14.9, |
| "mean_token_accuracy": 0.7683228138834238, |
| "num_tokens": 13575902.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.9128325123339891, |
| "epoch": 0.672, |
| "grad_norm": 17.5, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 14.587, |
| "mean_token_accuracy": 0.771899900585413, |
| "num_tokens": 13895997.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.9165764002129435, |
| "epoch": 0.688, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 14.7149, |
| "mean_token_accuracy": 0.7700499434024095, |
| "num_tokens": 14234888.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.9249953411519527, |
| "epoch": 0.704, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 14.9209, |
| "mean_token_accuracy": 0.7661528721451759, |
| "num_tokens": 14567908.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.9112705899402499, |
| "epoch": 0.72, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 14.5659, |
| "mean_token_accuracy": 0.7717954892665148, |
| "num_tokens": 14882927.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.90460856705904, |
| "epoch": 0.736, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 14.5339, |
| "mean_token_accuracy": 0.7712412856519222, |
| "num_tokens": 15217342.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.8897293049842119, |
| "epoch": 0.752, |
| "grad_norm": 15.5, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 14.2105, |
| "mean_token_accuracy": 0.7750069301575422, |
| "num_tokens": 15547710.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.9218722280114889, |
| "epoch": 0.768, |
| "grad_norm": 16.25, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 14.8733, |
| "mean_token_accuracy": 0.7659960601478815, |
| "num_tokens": 15877177.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.8870411489158869, |
| "epoch": 0.784, |
| "grad_norm": 13.9375, |
| "learning_rate": 1.86e-05, |
| "loss": 14.2657, |
| "mean_token_accuracy": 0.7761844106018543, |
| "num_tokens": 16219640.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9038290875032544, |
| "epoch": 0.8, |
| "grad_norm": 16.0, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 14.5034, |
| "mean_token_accuracy": 0.7724477723240852, |
| "num_tokens": 16548261.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 2.657470052719116, |
| "eval_biology_loss": 2.942234516143799, |
| "eval_biology_mean_token_accuracy": 0.49295870113372803, |
| "eval_biology_num_tokens": 16548261.0, |
| "eval_biology_runtime": 49.4094, |
| "eval_biology_samples_per_second": 10.12, |
| "eval_biology_steps_per_second": 2.53, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 1.2639175114631653, |
| "eval_chemistry_loss": 1.2617712020874023, |
| "eval_chemistry_mean_token_accuracy": 0.7095342454910278, |
| "eval_chemistry_num_tokens": 16548261.0, |
| "eval_chemistry_runtime": 60.9985, |
| "eval_chemistry_samples_per_second": 8.197, |
| "eval_chemistry_steps_per_second": 2.049, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.9355086851119995, |
| "eval_math_loss": 1.1333692073822021, |
| "eval_math_mean_token_accuracy": 0.7395946278572082, |
| "eval_math_num_tokens": 16548261.0, |
| "eval_math_runtime": 62.8632, |
| "eval_math_samples_per_second": 7.954, |
| "eval_math_steps_per_second": 1.988, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.9225423102378845, |
| "eval_physics_loss": 0.9129735231399536, |
| "eval_physics_mean_token_accuracy": 0.770587375164032, |
| "eval_physics_num_tokens": 16548261.0, |
| "eval_physics_runtime": 71.6493, |
| "eval_physics_samples_per_second": 6.978, |
| "eval_physics_steps_per_second": 1.745, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.890050390549004, |
| "epoch": 0.816, |
| "grad_norm": 14.8125, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 14.3791, |
| "mean_token_accuracy": 0.7720852673053742, |
| "num_tokens": 16870533.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.9036840088665485, |
| "epoch": 0.832, |
| "grad_norm": 16.5, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 14.5054, |
| "mean_token_accuracy": 0.7727169577032328, |
| "num_tokens": 17196371.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.8895360480993986, |
| "epoch": 0.848, |
| "grad_norm": 13.9375, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 14.2591, |
| "mean_token_accuracy": 0.7748707763850688, |
| "num_tokens": 17527066.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.8957800149917603, |
| "epoch": 0.864, |
| "grad_norm": 14.5, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 14.3928, |
| "mean_token_accuracy": 0.7738867543637753, |
| "num_tokens": 17859605.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.9031556732952595, |
| "epoch": 0.88, |
| "grad_norm": 16.875, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 14.5649, |
| "mean_token_accuracy": 0.7709400031715632, |
| "num_tokens": 18187315.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.8537686172872782, |
| "epoch": 0.896, |
| "grad_norm": 14.0, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 13.6764, |
| "mean_token_accuracy": 0.7827610898762941, |
| "num_tokens": 18536790.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.8935620501637459, |
| "epoch": 0.912, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 14.3654, |
| "mean_token_accuracy": 0.7719177346676588, |
| "num_tokens": 18869344.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.8791772209107875, |
| "epoch": 0.928, |
| "grad_norm": 16.625, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 14.1875, |
| "mean_token_accuracy": 0.774543008953333, |
| "num_tokens": 19199834.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.8686152957379818, |
| "epoch": 0.944, |
| "grad_norm": 15.875, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 13.88, |
| "mean_token_accuracy": 0.7793182540684939, |
| "num_tokens": 19525385.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.8704672519117593, |
| "epoch": 0.96, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 14.0634, |
| "mean_token_accuracy": 0.7773549720644951, |
| "num_tokens": 19856973.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 2.5744068441390993, |
| "eval_biology_loss": 2.907360315322876, |
| "eval_biology_mean_token_accuracy": 0.49524139547348023, |
| "eval_biology_num_tokens": 19856973.0, |
| "eval_biology_runtime": 49.0409, |
| "eval_biology_samples_per_second": 10.196, |
| "eval_biology_steps_per_second": 2.549, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 1.2219904832839965, |
| "eval_chemistry_loss": 1.2410826683044434, |
| "eval_chemistry_mean_token_accuracy": 0.7126172127723693, |
| "eval_chemistry_num_tokens": 19856973.0, |
| "eval_chemistry_runtime": 60.7859, |
| "eval_chemistry_samples_per_second": 8.226, |
| "eval_chemistry_steps_per_second": 2.056, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.9290778393745422, |
| "eval_math_loss": 1.1203620433807373, |
| "eval_math_mean_token_accuracy": 0.7419518904685974, |
| "eval_math_num_tokens": 19856973.0, |
| "eval_math_runtime": 62.3226, |
| "eval_math_samples_per_second": 8.023, |
| "eval_math_steps_per_second": 2.006, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.8969436359405517, |
| "eval_physics_loss": 0.8939215540885925, |
| "eval_physics_mean_token_accuracy": 0.7737585263252258, |
| "eval_physics_num_tokens": 19856973.0, |
| "eval_physics_runtime": 71.3008, |
| "eval_physics_samples_per_second": 7.013, |
| "eval_physics_steps_per_second": 1.753, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8691291939467192, |
| "epoch": 0.976, |
| "grad_norm": 16.0, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 13.9642, |
| "mean_token_accuracy": 0.7793323084712028, |
