{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.09808409076047865, "eval_steps": 500, "global_step": 1500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "entropy": 10.742113399505616, "epoch": 0.0003269469692015955, "grad_norm": 4.78125, "learning_rate": 2e-06, "loss": 10.776, "mean_token_accuracy": 0.0, "num_tokens": 10275.0, "step": 5 }, { "entropy": 10.742127990722656, "epoch": 0.000653893938403191, "grad_norm": 4.65625, "learning_rate": 4.5e-06, "loss": 10.7595, "mean_token_accuracy": 0.0, "num_tokens": 20293.0, "step": 10 }, { "entropy": 10.742136764526368, "epoch": 0.0009808409076047864, "grad_norm": 4.625, "learning_rate": 7e-06, "loss": 10.7523, "mean_token_accuracy": 0.0, "num_tokens": 30769.0, "step": 15 }, { "entropy": 10.742146301269532, "epoch": 0.001307787876806382, "grad_norm": 4.125, "learning_rate": 9.5e-06, "loss": 10.7225, "mean_token_accuracy": 0.00012004801537841558, "num_tokens": 41351.0, "step": 20 }, { "entropy": 10.742047691345215, "epoch": 0.0016347348460079775, "grad_norm": 4.46875, "learning_rate": 1.2e-05, "loss": 10.636, "mean_token_accuracy": 0.0034249894786626102, "num_tokens": 51137.0, "step": 25 }, { "entropy": 10.741790771484375, "epoch": 0.001961681815209573, "grad_norm": 3.453125, "learning_rate": 1.4500000000000002e-05, "loss": 10.5702, "mean_token_accuracy": 0.02547865901142359, "num_tokens": 61184.0, "step": 30 }, { "entropy": 10.740883350372314, "epoch": 0.0022886287844111684, "grad_norm": 3.25, "learning_rate": 1.7000000000000003e-05, "loss": 10.4596, "mean_token_accuracy": 0.04042313378304243, "num_tokens": 71499.0, "step": 35 }, { "entropy": 10.737741374969483, "epoch": 0.002615575753612764, "grad_norm": 2.578125, "learning_rate": 1.95e-05, "loss": 10.3621, "mean_token_accuracy": 0.040610473789274695, "num_tokens": 81761.0, "step": 40 }, { "entropy": 10.730296325683593, "epoch": 0.0029425227228143595, "grad_norm": 2.03125, "learning_rate": 2.2e-05, "loss": 10.3001, "mean_token_accuracy": 0.04410006906837225, "num_tokens": 93149.0, "step": 45 }, { "entropy": 10.71926326751709, "epoch": 0.003269469692015955, "grad_norm": 1.953125, "learning_rate": 2.4500000000000003e-05, "loss": 10.2165, "mean_token_accuracy": 0.04155384935438633, "num_tokens": 103201.0, "step": 50 }, { "entropy": 10.708568000793457, "epoch": 0.0035964166612175506, "grad_norm": 1.859375, "learning_rate": 2.7e-05, "loss": 10.1726, "mean_token_accuracy": 0.04020474925637245, "num_tokens": 114340.0, "step": 55 }, { "entropy": 10.701301193237304, "epoch": 0.003923363630419146, "grad_norm": 1.796875, "learning_rate": 2.95e-05, "loss": 10.1116, "mean_token_accuracy": 0.04064096435904503, "num_tokens": 125783.0, "step": 60 }, { "entropy": 10.697566413879395, "epoch": 0.004250310599620741, "grad_norm": 1.6796875, "learning_rate": 3.2e-05, "loss": 10.009, "mean_token_accuracy": 0.045371340587735176, "num_tokens": 135351.0, "step": 65 }, { "entropy": 10.687885189056397, "epoch": 0.004577257568822337, "grad_norm": 1.5234375, "learning_rate": 3.4500000000000005e-05, "loss": 10.0007, "mean_token_accuracy": 0.04735755994915962, "num_tokens": 146942.0, "step": 70 }, { "entropy": 10.67340784072876, "epoch": 0.004904204538023932, "grad_norm": 1.53125, "learning_rate": 3.7e-05, "loss": 9.9564, "mean_token_accuracy": 0.04602020643651485, "num_tokens": 157676.0, "step": 75 }, { "entropy": 10.66284465789795, "epoch": 0.005231151507225528, "grad_norm": 1.625, "learning_rate": 3.95e-05, "loss": 9.8556, "mean_token_accuracy": 0.04378781914710998, "num_tokens": 168608.0, "step": 80 }, { "entropy": 10.652789783477782, "epoch": 0.0055580984764271235, "grad_norm": 1.59375, "learning_rate": 4.2000000000000004e-05, "loss": 9.7935, "mean_token_accuracy": 0.04510065279901028, "num_tokens": 179268.0, "step": 85 }, { "entropy": 10.641815185546875, "epoch": 0.005885045445628719, "grad_norm": 1.3671875, "learning_rate": 4.45e-05, "loss": 9.7758, "mean_token_accuracy": 0.04471180438995361, "num_tokens": 189279.0, "step": 90 }, { "entropy": 10.630170726776123, "epoch": 0.006211992414830315, "grad_norm": 1.4375, "learning_rate": 4.7000000000000004e-05, "loss": 9.6597, "mean_token_accuracy": 0.051299701631069186, "num_tokens": 198791.0, "step": 95 }, { "entropy": 10.610001182556152, "epoch": 0.00653893938403191, "grad_norm": 1.59375, "learning_rate": 4.9500000000000004e-05, "loss": 9.6081, "mean_token_accuracy": 0.05303466804325581, "num_tokens": 208262.0, "step": 100 }, { "entropy": 10.579906845092774, "epoch": 0.006865886353233506, "grad_norm": 1.3828125, "learning_rate": 5.2e-05, "loss": 9.5352, "mean_token_accuracy": 0.056127872318029404, "num_tokens": 219359.0, "step": 105 }, { "entropy": 10.532975387573241, "epoch": 0.007192833322435101, "grad_norm": 1.3515625, "learning_rate": 5.45e-05, "loss": 9.3977, "mean_token_accuracy": 0.058200102671980856, "num_tokens": 229582.0, "step": 110 }, { "entropy": 10.477785778045654, "epoch": 0.007519780291636697, "grad_norm": 1.421875, "learning_rate": 5.7e-05, "loss": 9.3974, "mean_token_accuracy": 0.05185054987668991, "num_tokens": 239856.0, "step": 115 }, { "entropy": 10.430697154998779, "epoch": 0.007846727260838291, "grad_norm": 1.25, "learning_rate": 5.9499999999999996e-05, "loss": 9.2645, "mean_token_accuracy": 0.05604934468865395, "num_tokens": 249620.0, "step": 120 }, { "entropy": 10.360366725921631, "epoch": 0.008173674230039887, "grad_norm": 1.296875, "learning_rate": 6.2e-05, "loss": 9.2143, "mean_token_accuracy": 0.056453173980116844, "num_tokens": 260081.0, "step": 125 }, { "entropy": 10.311341571807862, "epoch": 0.008500621199241483, "grad_norm": 1.25, "learning_rate": 6.450000000000001e-05, "loss": 9.0411, "mean_token_accuracy": 0.0628997091203928, "num_tokens": 269889.0, "step": 130 }, { "entropy": 10.249684810638428, "epoch": 0.008827568168443078, "grad_norm": 1.1640625, "learning_rate": 6.7e-05, "loss": 9.0288, "mean_token_accuracy": 0.05760575719177723, "num_tokens": 279884.0, "step": 135 }, { "entropy": 10.16486234664917, "epoch": 0.009154515137644674, "grad_norm": 1.0859375, "learning_rate": 6.950000000000001e-05, "loss": 8.9747, "mean_token_accuracy": 0.05335375927388668, "num_tokens": 290532.0, "step": 140 }, { "entropy": 10.097373962402344, "epoch": 0.00948146210684627, "grad_norm": 1.09375, "learning_rate": 7.2e-05, "loss": 8.822, "mean_token_accuracy": 0.0569435752928257, "num_tokens": 300384.0, "step": 145 }, { "entropy": 9.977852058410644, "epoch": 0.009808409076047865, "grad_norm": 1.0390625, "learning_rate": 7.45e-05, "loss": 8.682, "mean_token_accuracy": 0.059730267897248265, "num_tokens": 310486.0, "step": 150 }, { "entropy": 9.833982563018798, "epoch": 0.01013535604524946, "grad_norm": 0.9296875, "learning_rate": 7.7e-05, "loss": 8.6837, "mean_token_accuracy": 0.05625330694019794, "num_tokens": 320405.0, "step": 155 }, { "entropy": 9.774449253082276, "epoch": 0.010462303014451056, "grad_norm": 0.8671875, "learning_rate": 7.950000000000001e-05, "loss": 8.6079, "mean_token_accuracy": 0.05718303620815277, "num_tokens": 331041.0, "step": 160 }, { "entropy": 9.63415584564209, "epoch": 0.010789249983652651, "grad_norm": 0.90234375, "learning_rate": 8.2e-05, "loss": 8.5797, "mean_token_accuracy": 0.05304349213838577, "num_tokens": 341014.0, "step": 165 }, { "entropy": 9.469839668273925, "epoch": 0.011116196952854247, "grad_norm": 0.90234375, "learning_rate": 8.450000000000001e-05, "loss": 8.5053, "mean_token_accuracy": 0.056027017533779144, "num_tokens": 352840.0, "step": 170 }, { "entropy": 9.372684097290039, "epoch": 0.011443143922055843, "grad_norm": 0.796875, "learning_rate": 8.7e-05, "loss": 8.4457, "mean_token_accuracy": 0.05765199661254883, "num_tokens": 362829.0, "step": 175 }, { "entropy": 9.20329122543335, "epoch": 0.011770090891257438, "grad_norm": 0.91796875, "learning_rate": 8.95e-05, "loss": 8.3589, "mean_token_accuracy": 0.05721031129360199, "num_tokens": 373330.0, "step": 180 }, { "entropy": 9.161731624603272, "epoch": 0.012097037860459034, "grad_norm": 0.87890625, "learning_rate": 9.2e-05, "loss": 8.3475, "mean_token_accuracy": 0.0628576211631298, "num_tokens": 383488.0, "step": 185 }, { "entropy": 9.003976440429687, "epoch": 0.01242398482966063, "grad_norm": 0.67578125, "learning_rate": 9.45e-05, "loss": 8.3301, "mean_token_accuracy": 