| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.3351694207978635, |
| "eval_steps": 100, |
| "global_step": 3000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.548704273160547, |
| "epoch": 0.004451121126133645, |
| "grad_norm": 1192.0, |
| "learning_rate": 1.8e-07, |
| "loss": 47.4036, |
| "mean_token_accuracy": 0.641148860566318, |
| "num_tokens": 63133.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.534384406823665, |
| "epoch": 0.00890224225226729, |
| "grad_norm": 1328.0, |
| "learning_rate": 3.8e-07, |
| "loss": 50.4685, |
| "mean_token_accuracy": 0.6258877433836461, |
| "num_tokens": 128968.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.537126874178648, |
| "epoch": 0.013353363378400934, |
| "grad_norm": 1320.0, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 49.4492, |
| "mean_token_accuracy": 0.6289473636075854, |
| "num_tokens": 193314.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.5729353995062411, |
| "epoch": 0.01780448450453458, |
| "grad_norm": 1012.0, |
| "learning_rate": 7.8e-07, |
| "loss": 49.5744, |
| "mean_token_accuracy": 0.6247817112132907, |
| "num_tokens": 257189.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.5564130786806345, |
| "epoch": 0.022255605630668224, |
| "grad_norm": 1096.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 44.4523, |
| "mean_token_accuracy": 0.6425775093957782, |
| "num_tokens": 324224.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.590752430446446, |
| "epoch": 0.026706726756801868, |
| "grad_norm": 816.0, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 43.7528, |
| "mean_token_accuracy": 0.6364219607785344, |
| "num_tokens": 394654.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.6452584797516465, |
| "epoch": 0.031157847882935515, |
| "grad_norm": 600.0, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 39.6529, |
| "mean_token_accuracy": 0.6429168622940779, |
| "num_tokens": 461658.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.7016998372972012, |
| "epoch": 0.03560896900906916, |
| "grad_norm": 450.0, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 35.508, |
| "mean_token_accuracy": 0.6540533849969506, |
| "num_tokens": 528285.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8241214359179139, |
| "epoch": 0.04006009013520281, |
| "grad_norm": 320.0, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 34.6569, |
| "mean_token_accuracy": 0.6461574586108327, |
| "num_tokens": 595336.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.0065209370106458, |
| "epoch": 0.04451121126133645, |
| "grad_norm": 208.0, |
| "learning_rate": 1.98e-06, |
| "loss": 31.2419, |
| "mean_token_accuracy": 0.6523716723546386, |
| "num_tokens": 663126.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04451121126133645, |
| "eval_biology_entropy": 1.5740653836727143, |
| "eval_biology_loss": 2.817919969558716, |
| "eval_biology_mean_token_accuracy": 0.5265991420149804, |
| "eval_biology_num_tokens": 663126.0, |
| "eval_biology_runtime": 49.0691, |
| "eval_biology_samples_per_second": 10.19, |
| "eval_biology_steps_per_second": 10.19, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04451121126133645, |
| "eval_chemistry_entropy": 1.1085504465103149, |
| "eval_chemistry_loss": 1.8567637205123901, |
| "eval_chemistry_mean_token_accuracy": 0.6607221029996871, |
| "eval_chemistry_num_tokens": 663126.0, |
| "eval_chemistry_runtime": 56.0546, |
| "eval_chemistry_samples_per_second": 8.92, |
| "eval_chemistry_steps_per_second": 8.92, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04451121126133645, |
| "eval_math_entropy": 0.7685319324731826, |
| "eval_math_loss": 1.6490864753723145, |
| "eval_math_mean_token_accuracy": 0.701621960401535, |
| "eval_math_num_tokens": 663126.0, |
| "eval_math_runtime": 57.7457, |
| "eval_math_samples_per_second": 8.659, |
| "eval_math_steps_per_second": 8.659, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04451121126133645, |
| "eval_physics_entropy": 0.9800559472441673, |
| "eval_physics_loss": 1.658416986465454, |
| "eval_physics_mean_token_accuracy": 0.6876975857019424, |
| "eval_physics_num_tokens": 663126.0, |
| "eval_physics_runtime": 64.5823, |
| "eval_physics_samples_per_second": 7.742, |
| "eval_physics_steps_per_second": 7.742, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.2310039116069675, |
| "epoch": 0.048962332387470095, |
| "grad_norm": 129.0, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 28.6754, |
| "mean_token_accuracy": 0.6632729232311249, |
| "num_tokens": 731060.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.4431365672498941, |
| "epoch": 0.053413453513603736, |
| "grad_norm": 86.0, |
| "learning_rate": 2.38e-06, |
| "loss": 26.3421, |
| "mean_token_accuracy": 0.6755226453766227, |
| "num_tokens": 801635.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.4806174896657467, |
| "epoch": 0.05786457463973738, |
| "grad_norm": 72.0, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 25.6649, |
| "mean_token_accuracy": 0.6807831462472678, |
| "num_tokens": 867260.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.511351894028485, |
| "epoch": 0.06231569576587103, |
| "grad_norm": 95.5, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 25.1648, |
| "mean_token_accuracy": 0.6787236794829369, |
| "num_tokens": 931344.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.4235187267884613, |
| "epoch": 0.06676681689200467, |
| "grad_norm": 57.25, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 23.0409, |
| "mean_token_accuracy": 0.6943066673353314, |
| "num_tokens": 993586.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.36321877297014, |
| "epoch": 0.07121793801813832, |
| "grad_norm": 52.75, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 22.2695, |
| "mean_token_accuracy": 0.6991008354350925, |
| "num_tokens": 1060812.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.2639451341703534, |
| "epoch": 0.07566905914427197, |
| "grad_norm": 55.0, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 20.5103, |
| "mean_token_accuracy": 0.7160237554460764, |
| "num_tokens": 1131832.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.3122306695207953, |
| "epoch": 0.08012018027040561, |
| "grad_norm": 43.25, |
| "learning_rate": 3.58e-06, |
| "loss": 20.6467, |
| "mean_token_accuracy": 0.7132720120251179, |
| "num_tokens": 1200650.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.321925420127809, |
| "epoch": 0.08457130139653925, |
| "grad_norm": 52.25, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 20.6542, |
| "mean_token_accuracy": 0.7144606521353125, |
| "num_tokens": 1265107.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.304779276624322, |
| "epoch": 0.0890224225226729, |
| "grad_norm": 51.75, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 19.8821, |
| "mean_token_accuracy": 0.7192140439525246, |
| "num_tokens": 1327380.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_biology_entropy": 1.7584778207540512, |
| "eval_biology_loss": 2.0029563903808594, |
| "eval_biology_mean_token_accuracy": 0.5894142254590988, |
| "eval_biology_num_tokens": 1327380.0, |
| "eval_biology_runtime": 48.4623, |
| "eval_biology_samples_per_second": 10.317, |
| "eval_biology_steps_per_second": 10.317, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_chemistry_entropy": 1.2313024162650108, |
| "eval_chemistry_loss": 1.23660409450531, |
| "eval_chemistry_mean_token_accuracy": 0.7191956076025963, |
| "eval_chemistry_num_tokens": 1327380.0, |
| "eval_chemistry_runtime": 56.1215, |
| "eval_chemistry_samples_per_second": 8.909, |
| "eval_chemistry_steps_per_second": 8.909, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_math_entropy": 0.9544051561951638, |
| "eval_math_loss": 1.1321452856063843, |
| "eval_math_mean_token_accuracy": 0.745829626917839, |
| "eval_math_num_tokens": 1327380.0, |
| "eval_math_runtime": 57.6893, |
| "eval_math_samples_per_second": 8.667, |
| "eval_math_steps_per_second": 8.667, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_physics_entropy": 1.097377147436142, |
| "eval_physics_loss": 1.1237763166427612, |
| "eval_physics_mean_token_accuracy": 0.7436373601555825, |
| "eval_physics_num_tokens": 1327380.0, |
| "eval_physics_runtime": 64.6254, |
| "eval_physics_samples_per_second": 7.737, |
| "eval_physics_steps_per_second": 7.737, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.2635900884866715, |
| "epoch": 0.09347354364880654, |
| "grad_norm": 54.5, |
| "learning_rate": 4.18e-06, |
| "loss": 20.5549, |
| "mean_token_accuracy": 0.7137404002249241, |
| "num_tokens": 1393379.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.253603531420231, |
| "epoch": 0.09792466477494019, |
| "grad_norm": 45.25, |
| "learning_rate": 4.38e-06, |
| "loss": 19.3171, |
| "mean_token_accuracy": 0.7269632073119283, |
| "num_tokens": 1460130.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.205259626172483, |
| "epoch": 0.10237578590107384, |
| "grad_norm": 50.5, |
| "learning_rate": 4.58e-06, |
| "loss": 18.3853, |
| "mean_token_accuracy": 0.7334358854219317, |
| "num_tokens": 1529183.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.2351684581488371, |
| "epoch": 0.10682690702720747, |
| "grad_norm": 61.5, |
| "learning_rate": 4.78e-06, |
| "loss": 19.0263, |
| "mean_token_accuracy": 0.723250618390739, |
| "num_tokens": 1597405.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.2351113758981227, |
| "epoch": 0.11127802815334112, |
| "grad_norm": 43.0, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 18.9974, |
| "mean_token_accuracy": 0.7199854323640466, |
| "num_tokens": 1661767.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.1816746497526764, |
| "epoch": 0.11572914927947477, |
| "grad_norm": 45.75, |
| "learning_rate": 5.18e-06, |
| "loss": 18.6797, |
| "mean_token_accuracy": 0.7243335520848632, |
| "num_tokens": 1728207.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.1266852594912051, |
| "epoch": 0.12018027040560841, |
| "grad_norm": 45.75, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 17.8377, |
| "mean_token_accuracy": 0.732505951821804, |
| "num_tokens": 1796055.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.1155648654326797, |
| "epoch": 0.12463139153174206, |
| "grad_norm": 49.0, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 17.5397, |
| "mean_token_accuracy": 0.7362749936059118, |
| "num_tokens": 1859684.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.1514084933325648, |
| "epoch": 0.1290825126578757, |
| "grad_norm": 51.0, |
| "learning_rate": 5.78e-06, |
| "loss": 18.5451, |
| "mean_token_accuracy": 0.723321282491088, |
| "num_tokens": 1921704.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.1451087817549706, |
| "epoch": 0.13353363378400934, |
| "grad_norm": 42.5, |
| "learning_rate": 5.98e-06, |
| "loss": 18.2063, |
| "mean_token_accuracy": 0.7288043411448598, |
| "num_tokens": 1985766.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.13353363378400934, |
| "eval_biology_entropy": 1.5927132781147957, |
| "eval_biology_loss": 1.7975378036499023, |
| "eval_biology_mean_token_accuracy": 0.6119548050761223, |
| "eval_biology_num_tokens": 1985766.0, |
| "eval_biology_runtime": 48.6056, |
| "eval_biology_samples_per_second": 10.287, |
| "eval_biology_steps_per_second": 10.287, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.13353363378400934, |
| "eval_chemistry_entropy": 1.0956866154670715, |
| "eval_chemistry_loss": 1.0781587362289429, |
| "eval_chemistry_mean_token_accuracy": 0.7388799833655357, |
| "eval_chemistry_num_tokens": 1985766.0, |
| "eval_chemistry_runtime": 55.9983, |
| "eval_chemistry_samples_per_second": 8.929, |
| "eval_chemistry_steps_per_second": 8.929, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.13353363378400934, |
| "eval_math_entropy": 0.8625128073692322, |
| "eval_math_loss": 1.0247634649276733, |
| "eval_math_mean_token_accuracy": 0.7569631568193436, |
| "eval_math_num_tokens": 1985766.0, |
| "eval_math_runtime": 57.7379, |
| "eval_math_samples_per_second": 8.66, |
| "eval_math_steps_per_second": 8.66, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.13353363378400934, |
| "eval_physics_entropy": 0.9960197100043297, |
| "eval_physics_loss": 0.993407666683197, |
| "eval_physics_mean_token_accuracy": 0.7602200539112091, |
| "eval_physics_num_tokens": 1985766.0, |
| "eval_physics_runtime": 64.6649, |
| "eval_physics_samples_per_second": 7.732, |
| "eval_physics_steps_per_second": 7.732, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.1051979778334498, |
| "epoch": 0.137984754910143, |
| "grad_norm": 49.5, |
| "learning_rate": 6.18e-06, |
| "loss": 17.5537, |
| "mean_token_accuracy": 0.7314090682193637, |
| "num_tokens": 2049041.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.0414688095450402, |
| "epoch": 0.14243587603627664, |
| "grad_norm": 37.75, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 16.669, |
| "mean_token_accuracy": 0.7476102706044913, |
| "num_tokens": 2118068.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.030502800643444, |
| "epoch": 0.14688699716241027, |
| "grad_norm": 46.5, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 16.4295, |
| "mean_token_accuracy": 0.7489048192277551, |
| "num_tokens": 2186387.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.0635908702388406, |
| "epoch": 0.15133811828854393, |
| "grad_norm": 42.0, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 17.235, |
| "mean_token_accuracy": 0.7380516286939383, |
| "num_tokens": 2252650.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.0262556517496706, |
| "epoch": 0.15578923941467757, |
| "grad_norm": 41.0, |
| "learning_rate": 6.98e-06, |
| "loss": 16.1234, |
| "mean_token_accuracy": 0.7529099363833666, |
| "num_tokens": 2320358.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.9883157461881638, |
| "epoch": 0.16024036054081123, |
| "grad_norm": 44.25, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 16.301, |
| "mean_token_accuracy": 0.7517775129526854, |
| "num_tokens": 2388824.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.9906315876170992, |
| "epoch": 0.16469148166694486, |
| "grad_norm": 39.75, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 15.5904, |
| "mean_token_accuracy": 0.760966083779931, |
| "num_tokens": 2456144.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.0959791645407677, |
| "epoch": 0.1691426027930785, |
| "grad_norm": 37.5, |
| "learning_rate": 7.58e-06, |
| "loss": 17.7416, |
| "mean_token_accuracy": 0.7330215321853757, |
| "num_tokens": 2515325.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.0152791179716587, |
| "epoch": 0.17359372391921216, |
| "grad_norm": 38.25, |
| "learning_rate": 7.78e-06, |
| "loss": 16.1438, |
| "mean_token_accuracy": 0.7489995956420898, |
| "num_tokens": 2580490.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.015568952821195, |
| "epoch": 0.1780448450453458, |
| "grad_norm": 32.5, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 16.3986, |
| "mean_token_accuracy": 0.7477670447900892, |
| "num_tokens": 2644330.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_biology_entropy": 1.4987669139504434, |
| "eval_biology_loss": 1.7285088300704956, |
| "eval_biology_mean_token_accuracy": 0.6236078375577927, |
| "eval_biology_num_tokens": 2644330.0, |
| "eval_biology_runtime": 48.385, |
| "eval_biology_samples_per_second": 10.334, |
| "eval_biology_steps_per_second": 10.334, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_chemistry_entropy": 1.0116422069072724, |
| "eval_chemistry_loss": 1.0169917345046997, |