| "num_tokens": 20190600.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.8796132244169712, |
| "epoch": 0.992, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 14.1458, |
| "mean_token_accuracy": 0.7762652188539505, |
| "num_tokens": 20518014.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8544556263834238, |
| "epoch": 1.008, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 13.629, |
| "mean_token_accuracy": 0.7817003551870585, |
| "num_tokens": 20859215.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.8545751355588436, |
| "epoch": 1.024, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 13.7525, |
| "mean_token_accuracy": 0.7807697676122188, |
| "num_tokens": 21186658.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.8538653265684844, |
| "epoch": 1.04, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 13.7341, |
| "mean_token_accuracy": 0.7808921810239553, |
| "num_tokens": 21517237.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8380249921232462, |
| "epoch": 1.056, |
| "grad_norm": 14.9375, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 13.4639, |
| "mean_token_accuracy": 0.7846374321728945, |
| "num_tokens": 21852528.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.855887845158577, |
| "epoch": 1.072, |
| "grad_norm": 14.125, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 13.7632, |
| "mean_token_accuracy": 0.7813815232366323, |
| "num_tokens": 22192380.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.8563600070774555, |
| "epoch": 1.088, |
| "grad_norm": 17.125, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 13.7002, |
| "mean_token_accuracy": 0.7817271586507559, |
| "num_tokens": 22517971.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.8393923103809356, |
| "epoch": 1.104, |
| "grad_norm": 14.0625, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 13.5135, |
| "mean_token_accuracy": 0.7835057020187378, |
| "num_tokens": 22848694.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.8503093957901001, |
| "epoch": 1.12, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 13.7396, |
| "mean_token_accuracy": 0.7802822068333626, |
| "num_tokens": 23183889.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 2.5127918043136597, |
| "eval_biology_loss": 2.924085855484009, |
| "eval_biology_mean_token_accuracy": 0.49510638427734377, |
| "eval_biology_num_tokens": 23183889.0, |
| "eval_biology_runtime": 49.4091, |
| "eval_biology_samples_per_second": 10.12, |
| "eval_biology_steps_per_second": 2.53, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 1.1826457905769348, |
| "eval_chemistry_loss": 1.23135244846344, |
| "eval_chemistry_mean_token_accuracy": 0.7140716204643249, |
| "eval_chemistry_num_tokens": 23183889.0, |
| "eval_chemistry_runtime": 61.3044, |
| "eval_chemistry_samples_per_second": 8.156, |
| "eval_chemistry_steps_per_second": 2.039, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.9012531743049622, |
| "eval_math_loss": 1.1168676614761353, |
| "eval_math_mean_token_accuracy": 0.742647081375122, |
| "eval_math_num_tokens": 23183889.0, |
| "eval_math_runtime": 62.8933, |
| "eval_math_samples_per_second": 7.95, |
| "eval_math_steps_per_second": 1.987, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.8671332702636719, |
| "eval_physics_loss": 0.8811478614807129, |
| "eval_physics_mean_token_accuracy": 0.7764565787315368, |
| "eval_physics_num_tokens": 23183889.0, |
| "eval_physics_runtime": 71.7862, |
| "eval_physics_samples_per_second": 6.965, |
| "eval_physics_steps_per_second": 1.741, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8146034182980657, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 13.6875, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 13.0963, |
| "mean_token_accuracy": 0.7886282972991466, |
| "num_tokens": 23523903.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.8425636734813452, |
| "epoch": 1.152, |
| "grad_norm": 16.875, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 13.54, |
| "mean_token_accuracy": 0.7844115447252988, |
| "num_tokens": 23848848.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.8615062309429049, |
| "epoch": 1.168, |
| "grad_norm": 16.25, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 13.7096, |
| "mean_token_accuracy": 0.7805542998015881, |
| "num_tokens": 24174589.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.8374113839119672, |
| "epoch": 1.184, |
| "grad_norm": 16.25, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 13.5026, |
| "mean_token_accuracy": 0.7833674903959036, |
| "num_tokens": 24495981.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.8216371892020107, |
| "epoch": 1.2, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 13.302, |
| "mean_token_accuracy": 0.786701825261116, |
| "num_tokens": 24826245.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.848210446164012, |
| "epoch": 1.216, |
| "grad_norm": 18.5, |
| "learning_rate": 1.66e-05, |
| "loss": 13.5667, |
| "mean_token_accuracy": 0.7812601257115602, |
| "num_tokens": 25147775.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8356876520439982, |
| "epoch": 1.232, |
| "grad_norm": 16.0, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 13.4138, |
| "mean_token_accuracy": 0.7863026071339846, |
| "num_tokens": 25480033.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.8342631004750729, |
| "epoch": 1.248, |
| "grad_norm": 16.5, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 13.3815, |
| "mean_token_accuracy": 0.7856004070490599, |
| "num_tokens": 25811978.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8264019176363945, |
| "epoch": 1.264, |
| "grad_norm": 16.125, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 13.3075, |
| "mean_token_accuracy": 0.7860219534486532, |
| "num_tokens": 26146299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.8564631534740329, |
| "epoch": 1.28, |
| "grad_norm": 16.125, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 13.8051, |
| "mean_token_accuracy": 0.778905876353383, |
| "num_tokens": 26478207.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 2.501225646018982, |
| "eval_biology_loss": 2.926992177963257, |
| "eval_biology_mean_token_accuracy": 0.49775544834136964, |
| "eval_biology_num_tokens": 26478207.0, |
| "eval_biology_runtime": 49.5165, |
| "eval_biology_samples_per_second": 10.098, |
| "eval_biology_steps_per_second": 2.524, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.1599062328338623, |
| "eval_chemistry_loss": 1.2233794927597046, |
| "eval_chemistry_mean_token_accuracy": 0.7164447598457336, |
| "eval_chemistry_num_tokens": 26478207.0, |
| "eval_chemistry_runtime": 61.2476, |
| "eval_chemistry_samples_per_second": 8.164, |
| "eval_chemistry_steps_per_second": 2.041, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.8935993061065673, |
| "eval_math_loss": 1.113796591758728, |
| "eval_math_mean_token_accuracy": 0.7438100085258484, |
| "eval_math_num_tokens": 26478207.0, |
| "eval_math_runtime": 63.571, |
| "eval_math_samples_per_second": 7.865, |
| "eval_math_steps_per_second": 1.966, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.8521192736625671, |
| "eval_physics_loss": 0.8711134791374207, |
| "eval_physics_mean_token_accuracy": 0.7784227828979492, |
| "eval_physics_num_tokens": 26478207.0, |
| "eval_physics_runtime": 71.9421, |
| "eval_physics_samples_per_second": 6.95, |
| "eval_physics_steps_per_second": 1.738, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8208133645355702, |
| "epoch": 1.296, |
| "grad_norm": 16.375, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 13.1842, |
| "mean_token_accuracy": 0.7876175127923488, |