0.06189408227801323, "num_tokens": 393936.0, "step": 190 }, { "entropy": 8.969250679016113, "epoch": 0.012750931798862225, "grad_norm": 0.91796875, "learning_rate": 9.7e-05, "loss": 8.4052, "mean_token_accuracy": 0.05606413669884205, "num_tokens": 404013.0, "step": 195 }, { "entropy": 8.889835643768311, "epoch": 0.01307787876806382, "grad_norm": 0.69921875, "learning_rate": 9.95e-05, "loss": 8.306, "mean_token_accuracy": 0.06066743396222592, "num_tokens": 414009.0, "step": 200 }, { "entropy": 8.863146209716797, "epoch": 0.013404825737265416, "grad_norm": 0.734375, "learning_rate": 0.000102, "loss": 8.3384, "mean_token_accuracy": 0.061011272668838504, "num_tokens": 424011.0, "step": 205 }, { "entropy": 8.82211093902588, "epoch": 0.013731772706467011, "grad_norm": 0.859375, "learning_rate": 0.00010449999999999999, "loss": 8.2994, "mean_token_accuracy": 0.05930745899677277, "num_tokens": 434985.0, "step": 210 }, { "entropy": 8.761214542388917, "epoch": 0.014058719675668607, "grad_norm": 0.87109375, "learning_rate": 0.000107, "loss": 8.3062, "mean_token_accuracy": 0.05960893705487251, "num_tokens": 446276.0, "step": 215 }, { "entropy": 8.744998741149903, "epoch": 0.014385666644870202, "grad_norm": 1.0859375, "learning_rate": 0.0001095, "loss": 8.2043, "mean_token_accuracy": 0.0643354371190071, "num_tokens": 455307.0, "step": 220 }, { "entropy": 8.685678005218506, "epoch": 0.014712613614071798, "grad_norm": 0.953125, "learning_rate": 0.000112, "loss": 8.292, "mean_token_accuracy": 0.06164254136383533, "num_tokens": 465253.0, "step": 225 }, { "entropy": 8.697282218933106, "epoch": 0.015039560583273394, "grad_norm": 0.83203125, "learning_rate": 0.0001145, "loss": 8.2881, "mean_token_accuracy": 0.058608485758304595, "num_tokens": 475588.0, "step": 230 }, { "entropy": 8.673004341125488, "epoch": 0.015366507552474989, "grad_norm": 0.93359375, "learning_rate": 0.00011700000000000001, "loss": 8.1716, "mean_token_accuracy": 0.06607786752283573, "num_tokens": 485356.0, "step": 235 }, { "entropy": 8.679743576049805, "epoch": 0.015693454521676583, "grad_norm": 0.90234375, "learning_rate": 0.00011949999999999999, "loss": 8.2042, "mean_token_accuracy": 0.06503303684294223, "num_tokens": 494798.0, "step": 240 }, { "entropy": 8.688579082489014, "epoch": 0.01602040149087818, "grad_norm": 0.78125, "learning_rate": 0.000122, "loss": 8.2481, "mean_token_accuracy": 0.066758993268013, "num_tokens": 504907.0, "step": 245 }, { "entropy": 8.61370620727539, "epoch": 0.016347348460079774, "grad_norm": 0.875, "learning_rate": 0.0001245, "loss": 8.27, "mean_token_accuracy": 0.06372609883546829, "num_tokens": 515219.0, "step": 250 }, { "entropy": 8.680877590179444, "epoch": 0.01667429542928137, "grad_norm": 0.953125, "learning_rate": 0.000127, "loss": 8.2037, "mean_token_accuracy": 0.06568075716495514, "num_tokens": 524467.0, "step": 255 }, { "entropy": 8.654908657073975, "epoch": 0.017001242398482965, "grad_norm": 0.9140625, "learning_rate": 0.0001295, "loss": 8.222, "mean_token_accuracy": 0.06575659550726413, "num_tokens": 533573.0, "step": 260 }, { "entropy": 8.531476593017578, "epoch": 0.01732818936768456, "grad_norm": 1.6171875, "learning_rate": 0.000132, "loss": 8.2469, "mean_token_accuracy": 0.06724704653024674, "num_tokens": 543373.0, "step": 265 }, { "entropy": 8.580636024475098, "epoch": 0.017655136336886156, "grad_norm": 0.88671875, "learning_rate": 0.00013450000000000002, "loss": 8.1985, "mean_token_accuracy": 0.06407700963318348, "num_tokens": 554398.0, "step": 270 }, { "entropy": 8.66346378326416, "epoch": 0.017982083306087752, "grad_norm": 0.9140625, "learning_rate": 0.00013700000000000002, "loss": 8.1736, "mean_token_accuracy": 0.06911301650106907, "num_tokens": 564362.0, "step": 275 }, { "entropy": 8.573775577545167, "epoch": 0.018309030275289347, "grad_norm": 0.796875, "learning_rate": 0.0001395, "loss": 8.2462, "mean_token_accuracy": 0.06476497501134873, "num_tokens": 575658.0, "step": 280 }, { "entropy": 8.612453079223632, "epoch": 0.018635977244490943, "grad_norm": 0.984375, "learning_rate": 0.00014199999999999998, "loss": 8.2019, "mean_token_accuracy": 0.0694070715457201, "num_tokens": 585744.0, "step": 285 }, { "entropy": 8.532073974609375, "epoch": 0.01896292421369254, "grad_norm": 0.8125, "learning_rate": 0.0001445, "loss": 8.1013, "mean_token_accuracy": 0.06488311924040317, "num_tokens": 595435.0, "step": 290 }, { "entropy": 8.593374729156494, "epoch": 0.019289871182894134, "grad_norm": 0.94140625, "learning_rate": 0.000147, "loss": 8.1323, "mean_token_accuracy": 0.06929427571594715, "num_tokens": 605374.0, "step": 295 }, { "entropy": 8.474758338928222, "epoch": 0.01961681815209573, "grad_norm": 0.95703125, "learning_rate": 0.0001495, "loss": 8.2146, "mean_token_accuracy": 0.06380412466824055, "num_tokens": 616418.0, "step": 300 }, { "entropy": 8.502184677124024, "epoch": 0.019943765121297325, "grad_norm": 1.1015625, "learning_rate": 0.000152, "loss": 8.1067, "mean_token_accuracy": 0.07269482649862766, "num_tokens": 625857.0, "step": 305 }, { "entropy": 8.525908470153809, "epoch": 0.02027071209049892, "grad_norm": 0.81640625, "learning_rate": 0.00015450000000000001, "loss": 8.1735, "mean_token_accuracy": 0.07054477296769619, "num_tokens": 636469.0, "step": 310 }, { "entropy": 8.518884563446045, "epoch": 0.020597659059700516, "grad_norm": 0.890625, "learning_rate": 0.000157, "loss": 8.1442, "mean_token_accuracy": 0.06903811804950237, "num_tokens": 648105.0, "step": 315 }, { "entropy": 8.525949001312256, "epoch": 0.02092460602890211, "grad_norm": 0.8984375, "learning_rate": 0.0001595, "loss": 8.0683, "mean_token_accuracy": 0.07465084679424763, "num_tokens": 657214.0, "step": 320 }, { "entropy": 8.480726528167725, "epoch": 0.021251552998103707, "grad_norm": 0.88671875, "learning_rate": 0.000162, "loss": 8.1398, "mean_token_accuracy": 0.06769934073090553, "num_tokens": 667228.0, "step": 325 }, { "entropy": 8.504579067230225, "epoch": 0.021578499967305303, "grad_norm": 0.90234375, "learning_rate": 0.00016450000000000001, "loss": 8.0862, "mean_token_accuracy": 0.07170318402349948, "num_tokens": 677412.0, "step": 330 }, { "entropy": 8.48094425201416, "epoch": 0.0219054469365069, "grad_norm": 0.91796875, "learning_rate": 0.00016700000000000002, "loss": 8.0583, "mean_token_accuracy": 0.07172369435429574, "num_tokens": 687893.0, "step": 335 }, { "entropy": 8.471054744720458, "epoch": 0.022232393905708494, "grad_norm": 0.91015625, "learning_rate": 0.00016950000000000003, "loss": 8.1798, "mean_token_accuracy": 0.06923119835555554, "num_tokens": 698497.0, "step": 340 }, { "entropy": 8.496150588989257, "epoch": 0.02255934087491009, "grad_norm": 1.03125, "learning_rate": 0.00017199999999999998, "loss": 8.063, "mean_token_accuracy": 0.07196387574076653, "num_tokens": 708593.0, "step": 345 }, { "entropy": 8.432998275756836, "epoch": 0.022886287844111685, "grad_norm": 0.93359375, "learning_rate": 0.00017449999999999999, "loss": 8.0788, "mean_token_accuracy": 0.07477995418012143, "num_tokens": 719781.0, "step": 350 }, { "entropy": 8.490119457244873, "epoch": 0.02321323481331328, "grad_norm": 0.875, "learning_rate": 0.000177, "loss": 8.088, "mean_token_accuracy": 0.0711702361702919, "num_tokens": 729946.0, "step": 355 }, { "entropy": 8.486604785919189, "epoch": 0.023540181782514876, "grad_norm": 0.87890625, "learning_rate": 0.0001795, "loss": 8.0756, "mean_token_accuracy": 0.07159654907882214, "num_tokens": 740542.0, "step": 360 }, { "entropy": 8.415905952453613, "epoch": 0.02386712875171647, "grad_norm": 1.3203125, "learning_rate": 0.000182, "loss": 7.9728, "mean_token_accuracy": 0.08065467663109302, "num_tokens": 750387.0, "step": 365 }, { "entropy": 8.41865062713623, "epoch": 0.024194075720918067, "grad_norm": 0.9375, "learning_rate": 0.0001845, "loss": 8.0256, "mean_token_accuracy": 0.07667291313409805, "num_tokens": 759708.0, "step": 370 }, { "entropy": 8.375955581665039, "epoch": 0.024521022690119663, "grad_norm": 1.0078125, "learning_rate": 0.000187, "loss": 7.9862, "mean_token_accuracy": 0.07728637531399726, "num_tokens": 770655.0, "step": 375 }, { "entropy": 8.447672843933105, "epoch": 0.02484796965932126, "grad_norm": 1.1640625, "learning_rate": 0.0001895, "loss": 8.0944, "mean_token_accuracy": 0.07428529076278209, "num_tokens": 