| "eval_chemistry_mean_token_accuracy": 0.7494727990031242, |
| "eval_chemistry_num_tokens": 2644330.0, |
| "eval_chemistry_runtime": 55.9816, |
| "eval_chemistry_samples_per_second": 8.932, |
| "eval_chemistry_steps_per_second": 8.932, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_math_entropy": 0.830569740831852, |
| "eval_math_loss": 0.9943051934242249, |
| "eval_math_mean_token_accuracy": 0.7620642938613892, |
| "eval_math_num_tokens": 2644330.0, |
| "eval_math_runtime": 60.3723, |
| "eval_math_samples_per_second": 8.282, |
| "eval_math_steps_per_second": 8.282, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_physics_entropy": 0.9355049659013748, |
| "eval_physics_loss": 0.94325852394104, |
| "eval_physics_mean_token_accuracy": 0.7684184198379517, |
| "eval_physics_num_tokens": 2644330.0, |
| "eval_physics_runtime": 65.391, |
| "eval_physics_samples_per_second": 7.646, |
| "eval_physics_steps_per_second": 7.646, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.9359495824202895, |
| "epoch": 0.18249596617147945, |
| "grad_norm": 37.5, |
| "learning_rate": 8.18e-06, |
| "loss": 15.074, |
| "mean_token_accuracy": 0.7657283775508403, |
| "num_tokens": 2713234.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.0310791170224547, |
| "epoch": 0.18694708729761308, |
| "grad_norm": 36.75, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 16.5812, |
| "mean_token_accuracy": 0.7489650597795844, |
| "num_tokens": 2783261.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.0061013143509627, |
| "epoch": 0.19139820842374672, |
| "grad_norm": 34.25, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 16.2064, |
| "mean_token_accuracy": 0.7534871533513069, |
| "num_tokens": 2850170.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.9983460938557982, |
| "epoch": 0.19584932954988038, |
| "grad_norm": 36.5, |
| "learning_rate": 8.78e-06, |
| "loss": 15.9251, |
| "mean_token_accuracy": 0.7519530924037099, |
| "num_tokens": 2913700.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.9914884481579065, |
| "epoch": 0.200300450676014, |
| "grad_norm": 43.25, |
| "learning_rate": 8.98e-06, |
| "loss": 16.0176, |
| "mean_token_accuracy": 0.7508705228567123, |
| "num_tokens": 2981212.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.9640501407906413, |
| "epoch": 0.20475157180214767, |
| "grad_norm": 43.75, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 15.5197, |
| "mean_token_accuracy": 0.7612664353102445, |
| "num_tokens": 3048973.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.9837653011083602, |
| "epoch": 0.2092026929282813, |
| "grad_norm": 32.75, |
| "learning_rate": 9.38e-06, |
| "loss": 16.0462, |
| "mean_token_accuracy": 0.7515783123672009, |
| "num_tokens": 3117033.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.9881603302434087, |
| "epoch": 0.21365381405441494, |
| "grad_norm": 35.5, |
| "learning_rate": 9.58e-06, |
| "loss": 15.6722, |
| "mean_token_accuracy": 0.7575565822422504, |
| "num_tokens": 3185255.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.960085684619844, |
| "epoch": 0.2181049351805486, |
| "grad_norm": 43.5, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 15.732, |
| "mean_token_accuracy": 0.7576891854405403, |
| "num_tokens": 3253121.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9170863745734096, |
| "epoch": 0.22255605630668224, |
| "grad_norm": 39.5, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 14.4657, |
| "mean_token_accuracy": 0.7695914417505264, |
| "num_tokens": 3322823.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.22255605630668224, |
| "eval_biology_entropy": 1.4002713676095009, |
| "eval_biology_loss": 1.6903640031814575, |
| "eval_biology_mean_token_accuracy": 0.6292068202495575, |
| "eval_biology_num_tokens": 3322823.0, |
| "eval_biology_runtime": 48.4234, |
| "eval_biology_samples_per_second": 10.326, |
| "eval_biology_steps_per_second": 10.326, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.22255605630668224, |
| "eval_chemistry_entropy": 0.9267486627101899, |
| "eval_chemistry_loss": 0.9770777821540833, |
| "eval_chemistry_mean_token_accuracy": 0.7565804206728935, |
| "eval_chemistry_num_tokens": 3322823.0, |
| "eval_chemistry_runtime": 56.3359, |
| "eval_chemistry_samples_per_second": 8.875, |
| "eval_chemistry_steps_per_second": 8.875, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.22255605630668224, |
| "eval_math_entropy": 0.7625217022895813, |
| "eval_math_loss": 0.9808707237243652, |
| "eval_math_mean_token_accuracy": 0.7647298012971878, |
| "eval_math_num_tokens": 3322823.0, |
| "eval_math_runtime": 57.8095, |
| "eval_math_samples_per_second": 8.649, |
| "eval_math_steps_per_second": 8.649, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.22255605630668224, |
| "eval_physics_entropy": 0.8588201667070389, |
| "eval_physics_loss": 0.910319983959198, |
| "eval_physics_mean_token_accuracy": 0.7743916043043136, |
| "eval_physics_num_tokens": 3322823.0, |
| "eval_physics_runtime": 64.6422, |
| "eval_physics_samples_per_second": 7.735, |
| "eval_physics_steps_per_second": 7.735, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.9297409053891897, |
| "epoch": 0.2270071774328159, |
| "grad_norm": 34.0, |
| "learning_rate": 1.018e-05, |
| "loss": 15.1678, |
| "mean_token_accuracy": 0.7620427209883929, |
| "num_tokens": 3393393.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.017524036206305, |
| "epoch": 0.23145829855894953, |
| "grad_norm": 41.75, |
| "learning_rate": 1.038e-05, |
| "loss": 16.5918, |
| "mean_token_accuracy": 0.7439933486282826, |
| "num_tokens": 3454750.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.9231159623712302, |
| "epoch": 0.2359094196850832, |
| "grad_norm": 43.5, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 14.7926, |
| "mean_token_accuracy": 0.7705453760921955, |
| "num_tokens": 3526914.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.0112015223130584, |
| "epoch": 0.24036054081121683, |
| "grad_norm": 42.25, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 16.3174, |
| "mean_token_accuracy": 0.7517922900617122, |
| "num_tokens": 3589449.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.9775515835732221, |
| "epoch": 0.24481166193735046, |
| "grad_norm": 37.5, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 15.5292, |
| "mean_token_accuracy": 0.7556810062378645, |
| "num_tokens": 3655092.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.9711876364424825, |
| "epoch": 0.24926278306348412, |
| "grad_norm": 42.75, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 15.8816, |
| "mean_token_accuracy": 0.7561034083366394, |
| "num_tokens": 3719113.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.9615386385470629, |
| "epoch": 0.25371390418961776, |
| "grad_norm": 33.75, |
| "learning_rate": 1.138e-05, |
| "loss": 15.6077, |
| "mean_token_accuracy": 0.7553648345172406, |
| "num_tokens": 3785282.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.9940539756789804, |
| "epoch": 0.2581650253157514, |
| "grad_norm": 31.5, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 15.7863, |
| "mean_token_accuracy": 0.758138832077384, |
| "num_tokens": 3852276.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.8664842249825597, |
| "epoch": 0.262616146441885, |
| "grad_norm": 32.75, |
| "learning_rate": 1.178e-05, |
| "loss": 14.4448, |
| "mean_token_accuracy": 0.7707123512402176, |
| "num_tokens": 3925649.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.9649652477353812, |
| "epoch": 0.2670672675680187, |
| "grad_norm": 39.25, |
| "learning_rate": 1.198e-05, |
| "loss": 15.1734, |
| "mean_token_accuracy": 0.7605622936040163, |
| "num_tokens": 3990505.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_biology_entropy": 1.3548224151134491, |
| "eval_biology_loss": 1.6628271341323853, |
| "eval_biology_mean_token_accuracy": 0.6329956727027893, |
| "eval_biology_num_tokens": 3990505.0, |
| "eval_biology_runtime": 48.5951, |
| "eval_biology_samples_per_second": 10.289, |
| "eval_biology_steps_per_second": 10.289, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_chemistry_entropy": 0.8806400309503079, |
| "eval_chemistry_loss": 0.9497475028038025, |
| "eval_chemistry_mean_token_accuracy": 0.7612205193042755, |
| "eval_chemistry_num_tokens": 3990505.0, |
| "eval_chemistry_runtime": 56.0415, |
| "eval_chemistry_samples_per_second": 8.922, |
| "eval_chemistry_steps_per_second": 8.922, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_math_entropy": 0.7478249241113663, |
| "eval_math_loss": 0.9696071743965149, |
| "eval_math_mean_token_accuracy": 0.7666747701764107, |
| "eval_math_num_tokens": 3990505.0, |
| "eval_math_runtime": 57.7058, |
| "eval_math_samples_per_second": 8.665, |
| "eval_math_steps_per_second": 8.665, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_physics_entropy": 0.8181254861354827, |
| "eval_physics_loss": 0.8910531997680664, |
| "eval_physics_mean_token_accuracy": 0.7779091320037842, |
| "eval_physics_num_tokens": 3990505.0, |
| "eval_physics_runtime": 64.595, |
| "eval_physics_samples_per_second": 7.741, |
| "eval_physics_steps_per_second": 7.741, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.9066566737368703, |
| "epoch": 0.27151838869415235, |
| "grad_norm": 35.5, |
| "learning_rate": 1.218e-05, |
| "loss": 14.8095, |
| "mean_token_accuracy": 0.7637650985270739, |
| "num_tokens": 4060657.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.8951275391504169, |
| "epoch": 0.275969509820286, |
| "grad_norm": 31.625, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 14.3937, |
| "mean_token_accuracy": 0.7713713925331831, |
| "num_tokens": 4133119.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.9309091791510582, |
| "epoch": 0.2804206309464196, |
| "grad_norm": 40.25, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 14.9362, |
| "mean_token_accuracy": 0.7651385851204395, |
| "num_tokens": 4200051.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.9745316403917968, |
| "epoch": 0.2848717520725533, |
| "grad_norm": 41.0, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 15.86, |
| "mean_token_accuracy": 0.7543578896671533, |
| "num_tokens": 4267403.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.9337521677836775, |
| "epoch": 0.28932287319868694, |
| "grad_norm": 39.0, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 14.9738, |
| "mean_token_accuracy": 0.7643170915544033, |
| "num_tokens": 4333268.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.9018357262946666, |
| "epoch": 0.29377399432482054, |
| "grad_norm": 53.5, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 14.5097, |
| "mean_token_accuracy": 0.7716417364776135, |
| "num_tokens": 4400242.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.9362477554008365, |
| "epoch": 0.2982251154509542, |
| "grad_norm": 34.5, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 15.1358, |
| "mean_token_accuracy": 0.7618589885532856, |
| "num_tokens": 4464714.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.8689518304541707, |
| "epoch": 0.30267623657708786, |
| "grad_norm": 41.25, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 14.3018, |
| "mean_token_accuracy": 0.7737269978970289, |
| "num_tokens": 4535458.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.9227319629862905, |
| "epoch": 0.3071273577032215, |
| "grad_norm": 38.25, |
| "learning_rate": 1.378e-05, |
| "loss": 14.6416, |
| "mean_token_accuracy": 0.7681682342663407, |
| "num_tokens": 4599099.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.865695190615952, |
| "epoch": 0.31157847882935513, |
| "grad_norm": 28.25, |
| "learning_rate": 1.398e-05, |
| "loss": 14.0258, |
| "mean_token_accuracy": 0.7762727830559015, |
| "num_tokens": 4662903.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.31157847882935513, |
| "eval_biology_entropy": 1.4728712896108627, |
| "eval_biology_loss": 1.6218550205230713, |
| "eval_biology_mean_token_accuracy": 0.6382006840705872, |
| "eval_biology_num_tokens": 4662903.0, |
| "eval_biology_runtime": 48.5605, |
| "eval_biology_samples_per_second": 10.296, |
| "eval_biology_steps_per_second": 10.296, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.31157847882935513, |
| "eval_chemistry_entropy": 0.9336612664461136, |
| "eval_chemistry_loss": 0.9250124096870422, |
| "eval_chemistry_mean_token_accuracy": 0.7656691336631775, |
| "eval_chemistry_num_tokens": 4662903.0, |
| "eval_chemistry_runtime": 56.0811, |
| "eval_chemistry_samples_per_second": 8.916, |
| "eval_chemistry_steps_per_second": 8.916, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.31157847882935513, |
| "eval_math_entropy": 0.7846916960477829, |
| "eval_math_loss": 0.9586912393569946, |
| "eval_math_mean_token_accuracy": 0.7681104638576508, |
| "eval_math_num_tokens": 4662903.0, |
| "eval_math_runtime": 57.7641, |
| "eval_math_samples_per_second": 8.656, |
| "eval_math_steps_per_second": 8.656, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.31157847882935513, |
| "eval_physics_entropy": 0.8758866502046585, |
| "eval_physics_loss": 0.8707027435302734, |
| "eval_physics_mean_token_accuracy": 0.7806350255012512, |
| "eval_physics_num_tokens": 4662903.0, |
| "eval_physics_runtime": 64.5913, |
| "eval_physics_samples_per_second": 7.741, |
| "eval_physics_steps_per_second": 7.741, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.9134258929640054, |
| "epoch": 0.3160295999554888, |
| "grad_norm": 42.0, |
| "learning_rate": 1.418e-05, |
| "loss": 15.0312, |
| "mean_token_accuracy": 0.7628084912896156, |
| "num_tokens": 4729166.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.97938490267843, |
| "epoch": 0.32048072108162245, |
| "grad_norm": 41.75, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 15.7738, |
| "mean_token_accuracy": 0.7518788330256939, |
| "num_tokens": 4796815.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.8446940638124942, |
| "epoch": 0.32493184220775606, |
| "grad_norm": 34.5, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 13.5574, |
| "mean_token_accuracy": 0.782823646441102, |
| "num_tokens": 4865172.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.9024957665242255, |
| "epoch": 0.3293829633338897, |
| "grad_norm": 33.75, |
| "learning_rate": 1.478e-05, |
| "loss": 14.8961, |
| "mean_token_accuracy": 0.763876224681735, |
| "num_tokens": 4932671.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.9568601671606303, |
| "epoch": 0.3338340844600234, |
| "grad_norm": 33.25, |
| "learning_rate": 1.498e-05, |
| "loss": 15.257, |
| "mean_token_accuracy": 0.7628579653799534, |
| "num_tokens": 4998710.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8733517214655876, |
| "epoch": 0.338285205586157, |
| "grad_norm": 30.75, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 14.1521, |
| "mean_token_accuracy": 0.7736187670379877, |
| "num_tokens": 5066948.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.9070882228203118, |
| "epoch": 0.34273632671229065, |
| "grad_norm": 35.25, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 14.7033, |
| "mean_token_accuracy": 0.7682277161628008, |
| "num_tokens": 5129950.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.9775535127148032, |
| "epoch": 0.3471874478384243, |
| "grad_norm": 34.5, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 15.453, |
| "mean_token_accuracy": 0.754616218432784, |
| "num_tokens": 5193519.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8829231994226575, |
| "epoch": 0.351638568964558, |
| "grad_norm": 41.75, |
| "learning_rate": 1.578e-05, |
| "loss": 14.6374, |
