| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.01568037131119265, |
| "eval_steps": 500, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 10.651023864746094, |
| "epoch": 0.00015680371311192648, |
| "grad_norm": 11.5625, |
| "learning_rate": 2e-06, |
| "loss": 10.8028, |
| "mean_token_accuracy": 6.958942394703626e-05, |
| "num_tokens": 14082.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 10.651004600524903, |
| "epoch": 0.00031360742622385296, |
| "grad_norm": 11.375, |
| "learning_rate": 4.5e-06, |
| "loss": 10.7633, |
| "mean_token_accuracy": 6.574622238986194e-05, |
| "num_tokens": 29900.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 10.650343418121338, |
| "epoch": 0.00047041113933577947, |
| "grad_norm": 8.4375, |
| "learning_rate": 7e-06, |
| "loss": 10.5185, |
| "mean_token_accuracy": 0.01316747289383784, |
| "num_tokens": 43394.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 10.64456491470337, |
| "epoch": 0.0006272148524477059, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.5e-06, |
| "loss": 10.2603, |
| "mean_token_accuracy": 0.03233455698937178, |
| "num_tokens": 58563.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 10.616307353973388, |
| "epoch": 0.0007840185655596325, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.2e-05, |
| "loss": 9.9833, |
| "mean_token_accuracy": 0.034109539538621905, |
| "num_tokens": 73923.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 10.588720226287842, |
| "epoch": 0.0009408222786715589, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.4500000000000002e-05, |
| "loss": 9.8466, |
| "mean_token_accuracy": 0.03426761133596301, |
| "num_tokens": 87652.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 10.595056247711181, |
| "epoch": 0.0010976259917834855, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.7000000000000003e-05, |
| "loss": 9.7741, |
| "mean_token_accuracy": 0.031784738413989544, |
| "num_tokens": 101369.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 10.59932107925415, |
| "epoch": 0.0012544297048954118, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.95e-05, |
| "loss": 9.6922, |
| "mean_token_accuracy": 0.03506205677986145, |
| "num_tokens": 114428.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 10.581777667999267, |
| "epoch": 0.0014112334180073384, |
| "grad_norm": 2.3125, |
| "learning_rate": 2.2e-05, |
| "loss": 9.706, |
| "mean_token_accuracy": 0.03386611118912697, |
| "num_tokens": 130472.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 10.577643585205077, |
| "epoch": 0.001568037131119265, |
| "grad_norm": 2.40625, |
| "learning_rate": 2.4500000000000003e-05, |
| "loss": 9.6852, |
| "mean_token_accuracy": 0.030213401652872562, |
| "num_tokens": 144077.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 10.59252758026123, |
| "epoch": 0.0017248408442311913, |
| "grad_norm": 2.15625, |
| "learning_rate": 2.7e-05, |
| "loss": 9.6373, |
| "mean_token_accuracy": 0.03242910895496607, |
| "num_tokens": 158941.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 10.579074954986572, |
| "epoch": 0.0018816445573431179, |
| "grad_norm": 2.140625, |
| "learning_rate": 2.95e-05, |
| "loss": 9.563, |
| "mean_token_accuracy": 0.02959225047379732, |
| "num_tokens": 174362.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 10.570305156707764, |