780263.0, "step": 380 }, { "entropy": 8.440955066680909, "epoch": 0.025174916628522854, "grad_norm": 0.859375, "learning_rate": 0.000192, "loss": 7.9377, "mean_token_accuracy": 0.08051865920424461, "num_tokens": 789667.0, "step": 385 }, { "entropy": 8.339350986480714, "epoch": 0.02550186359772445, "grad_norm": 0.96484375, "learning_rate": 0.0001945, "loss": 7.9736, "mean_token_accuracy": 0.07487879581749439, "num_tokens": 800099.0, "step": 390 }, { "entropy": 8.333793830871581, "epoch": 0.025828810566926045, "grad_norm": 1.1875, "learning_rate": 0.00019700000000000002, "loss": 7.9658, "mean_token_accuracy": 0.08434236496686935, "num_tokens": 809347.0, "step": 395 }, { "entropy": 8.389479732513427, "epoch": 0.02615575753612764, "grad_norm": 1.0234375, "learning_rate": 0.00019950000000000002, "loss": 7.9885, "mean_token_accuracy": 0.07460427135229111, "num_tokens": 819721.0, "step": 400 }, { "entropy": 8.356211471557618, "epoch": 0.026482704505329236, "grad_norm": 1.1328125, "learning_rate": 0.000202, "loss": 7.92, "mean_token_accuracy": 0.07878225892782212, "num_tokens": 830278.0, "step": 405 }, { "entropy": 8.31589298248291, "epoch": 0.02680965147453083, "grad_norm": 1.09375, "learning_rate": 0.00020449999999999998, "loss": 7.9775, "mean_token_accuracy": 0.07939173206686974, "num_tokens": 839760.0, "step": 410 }, { "entropy": 8.388703918457031, "epoch": 0.027136598443732427, "grad_norm": 0.859375, "learning_rate": 0.000207, "loss": 7.9903, "mean_token_accuracy": 0.07688882425427437, "num_tokens": 850778.0, "step": 415 }, { "entropy": 8.303010845184327, "epoch": 0.027463545412934023, "grad_norm": 1.125, "learning_rate": 0.0002095, "loss": 7.9098, "mean_token_accuracy": 0.080412358045578, "num_tokens": 860419.0, "step": 420 }, { "entropy": 8.438417148590087, "epoch": 0.027790492382135618, "grad_norm": 1.0546875, "learning_rate": 0.000212, "loss": 8.0202, "mean_token_accuracy": 0.08035937286913394, "num_tokens": 869939.0, "step": 425 }, { "entropy": 8.34555139541626, "epoch": 0.028117439351337214, "grad_norm": 0.9375, "learning_rate": 0.0002145, "loss": 7.9552, "mean_token_accuracy": 0.07891882248222828, "num_tokens": 880331.0, "step": 430 }, { "entropy": 8.413178730010987, "epoch": 0.02844438632053881, "grad_norm": 1.15625, "learning_rate": 0.00021700000000000002, "loss": 8.032, "mean_token_accuracy": 0.07255076095461846, "num_tokens": 891342.0, "step": 435 }, { "entropy": 8.292833995819091, "epoch": 0.028771333289740405, "grad_norm": 0.984375, "learning_rate": 0.0002195, "loss": 7.9823, "mean_token_accuracy": 0.07935103215277195, "num_tokens": 902253.0, "step": 440 }, { "entropy": 8.271970176696778, "epoch": 0.029098280258942, "grad_norm": 1.015625, "learning_rate": 0.000222, "loss": 7.9096, "mean_token_accuracy": 0.07985292673110962, "num_tokens": 913086.0, "step": 445 }, { "entropy": 8.302376174926758, "epoch": 0.029425227228143596, "grad_norm": 1.0234375, "learning_rate": 0.0002245, "loss": 7.9366, "mean_token_accuracy": 0.08083747625350952, "num_tokens": 925082.0, "step": 450 }, { "entropy": 8.271292591094971, "epoch": 0.02975217419734519, "grad_norm": 1.015625, "learning_rate": 0.00022700000000000002, "loss": 7.9089, "mean_token_accuracy": 0.07454345673322678, "num_tokens": 933955.0, "step": 455 }, { "entropy": 8.287126731872558, "epoch": 0.030079121166546787, "grad_norm": 0.94921875, "learning_rate": 0.00022950000000000002, "loss": 7.8699, "mean_token_accuracy": 0.07939782850444317, "num_tokens": 943951.0, "step": 460 }, { "entropy": 8.33047170639038, "epoch": 0.030406068135748383, "grad_norm": 0.9296875, "learning_rate": 0.00023200000000000003, "loss": 7.8629, "mean_token_accuracy": 0.081587253510952, "num_tokens": 955297.0, "step": 465 }, { "entropy": 8.157269191741943, "epoch": 0.030733015104949978, "grad_norm": 0.88671875, "learning_rate": 0.00023449999999999998, "loss": 7.9056, "mean_token_accuracy": 0.08038199990987778, "num_tokens": 967012.0, "step": 470 }, { "entropy": 8.305221843719483, "epoch": 0.031059962074151574, "grad_norm": 0.90234375, "learning_rate": 0.000237, "loss": 7.8177, "mean_token_accuracy": 0.08144455999135972, "num_tokens": 977956.0, "step": 475 }, { "entropy": 8.184438991546632, "epoch": 0.031386909043353166, "grad_norm": 1.046875, "learning_rate": 0.0002395, "loss": 7.8361, "mean_token_accuracy": 0.08141265660524369, "num_tokens": 987270.0, "step": 480 }, { "entropy": 8.198512172698974, "epoch": 0.03171385601255476, "grad_norm": 0.9453125, "learning_rate": 0.000242, "loss": 7.824, "mean_token_accuracy": 0.08384960517287254, "num_tokens": 998010.0, "step": 485 }, { "entropy": 8.246695518493652, "epoch": 0.03204080298175636, "grad_norm": 1.09375, "learning_rate": 0.0002445, "loss": 7.86, "mean_token_accuracy": 0.09095575362443924, "num_tokens": 1008173.0, "step": 490 }, { "entropy": 8.233383846282958, "epoch": 0.03236774995095795, "grad_norm": 1.0703125, "learning_rate": 0.000247, "loss": 7.8713, "mean_token_accuracy": 0.08157122023403644, "num_tokens": 1019037.0, "step": 495 }, { "entropy": 8.247837829589844, "epoch": 0.03269469692015955, "grad_norm": 1.140625, "learning_rate": 0.0002495, "loss": 7.8742, "mean_token_accuracy": 0.08232737332582474, "num_tokens": 1028822.0, "step": 500 }, { "entropy": 8.173187160491944, "epoch": 0.033021643889361144, "grad_norm": 1.109375, "learning_rate": 0.000252, "loss": 7.7428, "mean_token_accuracy": 0.08822857066988946, "num_tokens": 1038749.0, "step": 505 }, { "entropy": 8.270493030548096, "epoch": 0.03334859085856274, "grad_norm": 0.921875, "learning_rate": 0.0002545, "loss": 7.8827, "mean_token_accuracy": 0.07765771076083183, "num_tokens": 1049087.0, "step": 510 }, { "entropy": 8.135348224639893, "epoch": 0.033675537827764335, "grad_norm": 0.92578125, "learning_rate": 0.000257, "loss": 7.7678, "mean_token_accuracy": 0.08465715497732162, "num_tokens": 1060711.0, "step": 515 }, { "entropy": 8.170577144622802, "epoch": 0.03400248479696593, "grad_norm": 1.0859375, "learning_rate": 0.0002595, "loss": 7.855, "mean_token_accuracy": 0.08124198243021966, "num_tokens": 1070643.0, "step": 520 }, { "entropy": 8.271326732635497, "epoch": 0.034329431766167526, "grad_norm": 1.1015625, "learning_rate": 0.000262, "loss": 7.8145, "mean_token_accuracy": 0.08234022073447704, "num_tokens": 1080970.0, "step": 525 }, { "entropy": 8.16833953857422, "epoch": 0.03465637873536912, "grad_norm": 1.0546875, "learning_rate": 0.00026450000000000003, "loss": 7.7443, "mean_token_accuracy": 0.08425415754318237, "num_tokens": 1090833.0, "step": 530 }, { "entropy": 8.179855918884277, "epoch": 0.03498332570457072, "grad_norm": 1.0703125, "learning_rate": 0.00026700000000000004, "loss": 7.8173, "mean_token_accuracy": 0.08293091841042041, "num_tokens": 1100160.0, "step": 535 }, { "entropy": 8.109969568252563, "epoch": 0.03531027267377231, "grad_norm": 1.0625, "learning_rate": 0.00026950000000000005, "loss": 7.7834, "mean_token_accuracy": 0.08286757096648216, "num_tokens": 1110846.0, "step": 540 }, { "entropy": 8.109409475326538, "epoch": 0.03563721964297391, "grad_norm": 0.98828125, "learning_rate": 0.00027200000000000005, "loss": 7.685, "mean_token_accuracy": 0.0879479818046093, "num_tokens": 1119529.0, "step": 545 }, { "entropy": 8.154296016693115, "epoch": 0.035964166612175504, "grad_norm": 0.9765625, "learning_rate": 0.0002745, "loss": 7.757, "mean_token_accuracy": 0.0852863110601902, "num_tokens": 1129798.0, "step": 550 }, { "entropy": 8.109227848052978, "epoch": 0.0362911135813771, "grad_norm": 0.9140625, "learning_rate": 0.000277, "loss": 7.7885, "mean_token_accuracy": 0.08763574361801148, "num_tokens": 1141689.0, "step": 555 }, { "entropy": 8.144587326049805, "epoch": 0.036618060550578695, "grad_norm": 1.046875, "learning_rate": 0.0002795, "loss": 7.7427, "mean_token_accuracy": 0.08682122454047203, "num_tokens": 1152522.0, "step": 560 }, { "entropy": 8.103516864776612, "epoch": 0.03694500751978029, "grad_norm": 1.046875, "learning_rate": 0.00028199999999999997, "loss": 7.717, "mean_token_accuracy": 0.08920524194836617, "num_tokens": 1163135.0, "step": 565 }, { "entropy": 8.070560359954834, "epoch": 0.037271954488981886, "grad_norm": 1.0390625, "learning_rate": 0.0002845, "loss": 7.6697, "mean_token_accuracy": 0.09047451019287109, "num_tokens": 1172866.0, "step": 570 }, { "entropy": 8.063013648986816, "epoch": 0.03759890145818348, "grad_norm": 1.0546875, "learning_rate": 0.000287, "loss": 7.745, "mean_token_accuracy": 0.08657962679862977, "num_tokens": 1182924.0, "step": 575 }, { "entropy": 8.056052541732788, "epoch": 0.03792584842738508, "grad_norm": 0.93359375, "learning_rate": 0.0002895, "loss": 7.7485, "mean_token_accuracy": 0.0842316836118698, "num_tokens": 1194036.0, "step": 580 }, { "entropy": 8.069906187057494, "epoch": 0.03825279539658667, "grad_norm": 1.1328125, "learning_rate": 0.000292, "loss": 7.6946, "mean_token_accuracy": 0.0934490866959095, "num_tokens": 1205162.0, "step": 585 }, { "entropy": 8.092452239990234, "epoch": 0.03857974236578827, "grad_norm": 1.078125, "learning_rate": 0.0002945, "loss": 7.7078, "mean_token_accuracy": 0.08852062448859215, "num_tokens": 1214953.0, "step": 590 }, { "entropy": 8.008755350112915, "epoch": 0.038906689334989863, "grad_norm": 0.96484375, "learning_rate": 0.000297, "loss": 7.7271, "mean_token_accuracy": 0.08698445111513138, "num_tokens": 1225745.0, "step": 595 }, { "entropy": 8.161905765533447, "epoch": 0.03923363630419146, "grad_norm": 0.92578125, "learning_rate": 0.0002995, "loss": 7.7125, "mean_token_accuracy": 0.08865412771701812, "num_tokens": 1236076.0, "step": 600 }, { "entropy": 7.991028499603272, "epoch": 0.039560583273393055, "grad_norm": 1.0234375, "learning_rate": 0.000302, "loss": 7.7014, "mean_token_accuracy": 0.08888744562864304, "num_tokens": 1245843.0, "step": 605 }, { "entropy": 8.065514373779298, "epoch": 0.03988753024259465, "grad_norm": 1.109375, "learning_rate": 0.0003045, "loss": 7.714, "mean_token_accuracy": 0.08649160899221897, "num_tokens": 1255723.0, "step": 610 }, { "entropy": 8.174874591827393, "epoch": 0.040214477211796246, "grad_norm": 1.0546875, "learning_rate": 0.000307, "loss": 7.6046, "mean_token_accuracy": 0.09897364154458047, "num_tokens": 1265289.0, "step": 615 }, { "entropy": 7.849688863754272, "epoch": 0.04054142418099784, "grad_norm": 1.1953125, "learning_rate": 0.0003095, "loss": 7.7132, "mean_token_accuracy": 0.09138823449611663, "num_tokens": 1275495.0, "step": 620 }, { "entropy": 8.07530312538147, "epoch": 0.04086837115019944, "grad_norm": 1.25, "learning_rate": 0.000312, "loss": 7.711, "mean_token_accuracy": 0.08681774735450745, "num_tokens": 1286473.0, "step": 625 }, { "entropy": 8.080075311660767, "epoch": 0.04119531811940103, "grad_norm": 1.015625, "learning_rate": 0.0003145, "loss": 7.6718, "mean_token_accuracy": 0.0849516399204731, "num_tokens": 1296460.0, "step": 630 }, { "entropy": 8.104103088378906, "epoch": 0.04152226508860263, "grad_norm": 1.078125, "learning_rate": 0.000317, "loss": 7.6993, "mean_token_accuracy": 0.08389882594347, "num_tokens": 1306497.0, "step": 635 }, { "entropy": 7.975687408447266, "epoch": 0.04184921205780422, "grad_norm": 0.984375, "learning_rate": 0.0003195, "loss": 7.6485, "mean_token_accuracy": 0.09450172930955887, "num_tokens": 1318420.0, "step": 640 }, { "entropy": 7.9988954067230225, "epoch": 0.04217615902700582, "grad_norm": 1.1640625, "learning_rate": 0.000322, "loss": 7.6429, "mean_token_accuracy": 0.08192968890070915, "num_tokens": 1327612.0, "step": 645 }, { "entropy": 7.94122314453125, "epoch": 0.042503105996207415, "grad_norm": 1.0234375, "learning_rate": 0.00032450000000000003, "loss": 7.5709, "mean_token_accuracy": 0.09098407700657844, "num_tokens": 1337450.0, "step": 650 }, { "entropy": 7.95971908569336, "epoch": 0.04283005296540901, "grad_norm": 1.1015625, "learning_rate": 0.00032700000000000003, "loss": 7.6652, "mean_token_accuracy": 0.08624772727489471, "num_tokens": 1347953.0, "step": 655 }, { "entropy": 7.912782192230225, "epoch": 0.043156999934610606, "grad_norm": 1.0625, "learning_rate": 0.00032950000000000004, "loss": 7.5909, "mean_token_accuracy": 0.08983312919735909, "num_tokens": 1358009.0, "step": 660 }, { "entropy": 8.027462434768676, "epoch": 0.0434839469038122, "grad_norm": 0.9140625, "learning_rate": 0.00033200000000000005, "loss": 7.6471, "mean_token_accuracy": 0.08675070479512215, "num_tokens": 1367725.0, "step": 665 }, { "entropy": 7.954972267150879, "epoch": 0.0438108938730138, "grad_norm": 0.94140625, "learning_rate": 0.00033450000000000005, "loss": 7.7181, "mean_token_accuracy": 0.0875462144613266, "num_tokens": 1379334.0, "step": 670 }, { "entropy": 7.962738800048828, "epoch": 0.04413784084221539, "grad_norm": 1.1875, "learning_rate": 0.000337, "loss": 7.6282, "mean_token_accuracy": 0.09448311552405357, "num_tokens": 1389991.0, "step": 675 }, { "entropy": 7.954906797409057, "epoch": 0.04446478781141699, "grad_norm": 1.09375, "learning_rate": 0.0003395, "loss": 7.5957, "mean_token_accuracy": 0.08683520555496216, "num_tokens": 1400797.0, "step": 680 }, { "entropy": 7.937600803375244, "epoch": 0.04479173478061858, "grad_norm": 1.140625, "learning_rate": 0.000342, "loss": 7.6082, "mean_token_accuracy": 0.09097633212804794, "num_tokens": 1411805.0, "step": 685 }, { "entropy": 7.955043363571167, "epoch": 0.04511868174982018, "grad_norm": 1.0703125, "learning_rate": 0.00034449999999999997, "loss": 7.5461, "mean_token_accuracy": 0.09095973521471024, "num_tokens": 1421539.0, "step": 690 }, { "entropy": 7.900257730484009, "epoch": 0.045445628719021774, "grad_norm": 1.125, "learning_rate": 0.000347, "loss": 7.5068, "mean_token_accuracy": 0.09854376316070557, "num_tokens": 1432136.0, "step": 695 }, { "entropy": 7.864266300201416, "epoch": 0.04577257568822337, "grad_norm": 1.0, "learning_rate": 0.0003495, "loss": 7.6082, "mean_token_accuracy": 0.09226389080286027, "num_tokens": 1443471.0, "step": 700 }, { "entropy": 7.924344682693482, "epoch": 0.046099522657424966, "grad_norm": 0.98046875, "learning_rate": 0.000352, "loss": 7.5336, "mean_token_accuracy": 0.09339390024542808, "num_tokens": 1453503.0, "step": 705 }, { "entropy": 7.8929485321044925, "epoch": 0.04642646962662656, "grad_norm": 1.015625, "learning_rate": 0.0003545, "loss": 7.5331, "mean_token_accuracy": 0.0965921014547348, "num_tokens": 1463254.0, "step": 710 }, { "entropy": 8.004297828674316, "epoch": 0.04675341659582816, "grad_norm": 0.8984375, "learning_rate": 0.000357, "loss": 7.6325, "mean_token_accuracy": 0.0870884120464325, "num_tokens": 1472890.0, "step": 715 }, { "entropy": 7.844469308853149, "epoch": 0.04708036356502975, "grad_norm": 1.09375, "learning_rate": 0.0003595, "loss": 7.4847, "mean_token_accuracy": 0.10488657429814338, "num_tokens": 1483583.0, "step": 720 }, { "entropy": 7.928372240066528, "epoch": 0.04740731053423135, "grad_norm": 1.0625, "learning_rate": 0.000362, "loss": 7.6397, "mean_token_accuracy": 0.09026602432131767, "num_tokens": 1494278.0, "step": 725 }, { "entropy": 7.923847436904907, "epoch": 0.04773425750343294, "grad_norm": 1.1171875, "learning_rate": 0.0003645, "loss": 7.5448, "mean_token_accuracy": 0.09652285501360894, "num_tokens": 1504638.0, "step": 730 }, { "entropy": 7.885090017318726, "epoch": 0.04806120447263454, "grad_norm": 1.0703125, "learning_rate": 0.000367, "loss": 7.5522, "mean_token_accuracy": 0.092280051112175, "num_tokens": 1514581.0, "step": 735 }, { "entropy": 7.8557947158813475, "epoch": 0.048388151441836134, "grad_norm": 1.0546875, "learning_rate": 0.0003695, "loss": 7.5964, "mean_token_accuracy": 0.09087880104780197, "num_tokens": 1525455.0, "step": 740 }, { "entropy": 7.926615905761719, "epoch": 0.04871509841103773, "grad_norm": 0.9921875, "learning_rate": 0.000372, "loss": 7.5333, "mean_token_accuracy": 0.09311627596616745, "num_tokens": 1537267.0, "step": 745 }, { "entropy": 7.779848432540893, "epoch": 0.049042045380239326, "grad_norm": 1.1640625, "learning_rate": 0.0003745, "loss": 7.4671, "mean_token_accuracy": 0.09461563900113105, "num_tokens": 1547181.0, "step": 750 }, { "entropy": 7.773740196228028, "epoch": 0.04936899234944092, "grad_norm": 1.015625, "learning_rate": 0.000377, "loss": 7.3915, "mean_token_accuracy": 0.09637575298547744, "num_tokens": 1557267.0, "step": 755 }, { "entropy": 7.78641791343689, "epoch": 0.04969593931864252, "grad_norm": 1.140625, "learning_rate": 0.0003795, "loss": 7.4663, "mean_token_accuracy": 0.09401652291417122, "num_tokens": 1567762.0, "step": 760 }, { "entropy": 7.91934494972229, "epoch": 0.05002288628784411, "grad_norm": 0.90234375, "learning_rate": 0.000382, "loss": 