| "mean_token_accuracy": 0.7640287645161152, |
| "num_tokens": 5256503.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.9582383582368493, |
| "epoch": 0.3560896900906916, |
| "grad_norm": 40.25, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 15.6331, |
| "mean_token_accuracy": 0.7566823102533817, |
| "num_tokens": 5324751.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_biology_entropy": 1.554333319604397, |
| "eval_biology_loss": 1.5954139232635498, |
| "eval_biology_mean_token_accuracy": 0.6430081333518028, |
| "eval_biology_num_tokens": 5324751.0, |
| "eval_biology_runtime": 49.1621, |
| "eval_biology_samples_per_second": 10.17, |
| "eval_biology_steps_per_second": 10.17, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_chemistry_entropy": 0.9556231063604355, |
| "eval_chemistry_loss": 0.907728910446167, |
| "eval_chemistry_mean_token_accuracy": 0.7686411096453667, |
| "eval_chemistry_num_tokens": 5324751.0, |
| "eval_chemistry_runtime": 56.1564, |
| "eval_chemistry_samples_per_second": 8.904, |
| "eval_chemistry_steps_per_second": 8.904, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_math_entropy": 0.7810768990516662, |
| "eval_math_loss": 0.949228048324585, |
| "eval_math_mean_token_accuracy": 0.7701822088956833, |
| "eval_math_num_tokens": 5324751.0, |
| "eval_math_runtime": 57.7577, |
| "eval_math_samples_per_second": 8.657, |
| "eval_math_steps_per_second": 8.657, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_physics_entropy": 0.895748406291008, |
| "eval_physics_loss": 0.8605655431747437, |
| "eval_physics_mean_token_accuracy": 0.7827676599621772, |
| "eval_physics_num_tokens": 5324751.0, |
| "eval_physics_runtime": 64.6674, |
| "eval_physics_samples_per_second": 7.732, |
| "eval_physics_steps_per_second": 7.732, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8505169737152756, |
| "epoch": 0.36054081121682524, |
| "grad_norm": 41.0, |
| "learning_rate": 1.618e-05, |
| "loss": 13.7198, |
| "mean_token_accuracy": 0.7793542932718992, |
| "num_tokens": 5391664.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.8401029847562314, |
| "epoch": 0.3649919323429589, |
| "grad_norm": 33.25, |
| "learning_rate": 1.638e-05, |
| "loss": 13.3589, |
| "mean_token_accuracy": 0.7823891170322895, |
| "num_tokens": 5458519.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8939393721520901, |
| "epoch": 0.3694430534690925, |
| "grad_norm": 36.25, |
| "learning_rate": 1.658e-05, |
| "loss": 14.7591, |
| "mean_token_accuracy": 0.7666325703263283, |
| "num_tokens": 5523391.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8917801808565855, |
| "epoch": 0.37389417459522617, |
| "grad_norm": 30.375, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 14.1075, |
| "mean_token_accuracy": 0.7729504836723209, |
| "num_tokens": 5585508.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.8576251868158579, |
| "epoch": 0.37834529572135983, |
| "grad_norm": 38.25, |
| "learning_rate": 1.698e-05, |
| "loss": 14.3124, |
| "mean_token_accuracy": 0.774137607589364, |
| "num_tokens": 5650755.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8919966861605644, |
| "epoch": 0.38279641684749344, |
| "grad_norm": 33.75, |
| "learning_rate": 1.718e-05, |
| "loss": 14.1824, |
| "mean_token_accuracy": 0.7732574883848429, |
| "num_tokens": 5716321.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.9272915873676538, |
| "epoch": 0.3872475379736271, |
| "grad_norm": 42.25, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 14.8897, |
| "mean_token_accuracy": 0.7641183525323868, |
| "num_tokens": 5782698.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.8425593223422766, |
| "epoch": 0.39169865909976076, |
| "grad_norm": 30.0, |
| "learning_rate": 1.758e-05, |
| "loss": 13.8179, |
| "mean_token_accuracy": 0.7790724016726017, |
| "num_tokens": 5848889.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.8974415950477124, |
| "epoch": 0.3961497802258944, |
| "grad_norm": 34.25, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 14.5537, |
| "mean_token_accuracy": 0.7696893479675054, |
| "num_tokens": 5912919.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.868491224758327, |
| "epoch": 0.400600901352028, |
| "grad_norm": 29.5, |
| "learning_rate": 1.798e-05, |
| "loss": 14.1228, |
| "mean_token_accuracy": 0.7739411123096943, |
| "num_tokens": 5980229.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.400600901352028, |
| "eval_biology_entropy": 1.4719535266160966, |
| "eval_biology_loss": 1.5840708017349243, |
| "eval_biology_mean_token_accuracy": 0.6444517712593079, |
| "eval_biology_num_tokens": 5980229.0, |
| "eval_biology_runtime": 48.3845, |
| "eval_biology_samples_per_second": 10.334, |
| "eval_biology_steps_per_second": 10.334, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.400600901352028, |
| "eval_chemistry_entropy": 0.9006564086675644, |
| "eval_chemistry_loss": 0.8931591510772705, |
| "eval_chemistry_mean_token_accuracy": 0.771476098716259, |
| "eval_chemistry_num_tokens": 5980229.0, |
| "eval_chemistry_runtime": 55.9522, |
| "eval_chemistry_samples_per_second": 8.936, |
| "eval_chemistry_steps_per_second": 8.936, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.400600901352028, |
| "eval_math_entropy": 0.7833816863894463, |
| "eval_math_loss": 0.9406895637512207, |
| "eval_math_mean_token_accuracy": 0.7704955268502235, |
| "eval_math_num_tokens": 5980229.0, |
| "eval_math_runtime": 57.6393, |
| "eval_math_samples_per_second": 8.675, |
| "eval_math_steps_per_second": 8.675, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.400600901352028, |
| "eval_physics_entropy": 0.8495812131166458, |
| "eval_physics_loss": 0.8490710258483887, |
| "eval_physics_mean_token_accuracy": 0.7849877900481224, |
| "eval_physics_num_tokens": 5980229.0, |
| "eval_physics_runtime": 64.5379, |
| "eval_physics_samples_per_second": 7.747, |
| "eval_physics_steps_per_second": 7.747, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8243395276367664, |
| "epoch": 0.4050520224781617, |
| "grad_norm": 30.5, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 13.4007, |
| "mean_token_accuracy": 0.7826393615454436, |
| "num_tokens": 6048508.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.8578345064073801, |
| "epoch": 0.40950314360429535, |
| "grad_norm": 32.5, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 13.8874, |
| "mean_token_accuracy": 0.7774036221206189, |
| "num_tokens": 6114855.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.8919947864487767, |
| "epoch": 0.41395426473042896, |
| "grad_norm": 33.5, |
| "learning_rate": 1.858e-05, |
| "loss": 14.3682, |
| "mean_token_accuracy": 0.773763508349657, |
| "num_tokens": 6179050.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.877837923169136, |
| "epoch": 0.4184053858565626, |
| "grad_norm": 33.5, |
| "learning_rate": 1.878e-05, |
| "loss": 14.1955, |
| "mean_token_accuracy": 0.7732131693512201, |
| "num_tokens": 6244987.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.8329715754836797, |
| "epoch": 0.4228565069826963, |
| "grad_norm": 41.0, |
| "learning_rate": 1.898e-05, |
| "loss": 13.5939, |
| "mean_token_accuracy": 0.7785252593457699, |
| "num_tokens": 6310799.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8469648731872439, |
| "epoch": 0.4273076281088299, |
| "grad_norm": 36.75, |
| "learning_rate": 1.918e-05, |
| "loss": 13.6875, |
| "mean_token_accuracy": 0.7780090972781182, |
| "num_tokens": 6378152.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8630953669548035, |
| "epoch": 0.43175874923496355, |
| "grad_norm": 34.0, |
| "learning_rate": 1.938e-05, |
| "loss": 13.8631, |
| "mean_token_accuracy": 0.7757172901183367, |
| "num_tokens": 6442491.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.907545124553144, |
| "epoch": 0.4362098703610972, |
| "grad_norm": 28.75, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 14.7101, |
| "mean_token_accuracy": 0.766944982856512, |
| "num_tokens": 6510231.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.8721130631864071, |
| "epoch": 0.44066099148723087, |
| "grad_norm": 34.25, |
| "learning_rate": 1.978e-05, |
| "loss": 14.193, |
| "mean_token_accuracy": 0.7728719780221581, |
| "num_tokens": 6572858.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8767424447461962, |
| "epoch": 0.4451121126133645, |
| "grad_norm": 24.625, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 14.338, |
| "mean_token_accuracy": 0.7690305080264807, |
| "num_tokens": 6637273.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_biology_entropy": 1.4690953947901726, |
| "eval_biology_loss": 1.5712339878082275, |
| "eval_biology_mean_token_accuracy": 0.6474293968081475, |
| "eval_biology_num_tokens": 6637273.0, |
| "eval_biology_runtime": 48.4474, |
| "eval_biology_samples_per_second": 10.32, |
| "eval_biology_steps_per_second": 10.32, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_chemistry_entropy": 0.89704896864295, |
| "eval_chemistry_loss": 0.8829187154769897, |
| "eval_chemistry_mean_token_accuracy": 0.7738431564569473, |
| "eval_chemistry_num_tokens": 6637273.0, |
| "eval_chemistry_runtime": 56.1772, |
| "eval_chemistry_samples_per_second": 8.9, |
| "eval_chemistry_steps_per_second": 8.9, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_math_entropy": 0.7748903884291649, |
| "eval_math_loss": 0.9362199902534485, |
| "eval_math_mean_token_accuracy": 0.7724022324085236, |
| "eval_math_num_tokens": 6637273.0, |
| "eval_math_runtime": 57.7189, |
| "eval_math_samples_per_second": 8.663, |
| "eval_math_steps_per_second": 8.663, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_physics_entropy": 0.8675668377280236, |
| "eval_physics_loss": 0.8455091118812561, |
| "eval_physics_mean_token_accuracy": 0.7849516254663468, |
| "eval_physics_num_tokens": 6637273.0, |
| "eval_physics_runtime": 64.6865, |
| "eval_physics_samples_per_second": 7.73, |
| "eval_physics_steps_per_second": 7.73, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.8708374679088593, |
| "epoch": 0.44956323373949814, |
| "grad_norm": 33.25, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 13.9302, |
| "mean_token_accuracy": 0.774122205376625, |
| "num_tokens": 6700799.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8613243264146149, |
| "epoch": 0.4540143548656318, |
| "grad_norm": 32.0, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 14.124, |
| "mean_token_accuracy": 0.776879669725895, |
| "num_tokens": 6765798.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.8134710913524031, |
| "epoch": 0.4584654759917654, |
| "grad_norm": 31.625, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 13.3521, |
| "mean_token_accuracy": 0.7832869034260511, |
| "num_tokens": 6832774.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.890234874933958, |
| "epoch": 0.46291659711789906, |
| "grad_norm": 31.125, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 14.2521, |
| "mean_token_accuracy": 0.7710054367780685, |
| "num_tokens": 6896684.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.8423492605797946, |
| "epoch": 0.4673677182440327, |
| "grad_norm": 40.75, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 13.6826, |
| "mean_token_accuracy": 0.7855902209877967, |
| "num_tokens": 6964423.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8360794227570295, |
| "epoch": 0.4718188393701664, |
| "grad_norm": 27.0, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 13.722, |
| "mean_token_accuracy": 0.7778070000931621, |
| "num_tokens": 7033417.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.8722811192274094, |
| "epoch": 0.4762699604963, |
| "grad_norm": 35.0, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 14.2113, |
| "mean_token_accuracy": 0.7714831795543432, |
| "num_tokens": 7101274.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.8730784686282277, |
| "epoch": 0.48072108162243365, |
| "grad_norm": 28.5, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 14.0558, |
| "mean_token_accuracy": 0.7746166784316302, |
| "num_tokens": 7166608.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8250762648880482, |
| "epoch": 0.4851722027485673, |
| "grad_norm": 43.25, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 13.4239, |
| "mean_token_accuracy": 0.786403126642108, |
| "num_tokens": 7240958.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.8573660463094711, |
| "epoch": 0.4896233238747009, |
| "grad_norm": 32.5, |
| "learning_rate": 1.978e-05, |
| "loss": 13.9198, |
| "mean_token_accuracy": 0.7765211522579193, |
| "num_tokens": 7311968.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.4896233238747009, |
| "eval_biology_entropy": 1.4332191476225853, |
| "eval_biology_loss": 1.556909203529358, |
| "eval_biology_mean_token_accuracy": 0.6496381472945213, |
| "eval_biology_num_tokens": 7311968.0, |
| "eval_biology_runtime": 48.4694, |
| "eval_biology_samples_per_second": 10.316, |
| "eval_biology_steps_per_second": 10.316, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.4896233238747009, |
| "eval_chemistry_entropy": 0.8447221714258194, |
| "eval_chemistry_loss": 0.8686630129814148, |
| "eval_chemistry_mean_token_accuracy": 0.7762373806834221, |
| "eval_chemistry_num_tokens": 7311968.0, |
| "eval_chemistry_runtime": 56.0722, |
| "eval_chemistry_samples_per_second": 8.917, |
| "eval_chemistry_steps_per_second": 8.917, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.4896233238747009, |
| "eval_math_entropy": 0.7254823694527149, |
| "eval_math_loss": 0.937345564365387, |
| "eval_math_mean_token_accuracy": 0.7738885276317596, |
| "eval_math_num_tokens": 7311968.0, |
| "eval_math_runtime": 57.7068, |
| "eval_math_samples_per_second": 8.664, |
| "eval_math_steps_per_second": 8.664, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.4896233238747009, |
| "eval_physics_entropy": 0.8062724179625511, |
| "eval_physics_loss": 0.8333404660224915, |
| "eval_physics_mean_token_accuracy": 0.7875058565735817, |
| "eval_physics_num_tokens": 7311968.0, |
| "eval_physics_runtime": 65.4261, |
| "eval_physics_samples_per_second": 7.642, |
| "eval_physics_steps_per_second": 7.642, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.8604997573420405, |
| "epoch": 0.4940744450008346, |
| "grad_norm": 30.25, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 13.7608, |
| "mean_token_accuracy": 0.7766703650355339, |
| "num_tokens": 7378961.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.8193202435038984, |
| "epoch": 0.49852556612696824, |
| "grad_norm": 35.25, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 13.3273, |
| "mean_token_accuracy": 0.7826184341683984, |
| "num_tokens": 7444923.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.8790039600804448, |
| "epoch": 0.5029766872531019, |
| "grad_norm": 34.0, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 13.9384, |
| "mean_token_accuracy": 0.7745813885703683, |
| "num_tokens": 7509885.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.8574295949190855, |
| "epoch": 0.5074278083792355, |
| "grad_norm": 30.625, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 14.3156, |
| "mean_token_accuracy": 0.7717645790427923, |
| "num_tokens": 7576456.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.9021984472870826, |
| "epoch": 0.5118789295053692, |
| "grad_norm": 34.75, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 14.4928, |
| "mean_token_accuracy": 0.7711123242974282, |
| "num_tokens": 7641821.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.8312832812778652, |
| "epoch": 0.5163300506315028, |