| "epoch": 0.002038448270455044, |
| "grad_norm": 2.09375, |
| "learning_rate": 3.2e-05, |
| "loss": 9.4485, |
| "mean_token_accuracy": 0.033058703131973746, |
| "num_tokens": 190929.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 10.55489387512207, |
| "epoch": 0.002195251983566971, |
| "grad_norm": 2.015625, |
| "learning_rate": 3.4500000000000005e-05, |
| "loss": 9.3387, |
| "mean_token_accuracy": 0.033958965167403224, |
| "num_tokens": 203660.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 10.530017757415772, |
| "epoch": 0.0023520556966788973, |
| "grad_norm": 2.0625, |
| "learning_rate": 3.7e-05, |
| "loss": 9.2826, |
| "mean_token_accuracy": 0.027879416570067407, |
| "num_tokens": 218421.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 10.534087371826171, |
| "epoch": 0.0025088594097908237, |
| "grad_norm": 2.09375, |
| "learning_rate": 3.95e-05, |
| "loss": 9.1439, |
| "mean_token_accuracy": 0.03472804371267557, |
| "num_tokens": 233734.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 10.496679878234863, |
| "epoch": 0.0026656631229027505, |
| "grad_norm": 1.9765625, |
| "learning_rate": 4.2000000000000004e-05, |
| "loss": 9.0182, |
| "mean_token_accuracy": 0.02818337455391884, |
| "num_tokens": 247105.0, |
| "step": 85 |
| }, |
| { |
| "entropy": 10.467080688476562, |
| "epoch": 0.002822466836014677, |
| "grad_norm": 1.9140625, |
| "learning_rate": 4.45e-05, |
| "loss": 8.928, |
| "mean_token_accuracy": 0.03165087569504976, |
| "num_tokens": 262487.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 10.403542804718018, |
| "epoch": 0.002979270549126603, |
| "grad_norm": 1.9921875, |
| "learning_rate": 4.7000000000000004e-05, |
| "loss": 8.7777, |
| "mean_token_accuracy": 0.036813986487686634, |
| "num_tokens": 276705.0, |
| "step": 95 |
| }, |
| { |
| "entropy": 10.354411220550537, |
| "epoch": 0.00313607426223853, |
| "grad_norm": 1.9765625, |
| "learning_rate": 4.9500000000000004e-05, |
| "loss": 8.6775, |
| "mean_token_accuracy": 0.04005452003329992, |
| "num_tokens": 291031.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 10.269740104675293, |
| "epoch": 0.0032928779753504563, |
| "grad_norm": 1.890625, |
| "learning_rate": 5.2e-05, |
| "loss": 8.5069, |
| "mean_token_accuracy": 0.040437552519142626, |
| "num_tokens": 305444.0, |
| "step": 105 |
| }, |
| { |
| "entropy": 10.179180717468261, |
| "epoch": 0.0034496816884623826, |
| "grad_norm": 1.8671875, |
| "learning_rate": 5.45e-05, |
| "loss": 8.3811, |
| "mean_token_accuracy": 0.04589438661932945, |
| "num_tokens": 321168.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 10.070209217071532, |
| "epoch": 0.0036064854015743094, |
| "grad_norm": 1.7265625, |
| "learning_rate": 5.7e-05, |
| "loss": 8.1795, |
| "mean_token_accuracy": 0.05213871449232101, |
| "num_tokens": 334988.0, |
| "step": 115 |
| }, |
| { |
| "entropy": 9.922878646850586, |
| "epoch": 0.0037632891146862357, |
| "grad_norm": 1.6796875, |
| "learning_rate": 5.9499999999999996e-05, |
| "loss": 8.0559, |
| "mean_token_accuracy": 0.05347799621522427, |
| "num_tokens": 349049.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 9.7627742767334, |
| "epoch": 0.0039200928277981625, |
| "grad_norm": 1.765625, |
| "learning_rate": 6.2e-05, |