7.5892, "mean_token_accuracy": 0.08972916305065155, "num_tokens": 1579312.0, "step": 765 }, { "entropy": 7.819589614868164, "epoch": 0.05034983325704571, "grad_norm": 1.0546875, "learning_rate": 0.0003845, "loss": 7.4421, "mean_token_accuracy": 0.10521875619888306, "num_tokens": 1588752.0, "step": 770 }, { "entropy": 7.8574573516845705, "epoch": 0.0506767802262473, "grad_norm": 1.109375, "learning_rate": 0.00038700000000000003, "loss": 7.4474, "mean_token_accuracy": 0.09444977417588234, "num_tokens": 1599510.0, "step": 775 }, { "entropy": 7.748177242279053, "epoch": 0.0510037271954489, "grad_norm": 1.03125, "learning_rate": 0.00038950000000000003, "loss": 7.4229, "mean_token_accuracy": 0.09738618209958076, "num_tokens": 1609651.0, "step": 780 }, { "entropy": 7.896618747711182, "epoch": 0.051330674164650494, "grad_norm": 1.078125, "learning_rate": 0.00039200000000000004, "loss": 7.5672, "mean_token_accuracy": 0.08716193325817585, "num_tokens": 1619840.0, "step": 785 }, { "entropy": 7.841101741790771, "epoch": 0.05165762113385209, "grad_norm": 1.140625, "learning_rate": 0.00039450000000000005, "loss": 7.5263, "mean_token_accuracy": 0.09152752608060837, "num_tokens": 1630725.0, "step": 790 }, { "entropy": 7.733171272277832, "epoch": 0.051984568103053685, "grad_norm": 1.1484375, "learning_rate": 0.00039700000000000005, "loss": 7.3961, "mean_token_accuracy": 0.09614320546388626, "num_tokens": 1640535.0, "step": 795 }, { "entropy": 7.817097949981689, "epoch": 0.05231151507225528, "grad_norm": 1.0234375, "learning_rate": 0.0003995, "loss": 7.4086, "mean_token_accuracy": 0.09492226243019104, "num_tokens": 1650714.0, "step": 800 }, { "entropy": 7.741940069198608, "epoch": 0.05263846204145688, "grad_norm": 0.9296875, "learning_rate": 0.000402, "loss": 7.4748, "mean_token_accuracy": 0.09248731583356858, "num_tokens": 1661879.0, "step": 805 }, { "entropy": 7.823268270492553, "epoch": 0.05296540901065847, "grad_norm": 1.0546875, "learning_rate": 0.0004045, "loss": 7.403, "mean_token_accuracy": 0.09551335647702217, "num_tokens": 1672343.0, "step": 810 }, { "entropy": 7.712739515304565, "epoch": 0.05329235597986007, "grad_norm": 1.109375, "learning_rate": 0.00040699999999999997, "loss": 7.4106, "mean_token_accuracy": 0.09740137308835983, "num_tokens": 1682197.0, "step": 815 }, { "entropy": 7.708367252349854, "epoch": 0.05361930294906166, "grad_norm": 1.0, "learning_rate": 0.0004095, "loss": 7.4002, "mean_token_accuracy": 0.09502007588744163, "num_tokens": 1692182.0, "step": 820 }, { "entropy": 7.748778533935547, "epoch": 0.05394624991826326, "grad_norm": 1.1484375, "learning_rate": 0.000412, "loss": 7.4254, "mean_token_accuracy": 0.09064045622944832, "num_tokens": 1702646.0, "step": 825 }, { "entropy": 7.66537618637085, "epoch": 0.054273196887464854, "grad_norm": 1.0703125, "learning_rate": 0.0004145, "loss": 7.4296, "mean_token_accuracy": 0.0905129812657833, "num_tokens": 1712950.0, "step": 830 }, { "entropy": 7.710060453414917, "epoch": 0.05460014385666645, "grad_norm": 1.015625, "learning_rate": 0.000417, "loss": 7.3317, "mean_token_accuracy": 0.10179824382066727, "num_tokens": 1722799.0, "step": 835 }, { "entropy": 7.776296472549438, "epoch": 0.054927090825868045, "grad_norm": 1.0625, "learning_rate": 0.0004195, "loss": 7.4313, "mean_token_accuracy": 0.09205255433917045, "num_tokens": 1732687.0, "step": 840 }, { "entropy": 7.721476030349732, "epoch": 0.05525403779506964, "grad_norm": 1.015625, "learning_rate": 0.000422, "loss": 7.4183, "mean_token_accuracy": 0.09819221273064613, "num_tokens": 1743709.0, "step": 845 }, { "entropy": 7.701347494125367, "epoch": 0.055580984764271237, "grad_norm": 0.98046875, "learning_rate": 0.0004245, "loss": 7.3733, "mean_token_accuracy": 0.10262311398983001, "num_tokens": 1753129.0, "step": 850 }, { "entropy": 7.690869569778442, "epoch": 0.05590793173347283, "grad_norm": 1.09375, "learning_rate": 0.000427, "loss": 7.3682, "mean_token_accuracy": 0.09535444900393486, "num_tokens": 1764072.0, "step": 855 }, { "entropy": 7.708118629455567, "epoch": 0.05623487870267443, "grad_norm": 1.125, "learning_rate": 0.0004295, "loss": 7.3638, "mean_token_accuracy": 0.09914470463991165, "num_tokens": 1774015.0, "step": 860 }, { "entropy": 7.719418954849243, "epoch": 0.05656182567187602, "grad_norm": 1.109375, "learning_rate": 0.000432, "loss": 7.4128, "mean_token_accuracy": 0.0922047957777977, "num_tokens": 1784257.0, "step": 865 }, { "entropy": 7.6339034080505375, "epoch": 0.05688877264107762, "grad_norm": 1.0390625, "learning_rate": 0.0004345, "loss": 7.3383, "mean_token_accuracy": 0.09566345661878586, "num_tokens": 1793294.0, "step": 870 }, { "entropy": 7.739969205856323, "epoch": 0.057215719610279214, "grad_norm": 1.09375, "learning_rate": 0.000437, "loss": 7.3986, "mean_token_accuracy": 0.09636269062757492, "num_tokens": 1804034.0, "step": 875 }, { "entropy": 7.659945917129517, "epoch": 0.05754266657948081, "grad_norm": 1.3671875, "learning_rate": 0.0004395, "loss": 7.3202, "mean_token_accuracy": 0.10390383377671242, "num_tokens": 1812571.0, "step": 880 }, { "entropy": 7.643369054794311, "epoch": 0.057869613548682405, "grad_norm": 1.0078125, "learning_rate": 0.000442, "loss": 7.3716, "mean_token_accuracy": 0.09669788479804993, "num_tokens": 1822824.0, "step": 885 }, { "entropy": 7.761786460876465, "epoch": 0.058196560517884, "grad_norm": 1.09375, "learning_rate": 0.0004445, "loss": 7.4194, "mean_token_accuracy": 0.09798763766884803, "num_tokens": 1833237.0, "step": 890 }, { "entropy": 7.669866991043091, "epoch": 0.058523507487085596, "grad_norm": 1.15625, "learning_rate": 0.000447, "loss": 7.3896, "mean_token_accuracy": 0.09653202295303345, "num_tokens": 1843920.0, "step": 895 }, { "entropy": 7.630056381225586, "epoch": 0.05885045445628719, "grad_norm": 1.0390625, "learning_rate": 0.00044950000000000003, "loss": 7.411, "mean_token_accuracy": 0.09456825032830238, "num_tokens": 1854599.0, "step": 900 }, { "entropy": 7.694906568527221, "epoch": 0.05917740142548879, "grad_norm": 1.078125, "learning_rate": 0.00045200000000000004, "loss": 7.3294, "mean_token_accuracy": 0.0995859757065773, "num_tokens": 1863988.0, "step": 905 }, { "entropy": 7.561672687530518, "epoch": 0.05950434839469038, "grad_norm": 1.0703125, "learning_rate": 0.00045450000000000004, "loss": 7.3582, "mean_token_accuracy": 0.09936628043651581, "num_tokens": 1874249.0, "step": 910 }, { "entropy": 7.710276746749878, "epoch": 0.05983129536389198, "grad_norm": 1.078125, "learning_rate": 0.00045700000000000005, "loss": 7.3139, "mean_token_accuracy": 0.10104074403643608, "num_tokens": 1884221.0, "step": 915 }, { "entropy": 7.558821821212769, "epoch": 0.060158242333093574, "grad_norm": 0.96875, "learning_rate": 0.00045950000000000006, "loss": 7.3128, "mean_token_accuracy": 0.0991006389260292, "num_tokens": 1894287.0, "step": 920 }, { "entropy": 7.588020515441895, "epoch": 0.06048518930229517, "grad_norm": 1.15625, "learning_rate": 0.000462, "loss": 7.2791, "mean_token_accuracy": 0.10233154892921448, "num_tokens": 1903474.0, "step": 925 }, { "entropy": 7.6131902694702145, "epoch": 0.060812136271496765, "grad_norm": 1.1171875, "learning_rate": 0.0004645, "loss": 7.2966, "mean_token_accuracy": 0.10143398270010948, "num_tokens": 1912237.0, "step": 930 }, { "entropy": 7.64272894859314, "epoch": 0.06113908324069836, "grad_norm": 0.9296875, "learning_rate": 0.000467, "loss": 7.3396, "mean_token_accuracy": 0.09589844942092896, "num_tokens": 1922501.0, "step": 935 }, { "entropy": 7.628575706481934, "epoch": 0.061466030209899956, "grad_norm": 1.3046875, "learning_rate": 0.0004695, "loss": 7.3593, "mean_token_accuracy": 0.09694062098860741, "num_tokens": 1932910.0, "step": 940 }, { "entropy": 7.626384496688843, "epoch": 0.06179297717910155, "grad_norm": 0.89453125, "learning_rate": 0.000472, "loss": 7.4415, "mean_token_accuracy": 0.09250794872641563, "num_tokens": 1944408.0, "step": 945 }, { "entropy": 7.614904356002808, "epoch": 0.06211992414830315, "grad_norm": 1.046875, "learning_rate": 0.0004745, "loss": 7.34, "mean_token_accuracy": 0.09710650071501732, "num_tokens": 1955149.0, "step": 950 }, { "entropy": 7.6185708999633786, "epoch": 0.06244687111750474, "grad_norm": 1.1328125, "learning_rate": 0.000477, "loss": 7.2988, "mean_token_accuracy": 