| "num_tokens": 26806477.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.854189190082252, |
| "epoch": 1.312, |
| "grad_norm": 16.875, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 13.6651, |
| "mean_token_accuracy": 0.7810488305985928, |
| "num_tokens": 27137528.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8269255790859461, |
| "epoch": 1.328, |
| "grad_norm": 17.5, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 13.368, |
| "mean_token_accuracy": 0.7851872753351927, |
| "num_tokens": 27465859.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8443555533885956, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 16.5, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 13.5911, |
| "mean_token_accuracy": 0.7820561602711678, |
| "num_tokens": 27795568.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.8374204233288765, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 16.0, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 13.4833, |
| "mean_token_accuracy": 0.783146658167243, |
| "num_tokens": 28129988.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8366791510954499, |
| "epoch": 1.376, |
| "grad_norm": 17.75, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 13.3637, |
| "mean_token_accuracy": 0.7853023037314415, |
| "num_tokens": 28465903.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.8118171757087111, |
| "epoch": 1.392, |
| "grad_norm": 17.25, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 13.0363, |
| "mean_token_accuracy": 0.7896989852190017, |
| "num_tokens": 28801554.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.8223038006573915, |
| "epoch": 1.408, |
| "grad_norm": 15.875, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 13.2311, |
| "mean_token_accuracy": 0.7867654282599688, |
| "num_tokens": 29125114.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.8510244162753224, |
| "epoch": 1.424, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 13.7647, |
| "mean_token_accuracy": 0.7792895000427962, |
| "num_tokens": 29454334.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.805036261677742, |
| "epoch": 1.44, |
| "grad_norm": 15.25, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 12.9122, |
| "mean_token_accuracy": 0.7907555736601353, |
| "num_tokens": 29789253.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 2.507858594894409, |
| "eval_biology_loss": 2.8971524238586426, |
| "eval_biology_mean_token_accuracy": 0.49738200902938845, |
| "eval_biology_num_tokens": 29789253.0, |
| "eval_biology_runtime": 49.4122, |
| "eval_biology_samples_per_second": 10.119, |
| "eval_biology_steps_per_second": 2.53, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.1563323426246643, |
| "eval_chemistry_loss": 1.2145193815231323, |
| "eval_chemistry_mean_token_accuracy": 0.717510892868042, |
| "eval_chemistry_num_tokens": 29789253.0, |
| "eval_chemistry_runtime": 61.2472, |
| "eval_chemistry_samples_per_second": 8.164, |
| "eval_chemistry_steps_per_second": 2.041, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.8971080303192138, |
| "eval_math_loss": 1.1086454391479492, |
| "eval_math_mean_token_accuracy": 0.7439061784744263, |
| "eval_math_num_tokens": 29789253.0, |
| "eval_math_runtime": 62.8193, |
| "eval_math_samples_per_second": 7.959, |
| "eval_math_steps_per_second": 1.99, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.8453807921409607, |
| "eval_physics_loss": 0.8617435693740845, |
| "eval_physics_mean_token_accuracy": 0.7796971354484558, |
| "eval_physics_num_tokens": 29789253.0, |
| "eval_physics_runtime": 71.6142, |
| "eval_physics_samples_per_second": 6.982, |
| "eval_physics_steps_per_second": 1.745, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8077683765441179, |
| "epoch": 1.456, |
| "grad_norm": 13.8125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 13.0229, |
| "mean_token_accuracy": 0.7907570086419582, |
| "num_tokens": 30128576.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.8584844920784235, |
| "epoch": 1.472, |
| "grad_norm": 17.0, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 13.8325, |
| "mean_token_accuracy": 0.7795857060700655, |
| "num_tokens": 30452719.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.8337999247014523, |
| "epoch": 1.488, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 13.3892, |
| "mean_token_accuracy": 0.7849970065057278, |
| "num_tokens": 30782343.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.821822557784617, |
| "epoch": 1.504, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 13.1466, |
| "mean_token_accuracy": 0.7879386112093926, |
| "num_tokens": 31113238.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.8403494212776422, |
| "epoch": 1.52, |
| "grad_norm": 16.375, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 13.4805, |
| "mean_token_accuracy": 0.7833794906735421, |
| "num_tokens": 31444589.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8343268791213632, |
| "epoch": 1.536, |
| "grad_norm": 17.25, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 13.539, |
| "mean_token_accuracy": 0.7820737387984991, |
| "num_tokens": 31774137.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8169895254075528, |
| "epoch": 1.552, |
| "grad_norm": 16.625, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 12.9985, |
| "mean_token_accuracy": 0.7904236756265164, |
| "num_tokens": 32101776.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.8104152591899038, |
| "epoch": 1.568, |
| "grad_norm": 15.625, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 13.079, |
| "mean_token_accuracy": 0.7893446046859026, |
| "num_tokens": 32428021.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.8398273181170225, |
| "epoch": 1.584, |
| "grad_norm": 16.25, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 13.559, |
| "mean_token_accuracy": 0.7802358068525791, |
| "num_tokens": 32752735.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8366536511108279, |
| "epoch": 1.6, |
| "grad_norm": 15.5, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 13.3858, |
| "mean_token_accuracy": 0.7848577659577132, |
| "num_tokens": 33075822.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 2.4744019918441773, |
| "eval_biology_loss": 2.8814501762390137, |
| "eval_biology_mean_token_accuracy": 0.5001691384315491, |
| "eval_biology_num_tokens": 33075822.0, |
| "eval_biology_runtime": 49.6638, |
| "eval_biology_samples_per_second": 10.068, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.1388019576072692, |
| "eval_chemistry_loss": 1.2091480493545532, |
| "eval_chemistry_mean_token_accuracy": 0.7188141541481018, |
| "eval_chemistry_num_tokens": 33075822.0, |
| "eval_chemistry_runtime": 61.119, |
| "eval_chemistry_samples_per_second": 8.181, |
| "eval_chemistry_steps_per_second": 2.045, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.8872386133670807, |
| "eval_math_loss": 1.107426404953003, |
| "eval_math_mean_token_accuracy": 0.7446602578163147, |
| "eval_math_num_tokens": 33075822.0, |
| "eval_math_runtime": 62.8503, |
| "eval_math_samples_per_second": 7.955, |
| "eval_math_steps_per_second": 1.989, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.832112268447876, |
| "eval_physics_loss": 0.8559062480926514, |
| "eval_physics_mean_token_accuracy": 0.7812981810569763, |
| "eval_physics_num_tokens": 33075822.0, |
| "eval_physics_runtime": 71.6577, |
| "eval_physics_samples_per_second": 6.978, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.7955857511609793, |
| "epoch": 1.616, |
| "grad_norm": 16.125, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 12.8723, |
| "mean_token_accuracy": 0.7925955194979906, |