| "loss": 8.0302, |
| "mean_token_accuracy": 0.05416244864463806, |
| "num_tokens": 365497.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 9.585789394378661, |
| "epoch": 0.004076896540910088, |
| "grad_norm": 1.5, |
| "learning_rate": 6.450000000000001e-05, |
| "loss": 7.8073, |
| "mean_token_accuracy": 0.05755908451974392, |
| "num_tokens": 380324.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 9.368918418884277, |
| "epoch": 0.004233700254022015, |
| "grad_norm": 1.59375, |
| "learning_rate": 6.7e-05, |
| "loss": 7.6844, |
| "mean_token_accuracy": 0.06051859185099602, |
| "num_tokens": 393766.0, |
| "step": 135 |
| }, |
| { |
| "entropy": 9.158668899536133, |
| "epoch": 0.004390503967133942, |
| "grad_norm": 1.4296875, |
| "learning_rate": 6.950000000000001e-05, |
| "loss": 7.6272, |
| "mean_token_accuracy": 0.06358711272478104, |
| "num_tokens": 409369.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 8.945510005950927, |
| "epoch": 0.004547307680245868, |
| "grad_norm": 1.546875, |
| "learning_rate": 7.2e-05, |
| "loss": 7.5019, |
| "mean_token_accuracy": 0.06252376027405263, |
| "num_tokens": 425475.0, |
| "step": 145 |
| }, |
| { |
| "entropy": 8.711659908294678, |
| "epoch": 0.004704111393357795, |
| "grad_norm": 1.59375, |
| "learning_rate": 7.45e-05, |
| "loss": 7.3832, |
| "mean_token_accuracy": 0.06495344713330269, |
| "num_tokens": 440853.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 8.498833179473877, |
| "epoch": 0.0048609151064697214, |
| "grad_norm": 1.609375, |
| "learning_rate": 7.7e-05, |
| "loss": 7.3471, |
| "mean_token_accuracy": 0.06971911787986755, |
| "num_tokens": 457407.0, |
| "step": 155 |
| }, |
| { |
| "entropy": 8.355002212524415, |
| "epoch": 0.005017718819581647, |
| "grad_norm": 1.09375, |
| "learning_rate": 7.950000000000001e-05, |
| "loss": 7.2971, |
| "mean_token_accuracy": 0.06931432262063027, |
| "num_tokens": 472813.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 8.208080291748047, |
| "epoch": 0.005174522532693574, |
| "grad_norm": 1.5, |
| "learning_rate": 8.2e-05, |
| "loss": 7.3059, |
| "mean_token_accuracy": 0.06723905019462109, |
| "num_tokens": 489742.0, |
| "step": 165 |
| }, |
| { |
| "entropy": 8.143133544921875, |
| "epoch": 0.005331326245805501, |
| "grad_norm": 1.4375, |
| "learning_rate": 8.450000000000001e-05, |
| "loss": 7.2399, |
| "mean_token_accuracy": 0.06843472644686699, |
| "num_tokens": 503493.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 8.036779308319092, |
| "epoch": 0.005488129958917427, |
| "grad_norm": 1.1796875, |
| "learning_rate": 8.7e-05, |
| "loss": 7.1275, |
| "mean_token_accuracy": 0.07654446959495545, |
| "num_tokens": 520143.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 7.934807920455933, |
| "epoch": 0.005644933672029354, |
| "grad_norm": 1.3203125, |
| "learning_rate": 8.95e-05, |
| "loss": 7.232, |
| "mean_token_accuracy": 0.06283588744699956, |
| "num_tokens": 534776.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 7.953633880615234, |
| "epoch": 0.00580173738514128, |
| "grad_norm": 1.3359375, |
| "learning_rate": 9.2e-05, |
| "loss": 7.2097, |
| "mean_token_accuracy": 0.06803618557751179, |
| "num_tokens": 550256.0, |
| "step": 185 |
| }, |
| { |
| "entropy": 7.894899654388428, |
