| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8, |
| "eval_steps": 100, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.0750614512711762, |
| "epoch": 0.016, |
| "grad_norm": 162.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 41.6591, |
| "mean_token_accuracy": 0.6122298210859298, |
| "num_tokens": 338352.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.0839035604149103, |
| "epoch": 0.032, |
| "grad_norm": 155.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 41.6397, |
| "mean_token_accuracy": 0.6166786661371588, |
| "num_tokens": 671193.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.0894881710410118, |
| "epoch": 0.048, |
| "grad_norm": 132.0, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 41.0461, |
| "mean_token_accuracy": 0.6147234473377466, |
| "num_tokens": 1001008.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.138040168955922, |
| "epoch": 0.064, |
| "grad_norm": 126.0, |
| "learning_rate": 2.6e-06, |
| "loss": 41.8193, |
| "mean_token_accuracy": 0.6091248352080584, |
| "num_tokens": 1331208.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.1673120155930519, |
| "epoch": 0.08, |
| "grad_norm": 114.5, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 39.7083, |
| "mean_token_accuracy": 0.6180085053667426, |
| "num_tokens": 1667532.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.19383430108428, |
| "epoch": 0.096, |
| "grad_norm": 91.0, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 37.505, |
| "mean_token_accuracy": 0.6295596646144986, |
| "num_tokens": 2007176.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.354680097848177, |
| "epoch": 0.112, |
| "grad_norm": 78.5, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 37.0959, |
| "mean_token_accuracy": 0.6256176400929689, |
| "num_tokens": 2342259.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.6070489153265952, |
| "epoch": 0.128, |
| "grad_norm": 55.0, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 36.6385, |
| "mean_token_accuracy": 0.6223620869219303, |
| "num_tokens": 2663987.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.682814510539174, |
| "epoch": 0.144, |
| "grad_norm": 37.0, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 32.9488, |
| "mean_token_accuracy": 0.6449975268915296, |
| "num_tokens": 2997824.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.7247826386243106, |
| "epoch": 0.16, |
| "grad_norm": 32.0, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 30.8315, |
| "mean_token_accuracy": 0.6568072270601988, |
| "num_tokens": 3330597.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.8070496501922606, |
| "eval_biology_loss": 4.702436447143555, |
| "eval_biology_mean_token_accuracy": 0.3792178168296814, |
| "eval_biology_num_tokens": 3330597.0, |
| "eval_biology_runtime": 57.2423, |
| "eval_biology_samples_per_second": 8.735, |
| "eval_biology_steps_per_second": 2.184, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 2.312845594406128, |
| "eval_chemistry_loss": 2.5579307079315186, |
| "eval_chemistry_mean_token_accuracy": 0.5897436618804932, |
| "eval_chemistry_num_tokens": 3330597.0, |
| "eval_chemistry_runtime": 71.9394, |
| "eval_chemistry_samples_per_second": 6.95, |
| "eval_chemistry_steps_per_second": 1.738, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.3987004208564757, |
| "eval_math_loss": 1.9521548748016357, |
| "eval_math_mean_token_accuracy": 0.6628584680557251, |
| "eval_math_num_tokens": 3330597.0, |
| "eval_math_runtime": 74.5287, |
| "eval_math_samples_per_second": 6.709, |
| "eval_math_steps_per_second": 1.677, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.725432638168335, |
| "eval_physics_loss": 1.8779996633529663, |
| "eval_physics_mean_token_accuracy": 0.6590415091514588, |
| "eval_physics_num_tokens": 3330597.0, |
| "eval_physics_runtime": 85.2266, |
| "eval_physics_samples_per_second": 5.867, |
| "eval_physics_steps_per_second": 1.467, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.7152024820446967, |
| "epoch": 0.176, |
| "grad_norm": 26.875, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 29.6512, |
| "mean_token_accuracy": 0.6657887665554881, |
| "num_tokens": 3658264.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.622603354603052, |
| "epoch": 0.192, |
| "grad_norm": 19.375, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 28.3031, |
| "mean_token_accuracy": 0.6764084205031395, |
| "num_tokens": 3995568.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.555612001568079, |
| "epoch": 0.208, |
| "grad_norm": 19.875, |
| "learning_rate": 8.6e-06, |
| "loss": 26.6387, |
| "mean_token_accuracy": 0.6902163974940777, |
| "num_tokens": 4321436.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.512053081765771, |
| "epoch": 0.224, |
| "grad_norm": 17.875, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 26.0858, |
| "mean_token_accuracy": 0.694270471855998, |