| "grad_norm": 33.5, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 13.539, |
| "mean_token_accuracy": 0.7779583852738142, |
| "num_tokens": 7706502.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.9160518892109394, |
| "epoch": 0.5207811717576365, |
| "grad_norm": 35.0, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 14.7744, |
| "mean_token_accuracy": 0.7682293139398098, |
| "num_tokens": 7772353.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8255004599690438, |
| "epoch": 0.52523229288377, |
| "grad_norm": 26.625, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 13.6503, |
| "mean_token_accuracy": 0.7818126812577247, |
| "num_tokens": 7838516.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.8235707072541117, |
| "epoch": 0.5296834140099037, |
| "grad_norm": 42.0, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 13.328, |
| "mean_token_accuracy": 0.7810577813535928, |
| "num_tokens": 7902568.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.867654961720109, |
| "epoch": 0.5341345351360374, |
| "grad_norm": 29.5, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 13.7748, |
| "mean_token_accuracy": 0.7776078153401613, |
| "num_tokens": 7969704.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_biology_entropy": 1.37839106965065, |
| "eval_biology_loss": 1.5514307022094727, |
| "eval_biology_mean_token_accuracy": 0.6508332921862602, |
| "eval_biology_num_tokens": 7969704.0, |
| "eval_biology_runtime": 48.5631, |
| "eval_biology_samples_per_second": 10.296, |
| "eval_biology_steps_per_second": 10.296, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_chemistry_entropy": 0.8148281363546849, |
| "eval_chemistry_loss": 0.8604164719581604, |
| "eval_chemistry_mean_token_accuracy": 0.7772439575791359, |
| "eval_chemistry_num_tokens": 7969704.0, |
| "eval_chemistry_runtime": 56.0492, |
| "eval_chemistry_samples_per_second": 8.921, |
| "eval_chemistry_steps_per_second": 8.921, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_math_entropy": 0.7066318797171116, |
| "eval_math_loss": 0.9362333416938782, |
| "eval_math_mean_token_accuracy": 0.7748976874947547, |
| "eval_math_num_tokens": 7969704.0, |
| "eval_math_runtime": 57.7113, |
| "eval_math_samples_per_second": 8.664, |
| "eval_math_steps_per_second": 8.664, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_physics_entropy": 0.7838141260743141, |
| "eval_physics_loss": 0.8287554383277893, |
| "eval_physics_mean_token_accuracy": 0.7891261720657349, |
| "eval_physics_num_tokens": 7969704.0, |
| "eval_physics_runtime": 64.5774, |
| "eval_physics_samples_per_second": 7.743, |
| "eval_physics_steps_per_second": 7.743, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.823845097143203, |
| "epoch": 0.538585656262171, |
| "grad_norm": 31.375, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 13.3304, |
| "mean_token_accuracy": 0.7795415934175253, |
| "num_tokens": 8034765.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.8282254174351692, |
| "epoch": 0.5430367773883047, |
| "grad_norm": 26.25, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 13.8441, |
| "mean_token_accuracy": 0.7799333028495312, |
| "num_tokens": 8103869.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.8050381439737976, |
| "epoch": 0.5474878985144384, |
| "grad_norm": 28.25, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 12.7509, |
| "mean_token_accuracy": 0.7920488726347685, |
| "num_tokens": 8173386.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.8431567599065601, |
| "epoch": 0.551939019640572, |
| "grad_norm": 27.875, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 13.6451, |
| "mean_token_accuracy": 0.7757586792111397, |
| "num_tokens": 8242162.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.8362408608198166, |
| "epoch": 0.5563901407667056, |
| "grad_norm": 32.0, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 13.6867, |
| "mean_token_accuracy": 0.7783713828772306, |
| "num_tokens": 8308919.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.8537883767858148, |
| "epoch": 0.5608412618928392, |
| "grad_norm": 39.5, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 14.0357, |
| "mean_token_accuracy": 0.7729044172912836, |
| "num_tokens": 8365617.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.8582545002922416, |
| "epoch": 0.5652923830189729, |
| "grad_norm": 27.25, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 13.7589, |
| "mean_token_accuracy": 0.778778350353241, |
| "num_tokens": 8430728.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.8236037418246269, |
| "epoch": 0.5697435041451065, |
| "grad_norm": 25.125, |
| "learning_rate": 1.938e-05, |
| "loss": 13.2388, |
| "mean_token_accuracy": 0.784411870315671, |
| "num_tokens": 8497852.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.8496616246178746, |
| "epoch": 0.5741946252712402, |
| "grad_norm": 27.625, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 13.9613, |
| "mean_token_accuracy": 0.7750670064240694, |
| "num_tokens": 8558952.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.8457373520359397, |
| "epoch": 0.5786457463973739, |
| "grad_norm": 32.75, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 13.5433, |
| "mean_token_accuracy": 0.7806821152567863, |
| "num_tokens": 8622868.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.5786457463973739, |
| "eval_biology_entropy": 1.43521921479702, |
| "eval_biology_loss": 1.538223147392273, |
| "eval_biology_mean_token_accuracy": 0.6518344734311103, |
| "eval_biology_num_tokens": 8622868.0, |
| "eval_biology_runtime": 48.4265, |
| "eval_biology_samples_per_second": 10.325, |
| "eval_biology_steps_per_second": 10.325, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.5786457463973739, |
| "eval_chemistry_entropy": 0.8178334766924381, |
| "eval_chemistry_loss": 0.849298357963562, |
| "eval_chemistry_mean_token_accuracy": 0.7795388196110725, |
| "eval_chemistry_num_tokens": 8622868.0, |
| "eval_chemistry_runtime": 56.0158, |
| "eval_chemistry_samples_per_second": 8.926, |
| "eval_chemistry_steps_per_second": 8.926, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.5786457463973739, |
| "eval_math_entropy": 0.705706961274147, |
| "eval_math_loss": 0.9391148686408997, |
| "eval_math_mean_token_accuracy": 0.7748868613243103, |
| "eval_math_num_tokens": 8622868.0, |
| "eval_math_runtime": 57.6077, |
| "eval_math_samples_per_second": 8.679, |
| "eval_math_steps_per_second": 8.679, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.5786457463973739, |
| "eval_physics_entropy": 0.770459543287754, |
| "eval_physics_loss": 0.8210808038711548, |
| "eval_physics_mean_token_accuracy": 0.7902451878190041, |
| "eval_physics_num_tokens": 8622868.0, |
| "eval_physics_runtime": 64.5139, |
| "eval_physics_samples_per_second": 7.75, |
| "eval_physics_steps_per_second": 7.75, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.8071680258028209, |
| "epoch": 0.5830968675235075, |
| "grad_norm": 32.5, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 13.2153, |
| "mean_token_accuracy": 0.7843220077455044, |
| "num_tokens": 8694510.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.8942232267931104, |
| "epoch": 0.5875479886496411, |
| "grad_norm": 33.5, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 14.6327, |
| "mean_token_accuracy": 0.7653406778350472, |
| "num_tokens": 8757753.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.8597627034410834, |
| "epoch": 0.5919991097757747, |
| "grad_norm": 30.75, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 13.8741, |
| "mean_token_accuracy": 0.7796951614320278, |
| "num_tokens": 8823153.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.8664615718647838, |
| "epoch": 0.5964502309019084, |
| "grad_norm": 27.75, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 14.0396, |
| "mean_token_accuracy": 0.7731820710003376, |
| "num_tokens": 8886401.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.8158455387689173, |
| "epoch": 0.6009013520280421, |
| "grad_norm": 25.0, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 13.1708, |
| "mean_token_accuracy": 0.7833993557840586, |
| "num_tokens": 8958530.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.8429827799089253, |
| "epoch": 0.6053524731541757, |
| "grad_norm": 29.5, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 13.6607, |
| "mean_token_accuracy": 0.7796129040420056, |
| "num_tokens": 9024677.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.8648322455585002, |
| "epoch": 0.6098035942803094, |
| "grad_norm": 30.75, |
| "learning_rate": 1.918e-05, |
| "loss": 14.0503, |
| "mean_token_accuracy": 0.7712652169167995, |
| "num_tokens": 9090280.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.8274468386545777, |
| "epoch": 0.614254715406443, |
| "grad_norm": 31.0, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 13.5605, |
| "mean_token_accuracy": 0.7780856698751449, |
| "num_tokens": 9153771.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.7939455186016857, |
| "epoch": 0.6187058365325766, |
| "grad_norm": 28.5, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 12.6986, |
| "mean_token_accuracy": 0.7918769404292106, |
| "num_tokens": 9227117.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.8405648414045572, |
| "epoch": 0.6231569576587103, |
| "grad_norm": 31.25, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 13.9305, |
| "mean_token_accuracy": 0.7758563198149204, |
| "num_tokens": 9291760.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_biology_entropy": 1.4560990427136422, |
| "eval_biology_loss": 1.5385361909866333, |
| "eval_biology_mean_token_accuracy": 0.6521896251440048, |
| "eval_biology_num_tokens": 9291760.0, |
| "eval_biology_runtime": 48.3888, |
| "eval_biology_samples_per_second": 10.333, |
| "eval_biology_steps_per_second": 10.333, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_chemistry_entropy": 0.8454484658241272, |
| "eval_chemistry_loss": 0.8420565724372864, |
| "eval_chemistry_mean_token_accuracy": 0.7808461990356446, |
| "eval_chemistry_num_tokens": 9291760.0, |
| "eval_chemistry_runtime": 55.9558, |
| "eval_chemistry_samples_per_second": 8.936, |
| "eval_chemistry_steps_per_second": 8.936, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_math_entropy": 0.7586602506041527, |
| "eval_math_loss": 0.9284656643867493, |
| "eval_math_mean_token_accuracy": 0.7747275605797768, |
| "eval_math_num_tokens": 9291760.0, |
| "eval_math_runtime": 57.6406, |
| "eval_math_samples_per_second": 8.674, |
| "eval_math_steps_per_second": 8.674, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_physics_entropy": 0.8177245287299156, |
| "eval_physics_loss": 0.8168751001358032, |
| "eval_physics_mean_token_accuracy": 0.7908310489654541, |
| "eval_physics_num_tokens": 9291760.0, |
| "eval_physics_runtime": 64.5434, |
| "eval_physics_samples_per_second": 7.747, |
| "eval_physics_steps_per_second": 7.747, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.8965679749846458, |
| "epoch": 0.6276080787848439, |
| "grad_norm": 28.875, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 14.1354, |
| "mean_token_accuracy": 0.7691367525607348, |
| "num_tokens": 9357561.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.8075421938672662, |
| "epoch": 0.6320591999109776, |
| "grad_norm": 36.25, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 13.3559, |
| "mean_token_accuracy": 0.7846433013677597, |
| "num_tokens": 9425022.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.8154678157530725, |
| "epoch": 0.6365103210371112, |
| "grad_norm": 30.75, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 13.1969, |
| "mean_token_accuracy": 0.7874299377202988, |
| "num_tokens": 9492028.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.795359942317009, |
| "epoch": 0.6409614421632449, |
| "grad_norm": 33.5, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 12.7361, |
| "mean_token_accuracy": 0.7892100885510445, |
| "num_tokens": 9561091.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.8082486374303699, |
| "epoch": 0.6454125632893785, |
| "grad_norm": 36.25, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 13.4226, |
| "mean_token_accuracy": 0.7825187552720309, |
| "num_tokens": 9628383.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.7709562292322516, |
| "epoch": 0.6498636844155121, |
| "grad_norm": 24.875, |
| "learning_rate": 1.898e-05, |
| "loss": 12.3092, |
| "mean_token_accuracy": 0.7987042464315891, |
| "num_tokens": 9695723.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.8008807510137558, |
| "epoch": 0.6543148055416458, |
| "grad_norm": 27.75, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 13.1207, |
| "mean_token_accuracy": 0.7885788295418024, |
| "num_tokens": 9763393.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.8602127678692341, |
| "epoch": 0.6587659266677794, |
| "grad_norm": 30.625, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 13.8264, |
| "mean_token_accuracy": 0.7781572956591845, |
| "num_tokens": 9827272.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.8142364447936415, |
| "epoch": 0.6632170477939131, |
| "grad_norm": 25.75, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 13.1217, |
| "mean_token_accuracy": 0.783425547555089, |
| "num_tokens": 9892418.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.7871226992458105, |
| "epoch": 0.6676681689200468, |
| "grad_norm": 30.25, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 12.9626, |
| "mean_token_accuracy": 0.7883934937417507, |
| "num_tokens": 9955369.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6676681689200468, |
| "eval_biology_entropy": 1.4557775322794915, |
| "eval_biology_loss": 1.5257188081741333, |
| "eval_biology_mean_token_accuracy": 0.6541550720334053, |
| "eval_biology_num_tokens": 9955369.0, |
| "eval_biology_runtime": 48.3598, |
| "eval_biology_samples_per_second": 10.339, |
| "eval_biology_steps_per_second": 10.339, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6676681689200468, |
| "eval_chemistry_entropy": 0.8462375261187554, |
| "eval_chemistry_loss": 0.8326632976531982, |
| "eval_chemistry_mean_token_accuracy": 0.7831478416323662, |
| "eval_chemistry_num_tokens": 9955369.0, |
| "eval_chemistry_runtime": 56.166, |
| "eval_chemistry_samples_per_second": 8.902, |
| "eval_chemistry_steps_per_second": 8.902, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6676681689200468, |
| "eval_math_entropy": 0.7239785996079445, |
| "eval_math_loss": 0.9294689893722534, |
| "eval_math_mean_token_accuracy": 0.7758806052207947, |
| "eval_math_num_tokens": 9955369.0, |
| "eval_math_runtime": 57.7436, |
| "eval_math_samples_per_second": 8.659, |
| "eval_math_steps_per_second": 8.659, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6676681689200468, |
| "eval_physics_entropy": 0.80675083309412, |
| "eval_physics_loss": 0.8114446997642517, |
| "eval_physics_mean_token_accuracy": 0.7917336781024933, |
| "eval_physics_num_tokens": 9955369.0, |
| "eval_physics_runtime": 64.6166, |
| "eval_physics_samples_per_second": 7.738, |
| "eval_physics_steps_per_second": 7.738, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.800770258717239, |
| "epoch": 0.6721192900461804, |
| "grad_norm": 30.75, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 13.0324, |
| "mean_token_accuracy": 0.7872694227844477, |
| "num_tokens": 10026298.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.783286109380424, |
| "epoch": 0.676570411172314, |
| "grad_norm": 25.75, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 12.4817, |
| "mean_token_accuracy": 0.7961617294698954, |