| "epoch": 0.005958541098253206, |
| "grad_norm": 1.1796875, |
| "learning_rate": 9.45e-05, |
| "loss": 7.2178, |
| "mean_token_accuracy": 0.0686919379979372, |
| "num_tokens": 564797.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 7.844494724273682, |
| "epoch": 0.006115344811365133, |
| "grad_norm": 1.1953125, |
| "learning_rate": 9.7e-05, |
| "loss": 6.9952, |
| "mean_token_accuracy": 0.0732124924659729, |
| "num_tokens": 579721.0, |
| "step": 195 |
| }, |
| { |
| "entropy": 7.716418170928955, |
| "epoch": 0.00627214852447706, |
| "grad_norm": 1.359375, |
| "learning_rate": 9.95e-05, |
| "loss": 6.9564, |
| "mean_token_accuracy": 0.07807120829820632, |
| "num_tokens": 593431.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 7.715715551376343, |
| "epoch": 0.006428952237588986, |
| "grad_norm": 1.6796875, |
| "learning_rate": 0.000102, |
| "loss": 6.9708, |
| "mean_token_accuracy": 0.07868832051753998, |
| "num_tokens": 608176.0, |
| "step": 205 |
| }, |
| { |
| "entropy": 7.692095470428467, |
| "epoch": 0.0065857559507009125, |
| "grad_norm": 1.40625, |
| "learning_rate": 0.00010449999999999999, |
| "loss": 6.9833, |
| "mean_token_accuracy": 0.07725078389048576, |
| "num_tokens": 621784.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 7.700586032867432, |
| "epoch": 0.006742559663812839, |
| "grad_norm": 1.421875, |
| "learning_rate": 0.000107, |
| "loss": 6.9716, |
| "mean_token_accuracy": 0.07797399312257766, |
| "num_tokens": 634707.0, |
| "step": 215 |
| }, |
| { |
| "entropy": 7.623541641235351, |
| "epoch": 0.006899363376924765, |
| "grad_norm": 1.359375, |
| "learning_rate": 0.0001095, |
| "loss": 7.0494, |
| "mean_token_accuracy": 0.0720802016556263, |
| "num_tokens": 647835.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 7.631550598144531, |
| "epoch": 0.007056167090036692, |
| "grad_norm": 1.3125, |
| "learning_rate": 0.000112, |
| "loss": 6.9591, |
| "mean_token_accuracy": 0.07618656158447265, |
| "num_tokens": 662587.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 7.684398460388183, |
| "epoch": 0.007212970803148619, |
| "grad_norm": 1.3828125, |
| "learning_rate": 0.0001145, |
| "loss": 7.0602, |
| "mean_token_accuracy": 0.07651265673339366, |
| "num_tokens": 676329.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 7.561251878738403, |
| "epoch": 0.007369774516260545, |
| "grad_norm": 1.3203125, |
| "learning_rate": 0.00011700000000000001, |
| "loss": 6.953, |
| "mean_token_accuracy": 0.08274711072444915, |
| "num_tokens": 692541.0, |
| "step": 235 |
| }, |
| { |
| "entropy": 7.560929489135742, |
| "epoch": 0.0075265782293724715, |
| "grad_norm": 1.1953125, |
| "learning_rate": 0.00011949999999999999, |
| "loss": 7.0198, |
| "mean_token_accuracy": 0.07723658978939056, |
| "num_tokens": 706997.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 7.557625770568848, |
| "epoch": 0.007683381942484398, |
| "grad_norm": 1.3359375, |
| "learning_rate": 0.000122, |
| "loss": 6.9217, |
| "mean_token_accuracy": 0.08156572207808495, |
| "num_tokens": 720719.0, |
| "step": 245 |
| }, |
| { |
| "entropy": 7.592370510101318, |
| "epoch": 0.007840185655596325, |
| "grad_norm": 1.40625, |
| "learning_rate": 0.0001245, |
| "loss": 7.0018, |
| "mean_token_accuracy": 0.08049147836863994, |