| "num_tokens": 33411850.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8143383387476206, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 16.375, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 13.07, |
| "mean_token_accuracy": 0.7890652883797884, |
| "num_tokens": 33739456.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.7856693360954523, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 15.875, |
| "learning_rate": 1.46e-05, |
| "loss": 12.6467, |
| "mean_token_accuracy": 0.7940549373626709, |
| "num_tokens": 34072022.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.8198095491155982, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 17.375, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 13.1094, |
| "mean_token_accuracy": 0.7891215395182372, |
| "num_tokens": 34398959.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.7961690971627832, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 12.9049, |
| "mean_token_accuracy": 0.790962991118431, |
| "num_tokens": 34732284.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8210279244929553, |
| "epoch": 1.696, |
| "grad_norm": 16.375, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 13.1187, |
| "mean_token_accuracy": 0.788687152415514, |
| "num_tokens": 35053712.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.8245653146877885, |
| "epoch": 1.712, |
| "grad_norm": 17.125, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 13.3208, |
| "mean_token_accuracy": 0.784929183498025, |
| "num_tokens": 35393088.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.8105424832552671, |
| "epoch": 1.728, |
| "grad_norm": 16.0, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 12.9858, |
| "mean_token_accuracy": 0.7895865086466074, |
| "num_tokens": 35729007.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8141957949846983, |
| "epoch": 1.744, |
| "grad_norm": 16.75, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 13.0812, |
| "mean_token_accuracy": 0.7894516389816999, |
| "num_tokens": 36072489.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.8048153560608625, |
| "epoch": 1.76, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 12.9209, |
| "mean_token_accuracy": 0.7897520393133164, |
| "num_tokens": 36398285.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 2.445696955680847, |
| "eval_biology_loss": 2.8779242038726807, |
| "eval_biology_mean_token_accuracy": 0.5016539707183838, |
| "eval_biology_num_tokens": 36398285.0, |
| "eval_biology_runtime": 49.6599, |
| "eval_biology_samples_per_second": 10.068, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 1.1345578532218934, |
| "eval_chemistry_loss": 1.204408049583435, |
| "eval_chemistry_mean_token_accuracy": 0.71992467212677, |
| "eval_chemistry_num_tokens": 36398285.0, |
| "eval_chemistry_runtime": 61.246, |
| "eval_chemistry_samples_per_second": 8.164, |
| "eval_chemistry_steps_per_second": 2.041, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.8842824428081513, |
| "eval_math_loss": 1.1029928922653198, |
| "eval_math_mean_token_accuracy": 0.7455306906700134, |
| "eval_math_num_tokens": 36398285.0, |
| "eval_math_runtime": 62.8108, |
| "eval_math_samples_per_second": 7.96, |
| "eval_math_steps_per_second": 1.99, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.8283794307708741, |
| "eval_physics_loss": 0.8505285978317261, |
| "eval_physics_mean_token_accuracy": 0.7818370275497436, |
| "eval_physics_num_tokens": 36398285.0, |
| "eval_physics_runtime": 71.5963, |
| "eval_physics_samples_per_second": 6.984, |
| "eval_physics_steps_per_second": 1.746, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.8389779690653085, |
| "epoch": 1.776, |
| "grad_norm": 15.6875, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 13.4944, |
| "mean_token_accuracy": 0.7826780155301094, |
| "num_tokens": 36730479.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.8008179372176528, |
| "epoch": 1.792, |
| "grad_norm": 19.125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 12.9483, |
| "mean_token_accuracy": 0.7898946575820446, |
| "num_tokens": 37057425.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.7982527418062091, |
| "epoch": 1.808, |
| "grad_norm": 15.625, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 12.7713, |
| "mean_token_accuracy": 0.7935733944177628, |
| "num_tokens": 37395449.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.8088066600263119, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 12.9789, |
| "mean_token_accuracy": 0.7902951821684837, |
| "num_tokens": 37729535.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.8060135461390019, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 16.375, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 12.9445, |
| "mean_token_accuracy": 0.7909105230122805, |
| "num_tokens": 38057839.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.8027735522016883, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 13.0038, |
| "mean_token_accuracy": 0.7903423044830561, |
| "num_tokens": 38398033.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.8051119217649102, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 15.625, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 12.8759, |
| "mean_token_accuracy": 0.7893568992614746, |
| "num_tokens": 38729203.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8213391533121467, |
| "epoch": 1.888, |
| "grad_norm": 16.375, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 13.2453, |
| "mean_token_accuracy": 0.7860501617193222, |
| "num_tokens": 39058791.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.826800760999322, |
| "epoch": 1.904, |
| "grad_norm": 16.625, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 13.4803, |
| "mean_token_accuracy": 0.7845181468874216, |
| "num_tokens": 39386170.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.8467358741909266, |
| "epoch": 1.92, |
| "grad_norm": 16.375, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 13.4331, |
| "mean_token_accuracy": 0.784552700445056, |
| "num_tokens": 39710381.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 2.472013756752014, |
| "eval_biology_loss": 2.8643574714660645, |
| "eval_biology_mean_token_accuracy": 0.5028325555324554, |
| "eval_biology_num_tokens": 39710381.0, |
| "eval_biology_runtime": 49.4032, |
| "eval_biology_samples_per_second": 10.121, |
| "eval_biology_steps_per_second": 2.53, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 1.1285487518310546, |
| "eval_chemistry_loss": 1.199933648109436, |
| "eval_chemistry_mean_token_accuracy": 0.7208248000144959, |
| "eval_chemistry_num_tokens": 39710381.0, |
| "eval_chemistry_runtime": 61.2533, |
| "eval_chemistry_samples_per_second": 8.163, |
| "eval_chemistry_steps_per_second": 2.041, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.8806797494888305, |
| "eval_math_loss": 1.1004599332809448, |
| "eval_math_mean_token_accuracy": 0.7461761064529419, |
| "eval_math_num_tokens": 39710381.0, |
| "eval_math_runtime": 62.8098, |
| "eval_math_samples_per_second": 7.961, |
| "eval_math_steps_per_second": 1.99, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.8215924696922302, |
| "eval_physics_loss": 0.8454442024230957, |
| "eval_physics_mean_token_accuracy": 0.7830140724182129, |
| "eval_physics_num_tokens": 39710381.0, |
| "eval_physics_runtime": 71.6941, |
| "eval_physics_samples_per_second": 6.974, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.8093992147594691, |
| "epoch": 1.936, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 13.0771, |
| "mean_token_accuracy": 0.7870635632425547, |
| "num_tokens": 40043354.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.7759674414992332, |
| "epoch": 1.952, |
| "grad_norm": 15.375, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 12.4136, |
| "mean_token_accuracy": 0.7972983971238137, |
| "num_tokens": 40375790.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.8013758489862084, |
| "epoch": 1.968, |
| "grad_norm": 15.75, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 12.9987, |
| "mean_token_accuracy": 0.790325503051281, |
| "num_tokens": 40705451.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.8227595569565892, |
| "epoch": 1.984, |
| "grad_norm": 17.5, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 13.1906, |
| "mean_token_accuracy": 0.7863460905849934, |
| "num_tokens": 41040313.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.8205555606633425, |
| "epoch": 2.0, |
| "grad_norm": 15.625, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 13.1737, |
| "mean_token_accuracy": 0.7883093636482954, |
| "num_tokens": 41373608.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.7831195399165154, |
| "epoch": 2.016, |
| "grad_norm": 14.0, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 12.4894, |
| "mean_token_accuracy": 0.7959605794399977, |
| "num_tokens": 41708226.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.7830235229805111, |
| "epoch": 2.032, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 12.6023, |
| "mean_token_accuracy": 0.7948959324508905, |
| "num_tokens": 42045433.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.807098483107984, |
| "epoch": 2.048, |
| "grad_norm": 17.0, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 13.0344, |
| "mean_token_accuracy": 0.7881330821663142, |
| "num_tokens": 42372694.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.7877823824062944, |
| "epoch": 2.064, |
| "grad_norm": 15.25, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 12.5998, |
| "mean_token_accuracy": 0.7949316211044788, |
| "num_tokens": 42710240.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.8019804678857326, |
| "epoch": 2.08, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 12.9342, |
| "mean_token_accuracy": 0.7910612858831882, |
| "num_tokens": 43037844.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 2.445282099723816, |
| "eval_biology_loss": 2.8822696208953857, |
| "eval_biology_mean_token_accuracy": 0.5015360698699951, |
| "eval_biology_num_tokens": 43037844.0, |
| "eval_biology_runtime": 49.6427, |
| "eval_biology_samples_per_second": 10.072, |
| "eval_biology_steps_per_second": 2.518, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 1.1140752034187318, |
| "eval_chemistry_loss": 1.2005956172943115, |
| "eval_chemistry_mean_token_accuracy": 0.7206200876235962, |
| "eval_chemistry_num_tokens": 43037844.0, |
| "eval_chemistry_runtime": 61.3631, |
| "eval_chemistry_samples_per_second": 8.148, |
| "eval_chemistry_steps_per_second": 2.037, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.8749056966304779, |
| "eval_math_loss": 1.103084683418274, |
| "eval_math_mean_token_accuracy": 0.7457676644325256, |
| "eval_math_num_tokens": 43037844.0, |
| "eval_math_runtime": 62.8812, |
| "eval_math_samples_per_second": 7.951, |
| "eval_math_steps_per_second": 1.988, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.8129040551185608, |
| "eval_physics_loss": 0.842690110206604, |
| "eval_physics_mean_token_accuracy": 0.7835633087158204, |
| "eval_physics_num_tokens": 43037844.0, |
| "eval_physics_runtime": 72.1755, |
| "eval_physics_samples_per_second": 6.928, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.8027737921103835, |
| "epoch": 2.096, |
| "grad_norm": 16.625, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 12.8345, |
| "mean_token_accuracy": 0.7918097671121359, |
| "num_tokens": 43368675.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.7821535093709826, |
| "epoch": 2.112, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 12.5805, |
| "mean_token_accuracy": 0.7947387598454952, |
| "num_tokens": 43699544.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.7935595938935875, |
| "epoch": 2.128, |
| "grad_norm": 16.875, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 12.7363, |
| "mean_token_accuracy": 0.7916791636496783, |
| "num_tokens": 44012632.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.7835203887894749, |
| "epoch": 2.144, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 12.6582, |
| "mean_token_accuracy": 0.793102978542447, |
| "num_tokens": 44349732.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.8203705210238695, |
| "epoch": 2.16, |
| "grad_norm": 16.0, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 13.2168, |
| "mean_token_accuracy": 0.7861043442040682, |
| "num_tokens": 44680410.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.7885768229141832, |
| "epoch": 2.176, |
| "grad_norm": 14.9375, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 12.6388, |
| "mean_token_accuracy": 0.7946638122200966, |
| "num_tokens": 45017922.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.7834130380302667, |
| "epoch": 2.192, |
| "grad_norm": 15.75, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 12.5136, |
| "mean_token_accuracy": 0.7952086210250855, |
| "num_tokens": 45348179.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.7792341850697995, |
| "epoch": 2.208, |
| "grad_norm": 17.625, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 12.5501, |
| "mean_token_accuracy": 0.795701441168785, |
| "num_tokens": 45678397.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.773603293299675, |
| "epoch": 2.224, |
| "grad_norm": 14.5, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 12.47, |
| "mean_token_accuracy": 0.7955104906111956, |
| "num_tokens": 46010052.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.7809545382857322, |
| "epoch": 2.24, |
| "grad_norm": 15.125, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 12.5627, |
| "mean_token_accuracy": 0.794714093953371, |
| "num_tokens": 46347116.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 2.4254958543777465, |
| "eval_biology_loss": 2.8775830268859863, |
| "eval_biology_mean_token_accuracy": 0.5026782186031341, |
| "eval_biology_num_tokens": 46347116.0, |
| "eval_biology_runtime": 49.9543, |
| "eval_biology_samples_per_second": 10.009, |
| "eval_biology_steps_per_second": 2.502, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 1.1134962687492371, |
| "eval_chemistry_loss": 1.1991115808486938, |
| "eval_chemistry_mean_token_accuracy": 0.7207795276641845, |
| "eval_chemistry_num_tokens": 46347116.0, |
| "eval_chemistry_runtime": 61.3888, |
| "eval_chemistry_samples_per_second": 8.145, |
| "eval_chemistry_steps_per_second": 2.036, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.8740004925727844, |
| "eval_math_loss": 1.1019760370254517, |
| "eval_math_mean_token_accuracy": 0.7459057631492615, |
| "eval_math_num_tokens": 46347116.0, |
| "eval_math_runtime": 62.8571, |
| "eval_math_samples_per_second": 7.955, |
| "eval_math_steps_per_second": 1.989, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.8126016085147858, |
| "eval_physics_loss": 0.8402607440948486, |
| "eval_physics_mean_token_accuracy": 0.784248288154602, |
| "eval_physics_num_tokens": 46347116.0, |
| "eval_physics_runtime": 72.184, |
| "eval_physics_samples_per_second": 6.927, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.8127070046961308, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 16.625, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 13.0614, |
| "mean_token_accuracy": 0.7891584753990173, |
| "num_tokens": 46673074.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.7707240108400584, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 16.125, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 12.3774, |
| "mean_token_accuracy": 0.797768659889698, |
| "num_tokens": 47009386.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.8034002147614956, |
| "epoch": 2.288, |
| "grad_norm": 16.25, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 12.9193, |
| "mean_token_accuracy": 0.7906435623764991, |
| "num_tokens": 47344231.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.7937306514009833, |
| "epoch": 2.304, |
| "grad_norm": 17.25, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 12.7869, |
| "mean_token_accuracy": 0.7924283880740404, |
| "num_tokens": 47658392.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.7891798976808786, |
| "epoch": 2.32, |
| "grad_norm": 14.75, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 12.6378, |
| "mean_token_accuracy": 0.7924429185688495, |
| "num_tokens": 47995015.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.7635893626138568, |
| "epoch": 2.336, |
| "grad_norm": 15.375, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 12.3402, |
| "mean_token_accuracy": 0.7973897516727447, |
| "num_tokens": 48336708.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.7687528569251298, |
| "epoch": 2.352, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 12.2507, |
| "mean_token_accuracy": 0.7983826816082, |
| "num_tokens": 48665311.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.7886384373530746, |
| "epoch": 2.368, |
| "grad_norm": 15.6875, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 12.712, |
| "mean_token_accuracy": 0.7930802937597037, |
| "num_tokens": 49005334.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.8016523649916053, |
| "epoch": 2.384, |
| "grad_norm": 17.125, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 12.8426, |
| "mean_token_accuracy": 0.7910515271127224, |
| "num_tokens": 49331936.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.7944724131375551, |
| "epoch": 2.4, |
| "grad_norm": 16.25, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 12.7553, |
| "mean_token_accuracy": 0.7921236105263233, |
| "num_tokens": 49671048.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 2.4047589435577392, |
| "eval_biology_loss": 2.870595693588257, |
| "eval_biology_mean_token_accuracy": 0.5024100644588471, |
| "eval_biology_num_tokens": 49671048.0, |
| "eval_biology_runtime": 49.425, |
| "eval_biology_samples_per_second": 10.116, |
| "eval_biology_steps_per_second": 2.529, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 1.1087028155326843, |
| "eval_chemistry_loss": 1.1979060173034668, |
| "eval_chemistry_mean_token_accuracy": 0.7210436034202575, |
| "eval_chemistry_num_tokens": 49671048.0, |
| "eval_chemistry_runtime": 61.0853, |
| "eval_chemistry_samples_per_second": 8.185, |
| "eval_chemistry_steps_per_second": 2.046, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.8674321026802063, |
| "eval_math_loss": 1.102066159248352, |
| "eval_math_mean_token_accuracy": 0.7458969240188599, |
| "eval_math_num_tokens": 49671048.0, |
| "eval_math_runtime": 62.8397, |
| "eval_math_samples_per_second": 7.957, |
| "eval_math_steps_per_second": 1.989, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.8101383104324341, |
| "eval_physics_loss": 0.8381265997886658, |
| "eval_physics_mean_token_accuracy": 0.7847440228462219, |
| "eval_physics_num_tokens": 49671048.0, |
| "eval_physics_runtime": 72.2232, |
| "eval_physics_samples_per_second": 6.923, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.7810511577874422, |
| "epoch": 2.416, |
| "grad_norm": 15.6875, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 12.5807, |
| "mean_token_accuracy": 0.79508685618639, |
| "num_tokens": 50018182.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.7916204823181033, |
| "epoch": 2.432, |
| "grad_norm": 16.75, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 12.7028, |
| "mean_token_accuracy": 0.7943674992769957, |
| "num_tokens": 50341187.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.8073871158063411, |
| "epoch": 2.448, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 13.0377, |
| "mean_token_accuracy": 0.7892248127609491, |
| "num_tokens": 50667019.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8032200831919909, |
| "epoch": 2.464, |
| "grad_norm": 16.875, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 12.8177, |
| "mean_token_accuracy": 0.7923011161386967, |
| "num_tokens": 50991113.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.7860733466222882, |
| "epoch": 2.48, |
| "grad_norm": 16.125, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 12.7361, |
| "mean_token_accuracy": 0.7927486374974251, |
| "num_tokens": 51319624.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.8012511027976871, |
| "epoch": 2.496, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 12.8233, |
| "mean_token_accuracy": 0.7916884988546371, |
| "num_tokens": 51654633.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.7731666518375278, |
| "epoch": 2.512, |
| "grad_norm": 17.375, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 12.4696, |
| "mean_token_accuracy": 0.7960924245417118, |
| "num_tokens": 51980051.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.7793908750638365, |
| "epoch": 2.528, |
| "grad_norm": 14.875, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 12.5186, |
| "mean_token_accuracy": 0.7953244607895613, |
| "num_tokens": 52316109.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.7914063410833478, |
| "epoch": 2.544, |
| "grad_norm": 17.25, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 12.8145, |
| "mean_token_accuracy": 0.7913208685815334, |
| "num_tokens": 52634528.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.7939762134104967, |
| "epoch": 2.56, |
| "grad_norm": 17.0, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 12.6043, |
| "mean_token_accuracy": 0.7941447257995605, |
| "num_tokens": 52964035.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 2.3931925020217895, |
| "eval_biology_loss": 2.875809669494629, |
| "eval_biology_mean_token_accuracy": 0.5025460588932037, |
| "eval_biology_num_tokens": 52964035.0, |
| "eval_biology_runtime": 50.2947, |
| "eval_biology_samples_per_second": 9.941, |
| "eval_biology_steps_per_second": 2.485, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 1.0883408832550048, |
| "eval_chemistry_loss": 1.1965774297714233, |
| "eval_chemistry_mean_token_accuracy": 0.7214803085327148, |
| "eval_chemistry_num_tokens": 52964035.0, |
| "eval_chemistry_runtime": 61.7977, |
| "eval_chemistry_samples_per_second": 8.091, |
| "eval_chemistry_steps_per_second": 2.023, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.8583383002281189, |
| "eval_math_loss": 1.1013567447662354, |
| "eval_math_mean_token_accuracy": 0.7465548224449158, |
| "eval_math_num_tokens": 52964035.0, |
| "eval_math_runtime": 63.3224, |
| "eval_math_samples_per_second": 7.896, |
| "eval_math_steps_per_second": 1.974, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.7951563930511475, |
| "eval_physics_loss": 0.8360637426376343, |
| "eval_physics_mean_token_accuracy": 0.7851015777587891, |
| "eval_physics_num_tokens": 52964035.0, |
| "eval_physics_runtime": 71.5227, |
| "eval_physics_samples_per_second": 6.991, |
| "eval_physics_steps_per_second": 1.748, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.772414730861783, |
| "epoch": 2.576, |
| "grad_norm": 17.25, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 12.4145, |
| "mean_token_accuracy": 0.7962443709373475, |
| "num_tokens": 53293345.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.7868875458836555, |
| "epoch": 2.592, |