| "num_tokens": 10096065.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.7893949760124087, |
| "epoch": 0.6810215322984476, |
| "grad_norm": 30.0, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 13.1869, |
| "mean_token_accuracy": 0.785550700686872, |
| "num_tokens": 10160745.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.7694807301275433, |
| "epoch": 0.6854726534245813, |
| "grad_norm": 28.25, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 12.351, |
| "mean_token_accuracy": 0.7945820547640323, |
| "num_tokens": 10233806.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8161390135064721, |
| "epoch": 0.689923774550715, |
| "grad_norm": 31.75, |
| "learning_rate": 1.878e-05, |
| "loss": 13.1382, |
| "mean_token_accuracy": 0.7831036793068051, |
| "num_tokens": 10298845.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.7966067988425494, |
| "epoch": 0.6943748956768486, |
| "grad_norm": 26.0, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 12.9247, |
| "mean_token_accuracy": 0.7881739761680364, |
| "num_tokens": 10364601.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.8032784784212709, |
| "epoch": 0.6988260168029823, |
| "grad_norm": 31.0, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 13.22, |
| "mean_token_accuracy": 0.7806047774851322, |
| "num_tokens": 10434433.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.8202603160403669, |
| "epoch": 0.703277137929116, |
| "grad_norm": 31.875, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 13.3456, |
| "mean_token_accuracy": 0.7826265832409263, |
| "num_tokens": 10500891.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.8302266797982156, |
| "epoch": 0.7077282590552495, |
| "grad_norm": 30.125, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 13.3904, |
| "mean_token_accuracy": 0.7822956155985594, |
| "num_tokens": 10567784.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.8138553470373153, |
| "epoch": 0.7121793801813832, |
| "grad_norm": 30.25, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 13.1566, |
| "mean_token_accuracy": 0.7851062502712012, |
| "num_tokens": 10633325.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_biology_entropy": 1.3791690354943276, |
| "eval_biology_loss": 1.524488091468811, |
| "eval_biology_mean_token_accuracy": 0.6547593535780907, |
| "eval_biology_num_tokens": 10633325.0, |
| "eval_biology_runtime": 48.9722, |
| "eval_biology_samples_per_second": 10.21, |
| "eval_biology_steps_per_second": 10.21, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_chemistry_entropy": 0.8021684873104096, |
| "eval_chemistry_loss": 0.8266823291778564, |
| "eval_chemistry_mean_token_accuracy": 0.7833675542473792, |
| "eval_chemistry_num_tokens": 10633325.0, |
| "eval_chemistry_runtime": 56.3143, |
| "eval_chemistry_samples_per_second": 8.879, |
| "eval_chemistry_steps_per_second": 8.879, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_math_entropy": 0.7278057588338852, |
| "eval_math_loss": 0.9273984432220459, |
| "eval_math_mean_token_accuracy": 0.775624837398529, |
| "eval_math_num_tokens": 10633325.0, |
| "eval_math_runtime": 57.8846, |
| "eval_math_samples_per_second": 8.638, |
| "eval_math_steps_per_second": 8.638, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_physics_entropy": 0.7752277762889862, |
| "eval_physics_loss": 0.8078919649124146, |
| "eval_physics_mean_token_accuracy": 0.7920119677782058, |
| "eval_physics_num_tokens": 10633325.0, |
| "eval_physics_runtime": 64.6914, |
| "eval_physics_samples_per_second": 7.729, |
| "eval_physics_steps_per_second": 7.729, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.7584765239618718, |
| "epoch": 0.7166305013075168, |
| "grad_norm": 29.625, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 12.3764, |
| "mean_token_accuracy": 0.7947357408702374, |
| "num_tokens": 10699737.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.7241402980871499, |
| "epoch": 0.7210816224336505, |
| "grad_norm": 29.875, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 11.657, |
| "mean_token_accuracy": 0.8018363416194916, |
| "num_tokens": 10764319.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.8102777177467942, |
| "epoch": 0.7255327435597841, |
| "grad_norm": 28.0, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 13.1101, |
| "mean_token_accuracy": 0.7860476385802031, |
| "num_tokens": 10831617.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.7965463541448117, |
| "epoch": 0.7299838646859178, |
| "grad_norm": 33.5, |
| "learning_rate": 1.858e-05, |
| "loss": 13.0657, |
| "mean_token_accuracy": 0.7860200760886074, |
| "num_tokens": 10897916.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.7968934224918485, |
| "epoch": 0.7344349858120514, |
| "grad_norm": 31.25, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 12.7605, |
| "mean_token_accuracy": 0.7907386381179095, |
| "num_tokens": 10965483.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.815986966714263, |
| "epoch": 0.738886106938185, |
| "grad_norm": 28.625, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 13.4257, |
| "mean_token_accuracy": 0.7834169883280993, |
| "num_tokens": 11030485.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.7852886017411947, |
| "epoch": 0.7433372280643187, |
| "grad_norm": 30.75, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 12.7043, |
| "mean_token_accuracy": 0.7904564626514912, |
| "num_tokens": 11097108.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.7366841196082532, |
| "epoch": 0.7477883491904523, |
| "grad_norm": 28.0, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 12.0585, |
| "mean_token_accuracy": 0.80070744715631, |
| "num_tokens": 11165356.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.8185211919248104, |
| "epoch": 0.752239470316586, |
| "grad_norm": 24.75, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 13.2307, |
| "mean_token_accuracy": 0.7844700736925005, |
| "num_tokens": 11234429.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.8034698501229286, |
| "epoch": 0.7566905914427197, |
| "grad_norm": 30.0, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 12.8032, |
| "mean_token_accuracy": 0.7891142036765814, |
| "num_tokens": 11299223.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.7566905914427197, |
| "eval_biology_entropy": 1.372378243803978, |
| "eval_biology_loss": 1.5237188339233398, |
| "eval_biology_mean_token_accuracy": 0.6542273926138877, |
| "eval_biology_num_tokens": 11299223.0, |
| "eval_biology_runtime": 50.6031, |
| "eval_biology_samples_per_second": 9.881, |
| "eval_biology_steps_per_second": 9.881, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.7566905914427197, |
| "eval_chemistry_entropy": 0.7568316695690155, |
| "eval_chemistry_loss": 0.8222190737724304, |
| "eval_chemistry_mean_token_accuracy": 0.7850142307281495, |
| "eval_chemistry_num_tokens": 11299223.0, |
| "eval_chemistry_runtime": 55.9844, |
| "eval_chemistry_samples_per_second": 8.931, |
| "eval_chemistry_steps_per_second": 8.931, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.7566905914427197, |
| "eval_math_entropy": 0.6881917223036289, |
| "eval_math_loss": 0.9334322214126587, |
| "eval_math_mean_token_accuracy": 0.7758716726303101, |
| "eval_math_num_tokens": 11299223.0, |
| "eval_math_runtime": 57.6899, |
| "eval_math_samples_per_second": 8.667, |
| "eval_math_steps_per_second": 8.667, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.7566905914427197, |
| "eval_physics_entropy": 0.7379971377849579, |
| "eval_physics_loss": 0.8059815168380737, |
| "eval_physics_mean_token_accuracy": 0.7926979689002037, |
| "eval_physics_num_tokens": 11299223.0, |
| "eval_physics_runtime": 64.5278, |
| "eval_physics_samples_per_second": 7.749, |
| "eval_physics_steps_per_second": 7.749, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.7655514618381858, |
| "epoch": 0.7611417125688533, |
| "grad_norm": 27.5, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 12.6612, |
| "mean_token_accuracy": 0.7917376033961773, |
| "num_tokens": 11371200.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.8210326373577118, |
| "epoch": 0.7655928336949869, |
| "grad_norm": 24.0, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 13.2178, |
| "mean_token_accuracy": 0.7826757222414017, |
| "num_tokens": 11436799.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.7854057714343071, |
| "epoch": 0.7700439548211205, |
| "grad_norm": 33.25, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 12.7175, |
| "mean_token_accuracy": 0.7909122589975596, |
| "num_tokens": 11502965.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.818324554990977, |
| "epoch": 0.7744950759472542, |
| "grad_norm": 26.375, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 13.2921, |
| "mean_token_accuracy": 0.7835916109383106, |
| "num_tokens": 11567422.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.7833996049128473, |
| "epoch": 0.7789461970733879, |
| "grad_norm": 23.25, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 12.7544, |
| "mean_token_accuracy": 0.7901845004409551, |
| "num_tokens": 11637076.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.8559338381513953, |
| "epoch": 0.7833973181995215, |
| "grad_norm": 31.0, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 13.8373, |
| "mean_token_accuracy": 0.7730546899139881, |
| "num_tokens": 11703496.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.7425250586122274, |
| "epoch": 0.7878484393256552, |
| "grad_norm": 26.5, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 12.0498, |
| "mean_token_accuracy": 0.7987047124654054, |
| "num_tokens": 11772773.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.7746422350406647, |
| "epoch": 0.7922995604517888, |
| "grad_norm": 28.25, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 12.5952, |
| "mean_token_accuracy": 0.7921661175787449, |
| "num_tokens": 11840317.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8539435828104616, |
| "epoch": 0.7967506815779224, |
| "grad_norm": 31.625, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 13.9317, |
| "mean_token_accuracy": 0.7740429677069187, |
| "num_tokens": 11903376.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.802261015586555, |
| "epoch": 0.801201802704056, |
| "grad_norm": 31.125, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 12.8374, |
| "mean_token_accuracy": 0.7883880365639925, |
| "num_tokens": 11965530.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_biology_entropy": 1.3964615616202354, |
| "eval_biology_loss": 1.5156149864196777, |
| "eval_biology_mean_token_accuracy": 0.6555449724793434, |
| "eval_biology_num_tokens": 11965530.0, |
| "eval_biology_runtime": 48.406, |
| "eval_biology_samples_per_second": 10.329, |
| "eval_biology_steps_per_second": 10.329, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_chemistry_entropy": 0.7734425110518932, |
| "eval_chemistry_loss": 0.8168553709983826, |
| "eval_chemistry_mean_token_accuracy": 0.7861404958367347, |
| "eval_chemistry_num_tokens": 11965530.0, |
| "eval_chemistry_runtime": 55.979, |
| "eval_chemistry_samples_per_second": 8.932, |
| "eval_chemistry_steps_per_second": 8.932, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_math_entropy": 0.6790068586170673, |
| "eval_math_loss": 0.9259929060935974, |
| "eval_math_mean_token_accuracy": 0.7774025141000748, |
| "eval_math_num_tokens": 11965530.0, |
| "eval_math_runtime": 57.7166, |
| "eval_math_samples_per_second": 8.663, |
| "eval_math_steps_per_second": 8.663, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_physics_entropy": 0.7456672674119472, |
| "eval_physics_loss": 0.8008407950401306, |
| "eval_physics_mean_token_accuracy": 0.7940772858858108, |
| "eval_physics_num_tokens": 11965530.0, |
| "eval_physics_runtime": 64.5698, |
| "eval_physics_samples_per_second": 7.744, |
| "eval_physics_steps_per_second": 7.744, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.7921732081100344, |
| "epoch": 0.8056529238301897, |
| "grad_norm": 30.375, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 13.0154, |
| "mean_token_accuracy": 0.7864332607015967, |
| "num_tokens": 12031493.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.8018918853253126, |
| "epoch": 0.8101040449563234, |
| "grad_norm": 26.5, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 13.0744, |
| "mean_token_accuracy": 0.786062810011208, |
| "num_tokens": 12097917.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.8201096205040812, |
| "epoch": 0.814555166082457, |
| "grad_norm": 27.625, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 13.1547, |
| "mean_token_accuracy": 0.7859561391174793, |
| "num_tokens": 12163029.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.7941311306320131, |
| "epoch": 0.8190062872085907, |
| "grad_norm": 30.25, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 13.0577, |
| "mean_token_accuracy": 0.7833772595971823, |
| "num_tokens": 12224427.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.8423910791054368, |
| "epoch": 0.8234574083347244, |
| "grad_norm": 28.875, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 13.5246, |
| "mean_token_accuracy": 0.7849616762250662, |
| "num_tokens": 12291917.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.781013885512948, |
| "epoch": 0.8279085294608579, |
| "grad_norm": 26.875, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 12.6225, |
| "mean_token_accuracy": 0.7902223410084843, |
| "num_tokens": 12365332.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.7929084118455648, |
| "epoch": 0.8323596505869916, |
| "grad_norm": 24.0, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 12.8591, |
| "mean_token_accuracy": 0.7884174790233374, |
| "num_tokens": 12441423.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.7592666557990014, |
| "epoch": 0.8368107717131252, |
| "grad_norm": 29.25, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 12.1548, |
| "mean_token_accuracy": 0.7982564911246299, |
| "num_tokens": 12509124.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.7280074388720095, |
| "epoch": 0.8412618928392589, |
| "grad_norm": 24.5, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 11.9887, |
| "mean_token_accuracy": 0.799430638551712, |
| "num_tokens": 12575355.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.7936474638059735, |
| "epoch": 0.8457130139653926, |
| "grad_norm": 30.5, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 13.0008, |
| "mean_token_accuracy": 0.7857760902494192, |
| "num_tokens": 12642651.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.8457130139653926, |
| "eval_biology_entropy": 1.4316322038173674, |
| "eval_biology_loss": 1.5113598108291626, |
| "eval_biology_mean_token_accuracy": 0.6554486703276634, |
| "eval_biology_num_tokens": 12642651.0, |
| "eval_biology_runtime": 48.3436, |
| "eval_biology_samples_per_second": 10.343, |
| "eval_biology_steps_per_second": 10.343, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.8457130139653926, |
| "eval_chemistry_entropy": 0.8050240458548069, |
| "eval_chemistry_loss": 0.8119493126869202, |
| "eval_chemistry_mean_token_accuracy": 0.7857032447457314, |
| "eval_chemistry_num_tokens": 12642651.0, |
| "eval_chemistry_runtime": 55.9498, |
| "eval_chemistry_samples_per_second": 8.937, |
| "eval_chemistry_steps_per_second": 8.937, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.8457130139653926, |
| "eval_math_entropy": 0.7324284837841988, |
| "eval_math_loss": 0.9314535856246948, |
| "eval_math_mean_token_accuracy": 0.774106702029705, |
| "eval_math_num_tokens": 12642651.0, |
| "eval_math_runtime": 57.6443, |
| "eval_math_samples_per_second": 8.674, |
| "eval_math_steps_per_second": 8.674, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.8457130139653926, |
| "eval_physics_entropy": 0.7888250146508217, |
| "eval_physics_loss": 0.7979229688644409, |
| "eval_physics_mean_token_accuracy": 0.7938786220550537, |
| "eval_physics_num_tokens": 12642651.0, |
| "eval_physics_runtime": 64.5548, |
| "eval_physics_samples_per_second": 7.745, |
| "eval_physics_steps_per_second": 7.745, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.7670430541969836, |
| "epoch": 0.8501641350915262, |
| "grad_norm": 28.625, |
| "learning_rate": 1.798e-05, |
| "loss": 12.0248, |
| "mean_token_accuracy": 0.7999522622674704, |
| "num_tokens": 12710065.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.7739125786349177, |
| "epoch": 0.8546152562176598, |
| "grad_norm": 29.875, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 12.8297, |
| "mean_token_accuracy": 0.7889043107628823, |
| "num_tokens": 12778408.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7532942065969109, |
| "epoch": 0.8590663773437934, |
| "grad_norm": 32.0, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 12.2406, |
| "mean_token_accuracy": 0.7968380350619555, |
| "num_tokens": 12845949.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.7928705636411906, |
| "epoch": 0.8635174984699271, |
| "grad_norm": 34.0, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 12.7215, |
| "mean_token_accuracy": 0.79536547139287, |
| "num_tokens": 12910630.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.7868898861110211, |
| "epoch": 0.8679686195960608, |
| "grad_norm": 25.25, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 12.8065, |
| "mean_token_accuracy": 0.786562405899167, |
| "num_tokens": 12975379.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.7435561339370906, |
| "epoch": 0.8724197407221944, |
| "grad_norm": 25.75, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 12.1042, |
| "mean_token_accuracy": 0.7956599161028862, |
| "num_tokens": 13046473.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.8422315520234406, |
| "epoch": 0.8768708618483281, |
| "grad_norm": 26.75, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 13.6777, |
| "mean_token_accuracy": 0.7786285186186432, |
| "num_tokens": 13105019.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.834715377446264, |
| "epoch": 0.8813219829744617, |
| "grad_norm": 28.375, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 13.4349, |
| "mean_token_accuracy": 0.7819999866187572, |
| "num_tokens": 13167768.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7662934713065624, |
| "epoch": 0.8857731041005953, |
| "grad_norm": 27.0, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 12.4302, |
| "mean_token_accuracy": 0.7945365350693464, |
| "num_tokens": 13235899.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7850984619930387, |
| "epoch": 0.890224225226729, |
| "grad_norm": 24.75, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 12.5393, |
| "mean_token_accuracy": 0.791939041018486, |
| "num_tokens": 13301659.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_biology_entropy": 1.3711472821831703, |
| "eval_biology_loss": 1.5089635848999023, |
| "eval_biology_mean_token_accuracy": 0.6568370184898377, |
| "eval_biology_num_tokens": 13301659.0, |
| "eval_biology_runtime": 48.3008, |
| "eval_biology_samples_per_second": 10.352, |
| "eval_biology_steps_per_second": 10.352, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_chemistry_entropy": 0.7663880889117718, |
| "eval_chemistry_loss": 0.8071920871734619, |
| "eval_chemistry_mean_token_accuracy": 0.7873176004290581, |
| "eval_chemistry_num_tokens": 13301659.0, |
| "eval_chemistry_runtime": 56.4708, |
| "eval_chemistry_samples_per_second": 8.854, |
| "eval_chemistry_steps_per_second": 8.854, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_math_entropy": 0.6875064164400101, |
| "eval_math_loss": 0.9262266755104065, |
| "eval_math_mean_token_accuracy": 0.7766597915887833, |
| "eval_math_num_tokens": 13301659.0, |
| "eval_math_runtime": 57.7873, |
| "eval_math_samples_per_second": 8.652, |
| "eval_math_steps_per_second": 8.652, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_physics_entropy": 0.7452095941007137, |
| "eval_physics_loss": 0.7976797819137573, |
| "eval_physics_mean_token_accuracy": 0.7943488684892654, |
| "eval_physics_num_tokens": 13301659.0, |
| "eval_physics_runtime": 64.7143, |
| "eval_physics_samples_per_second": 7.726, |
| "eval_physics_steps_per_second": 7.726, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.757675726339221, |
| "epoch": 0.8946753463528626, |
| "grad_norm": 26.5, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 12.6318, |
| "mean_token_accuracy": 0.7919723853468895, |
| "num_tokens": 13368880.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.8309912864118815, |
| "epoch": 0.8991264674789963, |
| "grad_norm": 33.25, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 13.2221, |
| "mean_token_accuracy": 0.7854577481746674, |
| "num_tokens": 13434364.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.7711499891243875, |
| "epoch": 0.9035775886051299, |
| "grad_norm": 26.0, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 12.8187, |
| "mean_token_accuracy": 0.7917429141700267, |
| "num_tokens": 13499871.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.7541942525655031, |
| "epoch": 0.9080287097312636, |
| "grad_norm": 26.0, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 12.1386, |
| "mean_token_accuracy": 0.8008204121142626, |
| "num_tokens": 13570986.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.7135496724396944, |
| "epoch": 0.9124798308573973, |
| "grad_norm": 28.125, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 11.4035, |
| "mean_token_accuracy": 0.8091535605490208, |
| "num_tokens": 13639400.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.805735620483756, |
| "epoch": 0.9169309519835308, |
| "grad_norm": 26.125, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 13.2883, |
| "mean_token_accuracy": 0.784512160345912, |
| "num_tokens": 13706343.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.8566908406093716, |
| "epoch": 0.9213820731096645, |
| "grad_norm": 31.625, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 13.6462, |
| "mean_token_accuracy": 0.7791480910032987, |
| "num_tokens": 13772145.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.8253630014136434, |
| "epoch": 0.9258331942357981, |
| "grad_norm": 24.125, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 13.3293, |
| "mean_token_accuracy": 0.7829373314976692, |
| "num_tokens": 13833901.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.7992613664828241, |
| "epoch": 0.9302843153619318, |
| "grad_norm": 34.0, |
| "learning_rate": 1.758e-05, |
| "loss": 13.0225, |
| "mean_token_accuracy": 0.7886002086102962, |
| "num_tokens": 13890922.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.7736437637358904, |
| "epoch": 0.9347354364880655, |
| "grad_norm": 28.0, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 12.5875, |
| "mean_token_accuracy": 0.7917680319398641, |
| "num_tokens": 13954258.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.9347354364880655, |
| "eval_biology_entropy": 1.3883276455402374, |
| "eval_biology_loss": 1.500346064567566, |
| "eval_biology_mean_token_accuracy": 0.6576701752543449, |
| "eval_biology_num_tokens": 13954258.0, |
| "eval_biology_runtime": 48.6057, |
| "eval_biology_samples_per_second": 10.287, |
| "eval_biology_steps_per_second": 10.287, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.9347354364880655, |
| "eval_chemistry_entropy": 0.7874405336976051, |
| "eval_chemistry_loss": 0.8040579557418823, |
| "eval_chemistry_mean_token_accuracy": 0.7880919007658959, |
| "eval_chemistry_num_tokens": 13954258.0, |
| "eval_chemistry_runtime": 57.2321, |
| "eval_chemistry_samples_per_second": 8.736, |
| "eval_chemistry_steps_per_second": 8.736, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.9347354364880655, |
| "eval_math_entropy": 0.7219732905030251, |
| "eval_math_loss": 0.9228026270866394, |
| "eval_math_mean_token_accuracy": 0.7756567891836167, |
| "eval_math_num_tokens": 13954258.0, |
| "eval_math_runtime": 57.7165, |
| "eval_math_samples_per_second": 8.663, |
| "eval_math_steps_per_second": 8.663, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.9347354364880655, |
| "eval_physics_entropy": 0.7771098895072936, |
| "eval_physics_loss": 0.7936882376670837, |
| "eval_physics_mean_token_accuracy": 0.7947471792697907, |
| "eval_physics_num_tokens": 13954258.0, |
| "eval_physics_runtime": 64.7161, |
| "eval_physics_samples_per_second": 7.726, |
| "eval_physics_steps_per_second": 7.726, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.7432751135900617, |
| "epoch": 0.9391865576141991, |
| "grad_norm": 26.5, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 11.9833, |
| "mean_token_accuracy": 0.8013271361589431, |
| "num_tokens": 14024422.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.761086848191917, |
| "epoch": 0.9436376787403328, |
| "grad_norm": 32.5, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 12.4642, |
| "mean_token_accuracy": 0.7918149266391993, |
| "num_tokens": 14093888.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.8075856968760491, |
| "epoch": 0.9480887998664663, |
| "grad_norm": 41.5, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 12.9706, |
| "mean_token_accuracy": 0.786743215098977, |
| "num_tokens": 14160983.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.7668558545410633, |
| "epoch": 0.9525399209926, |
| "grad_norm": 31.5, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 12.6955, |
| "mean_token_accuracy": 0.79208915643394, |
| "num_tokens": 14222626.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.8176628946326673, |
| "epoch": 0.9569910421187336, |
| "grad_norm": 30.75, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 13.1126, |
| "mean_token_accuracy": 0.7838487748056651, |
| "num_tokens": 14287730.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.808471773006022, |
| "epoch": 0.9614421632448673, |
| "grad_norm": 28.5, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 13.0789, |
| "mean_token_accuracy": 0.7878662195056677, |
| "num_tokens": 14354918.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.7701966028660536, |
| "epoch": 0.965893284371001, |
| "grad_norm": 32.5, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 12.4855, |
| "mean_token_accuracy": 0.7922379944473505, |
| "num_tokens": 14427338.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.7810970747843384, |
| "epoch": 0.9703444054971346, |
| "grad_norm": 26.75, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 12.7269, |
| "mean_token_accuracy": 0.7883514825254678, |
| "num_tokens": 14493349.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.7944581422954797, |
| "epoch": 0.9747955266232682, |
| "grad_norm": 29.0, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 12.7584, |
| "mean_token_accuracy": 0.7889562319964171, |
| "num_tokens": 14555091.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.7532775722444057, |
| "epoch": 0.9792466477494018, |
| "grad_norm": 24.75, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 12.1562, |
| "mean_token_accuracy": 0.798464347794652, |
| "num_tokens": 14620244.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_biology_entropy": 1.4140792628526688, |
| "eval_biology_loss": 1.5017729997634888, |
| "eval_biology_mean_token_accuracy": 0.6577960692644119, |
| "eval_biology_num_tokens": 14620244.0, |
| "eval_biology_runtime": 48.8657, |
| "eval_biology_samples_per_second": 10.232, |
| "eval_biology_steps_per_second": 10.232, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_chemistry_entropy": 0.7828991089761257, |
| "eval_chemistry_loss": 0.8003749251365662, |
| "eval_chemistry_mean_token_accuracy": 0.7885151287317276, |
| "eval_chemistry_num_tokens": 14620244.0, |
| "eval_chemistry_runtime": 56.3139, |
| "eval_chemistry_samples_per_second": 8.879, |
| "eval_chemistry_steps_per_second": 8.879, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_math_entropy": 0.6969314276576042, |
| "eval_math_loss": 0.9252308011054993, |
| "eval_math_mean_token_accuracy": 0.7768602112531662, |
| "eval_math_num_tokens": 14620244.0, |
| "eval_math_runtime": 58.4013, |
| "eval_math_samples_per_second": 8.561, |
| "eval_math_steps_per_second": 8.561, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_physics_entropy": 0.7627284322977066, |
| "eval_physics_loss": 0.7935377955436707, |
| "eval_physics_mean_token_accuracy": 0.7945903204679489, |
| "eval_physics_num_tokens": 14620244.0, |
| "eval_physics_runtime": 64.5805, |
| "eval_physics_samples_per_second": 7.742, |
| "eval_physics_steps_per_second": 7.742, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.7444026693701744, |
| "epoch": 0.9836977688755355, |
| "grad_norm": 25.75, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 12.1322, |
| "mean_token_accuracy": 0.8003021962940693, |
| "num_tokens": 14688111.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.7629616463556885, |
| "epoch": 0.9881488900016692, |
| "grad_norm": 24.375, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 12.3807, |
| "mean_token_accuracy": 0.7964084807783365, |
| "num_tokens": 14760951.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.8271248353645205, |
| "epoch": 0.9926000111278028, |
| "grad_norm": 28.625, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 13.3952, |
| "mean_token_accuracy": 0.7831025589257479, |
| "num_tokens": 14823141.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.7894777454435825, |
| "epoch": 0.9970511322539365, |
| "grad_norm": 31.25, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 12.8312, |
| "mean_token_accuracy": 0.7900656949728727, |
| "num_tokens": 14886593.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.7630960840683478, |
| "epoch": 1.00133533633784, |
| "grad_norm": 34.0, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 11.6223, |
| "mean_token_accuracy": 0.7963757422063258, |
| "num_tokens": 14951602.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.7022327370941639, |
| "epoch": 1.0057864574639737, |
| "grad_norm": 27.25, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 11.3126, |
| "mean_token_accuracy": 0.8089025482535362, |
| "num_tokens": 15016941.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.6971630771644414, |
| "epoch": 1.0102375785901074, |
| "grad_norm": 30.25, |
| "learning_rate": 1.718e-05, |
| "loss": 11.3002, |
| "mean_token_accuracy": 0.8061123300343752, |
| "num_tokens": 15083858.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.6488590116612614, |
| "epoch": 1.014688699716241, |
| "grad_norm": 31.0, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 10.3384, |
| "mean_token_accuracy": 0.8197494275867939, |
| "num_tokens": 15152865.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.7088900838978589, |
| "epoch": 1.0191398208423748, |
| "grad_norm": 30.625, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 11.6142, |
| "mean_token_accuracy": 0.8018755106255412, |
| "num_tokens": 15218564.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.726468394510448, |
| "epoch": 1.0235909419685083, |
| "grad_norm": 35.0, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 11.6262, |
| "mean_token_accuracy": 0.8015792027115822, |
| "num_tokens": 15281217.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.0235909419685083, |
| "eval_biology_entropy": 1.2445224853754044, |
| "eval_biology_loss": 1.5320156812667847, |
| "eval_biology_mean_token_accuracy": 0.6555127831697464, |
| "eval_biology_num_tokens": 15281217.0, |
| "eval_biology_runtime": 48.912, |
| "eval_biology_samples_per_second": 10.222, |
| "eval_biology_steps_per_second": 10.222, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.0235909419685083, |
| "eval_chemistry_entropy": 0.696570024073124, |
| "eval_chemistry_loss": 0.8090431094169617, |
| "eval_chemistry_mean_token_accuracy": 0.7879877038598061, |
| "eval_chemistry_num_tokens": 15281217.0, |
| "eval_chemistry_runtime": 56.0981, |
| "eval_chemistry_samples_per_second": 8.913, |
| "eval_chemistry_steps_per_second": 8.913, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.0235909419685083, |
| "eval_math_entropy": 0.6726472755968571, |
| "eval_math_loss": 0.9390206336975098, |
| "eval_math_mean_token_accuracy": 0.7758509650230407, |
| "eval_math_num_tokens": 15281217.0, |
| "eval_math_runtime": 57.6927, |
| "eval_math_samples_per_second": 8.667, |
| "eval_math_steps_per_second": 8.667, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.0235909419685083, |
| "eval_physics_entropy": 0.7037301003932953, |
| "eval_physics_loss": 0.8007642030715942, |
| "eval_physics_mean_token_accuracy": 0.7939389154314995, |
| "eval_physics_num_tokens": 15281217.0, |
| "eval_physics_runtime": 64.7129, |
| "eval_physics_samples_per_second": 7.726, |
| "eval_physics_steps_per_second": 7.726, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.6810367662459612, |
| "epoch": 1.0280420630946419, |
| "grad_norm": 30.0, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 10.979, |
| "mean_token_accuracy": 0.8090169824659824, |
| "num_tokens": 15346128.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.70857991008088, |
| "epoch": 1.0324931842207756, |
| "grad_norm": 26.75, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 11.6241, |
| "mean_token_accuracy": 0.8005826868116855, |
| "num_tokens": 15412606.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.7052770711481571, |
| "epoch": 1.0369443053469092, |
| "grad_norm": 28.5, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 11.234, |
| "mean_token_accuracy": 0.808265782892704, |
| "num_tokens": 15484636.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.6550373506732285, |
| "epoch": 1.041395426473043, |
| "grad_norm": 31.625, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 10.5516, |
| "mean_token_accuracy": 0.8196316309273243, |
| "num_tokens": 15556398.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.6928601786494255, |
| "epoch": 1.0458465475991765, |
| "grad_norm": 38.0, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 11.1238, |
| "mean_token_accuracy": 0.8107457913458347, |
| "num_tokens": 15624248.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.6943131916224956, |
| "epoch": 1.0502976687253103, |
| "grad_norm": 32.5, |
| "learning_rate": 1.698e-05, |
| "loss": 11.3797, |
| "mean_token_accuracy": 0.8058217719197274, |
| "num_tokens": 15689184.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.7262950800359249, |
| "epoch": 1.0547487898514438, |
| "grad_norm": 37.5, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 11.6024, |
| "mean_token_accuracy": 0.8025738406926394, |
| "num_tokens": 15754370.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.6695299903862179, |
| "epoch": 1.0591999109775774, |
| "grad_norm": 28.125, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 11.0414, |
| "mean_token_accuracy": 0.8116515409201384, |
| "num_tokens": 15824975.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.7191860099323094, |
| "epoch": 1.0636510321037111, |
| "grad_norm": 31.625, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 11.4248, |
| "mean_token_accuracy": 0.8067556466907263, |
| "num_tokens": 15886149.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.6812492808327079, |
| "epoch": 1.0681021532298447, |
| "grad_norm": 29.125, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 11.0997, |
| "mean_token_accuracy": 0.810133233293891, |
| "num_tokens": 15956560.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.0681021532298447, |
| "eval_biology_entropy": 1.2292085955142975, |
| "eval_biology_loss": 1.5338053703308105, |
| "eval_biology_mean_token_accuracy": 0.6553138529062271, |
| "eval_biology_num_tokens": 15956560.0, |
| "eval_biology_runtime": 48.3169, |
| "eval_biology_samples_per_second": 10.348, |
| "eval_biology_steps_per_second": 10.348, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.0681021532298447, |
| "eval_chemistry_entropy": 0.6888745855689049, |
| "eval_chemistry_loss": 0.8065437078475952, |
| "eval_chemistry_mean_token_accuracy": 0.7879378204941749, |
| "eval_chemistry_num_tokens": 15956560.0, |
| "eval_chemistry_runtime": 55.9304, |
| "eval_chemistry_samples_per_second": 8.94, |
| "eval_chemistry_steps_per_second": 8.94, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.0681021532298447, |
| "eval_math_entropy": 0.6415176319479943, |
| "eval_math_loss": 0.9477712512016296, |
| "eval_math_mean_token_accuracy": 0.7750236385464668, |
| "eval_math_num_tokens": 15956560.0, |
| "eval_math_runtime": 57.6856, |
| "eval_math_samples_per_second": 8.668, |
| "eval_math_steps_per_second": 8.668, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.0681021532298447, |
| "eval_physics_entropy": 0.6823045939207077, |
| "eval_physics_loss": 0.8033751249313354, |
| "eval_physics_mean_token_accuracy": 0.7934308009147644, |
| "eval_physics_num_tokens": 15956560.0, |
| "eval_physics_runtime": 64.5401, |
| "eval_physics_samples_per_second": 7.747, |
| "eval_physics_steps_per_second": 7.747, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.6983694306574761, |
| "epoch": 1.0725532743559785, |
| "grad_norm": 27.5, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 11.2242, |
| "mean_token_accuracy": 0.8091347351670265, |
| "num_tokens": 16025633.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.6855760600417853, |
| "epoch": 1.077004395482112, |
| "grad_norm": 37.25, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 11.1355, |
| "mean_token_accuracy": 0.8072514422237873, |
| "num_tokens": 16094535.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.6998407649807632, |
| "epoch": 1.0814555166082458, |
| "grad_norm": 34.0, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 11.264, |
| "mean_token_accuracy": 0.8066206596791744, |
| "num_tokens": 16159781.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.6935182044282555, |
| "epoch": 1.0859066377343793, |
| "grad_norm": 32.0, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 11.2659, |
| "mean_token_accuracy": 0.8107964657247066, |
| "num_tokens": 16227695.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.7162637470290065, |
| "epoch": 1.090357758860513, |
| "grad_norm": 35.5, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 11.5333, |
| "mean_token_accuracy": 0.8014760285615921, |
| "num_tokens": 16293211.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.6996999345719814, |
| "epoch": 1.0948088799866467, |
| "grad_norm": 36.5, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 11.2896, |
| "mean_token_accuracy": 0.8079928819090128, |
| "num_tokens": 16361111.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.7367887786589563, |
| "epoch": 1.0992600011127802, |
| "grad_norm": 33.0, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 11.6919, |
| "mean_token_accuracy": 0.8021658387035131, |
| "num_tokens": 16424587.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.6744886384345591, |
| "epoch": 1.103711122238914, |
| "grad_norm": 30.375, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 11.0466, |
| "mean_token_accuracy": 0.8132975608110428, |
| "num_tokens": 16495082.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.7061602457426488, |
| "epoch": 1.1081622433650475, |
| "grad_norm": 35.25, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 11.4216, |
| "mean_token_accuracy": 0.8040795765817166, |
| "num_tokens": 16557596.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.6840292084030807, |
| "epoch": 1.1126133644911813, |
| "grad_norm": 25.0, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 11.0634, |
| "mean_token_accuracy": 0.8077531451359391, |
| "num_tokens": 16625999.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.1126133644911813, |
| "eval_biology_entropy": 1.2276055392622949, |
| "eval_biology_loss": 1.5302549600601196, |
| "eval_biology_mean_token_accuracy": 0.6561977040767669, |
| "eval_biology_num_tokens": 16625999.0, |
| "eval_biology_runtime": 48.4456, |
| "eval_biology_samples_per_second": 10.321, |
| "eval_biology_steps_per_second": 10.321, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.1126133644911813, |
| "eval_chemistry_entropy": 0.704082203745842, |
| "eval_chemistry_loss": 0.8006765246391296, |
| "eval_chemistry_mean_token_accuracy": 0.7888219217658043, |
| "eval_chemistry_num_tokens": 16625999.0, |
| "eval_chemistry_runtime": 56.1591, |
| "eval_chemistry_samples_per_second": 8.903, |
| "eval_chemistry_steps_per_second": 8.903, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.1126133644911813, |
| "eval_math_entropy": 0.6378613507449626, |
| "eval_math_loss": 0.9441410899162292, |
| "eval_math_mean_token_accuracy": 0.7765913832187653, |
| "eval_math_num_tokens": 16625999.0, |
| "eval_math_runtime": 57.6636, |
| "eval_math_samples_per_second": 8.671, |
| "eval_math_steps_per_second": 8.671, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.1126133644911813, |
| "eval_physics_entropy": 0.69436886754632, |
| "eval_physics_loss": 0.8003541827201843, |
| "eval_physics_mean_token_accuracy": 0.7943150995373726, |
| "eval_physics_num_tokens": 16625999.0, |
| "eval_physics_runtime": 64.6072, |
| "eval_physics_samples_per_second": 7.739, |
| "eval_physics_steps_per_second": 7.739, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.6595550235360861, |
| "epoch": 1.1170644856173149, |
| "grad_norm": 29.375, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 10.6827, |
| "mean_token_accuracy": 0.8140194948762656, |
| "num_tokens": 16697896.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.7656160462647676, |
| "epoch": 1.1215156067434484, |
| "grad_norm": 27.5, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 12.3143, |
| "mean_token_accuracy": 0.794426592066884, |
| "num_tokens": 16763539.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.6974834809079766, |
| "epoch": 1.1259667278695822, |
| "grad_norm": 34.0, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 11.2994, |
| "mean_token_accuracy": 0.8065015953034163, |
| "num_tokens": 16829702.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.6626461429521442, |
| "epoch": 1.1304178489957157, |
| "grad_norm": 25.75, |
| "learning_rate": 1.658e-05, |
| "loss": 10.6385, |
| "mean_token_accuracy": 0.8147155798971653, |
| "num_tokens": 16899120.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.676584966108203, |
| "epoch": 1.1348689701218495, |
| "grad_norm": 34.0, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 11.0909, |
| "mean_token_accuracy": 0.8109572473913431, |
| "num_tokens": 16964702.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.6975578533485531, |
| "epoch": 1.139320091247983, |
| "grad_norm": 36.75, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 11.3117, |
| "mean_token_accuracy": 0.8072918478399516, |
| "num_tokens": 17028582.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.6725874023512006, |
| "epoch": 1.1437712123741166, |
| "grad_norm": 30.5, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 10.5511, |
| "mean_token_accuracy": 0.8158095881342888, |
| "num_tokens": 17094349.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.7200934633612632, |
| "epoch": 1.1482223335002504, |
| "grad_norm": 38.0, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 11.8588, |
| "mean_token_accuracy": 0.7991140142083168, |
| "num_tokens": 17161099.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.6904380663298071, |
| "epoch": 1.152673454626384, |
| "grad_norm": 31.125, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 11.2058, |
| "mean_token_accuracy": 0.810056545957923, |
| "num_tokens": 17226592.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.7087613576091826, |
| "epoch": 1.1571245757525177, |
| "grad_norm": 28.5, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 11.5093, |
| "mean_token_accuracy": 0.8009733863174915, |
| "num_tokens": 17293076.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.1571245757525177, |
| "eval_biology_entropy": 1.218234146952629, |
| "eval_biology_loss": 1.5305497646331787, |
| "eval_biology_mean_token_accuracy": 0.6546677138805389, |
| "eval_biology_num_tokens": 17293076.0, |
| "eval_biology_runtime": 48.5075, |
| "eval_biology_samples_per_second": 10.308, |
| "eval_biology_steps_per_second": 10.308, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.1571245757525177, |
| "eval_chemistry_entropy": 0.7129946960806847, |
| "eval_chemistry_loss": 0.7968347072601318, |
| "eval_chemistry_mean_token_accuracy": 0.7894422112107277, |
| "eval_chemistry_num_tokens": 17293076.0, |
| "eval_chemistry_runtime": 56.0308, |
| "eval_chemistry_samples_per_second": 8.924, |
| "eval_chemistry_steps_per_second": 8.924, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.1571245757525177, |
| "eval_math_entropy": 0.6545038941800594, |
| "eval_math_loss": 0.9409072399139404, |
| "eval_math_mean_token_accuracy": 0.7757919223308564, |
| "eval_math_num_tokens": 17293076.0, |
| "eval_math_runtime": 58.2426, |
| "eval_math_samples_per_second": 8.585, |
| "eval_math_steps_per_second": 8.585, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.1571245757525177, |
| "eval_physics_entropy": 0.7075124041736126, |
| "eval_physics_loss": 0.7950467467308044, |
| "eval_physics_mean_token_accuracy": 0.7946779307126999, |
| "eval_physics_num_tokens": 17293076.0, |
| "eval_physics_runtime": 64.647, |
| "eval_physics_samples_per_second": 7.734, |
| "eval_physics_steps_per_second": 7.734, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.7032289003953338, |
| "epoch": 1.1615756968786513, |
| "grad_norm": 33.5, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 11.3473, |
| "mean_token_accuracy": 0.8045333504676819, |
| "num_tokens": 17360241.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.7421052502468228, |
| "epoch": 1.166026818004785, |
| "grad_norm": 38.75, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 11.957, |
| "mean_token_accuracy": 0.7979696854948998, |
| "num_tokens": 17424840.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.7015250721015036, |
| "epoch": 1.1704779391309186, |
| "grad_norm": 31.25, |
| "learning_rate": 1.638e-05, |
| "loss": 11.3944, |
| "mean_token_accuracy": 0.8029961306601763, |
| "num_tokens": 17493054.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.7256078576669097, |
| "epoch": 1.1749290602570523, |
| "grad_norm": 39.25, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 11.7414, |
| "mean_token_accuracy": 0.7999141335487365, |
| "num_tokens": 17557844.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.6931503497064113, |
| "epoch": 1.179380181383186, |
| "grad_norm": 33.0, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 11.2211, |
| "mean_token_accuracy": 0.8085327189415693, |
| "num_tokens": 17627019.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.7268060699105263, |
| "epoch": 1.1838313025093194, |