| "num_tokens": 734572.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 7.522368669509888, |
| "epoch": 0.007996989368708251, |
| "grad_norm": 1.2734375, |
| "learning_rate": 0.000127, |
| "loss": 6.8688, |
| "mean_token_accuracy": 0.08029020316898823, |
| "num_tokens": 751365.0, |
| "step": 255 |
| }, |
| { |
| "entropy": 7.458198308944702, |
| "epoch": 0.008153793081820177, |
| "grad_norm": 1.375, |
| "learning_rate": 0.0001295, |
| "loss": 6.8057, |
| "mean_token_accuracy": 0.08540241345763207, |
| "num_tokens": 765549.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 7.365204620361328, |
| "epoch": 0.008310596794932104, |
| "grad_norm": 1.1796875, |
| "learning_rate": 0.000132, |
| "loss": 6.7974, |
| "mean_token_accuracy": 0.0849799670279026, |
| "num_tokens": 780071.0, |
| "step": 265 |
| }, |
| { |
| "entropy": 7.453489208221436, |
| "epoch": 0.00846740050804403, |
| "grad_norm": 1.3125, |
| "learning_rate": 0.00013450000000000002, |
| "loss": 6.8488, |
| "mean_token_accuracy": 0.08373216427862644, |
| "num_tokens": 795328.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 7.333805704116822, |
| "epoch": 0.008624204221155956, |
| "grad_norm": 1.2890625, |
| "learning_rate": 0.00013700000000000002, |
| "loss": 6.759, |
| "mean_token_accuracy": 0.09399082660675048, |
| "num_tokens": 808404.0, |
| "step": 275 |
| }, |
| { |
| "entropy": 7.400871515274048, |
| "epoch": 0.008781007934267884, |
| "grad_norm": 1.4296875, |
| "learning_rate": 0.0001395, |
| "loss": 6.857, |
| "mean_token_accuracy": 0.08847371190786361, |
| "num_tokens": 823204.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 7.292252063751221, |
| "epoch": 0.00893781164737981, |
| "grad_norm": 1.5, |
| "learning_rate": 0.00014199999999999998, |
| "loss": 6.7636, |
| "mean_token_accuracy": 0.08776607438921928, |
| "num_tokens": 836276.0, |
| "step": 285 |
| }, |
| { |
| "entropy": 7.366798543930054, |
| "epoch": 0.009094615360491736, |
| "grad_norm": 1.4296875, |
| "learning_rate": 0.0001445, |
| "loss": 6.7562, |
| "mean_token_accuracy": 0.09157689064741134, |
| "num_tokens": 850090.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 7.297074222564698, |
| "epoch": 0.009251419073603663, |
| "grad_norm": 1.2890625, |
| "learning_rate": 0.000147, |
| "loss": 6.7845, |
| "mean_token_accuracy": 0.08689201548695565, |
| "num_tokens": 865931.0, |
| "step": 295 |
| }, |
| { |
| "entropy": 7.316424083709717, |
| "epoch": 0.00940822278671559, |
| "grad_norm": 1.390625, |
| "learning_rate": 0.0001495, |
| "loss": 6.7825, |
| "mean_token_accuracy": 0.08415777459740639, |
| "num_tokens": 879960.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 7.245685482025147, |
| "epoch": 0.009565026499827515, |
| "grad_norm": 1.125, |
| "learning_rate": 0.000152, |
| "loss": 6.7297, |
| "mean_token_accuracy": 0.08968461528420449, |
| "num_tokens": 894711.0, |
| "step": 305 |
| }, |
| { |
| "entropy": 7.20286979675293, |
| "epoch": 0.009721830212939443, |
| "grad_norm": 1.28125, |
| "learning_rate": 0.00015450000000000001, |
| "loss": 6.6488, |
| "mean_token_accuracy": 0.08769082948565483, |
| "num_tokens": 911461.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 7.2604657173156735, |
| "epoch": 0.009878633926051369, |
| "grad_norm": 1.515625, |
| "learning_rate": 0.000157, |
| "loss": 6.8178, |
| "mean_token_accuracy": 0.08734595999121667, |
| "num_tokens": 926618.0, |
| "step": 315 |
| }, |
| { |
| "entropy": 7.233215236663819, |
| "epoch": 0.010035437639163295, |
| "grad_norm": 1.3515625, |
| "learning_rate": 0.0001595, |
| "loss": 6.7365, |
| "mean_token_accuracy": 0.09149369075894356, |
| "num_tokens": 942310.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 7.098740816116333, |
| "epoch": 0.010192241352275222, |
| "grad_norm": 1.1953125, |
| "learning_rate": 0.000162, |
| "loss": 6.638, |
| "mean_token_accuracy": 0.09136478006839752, |
| "num_tokens": 958180.0, |
| "step": 325 |
| }, |
| { |
| "entropy": 7.278954648971558, |
| "epoch": 0.010349045065387148, |
| "grad_norm": 1.2265625, |
| "learning_rate": 0.00016450000000000001, |
| "loss": 6.7407, |
| "mean_token_accuracy": 0.08464238122105598, |
| "num_tokens": 973484.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 7.1389687061309814, |
| "epoch": 0.010505848778499074, |
| "grad_norm": 1.1328125, |
| "learning_rate": 0.00016700000000000002, |
| "loss": 6.5988, |
| "mean_token_accuracy": 0.09544083774089814, |
| "num_tokens": 988712.0, |
| "step": 335 |
| }, |
| { |
| "entropy": 7.1739630699157715, |
| "epoch": 0.010662652491611002, |
| "grad_norm": 1.203125, |
| "learning_rate": 0.00016950000000000003, |
| "loss": 6.6637, |
| "mean_token_accuracy": 0.0873688019812107, |
| "num_tokens": 1005316.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 7.009389734268188, |
| "epoch": 0.010819456204722928, |
| "grad_norm": 1.2734375, |
| "learning_rate": 0.00017199999999999998, |
| "loss": 6.5965, |
| "mean_token_accuracy": 0.09233827069401741, |
| "num_tokens": 1020624.0, |
| "step": 345 |
| }, |
| { |
| "entropy": 7.205262613296509, |
| "epoch": 0.010976259917834854, |
| "grad_norm": 1.3828125, |
| "learning_rate": 0.00017449999999999999, |
| "loss": 6.6497, |
| "mean_token_accuracy": 0.09442537128925324, |
| "num_tokens": 1035346.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 7.131465578079224, |
| "epoch": 0.011133063630946781, |
| "grad_norm": 1.5, |
| "learning_rate": 0.000177, |
| "loss": 6.6518, |
| "mean_token_accuracy": 0.08939319550991058, |
| "num_tokens": 1052396.0, |
| "step": 355 |
| }, |
| { |
| "entropy": 7.066566705703735, |
| "epoch": 0.011289867344058707, |
| "grad_norm": 1.21875, |
| "learning_rate": 0.0001795, |
| "loss": 6.6706, |
| "mean_token_accuracy": 0.08813088536262512, |
| "num_tokens": 1068158.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 7.096987533569336, |
| "epoch": 0.011446671057170633, |
| "grad_norm": 1.21875, |
| "learning_rate": 0.000182, |
| "loss": 6.532, |
| "mean_token_accuracy": 0.10468010231852531, |
| "num_tokens": 1081891.0, |
| "step": 365 |
| }, |
| { |
| "entropy": 6.936192178726197, |
| "epoch": 0.01160347477028256, |
| "grad_norm": 1.390625, |
| "learning_rate": 0.0001845, |
| "loss": 6.4988, |
| "mean_token_accuracy": 0.10371973067522049, |
| "num_tokens": 1094953.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 6.988458728790283, |
| "epoch": 0.011760278483394487, |
| "grad_norm": 1.375, |
| "learning_rate": 0.000187, |
| "loss": 6.4762, |
| "mean_token_accuracy": 0.09902411252260208, |
| "num_tokens": 1109923.0, |
| "step": 375 |
| }, |
| { |
| "entropy": 7.017359066009521, |
| "epoch": 0.011917082196506413, |
| "grad_norm": 1.265625, |
| "learning_rate": 0.0001895, |
| "loss": 6.5871, |
| "mean_token_accuracy": 0.09580302238464355, |
| "num_tokens": 1126987.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 6.971819591522217, |
| "epoch": 0.01207388590961834, |
| "grad_norm": 1.3671875, |
| "learning_rate": 0.000192, |
| "loss": 6.5812, |
| "mean_token_accuracy": 0.09617941230535507, |
| "num_tokens": 1139931.0, |
| "step": 385 |
| }, |
| { |
| "entropy": 6.953200578689575, |
| "epoch": 0.012230689622730266, |
| "grad_norm": 1.3203125, |
| "learning_rate": 0.0001945, |
| "loss": 6.4646, |
| "mean_token_accuracy": 0.10358070284128189, |
| "num_tokens": 1155721.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 6.9849913120269775, |
| "epoch": 0.012387493335842192, |
| "grad_norm": 1.1484375, |
| "learning_rate": 0.00019700000000000002, |
| "loss": 6.5469, |
| "mean_token_accuracy": 0.10082540512084961, |
| "num_tokens": 1171655.0, |
| "step": 395 |
| }, |
| { |
| "entropy": 6.938650178909302, |
| "epoch": 0.01254429704895412, |
| "grad_norm": 1.171875, |
| "learning_rate": 0.00019950000000000002, |
| "loss": 6.5281, |
| "mean_token_accuracy": 0.10091597810387612, |
| "num_tokens": 1186666.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 6.9116593360900875, |
| "epoch": 0.012701100762066046, |
| "grad_norm": 1.46875, |
| "learning_rate": 0.000202, |
| "loss": 6.4346, |
| "mean_token_accuracy": 0.09778324812650681, |
| "num_tokens": 1198283.0, |
| "step": 405 |
| }, |
| { |
| "entropy": 6.876869249343872, |
| "epoch": 0.012857904475177971, |
| "grad_norm": 1.2734375, |
| "learning_rate": 0.00020449999999999998, |
| "loss": 6.4735, |
| "mean_token_accuracy": 0.09719930961728096, |
| "num_tokens": 1212613.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 6.932867479324341, |
| "epoch": 0.013014708188289899, |
| "grad_norm": 1.6171875, |
| "learning_rate": 0.000207, |
| "loss": 6.4643, |
| "mean_token_accuracy": 0.10083647891879081, |
| "num_tokens": 1227515.0, |
| "step": 415 |
| }, |
| { |
| "entropy": 6.873914337158203, |
| "epoch": 0.013171511901401825, |
| "grad_norm": 1.40625, |
| "learning_rate": 0.0002095, |
| "loss": 6.4762, |
| "mean_token_accuracy": 0.1099552720785141, |
| "num_tokens": 1239022.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 6.844020032882691, |
| "epoch": 0.013328315614513751, |
| "grad_norm": 1.1875, |
| "learning_rate": 0.000212, |
| "loss": 6.4853, |
| "mean_token_accuracy": 0.09865991845726967, |
| "num_tokens": 1254653.0, |
| "step": 425 |
| }, |
| { |
| "entropy": 6.919657754898071, |
| "epoch": 0.013485119327625679, |
| "grad_norm": 1.1875, |
| "learning_rate": 0.0002145, |
| "loss": 6.4894, |
| "mean_token_accuracy": 0.09527975097298622, |
| "num_tokens": 1270766.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 6.905207824707031, |
| "epoch": 0.013641923040737605, |
| "grad_norm": 1.25, |
| "learning_rate": 0.00021700000000000002, |
| "loss": 6.4774, |
| "mean_token_accuracy": 0.10371477827429772, |
| "num_tokens": 1284742.0, |
| "step": 435 |
| }, |
| { |
| "entropy": 6.912538671493531, |
| "epoch": 0.01379872675384953, |
| "grad_norm": 1.234375, |
| "learning_rate": 0.0002195, |
| "loss": 6.4678, |
| "mean_token_accuracy": 0.10031345337629319, |
| "num_tokens": 1297881.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 6.800807285308838, |
| "epoch": 0.013955530466961458, |
| "grad_norm": 1.09375, |
| "learning_rate": 0.000222, |
| "loss": 6.4071, |
| "mean_token_accuracy": 0.10895167514681817, |
| "num_tokens": 1314727.0, |
| "step": 445 |
| }, |
| { |
| "entropy": 6.85034818649292, |
| "epoch": 0.014112334180073384, |
| "grad_norm": 1.2421875, |
| "learning_rate": 0.0002245, |
| "loss": 6.4329, |
| "mean_token_accuracy": 0.10191441029310226, |
| "num_tokens": 1332630.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 6.820253705978393, |
| "epoch": 0.01426913789318531, |
| "grad_norm": 1.1015625, |
| "learning_rate": 0.00022700000000000002, |
| "loss": 6.5253, |
| "mean_token_accuracy": 0.09508129432797433, |
| "num_tokens": 1348028.0, |
| "step": 455 |
| }, |
| { |
| "entropy": 6.8252214908599855, |
| "epoch": 0.014425941606297238, |
| "grad_norm": 1.2265625, |
| "learning_rate": 0.00022950000000000002, |
| "loss": 6.3729, |
| "mean_token_accuracy": 0.10276357978582382, |
| "num_tokens": 1363723.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 6.869743585586548, |
| "epoch": 0.014582745319409163, |
| "grad_norm": 1.625, |
| "learning_rate": 0.00023200000000000003, |
| "loss": 6.4717, |
| "mean_token_accuracy": 0.10454175770282745, |
| "num_tokens": 1378400.0, |
| "step": 465 |
| }, |
| { |
| "entropy": 6.816213941574096, |
| "epoch": 0.01473954903252109, |
| "grad_norm": 1.21875, |
| "learning_rate": 0.00023449999999999998, |
| "loss": 6.3712, |
| "mean_token_accuracy": 0.10962107256054879, |
| "num_tokens": 1393383.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 6.703394031524658, |
| "epoch": 0.014896352745633017, |
| "grad_norm": 1.3515625, |
| "learning_rate": 0.000237, |
| "loss": 6.3905, |
| "mean_token_accuracy": 0.10753661692142487, |
| "num_tokens": 1409609.0, |
| "step": 475 |
| }, |
| { |
| "entropy": 6.746933174133301, |
| "epoch": 0.015053156458744943, |
| "grad_norm": 1.28125, |
| "learning_rate": 0.0002395, |
| "loss": 6.3711, |
| "mean_token_accuracy": 0.109779604524374, |
| "num_tokens": 1423378.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 6.657280206680298, |
| "epoch": 0.015209960171856869, |
| "grad_norm": 1.28125, |
| "learning_rate": 0.000242, |
| "loss": 6.3545, |
| "mean_token_accuracy": 0.10648095905780793, |
| "num_tokens": 1437722.0, |
| "step": 485 |
| }, |
| { |
| "entropy": 6.752739667892456, |
| "epoch": 0.015366763884968796, |
| "grad_norm": 1.203125, |
| "learning_rate": 0.0002445, |
| "loss": 6.4004, |
| "mean_token_accuracy": 0.10294785127043724, |
| "num_tokens": 1452969.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 6.7490668296813965, |
| "epoch": 0.015523567598080722, |
| "grad_norm": 1.3515625, |
| "learning_rate": 0.000247, |
| "loss": 6.406, |
| "mean_token_accuracy": 0.10591119825839997, |
| "num_tokens": 1466014.0, |
| "step": 495 |
| }, |
| { |
| "entropy": 6.7571159362792965, |
| "epoch": 0.01568037131119265, |
| "grad_norm": 1.3671875, |
| "learning_rate": 0.0002495, |
| "loss": 6.4293, |
| "mean_token_accuracy": 0.1040202483534813, |
| "num_tokens": 1482087.0, |
| "step": 500 |
| } |
| ], |
| "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": 2615647555584000.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