0.10158349275588989, "num_tokens": 1964777.0, "step": 955 }, { "entropy": 7.558726787567139, "epoch": 0.06277381808670633, "grad_norm": 1.015625, "learning_rate": 0.0004795, "loss": 7.27, "mean_token_accuracy": 0.10379163399338723, "num_tokens": 1974781.0, "step": 960 }, { "entropy": 7.597839450836181, "epoch": 0.06310076505590793, "grad_norm": 1.015625, "learning_rate": 0.000482, "loss": 7.2502, "mean_token_accuracy": 0.10211163312196732, "num_tokens": 1984540.0, "step": 965 }, { "entropy": 7.517283391952515, "epoch": 0.06342771202510952, "grad_norm": 1.015625, "learning_rate": 0.0004845, "loss": 7.2998, "mean_token_accuracy": 0.09965328201651573, "num_tokens": 1995360.0, "step": 970 }, { "entropy": 7.5842241764068605, "epoch": 0.06375465899431113, "grad_norm": 1.0, "learning_rate": 0.000487, "loss": 7.2705, "mean_token_accuracy": 0.09944166019558906, "num_tokens": 2006299.0, "step": 975 }, { "entropy": 7.646463441848755, "epoch": 0.06408160596351271, "grad_norm": 1.0546875, "learning_rate": 0.0004895, "loss": 7.323, "mean_token_accuracy": 0.10111710727214814, "num_tokens": 2017482.0, "step": 980 }, { "entropy": 7.496323204040527, "epoch": 0.06440855293271432, "grad_norm": 1.0546875, "learning_rate": 0.000492, "loss": 7.3474, "mean_token_accuracy": 0.09670697450637818, "num_tokens": 2027669.0, "step": 985 }, { "entropy": 7.659954690933228, "epoch": 0.0647354999019159, "grad_norm": 1.0078125, "learning_rate": 0.0004945, "loss": 7.2417, "mean_token_accuracy": 0.10287103429436684, "num_tokens": 2037047.0, "step": 990 }, { "entropy": 7.5259521961212155, "epoch": 0.06506244687111751, "grad_norm": 1.1171875, "learning_rate": 0.000497, "loss": 7.2875, "mean_token_accuracy": 0.09639681577682495, "num_tokens": 2046813.0, "step": 995 }, { "entropy": 7.487096881866455, "epoch": 0.0653893938403191, "grad_norm": 1.0859375, "learning_rate": 0.0004995, "loss": 7.2426, "mean_token_accuracy": 0.10415883064270019, "num_tokens": 2056669.0, "step": 1000 }, { "entropy": 7.525420379638672, "epoch": 0.0657163408095207, "grad_norm": 1.0546875, "learning_rate": 0.000499998026082006, "loss": 7.2404, "mean_token_accuracy": 0.10298535376787185, "num_tokens": 2066786.0, "step": 1005 }, { "entropy": 7.535466623306275, "epoch": 0.06604328777872229, "grad_norm": 1.2421875, "learning_rate": 0.0004999900070995136, "loss": 7.2927, "mean_token_accuracy": 0.1004390373826027, "num_tokens": 2077248.0, "step": 1010 }, { "entropy": 7.534571361541748, "epoch": 0.06637023474792389, "grad_norm": 1.0078125, "learning_rate": 0.0004999758199023239, "loss": 7.302, "mean_token_accuracy": 0.09730483591556549, "num_tokens": 2087448.0, "step": 1015 }, { "entropy": 7.495874738693237, "epoch": 0.06669718171712548, "grad_norm": 0.98046875, "learning_rate": 0.0004999554648793858, "loss": 7.2048, "mean_token_accuracy": 0.10220806449651718, "num_tokens": 2098481.0, "step": 1020 }, { "entropy": 7.465099191665649, "epoch": 0.06702412868632708, "grad_norm": 1.0234375, "learning_rate": 0.0004999289425887425, "loss": 7.1987, "mean_token_accuracy": 0.10606111437082291, "num_tokens": 2108775.0, "step": 1025 }, { "entropy": 7.554659366607666, "epoch": 0.06735107565552867, "grad_norm": 1.171875, "learning_rate": 0.0004998962537575161, "loss": 7.2178, "mean_token_accuracy": 0.1030527487397194, "num_tokens": 2118485.0, "step": 1030 }, { "entropy": 7.381258583068847, "epoch": 0.06767802262473027, "grad_norm": 1.0703125, "learning_rate": 0.0004998573992818874, "loss": 7.1164, "mean_token_accuracy": 0.10998099818825721, "num_tokens": 2127567.0, "step": 1035 }, { "entropy": 7.3920176982879635, "epoch": 0.06800496959393186, "grad_norm": 0.9609375, "learning_rate": 0.0004998123802270715, "loss": 7.161, "mean_token_accuracy": 0.10730142295360565, "num_tokens": 2138449.0, "step": 1040 }, { "entropy": 7.573119211196899, "epoch": 0.06833191656313346, "grad_norm": 1.1171875, "learning_rate": 0.0004997611978272886, "loss": 7.2895, "mean_token_accuracy": 0.09891922920942306, "num_tokens": 2148644.0, "step": 1045 }, { "entropy": 7.440708923339844, "epoch": 0.06865886353233505, "grad_norm": 1.109375, "learning_rate": 0.0004997038534857298, "loss": 7.2655, "mean_token_accuracy": 0.10525748580694198, "num_tokens": 2158412.0, "step": 1050 }, { "entropy": 7.638814115524292, "epoch": 0.06898581050153665, "grad_norm": 0.96484375, "learning_rate": 0.0004996403487745194, "loss": 7.2549, "mean_token_accuracy": 0.0952314555644989, "num_tokens": 2167762.0, "step": 1055 }, { "entropy": 7.516493034362793, "epoch": 0.06931275747073824, "grad_norm": 0.9453125, "learning_rate": 0.000499570685434671, "loss": 7.3654, "mean_token_accuracy": 0.09759802371263504, "num_tokens": 2178848.0, "step": 1060 }, { "entropy": 7.492746496200562, "epoch": 0.06963970443993985, "grad_norm": 0.98046875, "learning_rate": 0.0004994948653760405, "loss": 7.2098, "mean_token_accuracy": 0.10332456231117249, "num_tokens": 2189422.0, "step": 1065 }, { "entropy": 7.500653219223023, "epoch": 0.06996665140914143, "grad_norm": 1.03125, "learning_rate": 0.0004994128906772729, "loss": 7.1913, "mean_token_accuracy": 0.1038860872387886, "num_tokens": 2198982.0, "step": 1070 }, { "entropy": 7.364135599136352, "epoch": 0.07029359837834304, "grad_norm": 1.078125, "learning_rate": 0.000499324763585746, "loss": 7.2397, "mean_token_accuracy": 0.10290840938687325, "num_tokens": 2208568.0, "step": 1075 }, { "entropy": 7.583868932723999, "epoch": 0.07062054534754462, "grad_norm": 0.9609375, "learning_rate": 0.0004992304865175085, "loss": 7.1989, "mean_token_accuracy": 0.10371176674962043, "num_tokens": 2218892.0, "step": 1080 }, { "entropy": 7.379058122634888, "epoch": 0.07094749231674623, "grad_norm": 1.0, "learning_rate": 0.0004991300620572138, "loss": 7.1566, "mean_token_accuracy": 0.10587683096528053, "num_tokens": 2229398.0, "step": 1085 }, { "entropy": 7.411726665496826, "epoch": 0.07127443928594782, "grad_norm": 1.2578125, "learning_rate": 0.0004990234929580494, "loss": 7.2191, "mean_token_accuracy": 0.10308944508433342, "num_tokens": 2239584.0, "step": 1090 }, { "entropy": 7.556722116470337, "epoch": 0.07160138625514942, "grad_norm": 0.90625, "learning_rate": 0.0004989107821416609, "loss": 7.1501, "mean_token_accuracy": 0.1037745475769043, "num_tokens": 2250520.0, "step": 1095 }, { "entropy": 7.374285364151001, "epoch": 0.07192833322435101, "grad_norm": 0.99609375, "learning_rate": 0.0004987919326980723, "loss": 7.1537, "mean_token_accuracy": 0.10868279710412025, "num_tokens": 2260427.0, "step": 1100 }, { "entropy": 7.417365503311157, "epoch": 0.07225528019355261, "grad_norm": 0.95703125, "learning_rate": 0.0004986669478856011, "loss": 7.2346, "mean_token_accuracy": 0.09781083241105079, "num_tokens": 2270578.0, "step": 1105 }, { "entropy": 7.451626873016357, "epoch": 0.0725822271627542, "grad_norm": 1.0859375, "learning_rate": 0.0004985358311307688, "loss": 7.1794, "mean_token_accuracy": 0.10275761559605598, "num_tokens": 2280884.0, "step": 1110 }, { "entropy": 7.398854780197143, "epoch": 0.0729091741319558, "grad_norm": 1.0234375, "learning_rate": 0.0004983985860282081, "loss": 7.1681, "mean_token_accuracy": 0.10323931500315667, "num_tokens": 2290267.0, "step": 1115 }, { "entropy": 7.45128321647644, "epoch": 0.07323612110115739, "grad_norm": 1.125, "learning_rate": 0.0004982552163405623, "loss": 7.2119, "mean_token_accuracy": 0.09956866279244422, "num_tokens": 2300847.0, "step": 1120 }, { "entropy": 7.446847820281983, "epoch": 0.07356306807035899, "grad_norm": 1.0390625, "learning_rate": 0.0004981057259983839, "loss": 7.1101, "mean_token_accuracy": 0.10652703419327736, "num_tokens": 2311129.0, "step": 1125 }, { "entropy": 7.407537746429443, "epoch": 0.07389001503956058, "grad_norm": 0.9765625, "learning_rate": 0.0004979501191000262, "loss": 7.2093, "mean_token_accuracy": 0.10527276918292046, "num_tokens": 2322101.0, "step": 1130 }, { "entropy": 7.368101644515991, "epoch": 0.07421696200876218, "grad_norm": 0.99609375, "learning_rate": 0.0004977883999115311, "loss": 7.0926, "mean_token_accuracy": 0.10808824151754379, "num_tokens": 2332639.0, "step": 1135 }, { "entropy": 7.376673889160156, "epoch": 0.07454390897796377, "grad_norm": 0.91796875, "learning_rate": 0.0004976205728665113, "loss": 7.2022, "mean_token_accuracy": 0.10557337999343872, "num_tokens": 