| "grad_norm": 27.25, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 11.7197, |
| "mean_token_accuracy": 0.8022879388183355, |
| "num_tokens": 17692635.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.7190582377836108, |
| "epoch": 1.1882824236354532, |
| "grad_norm": 24.5, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 11.5057, |
| "mean_token_accuracy": 0.8022625509649515, |
| "num_tokens": 17760815.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.6894477564841509, |
| "epoch": 1.1927335447615868, |
| "grad_norm": 32.0, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 11.4586, |
| "mean_token_accuracy": 0.8065251305699348, |
| "num_tokens": 17826556.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.7025763915851713, |
| "epoch": 1.1971846658877205, |
| "grad_norm": 30.5, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 11.1576, |
| "mean_token_accuracy": 0.8117593966424466, |
| "num_tokens": 17890493.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.6598136224783957, |
| "epoch": 1.201635787013854, |
| "grad_norm": 34.0, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 10.7716, |
| "mean_token_accuracy": 0.8109393406659364, |
| "num_tokens": 17955924.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.201635787013854, |
| "eval_biology_entropy": 1.2750364818573, |
| "eval_biology_loss": 1.5298259258270264, |
| "eval_biology_mean_token_accuracy": 0.6556756514310836, |
| "eval_biology_num_tokens": 17955924.0, |
| "eval_biology_runtime": 48.3147, |
| "eval_biology_samples_per_second": 10.349, |
| "eval_biology_steps_per_second": 10.349, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.201635787013854, |
| "eval_chemistry_entropy": 0.7189217019379139, |
| "eval_chemistry_loss": 0.796171247959137, |
| "eval_chemistry_mean_token_accuracy": 0.7899833571314812, |
| "eval_chemistry_num_tokens": 17955924.0, |
| "eval_chemistry_runtime": 55.9788, |
| "eval_chemistry_samples_per_second": 8.932, |
| "eval_chemistry_steps_per_second": 8.932, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.201635787013854, |
| "eval_math_entropy": 0.6661258788108826, |
| "eval_math_loss": 0.9415155053138733, |
| "eval_math_mean_token_accuracy": 0.7756984531879425, |
| "eval_math_num_tokens": 17955924.0, |
| "eval_math_runtime": 57.6775, |
| "eval_math_samples_per_second": 8.669, |
| "eval_math_steps_per_second": 8.669, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.201635787013854, |
| "eval_physics_entropy": 0.7224332968890667, |
| "eval_physics_loss": 0.7987887859344482, |
| "eval_physics_mean_token_accuracy": 0.7940054004192352, |
| "eval_physics_num_tokens": 17955924.0, |
| "eval_physics_runtime": 64.5962, |
| "eval_physics_samples_per_second": 7.74, |
| "eval_physics_steps_per_second": 7.74, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.6855649249628186, |
| "epoch": 1.2060869081399876, |
| "grad_norm": 36.25, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 10.8523, |
| "mean_token_accuracy": 0.811124250292778, |
| "num_tokens": 18024547.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.6900663454085588, |
| "epoch": 1.2105380292661214, |
| "grad_norm": 35.25, |
| "learning_rate": 1.618e-05, |
| "loss": 11.4517, |
| "mean_token_accuracy": 0.8064537368714809, |
| "num_tokens": 18093886.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.7037728901952505, |
| "epoch": 1.214989150392255, |
| "grad_norm": 40.5, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 11.2797, |
| "mean_token_accuracy": 0.8061616580933333, |
| "num_tokens": 18166010.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.7059247604571283, |
| "epoch": 1.2194402715183887, |
| "grad_norm": 34.75, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 11.1789, |
| "mean_token_accuracy": 0.8107867985963821, |
| "num_tokens": 18236214.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.6619722669944167, |
| "epoch": 1.2238913926445223, |
| "grad_norm": 27.625, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 10.8868, |
| "mean_token_accuracy": 0.8131192527711392, |
| "num_tokens": 18300911.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.7543207122012973, |
| "epoch": 1.228342513770656, |
| "grad_norm": 36.75, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 12.1683, |
| "mean_token_accuracy": 0.7948378812521696, |
| "num_tokens": 18361378.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.7237727696076035, |
| "epoch": 1.2327936348967896, |
| "grad_norm": 30.5, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 11.7555, |
| "mean_token_accuracy": 0.8012358117848635, |
| "num_tokens": 18426528.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.6506179778836667, |
| "epoch": 1.2372447560229234, |
| "grad_norm": 32.25, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 10.5433, |
| "mean_token_accuracy": 0.8168023493140936, |
| "num_tokens": 18495853.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.673209915868938, |
| "epoch": 1.241695877149057, |
| "grad_norm": 29.5, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 10.9619, |
| "mean_token_accuracy": 0.8100661238655448, |
| "num_tokens": 18561871.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.7303794099017977, |
| "epoch": 1.2461469982751905, |
| "grad_norm": 32.0, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 11.7043, |
| "mean_token_accuracy": 0.803914862498641, |
| "num_tokens": 18632874.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.2461469982751905, |
| "eval_biology_entropy": 1.2309754735827445, |
| "eval_biology_loss": 1.5292385816574097, |
| "eval_biology_mean_token_accuracy": 0.656067096889019, |
| "eval_biology_num_tokens": 18632874.0, |
| "eval_biology_runtime": 48.3693, |
| "eval_biology_samples_per_second": 10.337, |
| "eval_biology_steps_per_second": 10.337, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.2461469982751905, |
| "eval_chemistry_entropy": 0.7238665162920952, |
| "eval_chemistry_loss": 0.7939327359199524, |
| "eval_chemistry_mean_token_accuracy": 0.7904411465525627, |
| "eval_chemistry_num_tokens": 18632874.0, |
| "eval_chemistry_runtime": 56.0588, |
| "eval_chemistry_samples_per_second": 8.919, |
| "eval_chemistry_steps_per_second": 8.919, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.2461469982751905, |
| "eval_math_entropy": 0.6524332526922226, |
| "eval_math_loss": 0.9418564438819885, |
| "eval_math_mean_token_accuracy": 0.7766549957990646, |
| "eval_math_num_tokens": 18632874.0, |
| "eval_math_runtime": 57.7209, |
| "eval_math_samples_per_second": 8.662, |
| "eval_math_steps_per_second": 8.662, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.2461469982751905, |
| "eval_physics_entropy": 0.716444758027792, |
| "eval_physics_loss": 0.794815182685852, |
| "eval_physics_mean_token_accuracy": 0.7951010932922363, |
| "eval_physics_num_tokens": 18632874.0, |
| "eval_physics_runtime": 64.5966, |
| "eval_physics_samples_per_second": 7.74, |
| "eval_physics_steps_per_second": 7.74, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.6917299875058234, |
| "epoch": 1.2505981194013243, |
| "grad_norm": 30.75, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 11.2511, |
| "mean_token_accuracy": 0.8058730617165566, |
| "num_tokens": 18705033.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.7333694417960942, |
| "epoch": 1.2550492405274578, |
| "grad_norm": 28.0, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 11.6592, |
| "mean_token_accuracy": 0.8026581913232803, |
| "num_tokens": 18775027.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.680407952517271, |
| "epoch": 1.2595003616535916, |
| "grad_norm": 39.0, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 11.1584, |
| "mean_token_accuracy": 0.8098192904144526, |
| "num_tokens": 18840707.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.6915396029129625, |
| "epoch": 1.2639514827797251, |
| "grad_norm": 28.25, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 11.0018, |
| "mean_token_accuracy": 0.808440937101841, |
| "num_tokens": 18907324.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.6727678737603128, |
| "epoch": 1.2684026039058587, |
| "grad_norm": 27.5, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 11.0561, |
| "mean_token_accuracy": 0.8090311955660582, |
| "num_tokens": 18980315.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.7300252987071871, |
| "epoch": 1.2728537250319925, |
| "grad_norm": 29.625, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 11.8477, |
| "mean_token_accuracy": 0.7992782685905695, |
| "num_tokens": 19047920.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.7177799047902227, |
| "epoch": 1.277304846158126, |
| "grad_norm": 25.0, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 11.4237, |
| "mean_token_accuracy": 0.8047383703291416, |
| "num_tokens": 19116973.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.691168091353029, |
| "epoch": 1.2817559672842598, |
| "grad_norm": 29.125, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 11.1768, |
| "mean_token_accuracy": 0.809157208353281, |
| "num_tokens": 19180813.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.6816177747212351, |
| "epoch": 1.2862070884103933, |
| "grad_norm": 28.75, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 11.0348, |
| "mean_token_accuracy": 0.808470681682229, |
| "num_tokens": 19257345.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.717674554977566, |
| "epoch": 1.2906582095365269, |
| "grad_norm": 32.5, |
| "learning_rate": 1.578e-05, |
| "loss": 11.7867, |
| "mean_token_accuracy": 0.8008464775979519, |
| "num_tokens": 19324641.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.2906582095365269, |
| "eval_biology_entropy": 1.2327818762660026, |
| "eval_biology_loss": 1.5377376079559326, |
| "eval_biology_mean_token_accuracy": 0.6552961577773094, |
| "eval_biology_num_tokens": 19324641.0, |
| "eval_biology_runtime": 48.4145, |
| "eval_biology_samples_per_second": 10.327, |
| "eval_biology_steps_per_second": 10.327, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.2906582095365269, |
| "eval_chemistry_entropy": 0.7290274969637394, |
| "eval_chemistry_loss": 0.7913830280303955, |
| "eval_chemistry_mean_token_accuracy": 0.7914319233894348, |
| "eval_chemistry_num_tokens": 19324641.0, |
| "eval_chemistry_runtime": 55.9996, |
| "eval_chemistry_samples_per_second": 8.929, |
| "eval_chemistry_steps_per_second": 8.929, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.2906582095365269, |
| "eval_math_entropy": 0.671364810436964, |
| "eval_math_loss": 0.9436923861503601, |
| "eval_math_mean_token_accuracy": 0.7749842387437821, |
| "eval_math_num_tokens": 19324641.0, |
| "eval_math_runtime": 57.6891, |
| "eval_math_samples_per_second": 8.667, |
| "eval_math_steps_per_second": 8.667, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.2906582095365269, |
| "eval_physics_entropy": 0.7204897730350495, |
| "eval_physics_loss": 0.797059178352356, |
| "eval_physics_mean_token_accuracy": 0.7948838269710541, |
| "eval_physics_num_tokens": 19324641.0, |
| "eval_physics_runtime": 64.7427, |
| "eval_physics_samples_per_second": 7.723, |
| "eval_physics_steps_per_second": 7.723, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.6798546722158789, |
| "epoch": 1.2951093306626607, |
| "grad_norm": 33.25, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 10.9395, |
| "mean_token_accuracy": 0.8126216687262058, |
| "num_tokens": 19391043.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.672836896032095, |
| "epoch": 1.2995604517887944, |
| "grad_norm": 34.0, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 10.7236, |
| "mean_token_accuracy": 0.8174181647598744, |
| "num_tokens": 19461093.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.6202881710603834, |
| "epoch": 1.304011572914928, |
| "grad_norm": 33.5, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 10.3584, |
| "mean_token_accuracy": 0.8210928410291671, |
| "num_tokens": 19527736.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.7194889483973383, |
| "epoch": 1.3084626940410615, |
| "grad_norm": 27.25, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 11.2977, |
| "mean_token_accuracy": 0.807739894837141, |
| "num_tokens": 19595645.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.6857993675395846, |
| "epoch": 1.3129138151671953, |
| "grad_norm": 26.375, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 11.2004, |
| "mean_token_accuracy": 0.8103679992258549, |
| "num_tokens": 19660322.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.7261745985597372, |
| "epoch": 1.3173649362933288, |
| "grad_norm": 32.25, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 11.6682, |
| "mean_token_accuracy": 0.799722233042121, |
| "num_tokens": 19725399.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.7061497126705945, |
| "epoch": 1.3218160574194626, |
| "grad_norm": 35.75, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 11.3375, |
| "mean_token_accuracy": 0.8046855855733156, |
| "num_tokens": 19791328.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.6635642904788256, |
| "epoch": 1.3262671785455962, |
| "grad_norm": 36.0, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 10.8223, |
| "mean_token_accuracy": 0.8167481131851673, |
| "num_tokens": 19857430.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.6782514818012715, |
| "epoch": 1.3307182996717297, |
| "grad_norm": 36.75, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 10.6277, |
| "mean_token_accuracy": 0.8130542360246181, |
| "num_tokens": 19922864.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.6839352380484343, |
| "epoch": 1.3351694207978635, |
| "grad_norm": 40.5, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 11.4643, |
| "mean_token_accuracy": 0.8028521131724119, |
| "num_tokens": 19988673.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.3351694207978635, |
| "eval_biology_entropy": 1.1549335314035416, |
| "eval_biology_loss": 1.5521533489227295, |
| "eval_biology_mean_token_accuracy": 0.654981644809246, |
| "eval_biology_num_tokens": 19988673.0, |
| "eval_biology_runtime": 48.3548, |
| "eval_biology_samples_per_second": 10.34, |
| "eval_biology_steps_per_second": 10.34, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.3351694207978635, |
| "eval_chemistry_entropy": 0.7007716604173183, |
| "eval_chemistry_loss": 0.7918070554733276, |
| "eval_chemistry_mean_token_accuracy": 0.7912032285928726, |
| "eval_chemistry_num_tokens": 19988673.0, |
| "eval_chemistry_runtime": 56.1386, |
| "eval_chemistry_samples_per_second": 8.907, |
| "eval_chemistry_steps_per_second": 8.907, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.3351694207978635, |
| "eval_math_entropy": 0.6557615910470486, |
| "eval_math_loss": 0.9451307654380798, |
| "eval_math_mean_token_accuracy": 0.7748154906630516, |
| "eval_math_num_tokens": 19988673.0, |
| "eval_math_runtime": 57.6345, |
| "eval_math_samples_per_second": 8.675, |
| "eval_math_steps_per_second": 8.675, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.3351694207978635, |
| "eval_physics_entropy": 0.7059628334641457, |
| "eval_physics_loss": 0.7964131832122803, |
| "eval_physics_mean_token_accuracy": 0.7947819554805755, |
| "eval_physics_num_tokens": 19988673.0, |
| "eval_physics_runtime": 64.5389, |
| "eval_physics_samples_per_second": 7.747, |
| "eval_physics_steps_per_second": 7.747, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "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": 1.3403278025521495e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