| "grad_norm": 14.0625, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 12.6335, |
| "mean_token_accuracy": 0.7949258577078581, |
| "num_tokens": 53620896.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.7584911126643419, |
| "epoch": 2.608, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 12.2908, |
| "mean_token_accuracy": 0.798742862045765, |
| "num_tokens": 53950989.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.782053854689002, |
| "epoch": 2.624, |
| "grad_norm": 15.875, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 12.5859, |
| "mean_token_accuracy": 0.7945446480065584, |
| "num_tokens": 54282732.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.781876846589148, |
| "epoch": 2.64, |
| "grad_norm": 16.75, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 12.5738, |
| "mean_token_accuracy": 0.7951088670641184, |
| "num_tokens": 54609748.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.816822605766356, |
| "epoch": 2.656, |
| "grad_norm": 16.875, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 13.1814, |
| "mean_token_accuracy": 0.787245037406683, |
| "num_tokens": 54953583.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.7931226765736937, |
| "epoch": 2.672, |
| "grad_norm": 16.5, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 12.7541, |
| "mean_token_accuracy": 0.7924020521342754, |
| "num_tokens": 55281945.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.7660689847543836, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 16.125, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 12.247, |
| "mean_token_accuracy": 0.7995080798864365, |
| "num_tokens": 55612391.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.7789125647395849, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 15.1875, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 12.6456, |
| "mean_token_accuracy": 0.7945462424308062, |
| "num_tokens": 55949791.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.78773049749434, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 18.0, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 12.5144, |
| "mean_token_accuracy": 0.7949357461184263, |
| "num_tokens": 56282204.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 2.3543988103866575, |
| "eval_biology_loss": 2.8883936405181885, |
| "eval_biology_mean_token_accuracy": 0.5017733733654022, |
| "eval_biology_num_tokens": 56282204.0, |
| "eval_biology_runtime": 49.363, |
| "eval_biology_samples_per_second": 10.129, |
| "eval_biology_steps_per_second": 2.532, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 1.0755338530540466, |
| "eval_chemistry_loss": 1.1958928108215332, |
| "eval_chemistry_mean_token_accuracy": 0.7213905358314514, |
| "eval_chemistry_num_tokens": 56282204.0, |
| "eval_chemistry_runtime": 61.2095, |
| "eval_chemistry_samples_per_second": 8.169, |
| "eval_chemistry_steps_per_second": 2.042, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.8536946845054626, |
| "eval_math_loss": 1.1009058952331543, |
| "eval_math_mean_token_accuracy": 0.7464403495788574, |
| "eval_math_num_tokens": 56282204.0, |
| "eval_math_runtime": 62.7708, |
| "eval_math_samples_per_second": 7.965, |
| "eval_math_steps_per_second": 1.991, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.7884848446846008, |
| "eval_physics_loss": 0.8343402147293091, |
| "eval_physics_mean_token_accuracy": 0.7853550724983215, |
| "eval_physics_num_tokens": 56282204.0, |
| "eval_physics_runtime": 71.5298, |
| "eval_physics_samples_per_second": 6.99, |
| "eval_physics_steps_per_second": 1.748, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.8021048182621598, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 16.625, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 13.0577, |
| "mean_token_accuracy": 0.788362230360508, |
| "num_tokens": 56605478.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.8005984915420413, |
| "epoch": 2.752, |
| "grad_norm": 15.8125, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 12.8355, |
| "mean_token_accuracy": 0.7925169125199318, |
| "num_tokens": 56935874.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.7718434376642108, |
| "epoch": 2.768, |
| "grad_norm": 16.75, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 12.3835, |
| "mean_token_accuracy": 0.7968426302075386, |
| "num_tokens": 57270552.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.7870646607130766, |
| "epoch": 2.784, |
| "grad_norm": 16.875, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 12.6738, |
| "mean_token_accuracy": 0.7931245800107718, |
| "num_tokens": 57602968.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.8263678561896086, |
| "epoch": 2.8, |
| "grad_norm": 17.375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 13.3692, |
| "mean_token_accuracy": 0.7837320301681757, |
| "num_tokens": 57925886.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.7726477351039648, |
| "epoch": 2.816, |
| "grad_norm": 15.8125, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 12.2578, |
| "mean_token_accuracy": 0.7985608454793691, |
| "num_tokens": 58262960.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.7880838630720973, |
| "epoch": 2.832, |
| "grad_norm": 16.125, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 12.8104, |
| "mean_token_accuracy": 0.7908314753323793, |
| "num_tokens": 58586324.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.7903873866423965, |
| "epoch": 2.848, |
| "grad_norm": 15.125, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 12.6606, |
| "mean_token_accuracy": 0.7933553613722324, |
| "num_tokens": 58912949.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.7762313047423959, |
| "epoch": 2.864, |
| "grad_norm": 14.875, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 12.5161, |
| "mean_token_accuracy": 0.7964224554598331, |
| "num_tokens": 59252494.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.7852193580940365, |
| "epoch": 2.88, |
| "grad_norm": 18.375, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 12.6227, |
| "mean_token_accuracy": 0.7941090241074562, |
| "num_tokens": 59574951.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 2.404080862045288, |
| "eval_biology_loss": 2.863299608230591, |
| "eval_biology_mean_token_accuracy": 0.5036493399143219, |
| "eval_biology_num_tokens": 59574951.0, |
| "eval_biology_runtime": 49.4474, |
| "eval_biology_samples_per_second": 10.112, |
| "eval_biology_steps_per_second": 2.528, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 1.092514799118042, |
| "eval_chemistry_loss": 1.1931474208831787, |
| "eval_chemistry_mean_token_accuracy": 0.7216541547775268, |
| "eval_chemistry_num_tokens": 59574951.0, |
| "eval_chemistry_runtime": 61.319, |
| "eval_chemistry_samples_per_second": 8.154, |
| "eval_chemistry_steps_per_second": 2.039, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.8589861969947815, |
| "eval_math_loss": 1.09934401512146, |
| "eval_math_mean_token_accuracy": 0.7468600268363953, |
| "eval_math_num_tokens": 59574951.0, |
| "eval_math_runtime": 62.8842, |
| "eval_math_samples_per_second": 7.951, |
| "eval_math_steps_per_second": 1.988, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.7957947330474854, |
| "eval_physics_loss": 0.832846462726593, |
| "eval_physics_mean_token_accuracy": 0.7856108026504517, |
| "eval_physics_num_tokens": 59574951.0, |
| "eval_physics_runtime": 71.6687, |
| "eval_physics_samples_per_second": 6.977, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.7800151167437435, |
| "epoch": 2.896, |
| "grad_norm": 16.375, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 12.5659, |
| "mean_token_accuracy": 0.7957312397658824, |