| "num_tokens": 4648491.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.4869094472378492, |
| "epoch": 0.24, |
| "grad_norm": 19.75, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 25.4417, |
| "mean_token_accuracy": 0.6989999961107969, |
| "num_tokens": 4986175.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.4632187210023404, |
| "epoch": 0.256, |
| "grad_norm": 19.375, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 25.0091, |
| "mean_token_accuracy": 0.7029873106628657, |
| "num_tokens": 5329320.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.4810095187276602, |
| "epoch": 0.272, |
| "grad_norm": 20.25, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 25.2607, |
| "mean_token_accuracy": 0.7004018183797598, |
| "num_tokens": 5658796.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.4340823762118817, |
| "epoch": 0.288, |
| "grad_norm": 18.25, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 24.6883, |
| "mean_token_accuracy": 0.7079170107841491, |
| "num_tokens": 5980201.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.4303428437560797, |
| "epoch": 0.304, |
| "grad_norm": 22.125, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 24.4539, |
| "mean_token_accuracy": 0.7080751962959766, |
| "num_tokens": 6298858.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.399438175559044, |
| "epoch": 0.32, |
| "grad_norm": 18.875, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 23.9145, |
| "mean_token_accuracy": 0.7128454186022282, |
| "num_tokens": 6622798.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.444120632171631, |
| "eval_biology_loss": 4.076927661895752, |
| "eval_biology_mean_token_accuracy": 0.42938668751716613, |
| "eval_biology_num_tokens": 6622798.0, |
| "eval_biology_runtime": 57.3705, |
| "eval_biology_samples_per_second": 8.715, |
| "eval_biology_steps_per_second": 2.179, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 1.8959890480041504, |
| "eval_chemistry_loss": 2.08449649810791, |
| "eval_chemistry_mean_token_accuracy": 0.6399502665996551, |
| "eval_chemistry_num_tokens": 6622798.0, |
| "eval_chemistry_runtime": 71.8716, |
| "eval_chemistry_samples_per_second": 6.957, |
| "eval_chemistry_steps_per_second": 1.739, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.2845887546539307, |
| "eval_math_loss": 1.6543570756912231, |
| "eval_math_mean_token_accuracy": 0.6954182968139648, |
| "eval_math_num_tokens": 6622798.0, |
| "eval_math_runtime": 74.7507, |
| "eval_math_samples_per_second": 6.689, |
| "eval_math_steps_per_second": 1.672, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.3866976137161255, |
| "eval_physics_loss": 1.4872251749038696, |
| "eval_physics_mean_token_accuracy": 0.7125752453804016, |
| "eval_physics_num_tokens": 6622798.0, |
| "eval_physics_runtime": 85.9206, |
| "eval_physics_samples_per_second": 5.819, |
| "eval_physics_steps_per_second": 1.455, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.3804268836975098, |
| "epoch": 0.336, |
| "grad_norm": 18.5, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 23.6643, |
| "mean_token_accuracy": 0.714659295603633, |
| "num_tokens": 6953611.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.3354483414441347, |
| "epoch": 0.352, |
| "grad_norm": 18.125, |
| "learning_rate": 1.46e-05, |
| "loss": 22.9062, |
| "mean_token_accuracy": 0.722674485296011, |
| "num_tokens": 7290378.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.3261277750134468, |
| "epoch": 0.368, |
| "grad_norm": 18.5, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 22.7576, |
| "mean_token_accuracy": 0.7249834384769202, |
| "num_tokens": 7621082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.3795554626733064, |
| "epoch": 0.384, |
| "grad_norm": 17.875, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 23.5828, |
| "mean_token_accuracy": 0.7127871386706829, |
| "num_tokens": 7955570.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.3256618689745665, |
| "epoch": 0.4, |
| "grad_norm": 20.0, |
| "learning_rate": 1.66e-05, |
| "loss": 22.7361, |
| "mean_token_accuracy": 0.7222114279866219, |
| "num_tokens": 8291049.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.298596952483058, |
| "epoch": 0.416, |
| "grad_norm": 22.0, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 22.2416, |
| "mean_token_accuracy": 0.7284637104719878, |
| "num_tokens": 8627310.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.311914747953415, |
| "epoch": 0.432, |
| "grad_norm": 19.375, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 22.4984, |
| "mean_token_accuracy": 0.7248033240437508, |
| "num_tokens": 8958371.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.2482076751068234, |
| "epoch": 0.448, |
| "grad_norm": 18.0, |
| "learning_rate": 1.86e-05, |
| "loss": 21.4775, |
| "mean_token_accuracy": 0.7356370355933904, |
| "num_tokens": 9294388.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.2598353076726199, |
| "epoch": 0.464, |
| "grad_norm": 20.5, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 21.5863, |
| "mean_token_accuracy": 0.7334068417549133, |
| "num_tokens": 9620366.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.2661810796707869, |
| "epoch": 0.48, |
| "grad_norm": 19.5, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 21.8336, |
| "mean_token_accuracy": 0.7298611674457789, |
| "num_tokens": 9955431.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.3976770553588866, |
| "eval_biology_loss": 3.904201030731201, |
| "eval_biology_mean_token_accuracy": 0.43922377967834475, |
| "eval_biology_num_tokens": 9955431.0, |
| "eval_biology_runtime": 57.4472, |
| "eval_biology_samples_per_second": 8.704, |
| "eval_biology_steps_per_second": 2.176, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 1.8206102285385133, |
| "eval_chemistry_loss": 1.956045389175415, |
| "eval_chemistry_mean_token_accuracy": 0.6531638257503509, |
| "eval_chemistry_num_tokens": 9955431.0, |
| "eval_chemistry_runtime": 71.9347, |
| "eval_chemistry_samples_per_second": 6.951, |
| "eval_chemistry_steps_per_second": 1.738, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.2509758601188659, |
| "eval_math_loss": 1.5806307792663574, |
| "eval_math_mean_token_accuracy": 0.7045728435516357, |
| "eval_math_num_tokens": 9955431.0, |
| "eval_math_runtime": 75.0056, |
| "eval_math_samples_per_second": 6.666, |
| "eval_math_steps_per_second": 1.667, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.304166310787201, |
| "eval_physics_loss": 1.3740965127944946, |
| "eval_physics_mean_token_accuracy": 0.7282451548576355, |
| "eval_physics_num_tokens": 9955431.0, |
| "eval_physics_runtime": 85.2346, |
| "eval_physics_samples_per_second": 5.866, |
| "eval_physics_steps_per_second": 1.467, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.2894000235944987, |
| "epoch": 0.496, |
| "grad_norm": 18.875, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 22.125, |
| "mean_token_accuracy": 0.7272629871964454, |
| "num_tokens": 10289883.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.258072443306446, |
| "epoch": 0.512, |
| "grad_norm": 20.5, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 21.7519, |
| "mean_token_accuracy": 0.7323984511196613, |
| "num_tokens": 10619468.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.2734836369752884, |
| "epoch": 0.528, |
| "grad_norm": 16.375, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 21.9114, |
| "mean_token_accuracy": 0.7292184431105853, |
| "num_tokens": 10944904.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.2483941707760096, |
| "epoch": 0.544, |
| "grad_norm": 19.25, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 21.5727, |
| "mean_token_accuracy": 0.7335125673562288, |
| "num_tokens": 11266409.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.2357387099415065, |
| "epoch": 0.56, |
| "grad_norm": 16.625, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 21.1173, |
| "mean_token_accuracy": 0.7381447162479162, |
| "num_tokens": 11605544.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.2104782849550246, |
| "epoch": 0.576, |
| "grad_norm": 19.25, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 20.9845, |
| "mean_token_accuracy": 0.740512777864933, |
| "num_tokens": 11931715.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.2183124080300332, |
| "epoch": 0.592, |
| "grad_norm": 19.375, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 21.0928, |
| "mean_token_accuracy": 0.7366720300167799, |
| "num_tokens": 12257243.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.2145392388105392, |
| "epoch": 0.608, |
| "grad_norm": 16.625, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 20.8598, |
| "mean_token_accuracy": 0.740410690754652, |
| "num_tokens": 12583154.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.2197224445641042, |
| "epoch": 0.624, |
| "grad_norm": 17.5, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 21.1959, |
| "mean_token_accuracy": 0.7364733774214983, |
| "num_tokens": 12905392.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.2305814437568188, |
| "epoch": 0.64, |
| "grad_norm": 18.625, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 21.2803, |
| "mean_token_accuracy": 0.7369419563561678, |
| "num_tokens": 13232198.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.2428079328536987, |
| "eval_biology_loss": 3.81221866607666, |
| "eval_biology_mean_token_accuracy": 0.44962965536117555, |
| "eval_biology_num_tokens": 13232198.0, |
| "eval_biology_runtime": 57.9921, |
| "eval_biology_samples_per_second": 8.622, |
| "eval_biology_steps_per_second": 2.155, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 1.71326642370224, |
| "eval_chemistry_loss": 1.8857945203781128, |
| "eval_chemistry_mean_token_accuracy": 0.6629956634044647, |
| "eval_chemistry_num_tokens": 13232198.0, |
| "eval_chemistry_runtime": 72.1018, |
| "eval_chemistry_samples_per_second": 6.935, |
| "eval_chemistry_steps_per_second": 1.734, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 1.1944113321304322, |
| "eval_math_loss": 1.5440242290496826, |
| "eval_math_mean_token_accuracy": 0.7086545734405517, |
| "eval_math_num_tokens": 13232198.0, |
| "eval_math_runtime": 75.1501, |
| "eval_math_samples_per_second": 6.653, |
| "eval_math_steps_per_second": 1.663, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.211931779384613, |
| "eval_physics_loss": 1.3088711500167847, |
| "eval_physics_mean_token_accuracy": 0.737203718662262, |
| "eval_physics_num_tokens": 13232198.0, |
| "eval_physics_runtime": 85.7441, |
| "eval_physics_samples_per_second": 5.831, |
| "eval_physics_steps_per_second": 1.458, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.2212170030921699, |
| "epoch": 0.656, |
| "grad_norm": 17.875, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 20.9332, |
| "mean_token_accuracy": 0.7387504555284977, |
| "num_tokens": 13575902.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.1882116444408894, |
| "epoch": 0.672, |
| "grad_norm": 20.25, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 20.6082, |
| "mean_token_accuracy": 0.743332651630044, |
| "num_tokens": 13895997.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.207483420521021, |
| "epoch": 0.688, |
| "grad_norm": 18.0, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 20.654, |
| "mean_token_accuracy": 0.7427127715200186, |
| "num_tokens": 14234888.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.2052442319691181, |
| "epoch": 0.704, |
| "grad_norm": 19.0, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 20.8893, |
| "mean_token_accuracy": 0.7387381002306939, |
| "num_tokens": 14567908.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.181990123167634, |
| "epoch": 0.72, |
| "grad_norm": 21.625, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 20.4926, |
| "mean_token_accuracy": 0.7439979754388333, |
| "num_tokens": 14882927.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.1967087093740703, |
| "epoch": 0.736, |
| "grad_norm": 20.375, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 20.5403, |
| "mean_token_accuracy": 0.7425415780395269, |
| "num_tokens": 15217342.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.1514527130872012, |
| "epoch": 0.752, |
| "grad_norm": 16.75, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 19.9152, |
| "mean_token_accuracy": 0.7488316483795643, |
| "num_tokens": 15547710.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.2158014349639417, |
| "epoch": 0.768, |
| "grad_norm": 21.375, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 21.052, |
| "mean_token_accuracy": 0.7373728647828102, |
| "num_tokens": 15877177.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.1517657596617936, |
| "epoch": 0.784, |
| "grad_norm": 16.125, |
| "learning_rate": 1.86e-05, |
| "loss": 19.9164, |
| "mean_token_accuracy": 0.7503607623279095, |
| "num_tokens": 16219640.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.180898468568921, |
| "epoch": 0.8, |
| "grad_norm": 21.75, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 20.3875, |
| "mean_token_accuracy": 0.7441588096320629, |
| "num_tokens": 16548261.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 3.2802926797866823, |
| "eval_biology_loss": 3.785775899887085, |
| "eval_biology_mean_token_accuracy": 0.4522908911705017, |
| "eval_biology_num_tokens": 16548261.0, |
| "eval_biology_runtime": 58.1484, |
| "eval_biology_samples_per_second": 8.599, |
| "eval_biology_steps_per_second": 2.15, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 1.712562572002411, |
| "eval_chemistry_loss": 1.8553112745285034, |
| "eval_chemistry_mean_token_accuracy": 0.6659346613883972, |
| "eval_chemistry_num_tokens": 16548261.0, |
| "eval_chemistry_runtime": 71.5095, |
| "eval_chemistry_samples_per_second": 6.992, |
| "eval_chemistry_steps_per_second": 1.748, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 1.1794157824516296, |
| "eval_math_loss": 1.5209077596664429, |
| "eval_math_mean_token_accuracy": 0.7121200709342956, |
| "eval_math_num_tokens": 16548261.0, |
| "eval_math_runtime": 74.5115, |
| "eval_math_samples_per_second": 6.71, |
| "eval_math_steps_per_second": 1.678, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.1884380350112915, |
| "eval_physics_loss": 1.2718405723571777, |
| "eval_physics_mean_token_accuracy": 0.742140392780304, |
| "eval_physics_num_tokens": 16548261.0, |
| "eval_physics_runtime": 85.7525, |
| "eval_physics_samples_per_second": 5.831, |
| "eval_physics_steps_per_second": 1.458, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.652179086470234e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