2343508.0, "step": 1140 }, { "entropy": 7.451087093353271, "epoch": 0.07487085594716537, "grad_norm": 0.90234375, "learning_rate": 0.0004974466425660307, "loss": 7.1134, "mean_token_accuracy": 0.10346391946077346, "num_tokens": 2353384.0, "step": 1145 }, { "entropy": 7.298187637329102, "epoch": 0.07519780291636696, "grad_norm": 1.0625, "learning_rate": 0.0004972666137784759, "loss": 7.096, "mean_token_accuracy": 0.11667231395840645, "num_tokens": 2363462.0, "step": 1150 }, { "entropy": 7.351812505722046, "epoch": 0.07552474988556857, "grad_norm": 0.93359375, "learning_rate": 0.0004970804914394271, "loss": 7.0795, "mean_token_accuracy": 0.10770620107650757, "num_tokens": 2373474.0, "step": 1155 }, { "entropy": 7.337015533447266, "epoch": 0.07585169685477015, "grad_norm": 1.1640625, "learning_rate": 0.0004968882806515225, "loss": 7.1254, "mean_token_accuracy": 0.10191202014684678, "num_tokens": 2385199.0, "step": 1160 }, { "entropy": 7.382783222198486, "epoch": 0.07617864382397176, "grad_norm": 1.078125, "learning_rate": 0.0004966899866843177, "loss": 7.1694, "mean_token_accuracy": 0.10132893994450569, "num_tokens": 2395798.0, "step": 1165 }, { "entropy": 7.443328046798706, "epoch": 0.07650559079317334, "grad_norm": 1.1328125, "learning_rate": 0.000496485614974142, "loss": 7.173, "mean_token_accuracy": 0.1050618901848793, "num_tokens": 2405536.0, "step": 1170 }, { "entropy": 7.3729442119598385, "epoch": 0.07683253776237495, "grad_norm": 1.0859375, "learning_rate": 0.0004962751711239492, "loss": 7.1101, "mean_token_accuracy": 0.11114847138524056, "num_tokens": 2415326.0, "step": 1175 }, { "entropy": 7.311748790740967, "epoch": 0.07715948473157654, "grad_norm": 1.1015625, "learning_rate": 0.0004960586609031636, "loss": 7.1065, "mean_token_accuracy": 0.10614153444766998, "num_tokens": 2426177.0, "step": 1180 }, { "entropy": 7.344895076751709, "epoch": 0.07748643170077814, "grad_norm": 1.046875, "learning_rate": 0.0004958360902475224, "loss": 7.1002, "mean_token_accuracy": 0.11174782365560532, "num_tokens": 2436308.0, "step": 1185 }, { "entropy": 7.365533924102783, "epoch": 0.07781337866997973, "grad_norm": 1.0234375, "learning_rate": 0.0004956074652589125, "loss": 7.1125, "mean_token_accuracy": 0.10806370452046395, "num_tokens": 2446866.0, "step": 1190 }, { "entropy": 7.348533725738525, "epoch": 0.07814032563918133, "grad_norm": 0.9453125, "learning_rate": 0.0004953727922052035, "loss": 7.1028, "mean_token_accuracy": 0.10454007759690284, "num_tokens": 2457999.0, "step": 1195 }, { "entropy": 7.344534254074096, "epoch": 0.07846727260838292, "grad_norm": 1.0859375, "learning_rate": 0.0004951320775200756, "loss": 7.0736, "mean_token_accuracy": 0.10834787487983703, "num_tokens": 2468333.0, "step": 1200 }, { "entropy": 7.31992015838623, "epoch": 0.07879421957758452, "grad_norm": 1.125, "learning_rate": 0.0004948853278028436, "loss": 7.2052, "mean_token_accuracy": 0.10361588671803475, "num_tokens": 2478288.0, "step": 1205 }, { "entropy": 7.396023082733154, "epoch": 0.07912116654678611, "grad_norm": 0.9921875, "learning_rate": 0.0004946325498182755, "loss": 7.097, "mean_token_accuracy": 0.11261972859501838, "num_tokens": 2487845.0, "step": 1210 }, { "entropy": 7.318902063369751, "epoch": 0.07944811351598771, "grad_norm": 1.0546875, "learning_rate": 0.0004943737504964076, "loss": 7.0849, "mean_token_accuracy": 0.10493758395314216, "num_tokens": 2498419.0, "step": 1215 }, { "entropy": 7.292612457275391, "epoch": 0.0797750604851893, "grad_norm": 0.96875, "learning_rate": 0.000494108936932354, "loss": 7.1059, "mean_token_accuracy": 0.10675550177693367, "num_tokens": 2509124.0, "step": 1220 }, { "entropy": 7.438989686965942, "epoch": 0.0801020074543909, "grad_norm": 1.109375, "learning_rate": 0.0004938381163861124, "loss": 7.0458, "mean_token_accuracy": 0.1076106570661068, "num_tokens": 2519255.0, "step": 1225 }, { "entropy": 7.25689377784729, "epoch": 0.08042895442359249, "grad_norm": 0.953125, "learning_rate": 0.0004935612962823645, "loss": 7.2158, "mean_token_accuracy": 0.09847642555832863, "num_tokens": 2530576.0, "step": 1230 }, { "entropy": 7.3991516590118405, "epoch": 0.0807559013927941, "grad_norm": 1.03125, "learning_rate": 0.0004932784842102739, "loss": 7.0837, "mean_token_accuracy": 0.1105932466685772, "num_tokens": 2542632.0, "step": 1235 }, { "entropy": 7.316380882263184, "epoch": 0.08108284836199568, "grad_norm": 1.15625, "learning_rate": 0.0004929896879232758, "loss": 7.0521, "mean_token_accuracy": 0.11338271498680115, "num_tokens": 2552960.0, "step": 1240 }, { "entropy": 7.354037237167359, "epoch": 0.08140979533119728, "grad_norm": 1.125, "learning_rate": 0.0004926949153388668, "loss": 7.0537, "mean_token_accuracy": 0.11079110056161881, "num_tokens": 2563214.0, "step": 1245 }, { "entropy": 7.210246610641479, "epoch": 0.08173674230039887, "grad_norm": 0.9453125, "learning_rate": 0.0004923941745383859, "loss": 7.046, "mean_token_accuracy": 0.10845032036304474, "num_tokens": 2573712.0, "step": 1250 }, { "entropy": 7.2730066776275635, "epoch": 0.08206368926960048, "grad_norm": 1.0625, "learning_rate": 0.000492087473766794, "loss": 6.9945, "mean_token_accuracy": 0.11024141758680343, "num_tokens": 2583024.0, "step": 1255 }, { "entropy": 7.286072540283203, "epoch": 0.08239063623880206, "grad_norm": 1.1484375, "learning_rate": 0.000491774821432448, "loss": 7.0445, "mean_token_accuracy": 0.11110128611326217, "num_tokens": 2591816.0, "step": 1260 }, { "entropy": 7.352578067779541, "epoch": 0.08271758320800367, "grad_norm": 0.92578125, "learning_rate": 0.0004914562261068693, "loss": 7.0586, "mean_token_accuracy": 0.11311088278889656, "num_tokens": 2602895.0, "step": 1265 }, { "entropy": 7.333470106124878, "epoch": 0.08304453017720526, "grad_norm": 1.015625, "learning_rate": 0.0004911316965245098, "loss": 7.0935, "mean_token_accuracy": 0.10813801214098931, "num_tokens": 2613537.0, "step": 1270 }, { "entropy": 7.226766681671142, "epoch": 0.08337147714640686, "grad_norm": 0.91796875, "learning_rate": 0.000490801241582512, "loss": 7.0326, "mean_token_accuracy": 0.11485574468970298, "num_tokens": 2623282.0, "step": 1275 }, { "entropy": 7.2596677303314205, "epoch": 0.08369842411560845, "grad_norm": 0.9609375, "learning_rate": 0.000490464870340465, "loss": 6.949, "mean_token_accuracy": 0.11277191489934921, "num_tokens": 2634237.0, "step": 1280 }, { "entropy": 7.214026975631714, "epoch": 0.08402537108481005, "grad_norm": 1.03125, "learning_rate": 0.0004901225920201563, "loss": 7.0452, "mean_token_accuracy": 0.1102260746061802, "num_tokens": 2643427.0, "step": 1285 }, { "entropy": 7.3286186218261715, "epoch": 0.08435231805401164, "grad_norm": 0.8984375, "learning_rate": 0.000489774416005319, "loss": 7.1127, "mean_token_accuracy": 0.10841076076030731, "num_tokens": 2653876.0, "step": 1290 }, { "entropy": 7.323719882965088, "epoch": 0.08467926502321324, "grad_norm": 1.015625, "learning_rate": 0.0004894203518413742, "loss": 7.0249, "mean_token_accuracy": 0.11151918098330497, "num_tokens": 2663265.0, "step": 1295 }, { "entropy": 7.252338171005249, "epoch": 0.08500621199241483, "grad_norm": 1.0546875, "learning_rate": 0.0004890604092351701, "loss": 6.9932, "mean_token_accuracy": 0.11705144569277763, "num_tokens": 2673289.0, "step": 1300 }, { "entropy": 7.213364362716675, "epoch": 0.08533315896161643, "grad_norm": 1.03125, "learning_rate": 0.000488694598054715, "loss": 6.9296, "mean_token_accuracy": 0.11891363188624382, "num_tokens": 2681843.0, "step": 1305 }, { "entropy": 7.257659196853638, "epoch": 0.08566010593081802, "grad_norm": 0.9609375, "learning_rate": 0.0004883229283289071, "loss": 7.054, "mean_token_accuracy": 0.10866474732756615, "num_tokens": 2692509.0, "step": 1310 }, { "entropy": 7.339709758758545, "epoch": 0.08598705290001962, "grad_norm": 0.9453125, "learning_rate": 0.00048794541024725993, "loss": 7.0904, "mean_token_accuracy": 0.11010322123765945, "num_tokens": 2703847.0, "step": 1315 }, { "entropy": 7.231632041931152, "epoch": 0.08631399986922121, "grad_norm": 0.953125, "learning_rate": 0.0004875620541596221, "loss": 6.9653, "mean_token_accuracy": 0.11453013718128205, "num_tokens": 2714173.0, "step": 1320 }, { "entropy": 7.22537088394165, "epoch": 0.08664094683842281, "grad_norm": 1.046875, "learning_rate": 