| "num_tokens": 59903274.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.7794828902930021, |
| "epoch": 2.912, |
| "grad_norm": 16.75, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 12.4792, |
| "mean_token_accuracy": 0.7961793307214975, |
| "num_tokens": 60234203.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.7716775756329298, |
| "epoch": 2.928, |
| "grad_norm": 16.5, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 12.4507, |
| "mean_token_accuracy": 0.7971672754734754, |
| "num_tokens": 60559123.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.7751465419307351, |
| "epoch": 2.944, |
| "grad_norm": 17.875, |
| "learning_rate": 8.6e-06, |
| "loss": 12.4246, |
| "mean_token_accuracy": 0.7971710093319416, |
| "num_tokens": 60893423.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.7817971961572766, |
| "epoch": 2.96, |
| "grad_norm": 15.4375, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 12.6213, |
| "mean_token_accuracy": 0.7947664484381676, |
| "num_tokens": 61228692.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.7864448856562376, |
| "epoch": 2.976, |
| "grad_norm": 15.8125, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 12.6128, |
| "mean_token_accuracy": 0.7931821491569281, |
| "num_tokens": 61560852.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.7615644473582506, |
| "epoch": 2.992, |
| "grad_norm": 16.125, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 12.2528, |
| "mean_token_accuracy": 0.7986856568604708, |
| "num_tokens": 61898951.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.7641174843534827, |
| "epoch": 3.008, |
| "grad_norm": 15.0625, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 12.2561, |
| "mean_token_accuracy": 0.7995043728500605, |
| "num_tokens": 62223875.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.7743637939915061, |
| "epoch": 3.024, |
| "grad_norm": 15.875, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 12.3679, |
| "mean_token_accuracy": 0.7970022208988666, |
| "num_tokens": 62553166.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.7582010868936777, |
| "epoch": 3.04, |
| "grad_norm": 16.125, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 12.2155, |
| "mean_token_accuracy": 0.7990330256521702, |
| "num_tokens": 62893113.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 2.399233222961426, |
| "eval_biology_loss": 2.867366313934326, |
| "eval_biology_mean_token_accuracy": 0.5034214525222779, |
| "eval_biology_num_tokens": 62893113.0, |
| "eval_biology_runtime": 49.4064, |
| "eval_biology_samples_per_second": 10.12, |
| "eval_biology_steps_per_second": 2.53, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 1.089496678829193, |
| "eval_chemistry_loss": 1.1933289766311646, |
| "eval_chemistry_mean_token_accuracy": 0.7224055013656616, |
| "eval_chemistry_num_tokens": 62893113.0, |
| "eval_chemistry_runtime": 61.2552, |
| "eval_chemistry_samples_per_second": 8.163, |
| "eval_chemistry_steps_per_second": 2.041, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.8592811126708985, |
| "eval_math_loss": 1.1005762815475464, |
| "eval_math_mean_token_accuracy": 0.7469019837379456, |
| "eval_math_num_tokens": 62893113.0, |
| "eval_math_runtime": 62.8086, |
| "eval_math_samples_per_second": 7.961, |
| "eval_math_steps_per_second": 1.99, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.7926631271839142, |
| "eval_physics_loss": 0.8320626020431519, |
| "eval_physics_mean_token_accuracy": 0.7857998023033143, |
| "eval_physics_num_tokens": 62893113.0, |
| "eval_physics_runtime": 71.555, |
| "eval_physics_samples_per_second": 6.988, |
| "eval_physics_steps_per_second": 1.747, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.7836578948423266, |
| "epoch": 3.056, |
| "grad_norm": 16.75, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 12.5565, |
| "mean_token_accuracy": 0.7944340173155069, |
| "num_tokens": 63218088.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.7783920112997293, |
| "epoch": 3.072, |
| "grad_norm": 17.5, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 12.4945, |
| "mean_token_accuracy": 0.7946112662553787, |
| "num_tokens": 63551201.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7899106752127409, |
| "epoch": 3.088, |
| "grad_norm": 16.625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 12.7648, |
| "mean_token_accuracy": 0.7915248584002257, |
| "num_tokens": 63876362.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.7593935616314411, |
| "epoch": 3.104, |
| "grad_norm": 16.375, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 12.1082, |
| "mean_token_accuracy": 0.8004607565701007, |
| "num_tokens": 64208171.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.7588531570509076, |
| "epoch": 3.12, |
| "grad_norm": 16.375, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 12.1557, |
| "mean_token_accuracy": 0.8000247534364462, |
| "num_tokens": 64544493.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.7889568522572518, |
| "epoch": 3.136, |
| "grad_norm": 16.375, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 12.774, |
| "mean_token_accuracy": 0.7912482611835003, |
| "num_tokens": 64872488.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.7564845994114876, |
| "epoch": 3.152, |
| "grad_norm": 15.8125, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 12.089, |
| "mean_token_accuracy": 0.8013048619031906, |
| "num_tokens": 65201124.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.7751831419765949, |
| "epoch": 3.168, |
| "grad_norm": 15.125, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 12.5378, |
| "mean_token_accuracy": 0.7958132576197385, |
| "num_tokens": 65525868.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7631789870560169, |
| "epoch": 3.184, |
| "grad_norm": 16.75, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 12.2134, |
| "mean_token_accuracy": 0.8015536881983281, |
| "num_tokens": 65861603.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7747107265517116, |
| "epoch": 3.2, |
| "grad_norm": 15.75, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 12.494, |
| "mean_token_accuracy": 0.7962346155196428, |
| "num_tokens": 66193675.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 2.365761664390564, |
| "eval_biology_loss": 2.876070022583008, |
| "eval_biology_mean_token_accuracy": 0.5039477293491363, |
| "eval_biology_num_tokens": 66193675.0, |
| "eval_biology_runtime": 49.4253, |
| "eval_biology_samples_per_second": 10.116, |
| "eval_biology_steps_per_second": 2.529, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 1.0774572858810425, |
| "eval_chemistry_loss": 1.1938129663467407, |
| "eval_chemistry_mean_token_accuracy": 0.722421464920044, |
| "eval_chemistry_num_tokens": 66193675.0, |
| "eval_chemistry_runtime": 61.2795, |
| "eval_chemistry_samples_per_second": 8.159, |
| "eval_chemistry_steps_per_second": 2.04, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.8556913251876831, |
| "eval_math_loss": 1.1010321378707886, |
| "eval_math_mean_token_accuracy": 0.746727294921875, |
| "eval_math_num_tokens": 66193675.0, |
| "eval_math_runtime": 63.1, |
| "eval_math_samples_per_second": 7.924, |
| "eval_math_steps_per_second": 1.981, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.7872861375808716, |
| "eval_physics_loss": 0.831552267074585, |
| "eval_physics_mean_token_accuracy": 0.7860952916145325, |
| "eval_physics_num_tokens": 66193675.0, |
| "eval_physics_runtime": 71.959, |
| "eval_physics_samples_per_second": 6.948, |
| "eval_physics_steps_per_second": 1.737, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.614072648666411e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