0.00048717287057589454, "loss": 6.9957, "mean_token_accuracy": 0.11089121177792549, "num_tokens": 2724315.0, "step": 1325 }, { "entropy": 7.243461084365845, "epoch": 0.0869678938076244, "grad_norm": 0.94921875, "learning_rate": 0.0004867778701657417, "loss": 7.0472, "mean_token_accuracy": 0.10567675605416298, "num_tokens": 2734738.0, "step": 1330 }, { "entropy": 7.266382265090942, "epoch": 0.087294840776826, "grad_norm": 1.140625, "learning_rate": 0.00048637706375829955, "loss": 6.9696, "mean_token_accuracy": 0.11352119892835617, "num_tokens": 2744565.0, "step": 1335 }, { "entropy": 7.109191370010376, "epoch": 0.0876217877460276, "grad_norm": 1.0390625, "learning_rate": 0.000485970462341878, "loss": 7.0196, "mean_token_accuracy": 0.1136182777583599, "num_tokens": 2754474.0, "step": 1340 }, { "entropy": 7.3075309753417965, "epoch": 0.0879487347152292, "grad_norm": 1.109375, "learning_rate": 0.00048555807706366044, "loss": 6.8976, "mean_token_accuracy": 0.1176247052848339, "num_tokens": 2763350.0, "step": 1345 }, { "entropy": 7.16860146522522, "epoch": 0.08827568168443078, "grad_norm": 1.140625, "learning_rate": 0.00048513991922939756, "loss": 6.9439, "mean_token_accuracy": 0.1173409454524517, "num_tokens": 2774148.0, "step": 1350 }, { "entropy": 7.156577491760254, "epoch": 0.08860262865363239, "grad_norm": 0.9765625, "learning_rate": 0.00048471600030309744, "loss": 6.9948, "mean_token_accuracy": 0.11294468194246292, "num_tokens": 2784151.0, "step": 1355 }, { "entropy": 7.252249240875244, "epoch": 0.08892957562283398, "grad_norm": 1.0, "learning_rate": 0.00048428633190671186, "loss": 6.9962, "mean_token_accuracy": 0.10971863046288491, "num_tokens": 2796049.0, "step": 1360 }, { "entropy": 7.211243677139282, "epoch": 0.08925652259203558, "grad_norm": 1.0546875, "learning_rate": 0.0004838509258198167, "loss": 7.014, "mean_token_accuracy": 0.10978813543915748, "num_tokens": 2805739.0, "step": 1365 }, { "entropy": 7.197784328460694, "epoch": 0.08958346956123717, "grad_norm": 1.046875, "learning_rate": 0.00048340979397929, "loss": 6.9708, "mean_token_accuracy": 0.10899091139435768, "num_tokens": 2815889.0, "step": 1370 }, { "entropy": 7.310032606124878, "epoch": 0.08991041653043877, "grad_norm": 1.4375, "learning_rate": 0.00048296294847898386, "loss": 7.1525, "mean_token_accuracy": 0.10318199545145035, "num_tokens": 2828732.0, "step": 1375 }, { "entropy": 7.255441331863404, "epoch": 0.09023736349964036, "grad_norm": 1.5625, "learning_rate": 0.0004825104015693934, "loss": 7.0177, "mean_token_accuracy": 0.10567167475819587, "num_tokens": 2839229.0, "step": 1380 }, { "entropy": 7.114026641845703, "epoch": 0.09056431046884196, "grad_norm": 1.09375, "learning_rate": 0.0004820521656573208, "loss": 6.9538, "mean_token_accuracy": 0.109841088950634, "num_tokens": 2849466.0, "step": 1385 }, { "entropy": 7.316811037063599, "epoch": 0.09089125743804355, "grad_norm": 1.0234375, "learning_rate": 0.00048158825330553505, "loss": 7.0207, "mean_token_accuracy": 0.10393416360020638, "num_tokens": 2859228.0, "step": 1390 }, { "entropy": 7.131655216217041, "epoch": 0.09121820440724515, "grad_norm": 1.1640625, "learning_rate": 0.00048111867723242763, "loss": 6.9467, "mean_token_accuracy": 0.11996429413557053, "num_tokens": 2868648.0, "step": 1395 }, { "entropy": 7.201031827926636, "epoch": 0.09154515137644674, "grad_norm": 0.98828125, "learning_rate": 0.0004806434503116637, "loss": 6.9255, "mean_token_accuracy": 0.11738296672701835, "num_tokens": 2879010.0, "step": 1400 }, { "entropy": 7.099094676971435, "epoch": 0.09187209834564834, "grad_norm": 1.0078125, "learning_rate": 0.0004801625855718296, "loss": 6.9033, "mean_token_accuracy": 0.1191631592810154, "num_tokens": 2890088.0, "step": 1405 }, { "entropy": 7.2180914878845215, "epoch": 0.09219904531484993, "grad_norm": 0.92578125, "learning_rate": 0.00047967609619607477, "loss": 7.0336, "mean_token_accuracy": 0.10593287870287896, "num_tokens": 2900798.0, "step": 1410 }, { "entropy": 7.279546785354614, "epoch": 0.09252599228405153, "grad_norm": 1.03125, "learning_rate": 0.0004791839955217513, "loss": 6.9909, "mean_token_accuracy": 0.1094129055738449, "num_tokens": 2912236.0, "step": 1415 }, { "entropy": 7.122265863418579, "epoch": 0.09285293925325312, "grad_norm": 1.0234375, "learning_rate": 0.00047868629704004786, "loss": 6.9169, "mean_token_accuracy": 0.11205016151070595, "num_tokens": 2922199.0, "step": 1420 }, { "entropy": 7.257745742797852, "epoch": 0.09317988622245472, "grad_norm": 0.953125, "learning_rate": 0.00047818301439561965, "loss": 6.9356, "mean_token_accuracy": 0.11461735665798187, "num_tokens": 2932379.0, "step": 1425 }, { "entropy": 7.132466983795166, "epoch": 0.09350683319165631, "grad_norm": 0.96875, "learning_rate": 0.00047767416138621454, "loss": 6.9356, "mean_token_accuracy": 0.10910105928778649, "num_tokens": 2942046.0, "step": 1430 }, { "entropy": 7.330983066558838, "epoch": 0.0938337801608579, "grad_norm": 1.0390625, "learning_rate": 0.000477159751962295, "loss": 6.9985, "mean_token_accuracy": 0.10300733298063278, "num_tokens": 2952468.0, "step": 1435 }, { "entropy": 7.150661563873291, "epoch": 0.0941607271300595, "grad_norm": 1.078125, "learning_rate": 0.00047663980022665507, "loss": 6.9525, "mean_token_accuracy": 0.11100440993905067, "num_tokens": 2962510.0, "step": 1440 }, { "entropy": 7.17984414100647, "epoch": 0.0944876740992611, "grad_norm": 0.90625, "learning_rate": 0.00047611432043403437, "loss": 6.9557, "mean_token_accuracy": 0.11558201462030411, "num_tokens": 2972725.0, "step": 1445 }, { "entropy": 7.108048772811889, "epoch": 0.0948146210684627, "grad_norm": 1.125, "learning_rate": 0.0004755833269907267, "loss": 6.9104, "mean_token_accuracy": 0.11575181782245636, "num_tokens": 2982761.0, "step": 1450 }, { "entropy": 7.144851493835449, "epoch": 0.09514156803766428, "grad_norm": 1.03125, "learning_rate": 0.0004750468344541857, "loss": 6.7885, "mean_token_accuracy": 0.11536038592457772, "num_tokens": 2992175.0, "step": 1455 }, { "entropy": 7.101435041427612, "epoch": 0.09546851500686589, "grad_norm": 0.97265625, "learning_rate": 0.00047450485753262525, "loss": 6.8631, "mean_token_accuracy": 0.12131781205534935, "num_tokens": 3002593.0, "step": 1460 }, { "entropy": 7.14970703125, "epoch": 0.09579546197606748, "grad_norm": 0.94921875, "learning_rate": 0.00047395741108461633, "loss": 6.8795, "mean_token_accuracy": 0.12142410352826119, "num_tokens": 3013251.0, "step": 1465 }, { "entropy": 7.122796583175659, "epoch": 0.09612240894526908, "grad_norm": 1.0078125, "learning_rate": 0.00047340451011867985, "loss": 6.8728, "mean_token_accuracy": 0.11778575032949448, "num_tokens": 3023290.0, "step": 1470 }, { "entropy": 7.1415787696838375, "epoch": 0.09644935591447067, "grad_norm": 1.0390625, "learning_rate": 0.00047284616979287515, "loss": 6.8188, "mean_token_accuracy": 0.12416595816612244, "num_tokens": 3032377.0, "step": 1475 }, { "entropy": 7.1211080074310305, "epoch": 0.09677630288367227, "grad_norm": 1.015625, "learning_rate": 0.00047228240541438433, "loss": 6.9079, "mean_token_accuracy": 0.11739139705896377, "num_tokens": 3043273.0, "step": 1480 }, { "entropy": 7.1460123538970945, "epoch": 0.09710324985287386, "grad_norm": 0.92578125, "learning_rate": 0.00047171323243909257, "loss": 6.9052, "mean_token_accuracy": 0.11405683010816574, "num_tokens": 3053696.0, "step": 1485 }, { "entropy": 7.023873901367187, "epoch": 0.09743019682207546, "grad_norm": 1.0078125, "learning_rate": 0.00047113866647116457, "loss": 6.9174, "mean_token_accuracy": 0.11818796917796134, "num_tokens": 3063742.0, "step": 1490 }, { "entropy": 7.20653715133667, "epoch": 0.09775714379127705, "grad_norm": 1.0078125, "learning_rate": 0.0004705587232626164, "loss": 6.8514, "mean_token_accuracy": 0.11485515907406807, "num_tokens": 3073153.0, "step": 1495 }, { "entropy": 7.071482133865357, "epoch": 0.09808409076047865, "grad_norm": 1.0234375, "learning_rate": 0.00046997341871288424, "loss": 6.8514, "mean_token_accuracy": 0.12352146431803704, "num_tokens": 3082567.0, "step": 1500 } ], "logging_steps": 5, "max_steps": 4000, "num_input_tokens_seen": 0, "num_train_epochs": 1, "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": 719605767536640.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }