| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.813749196380377, |
| "epoch": 0.016, |
| "grad_norm": 21.625, |
| "learning_rate": 1.8e-07, |
| "loss": 2.3527, |
| "mean_token_accuracy": 0.5754138542339206, |
| "num_tokens": 195510.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.7928078502416611, |
| "epoch": 0.032, |
| "grad_norm": 16.75, |
| "learning_rate": 3.8e-07, |
| "loss": 2.3363, |
| "mean_token_accuracy": 0.5766932301223278, |
| "num_tokens": 391008.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.8388511516153812, |
| "epoch": 0.048, |
| "grad_norm": 17.25, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 2.3414, |
| "mean_token_accuracy": 0.5755490336567164, |
| "num_tokens": 590136.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.853544193506241, |
| "epoch": 0.064, |
| "grad_norm": 12.5, |
| "learning_rate": 7.8e-07, |
| "loss": 2.2941, |
| "mean_token_accuracy": 0.577743387594819, |
| "num_tokens": 791432.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.9382900901138782, |
| "epoch": 0.08, |
| "grad_norm": 12.3125, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 2.2698, |
| "mean_token_accuracy": 0.5786384884268045, |
| "num_tokens": 990342.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 2.012592934817076, |
| "epoch": 0.096, |
| "grad_norm": 7.59375, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 2.2841, |
| "mean_token_accuracy": 0.5789956836029887, |
| "num_tokens": 1182339.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.020173179358244, |
| "epoch": 0.112, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 2.1881, |
| "mean_token_accuracy": 0.5869531789794564, |
| "num_tokens": 1386193.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 2.042734081298113, |
| "epoch": 0.128, |
| "grad_norm": 5.75, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 2.1864, |
| "mean_token_accuracy": 0.5859880622476339, |
| "num_tokens": 1583261.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.9902476705610752, |
| "epoch": 0.144, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 2.1581, |
| "mean_token_accuracy": 0.5905994104221464, |
| "num_tokens": 1774826.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 2.038115797191858, |
| "epoch": 0.16, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.98e-06, |
| "loss": 2.1863, |
| "mean_token_accuracy": 0.5856570154428482, |
| "num_tokens": 1971310.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.9723965148925782, |
| "eval_biology_loss": 2.1413300037384033, |
| "eval_biology_mean_token_accuracy": 0.6000847628116608, |
| "eval_biology_num_tokens": 1971310.0, |
| "eval_biology_runtime": 38.6804, |
| "eval_biology_samples_per_second": 12.926, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 0.6801536810398102, |
| "eval_physics_loss": 0.7345870733261108, |
| "eval_physics_mean_token_accuracy": 0.8230011963844299, |
| "eval_physics_num_tokens": 1971310.0, |
| "eval_physics_runtime": 56.9468, |
| "eval_physics_samples_per_second": 8.78, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.9897258900105954, |
| "epoch": 0.176, |
| "grad_norm": 4.90625, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 2.1203, |
| "mean_token_accuracy": 0.5903370209038258, |
| "num_tokens": 2169453.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.9791964955627919, |
| "epoch": 0.192, |
| "grad_norm": 5.8125, |
| "learning_rate": 2.38e-06, |
| "loss": 2.1168, |
| "mean_token_accuracy": 0.5923390585929156, |
| "num_tokens": 2366844.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.9716377504169942, |
| "epoch": 0.208, |
| "grad_norm": 5.1875, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 2.1059, |
| "mean_token_accuracy": 0.5896596999838948, |
| "num_tokens": 2559708.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.9294324100017548, |
| "epoch": 0.224, |
| "grad_norm": 5.53125, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 2.0638, |
| "mean_token_accuracy": 0.6001523662358522, |
| "num_tokens": 2756533.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.9123736470937729, |
| "epoch": 0.24, |
| "grad_norm": 5.0625, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 2.0241, |
| "mean_token_accuracy": 0.6036131344735622, |
| "num_tokens": 2948600.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.9347217477858067, |
| "epoch": 0.256, |
| "grad_norm": 5.15625, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 2.0531, |
| "mean_token_accuracy": 0.5974894173443317, |
| "num_tokens": 3141473.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.9242084234952928, |
| "epoch": 0.272, |
| "grad_norm": 5.09375, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 2.0261, |
| "mean_token_accuracy": 0.5972955338656902, |
| "num_tokens": 3337285.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.8868879064917565, |
| "epoch": 0.288, |
| "grad_norm": 5.375, |
| "learning_rate": 3.58e-06, |
| "loss": 1.9858, |
| "mean_token_accuracy": 0.6028371442109346, |
| "num_tokens": 3541221.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.8624922472983598, |
| "epoch": 0.304, |
| "grad_norm": 4.71875, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 1.943, |
| "mean_token_accuracy": 0.6070948902517557, |
| "num_tokens": 3734952.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.881417414546013, |
| "epoch": 0.32, |
| "grad_norm": 5.21875, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 1.9609, |
| "mean_token_accuracy": 0.603725228458643, |
| "num_tokens": 3922209.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.8500013475418091, |
| "eval_biology_loss": 1.9382704496383667, |
| "eval_biology_mean_token_accuracy": 0.6156538977622986, |
| "eval_biology_num_tokens": 3922209.0, |
| "eval_biology_runtime": 38.6714, |
| "eval_biology_samples_per_second": 12.929, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 0.6814348955154419, |
| "eval_physics_loss": 0.7077023386955261, |
| "eval_physics_mean_token_accuracy": 0.8221852016448975, |
| "eval_physics_num_tokens": 3922209.0, |
| "eval_physics_runtime": 56.9406, |
| "eval_physics_samples_per_second": 8.781, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.8236718311905862, |
| "epoch": 0.336, |
| "grad_norm": 5.125, |
| "learning_rate": 4.18e-06, |
| "loss": 1.9029, |
| "mean_token_accuracy": 0.6111209694296121, |
| "num_tokens": 4115941.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.851329033076763, |
| "epoch": 0.352, |
| "grad_norm": 5.0, |
| "learning_rate": 4.38e-06, |
| "loss": 1.8904, |
| "mean_token_accuracy": 0.6098318418487907, |
| "num_tokens": 4308755.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.8281462915241717, |
| "epoch": 0.368, |
| "grad_norm": 4.65625, |
| "learning_rate": 4.58e-06, |
| "loss": 1.8692, |
| "mean_token_accuracy": 0.6131063066422939, |
| "num_tokens": 4505910.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.831401702016592, |
| "epoch": 0.384, |
| "grad_norm": 4.96875, |
| "learning_rate": 4.78e-06, |
| "loss": 1.8357, |
| "mean_token_accuracy": 0.6127742934972048, |
| "num_tokens": 4696101.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.816836379468441, |
| "epoch": 0.4, |
| "grad_norm": 4.75, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 1.8158, |
| "mean_token_accuracy": 0.614238902926445, |
| "num_tokens": 4889445.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.8351925887167453, |
| "epoch": 0.416, |
| "grad_norm": 4.5625, |
| "learning_rate": 5.18e-06, |
| "loss": 1.8225, |
| "mean_token_accuracy": 0.6112797066569329, |
| "num_tokens": 5087624.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.8272491071373225, |
| "epoch": 0.432, |
| "grad_norm": 5.5, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 1.8098, |
| "mean_token_accuracy": 0.6129346057772637, |
| "num_tokens": 5273794.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.8296721816062926, |
| "epoch": 0.448, |
| "grad_norm": 5.46875, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 1.7925, |
| "mean_token_accuracy": 0.6138202898204327, |
| "num_tokens": 5463288.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.7533989433199166, |
| "epoch": 0.464, |
| "grad_norm": 5.09375, |
| "learning_rate": 5.78e-06, |
| "loss": 1.7277, |
| "mean_token_accuracy": 0.625254824757576, |
| "num_tokens": 5656827.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.7856548748910428, |
| "epoch": 0.48, |
| "grad_norm": 4.75, |
| "learning_rate": 5.98e-06, |
| "loss": 1.7492, |
| "mean_token_accuracy": 0.6199217233806849, |
| "num_tokens": 5853311.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.766937505722046, |
| "eval_biology_loss": 1.737308144569397, |
| "eval_biology_mean_token_accuracy": 0.6296310710906983, |
| "eval_biology_num_tokens": 5853311.0, |
| "eval_biology_runtime": 38.676, |
| "eval_biology_samples_per_second": 12.928, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 0.699759515285492, |
| "eval_physics_loss": 0.6600434184074402, |
| "eval_physics_mean_token_accuracy": 0.8216852688789368, |
| "eval_physics_num_tokens": 5853311.0, |
| "eval_physics_runtime": 56.9184, |
| "eval_physics_samples_per_second": 8.785, |
| "eval_physics_steps_per_second": 2.196, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.6893749851733446, |
| "epoch": 0.496, |
| "grad_norm": 5.125, |
| "learning_rate": 6.18e-06, |
| "loss": 1.6529, |
| "mean_token_accuracy": 0.6324049789458513, |
| "num_tokens": 6049788.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.777088150382042, |
| "epoch": 0.512, |
| "grad_norm": 5.0625, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 1.7373, |
| "mean_token_accuracy": 0.6215202666819095, |
| "num_tokens": 6243825.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.7654731079936028, |
| "epoch": 0.528, |
| "grad_norm": 4.75, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 1.7104, |
| "mean_token_accuracy": 0.6297121163457632, |
| "num_tokens": 6434188.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.7458826407790184, |
| "epoch": 0.544, |
| "grad_norm": 4.59375, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 1.7134, |
| "mean_token_accuracy": 0.6295231115072966, |
| "num_tokens": 6629832.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.7250068325549364, |
| "epoch": 0.56, |
| "grad_norm": 4.375, |
| "learning_rate": 6.98e-06, |
| "loss": 1.6747, |
| "mean_token_accuracy": 0.6324008051306009, |
| "num_tokens": 6824668.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.6957500472664833, |
| "epoch": 0.576, |
| "grad_norm": 4.8125, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 1.6528, |
| "mean_token_accuracy": 0.6356299221515656, |
| "num_tokens": 7025923.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.7250152129679919, |
| "epoch": 0.592, |
| "grad_norm": 4.375, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 1.6596, |
| "mean_token_accuracy": 0.63132487796247, |
| "num_tokens": 7218224.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.7018816463649273, |
| "epoch": 0.608, |
| "grad_norm": 5.1875, |
| "learning_rate": 7.58e-06, |
| "loss": 1.664, |
| "mean_token_accuracy": 0.6356819782406091, |
| "num_tokens": 7421216.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.7107378408312797, |
| "epoch": 0.624, |
| "grad_norm": 4.78125, |
| "learning_rate": 7.78e-06, |
| "loss": 1.6546, |
| "mean_token_accuracy": 0.6328394934535027, |
| "num_tokens": 7617258.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.7047107808291913, |
| "epoch": 0.64, |
| "grad_norm": 4.65625, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 1.636, |
| "mean_token_accuracy": 0.6354850973933935, |
| "num_tokens": 7806173.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.6343248481750487, |
| "eval_biology_loss": 1.6292321681976318, |
| "eval_biology_mean_token_accuracy": 0.6456177935600281, |
| "eval_biology_num_tokens": 7806173.0, |
| "eval_biology_runtime": 38.6524, |
| "eval_biology_samples_per_second": 12.936, |
| "eval_biology_steps_per_second": 3.234, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.6857430481910706, |
| "eval_physics_loss": 0.6523167490959167, |
| "eval_physics_mean_token_accuracy": 0.8234750027656556, |
| "eval_physics_num_tokens": 7806173.0, |
| "eval_physics_runtime": 56.9086, |
| "eval_physics_samples_per_second": 8.786, |
| "eval_physics_steps_per_second": 2.197, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.6953211184591055, |
| "epoch": 0.656, |
| "grad_norm": 4.8125, |
| "learning_rate": 8.18e-06, |
| "loss": 1.6492, |
| "mean_token_accuracy": 0.635795733332634, |
| "num_tokens": 8000514.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.7117650181055069, |
| "epoch": 0.672, |
| "grad_norm": 4.90625, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 1.651, |
| "mean_token_accuracy": 0.6328964576125145, |
| "num_tokens": 8187884.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.6630911942571402, |
| "epoch": 0.688, |
| "grad_norm": 5.0, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 1.6129, |
| "mean_token_accuracy": 0.6399691045284271, |
| "num_tokens": 8390643.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.6603879615664483, |
| "epoch": 0.704, |
| "grad_norm": 4.8125, |
| "learning_rate": 8.78e-06, |
| "loss": 1.6006, |
| "mean_token_accuracy": 0.6405742786824703, |
| "num_tokens": 8583266.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.5886821828782558, |
| "epoch": 0.72, |
| "grad_norm": 4.53125, |
| "learning_rate": 8.98e-06, |
| "loss": 1.5254, |
| "mean_token_accuracy": 0.6530824847519398, |
| "num_tokens": 8786233.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.6262568928301335, |
| "epoch": 0.736, |
| "grad_norm": 5.0625, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 1.5597, |
| "mean_token_accuracy": 0.6445462703704834, |
| "num_tokens": 8982898.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.6295693822205066, |
| "epoch": 0.752, |
| "grad_norm": 4.90625, |
| "learning_rate": 9.38e-06, |
| "loss": 1.5648, |
| "mean_token_accuracy": 0.6450414888560771, |
| "num_tokens": 9174361.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.5759615428745746, |
| "epoch": 0.768, |
| "grad_norm": 4.4375, |
| "learning_rate": 9.58e-06, |
| "loss": 1.5144, |
| "mean_token_accuracy": 0.653152122348547, |
| "num_tokens": 9373430.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.6083305418491363, |
| "epoch": 0.784, |
| "grad_norm": 4.40625, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 1.5488, |
| "mean_token_accuracy": 0.6459381774067878, |
| "num_tokens": 9570409.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.6188334189355373, |
| "epoch": 0.8, |
| "grad_norm": 4.78125, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 1.5541, |
| "mean_token_accuracy": 0.6456099949777127, |
| "num_tokens": 9766360.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.570951623916626, |
| "eval_biology_loss": 1.5443849563598633, |
| "eval_biology_mean_token_accuracy": 0.6565022644996643, |
| "eval_biology_num_tokens": 9766360.0, |
| "eval_biology_runtime": 38.6538, |
| "eval_biology_samples_per_second": 12.935, |
| "eval_biology_steps_per_second": 3.234, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.6805566384792328, |
| "eval_physics_loss": 0.6488668322563171, |
| "eval_physics_mean_token_accuracy": 0.8232229099273681, |
| "eval_physics_num_tokens": 9766360.0, |
| "eval_physics_runtime": 56.9177, |
| "eval_physics_samples_per_second": 8.785, |
| "eval_physics_steps_per_second": 2.196, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.5766926735639573, |
| "epoch": 0.816, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.018e-05, |
| "loss": 1.5247, |
| "mean_token_accuracy": 0.6521722257137299, |
| "num_tokens": 9963834.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.559133169800043, |
| "epoch": 0.832, |
| "grad_norm": 4.5, |
| "learning_rate": 1.038e-05, |
| "loss": 1.5075, |
| "mean_token_accuracy": 0.6519753206521273, |
| "num_tokens": 10160950.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.5535101789981127, |
| "epoch": 0.848, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 1.496, |
| "mean_token_accuracy": 0.6550566371530294, |
| "num_tokens": 10363131.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.573872973024845, |
| "epoch": 0.864, |
| "grad_norm": 4.375, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 1.5151, |
| "mean_token_accuracy": 0.6522478803992271, |
| "num_tokens": 10558003.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.5461064357310534, |
| "epoch": 0.88, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 1.4907, |
| "mean_token_accuracy": 0.6562161970883608, |
| "num_tokens": 10754470.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.5923856895416975, |
| "epoch": 0.896, |
| "grad_norm": 4.5, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 1.5274, |
| "mean_token_accuracy": 0.6506992731243372, |
| "num_tokens": 10948943.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.546384008973837, |
| "epoch": 0.912, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.138e-05, |
| "loss": 1.4879, |
| "mean_token_accuracy": 0.655264026299119, |
| "num_tokens": 11142613.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.5580323040485382, |
| "epoch": 0.928, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 1.4914, |
| "mean_token_accuracy": 0.6533704947680234, |
| "num_tokens": 11336779.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.4954375229775905, |
| "epoch": 0.944, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.178e-05, |
| "loss": 1.4496, |
| "mean_token_accuracy": 0.6618651241064072, |
| "num_tokens": 11536381.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.521157095208764, |
| "epoch": 0.96, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.198e-05, |
| "loss": 1.4654, |
| "mean_token_accuracy": 0.6575134672224522, |
| "num_tokens": 11735527.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.4938988003730773, |
| "eval_biology_loss": 1.4810906648635864, |
| "eval_biology_mean_token_accuracy": 0.6647453079223633, |
| "eval_biology_num_tokens": 11735527.0, |
| "eval_biology_runtime": 38.6461, |
| "eval_biology_samples_per_second": 12.938, |
| "eval_biology_steps_per_second": 3.234, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.6846246910095215, |
| "eval_physics_loss": 0.6483092308044434, |
| "eval_physics_mean_token_accuracy": 0.8228992919921875, |
| "eval_physics_num_tokens": 11735527.0, |
| "eval_physics_runtime": 57.2247, |
| "eval_physics_samples_per_second": 8.737, |
| "eval_physics_steps_per_second": 2.184, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.4897711608558892, |
| "epoch": 0.976, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.218e-05, |
| "loss": 1.4364, |
| "mean_token_accuracy": 0.6634432837367058, |
| "num_tokens": 11930935.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.5174701895564795, |
| "epoch": 0.992, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 1.4657, |
| "mean_token_accuracy": 0.6584253456443548, |
| "num_tokens": 12129585.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.4917617563158274, |
| "epoch": 1.008, |
| "grad_norm": 4.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 1.4435, |
| "mean_token_accuracy": 0.6625992625951767, |
| "num_tokens": 12325847.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.429138394445181, |
| "epoch": 1.024, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 1.3693, |
| "mean_token_accuracy": 0.6733787953853607, |
| "num_tokens": 12530820.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.4665595509111882, |
| "epoch": 1.04, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 1.4151, |
| "mean_token_accuracy": 0.6643557522445918, |
| "num_tokens": 12725384.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.4320495713502168, |
| "epoch": 1.056, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 1.3639, |
| "mean_token_accuracy": 0.6728409513831138, |
| "num_tokens": 12924592.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.456987388432026, |
| "epoch": 1.072, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 1.3964, |
| "mean_token_accuracy": 0.6672081802040338, |
| "num_tokens": 13112848.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.4462645407766104, |
| "epoch": 1.088, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 1.3914, |
| "mean_token_accuracy": 0.6701537113636732, |
| "num_tokens": 13305725.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.439976465329528, |
| "epoch": 1.104, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.378e-05, |
| "loss": 1.3853, |
| "mean_token_accuracy": 0.6701969560235739, |
| "num_tokens": 13498685.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.4542303152382374, |
| "epoch": 1.12, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.398e-05, |
| "loss": 1.3835, |
| "mean_token_accuracy": 0.6701514054089784, |
| "num_tokens": 13682146.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.3601415209770202, |
| "eval_biology_loss": 1.4304583072662354, |
| "eval_biology_mean_token_accuracy": 0.6715204911231994, |
| "eval_biology_num_tokens": 13682146.0, |
| "eval_biology_runtime": 38.6654, |
| "eval_biology_samples_per_second": 12.931, |
| "eval_biology_steps_per_second": 3.233, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.65302716588974, |
| "eval_physics_loss": 0.6474171876907349, |
| "eval_physics_mean_token_accuracy": 0.8226029920578003, |
| "eval_physics_num_tokens": 13682146.0, |
| "eval_physics_runtime": 56.8929, |
| "eval_physics_samples_per_second": 8.788, |
| "eval_physics_steps_per_second": 2.197, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.4230232119560242, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.418e-05, |
| "loss": 1.3747, |
| "mean_token_accuracy": 0.6713559620082379, |
| "num_tokens": 13876656.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.398179117962718, |
| "epoch": 1.152, |
| "grad_norm": 3.828125, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 1.3438, |
| "mean_token_accuracy": 0.675188298150897, |
| "num_tokens": 14086104.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.4270055878907442, |
| "epoch": 1.168, |
| "grad_norm": 4.375, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 1.373, |
| "mean_token_accuracy": 0.6712808165699243, |
| "num_tokens": 14274520.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.4195893313735723, |
| "epoch": 1.184, |
| "grad_norm": 4.0, |
| "learning_rate": 1.478e-05, |
| "loss": 1.3657, |
| "mean_token_accuracy": 0.6732822258025408, |
| "num_tokens": 14473623.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.4137178607285024, |
| "epoch": 1.2, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.498e-05, |
| "loss": 1.3527, |
| "mean_token_accuracy": 0.6739591065794229, |
| "num_tokens": 14661238.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.3734467361122369, |
| "epoch": 1.216, |
| "grad_norm": 4.25, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 1.3173, |
| "mean_token_accuracy": 0.6782099220901727, |
| "num_tokens": 14865915.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.3874890733510257, |
| "epoch": 1.232, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 1.3318, |
| "mean_token_accuracy": 0.6802723795175553, |
| "num_tokens": 15055682.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.3982337843626738, |
| "epoch": 1.248, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 1.3481, |
| "mean_token_accuracy": 0.6750583201646805, |
| "num_tokens": 15249576.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.4128487702459096, |
| "epoch": 1.264, |
| "grad_norm": 3.78125, |
| "learning_rate": 1.578e-05, |
| "loss": 1.3488, |
| "mean_token_accuracy": 0.6755638718605042, |
| "num_tokens": 15446149.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.4168333638459445, |
| "epoch": 1.28, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 1.3691, |
| "mean_token_accuracy": 0.6701620414853096, |
| "num_tokens": 15639317.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 1.3710014867782592, |
| "eval_biology_loss": 1.3889533281326294, |
| "eval_biology_mean_token_accuracy": 0.6776719264984131, |
| "eval_biology_num_tokens": 15639317.0, |
| "eval_biology_runtime": 38.6779, |
| "eval_biology_samples_per_second": 12.927, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.6683794283866882, |
| "eval_physics_loss": 0.6475260257720947, |
| "eval_physics_mean_token_accuracy": 0.8223769798278808, |
| "eval_physics_num_tokens": 15639317.0, |
| "eval_physics_runtime": 56.9416, |
| "eval_physics_samples_per_second": 8.781, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.415640415251255, |
| "epoch": 1.296, |
| "grad_norm": 3.890625, |
| "learning_rate": 1.618e-05, |
| "loss": 1.3559, |
| "mean_token_accuracy": 0.6743101593106985, |
| "num_tokens": 15835805.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.4065573081374168, |
| "epoch": 1.312, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.638e-05, |
| "loss": 1.3441, |
| "mean_token_accuracy": 0.6742365792393684, |
| "num_tokens": 16028375.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.4003059070557355, |
| "epoch": 1.328, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.658e-05, |
| "loss": 1.3416, |
| "mean_token_accuracy": 0.6751531921327114, |
| "num_tokens": 16230555.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.3695629570633172, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 3.46875, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 1.323, |
| "mean_token_accuracy": 0.6816294804215431, |
| "num_tokens": 16436297.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.3895002137869596, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.698e-05, |
| "loss": 1.3223, |
| "mean_token_accuracy": 0.678603322058916, |
| "num_tokens": 16630563.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.4141130354255438, |
| "epoch": 1.376, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.718e-05, |
| "loss": 1.3514, |
| "mean_token_accuracy": 0.6730918735265732, |
| "num_tokens": 16822602.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.3758564725518228, |
| "epoch": 1.392, |
| "grad_norm": 3.640625, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 1.3266, |
| "mean_token_accuracy": 0.6803319718688726, |
| "num_tokens": 17015750.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.3721271395683288, |
| "epoch": 1.408, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.758e-05, |
| "loss": 1.3173, |
| "mean_token_accuracy": 0.6792092099785805, |
| "num_tokens": 17207248.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.378094071522355, |
| "epoch": 1.424, |
| "grad_norm": 3.78125, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 1.3327, |
| "mean_token_accuracy": 0.6793139282613992, |
| "num_tokens": 17403869.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.3699288651347161, |
| "epoch": 1.44, |
| "grad_norm": 4.5, |
| "learning_rate": 1.798e-05, |
| "loss": 1.3147, |
| "mean_token_accuracy": 0.6776431813836098, |
| "num_tokens": 17597428.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 1.3595392904281616, |
| "eval_biology_loss": 1.3485888242721558, |
| "eval_biology_mean_token_accuracy": 0.6834584150314331, |
| "eval_biology_num_tokens": 17597428.0, |
| "eval_biology_runtime": 38.682, |
| "eval_biology_samples_per_second": 12.926, |
| "eval_biology_steps_per_second": 3.231, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.6735630776882172, |
| "eval_physics_loss": 0.6497707366943359, |
| "eval_physics_mean_token_accuracy": 0.8218984279632569, |
| "eval_physics_num_tokens": 17597428.0, |
| "eval_physics_runtime": 56.9341, |
| "eval_physics_samples_per_second": 8.782, |
| "eval_physics_steps_per_second": 2.196, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.3743872378021478, |
| "epoch": 1.456, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.3244, |
| "mean_token_accuracy": 0.6799841322004795, |
| "num_tokens": 17798083.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.3584794465452432, |
| "epoch": 1.472, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.2994, |
| "mean_token_accuracy": 0.6824300821870566, |
| "num_tokens": 17993802.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.3503762260079384, |
| "epoch": 1.488, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.858e-05, |
| "loss": 1.3014, |
| "mean_token_accuracy": 0.6844853088259697, |
| "num_tokens": 18185570.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.3729203592985868, |
| "epoch": 1.504, |
| "grad_norm": 3.46875, |
| "learning_rate": 1.878e-05, |
| "loss": 1.3216, |
| "mean_token_accuracy": 0.6780106417834759, |
| "num_tokens": 18377903.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.3659293711185456, |
| "epoch": 1.52, |
| "grad_norm": 3.46875, |
| "learning_rate": 1.898e-05, |
| "loss": 1.318, |
| "mean_token_accuracy": 0.6781526006758213, |
| "num_tokens": 18579284.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.366529828310013, |
| "epoch": 1.536, |
| "grad_norm": 4.5, |
| "learning_rate": 1.918e-05, |
| "loss": 1.3038, |
| "mean_token_accuracy": 0.6802028331905603, |
| "num_tokens": 18765324.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.3478929094970227, |
| "epoch": 1.552, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.938e-05, |
| "loss": 1.2881, |
| "mean_token_accuracy": 0.68250414095819, |
| "num_tokens": 18966382.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.339873855561018, |
| "epoch": 1.568, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.2915, |
| "mean_token_accuracy": 0.6844375465065241, |
| "num_tokens": 19160509.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.3251274175941945, |
| "epoch": 1.584, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.978e-05, |
| "loss": 1.2752, |
| "mean_token_accuracy": 0.6852322082966567, |
| "num_tokens": 19354662.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.384224559739232, |
| "epoch": 1.6, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.3287, |
| "mean_token_accuracy": 0.6757747534662485, |
| "num_tokens": 19543088.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 1.337540853023529, |
| "eval_biology_loss": 1.3165961503982544, |
| "eval_biology_mean_token_accuracy": 0.6879782662391662, |
| "eval_biology_num_tokens": 19543088.0, |
| "eval_biology_runtime": 38.6581, |
| "eval_biology_samples_per_second": 12.934, |
| "eval_biology_steps_per_second": 3.233, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.6811050243377685, |
| "eval_physics_loss": 0.6504300832748413, |
| "eval_physics_mean_token_accuracy": 0.8209615325927735, |
| "eval_physics_num_tokens": 19543088.0, |
| "eval_physics_runtime": 56.8501, |
| "eval_physics_samples_per_second": 8.795, |
| "eval_physics_steps_per_second": 2.199, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.3300264578312635, |
| "epoch": 1.616, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.2769, |
| "mean_token_accuracy": 0.685854634270072, |
| "num_tokens": 19743506.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.3689603984355927, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 1.3153, |
| "mean_token_accuracy": 0.6796921446919442, |
| "num_tokens": 19937567.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.3543198898434639, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 1.2901, |
| "mean_token_accuracy": 0.6821238681674003, |
| "num_tokens": 20129743.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.3178342673927546, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 3.640625, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 1.2644, |
| "mean_token_accuracy": 0.6889090698212386, |
| "num_tokens": 20331548.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.3326357528567314, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 1.2834, |
| "mean_token_accuracy": 0.6844820819795132, |
| "num_tokens": 20524247.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.327241437882185, |
| "epoch": 1.696, |
| "grad_norm": 3.078125, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 1.2708, |
| "mean_token_accuracy": 0.6858115192502737, |
| "num_tokens": 20718121.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.3415468953549863, |
| "epoch": 1.712, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 1.2797, |
| "mean_token_accuracy": 0.684544776007533, |
| "num_tokens": 20921419.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.320882337912917, |
| "epoch": 1.728, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 1.267, |
| "mean_token_accuracy": 0.6839414596557617, |
| "num_tokens": 21118581.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.2898854076862336, |
| "epoch": 1.744, |
| "grad_norm": 3.5625, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 1.2375, |
| "mean_token_accuracy": 0.6910523246973753, |
| "num_tokens": 21315128.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.31239253282547, |
| "epoch": 1.76, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.978e-05, |
| "loss": 1.2603, |
| "mean_token_accuracy": 0.6899993035942316, |
| "num_tokens": 21511066.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 1.2963400874137878, |
| "eval_biology_loss": 1.2855476140975952, |
| "eval_biology_mean_token_accuracy": 0.6923068904876709, |
| "eval_biology_num_tokens": 21511066.0, |
| "eval_biology_runtime": 38.6761, |
| "eval_biology_samples_per_second": 12.928, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.6632856893539428, |
| "eval_physics_loss": 0.6495886445045471, |
| "eval_physics_mean_token_accuracy": 0.8213236899375915, |
| "eval_physics_num_tokens": 21511066.0, |
| "eval_physics_runtime": 56.9409, |
| "eval_physics_samples_per_second": 8.781, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.3157803509384394, |
| "epoch": 1.776, |
| "grad_norm": 3.59375, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 1.2599, |
| "mean_token_accuracy": 0.6879700906574726, |
| "num_tokens": 21704368.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.2888959780335427, |
| "epoch": 1.792, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 1.2331, |
| "mean_token_accuracy": 0.6911204513162374, |
| "num_tokens": 21905784.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.3017573833465577, |
| "epoch": 1.808, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 1.2521, |
| "mean_token_accuracy": 0.6918430212885142, |
| "num_tokens": 22094159.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.329062307998538, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 3.078125, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 1.2592, |
| "mean_token_accuracy": 0.684392886608839, |
| "num_tokens": 22290654.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.339531321078539, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 1.2846, |
| "mean_token_accuracy": 0.6835382975637913, |
| "num_tokens": 22485675.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.295450346916914, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 1.2508, |
| "mean_token_accuracy": 0.6886295765638352, |
| "num_tokens": 22689900.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.2797737825661897, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 2.90625, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 1.2204, |
| "mean_token_accuracy": 0.6948459330946207, |
| "num_tokens": 22887275.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.3289342291653157, |
| "epoch": 1.888, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 1.2744, |
| "mean_token_accuracy": 0.6839813269674778, |
| "num_tokens": 23081241.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.2998227212578057, |
| "epoch": 1.904, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.2522, |
| "mean_token_accuracy": 0.6898551240563393, |
| "num_tokens": 23276417.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.300378565862775, |
| "epoch": 1.92, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 1.2471, |
| "mean_token_accuracy": 0.6882933590561151, |
| "num_tokens": 23476340.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 1.2630186223983764, |
| "eval_biology_loss": 1.2595529556274414, |
| "eval_biology_mean_token_accuracy": 0.6963475484848023, |
| "eval_biology_num_tokens": 23476340.0, |
| "eval_biology_runtime": 38.6878, |
| "eval_biology_samples_per_second": 12.924, |
| "eval_biology_steps_per_second": 3.231, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.6685261228084565, |
| "eval_physics_loss": 0.6503589749336243, |
| "eval_physics_mean_token_accuracy": 0.8208199772834778, |
| "eval_physics_num_tokens": 23476340.0, |
| "eval_physics_runtime": 56.9337, |
| "eval_physics_samples_per_second": 8.782, |
| "eval_physics_steps_per_second": 2.196, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.2681011691689492, |
| "epoch": 1.936, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 1.2121, |
| "mean_token_accuracy": 0.6959899343550205, |
| "num_tokens": 23673208.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.29563331566751, |
| "epoch": 1.952, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 1.2335, |
| "mean_token_accuracy": 0.6913977723568678, |
| "num_tokens": 23870993.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.315883708000183, |
| "epoch": 1.968, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 1.269, |
| "mean_token_accuracy": 0.6871387537568807, |
| "num_tokens": 24066234.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.2918797850608825, |
| "epoch": 1.984, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 1.2409, |
| "mean_token_accuracy": 0.690430635213852, |
| "num_tokens": 24262389.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.2895139936357736, |
| "epoch": 2.0, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 1.2465, |
| "mean_token_accuracy": 0.6922377996146679, |
| "num_tokens": 24455544.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 1.1053394179791212, |
| "epoch": 2.016, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 1.0366, |
| "mean_token_accuracy": 0.7227501507848502, |
| "num_tokens": 24645425.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 1.0759795285761355, |
| "epoch": 2.032, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 1.0324, |
| "mean_token_accuracy": 0.7241251844912767, |
| "num_tokens": 24835956.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 1.065844392031431, |
| "epoch": 2.048, |
| "grad_norm": 3.59375, |
| "learning_rate": 1.938e-05, |
| "loss": 0.9998, |
| "mean_token_accuracy": 0.7277645654976368, |
| "num_tokens": 25036470.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 1.0796132825315, |
| "epoch": 2.064, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 1.0147, |
| "mean_token_accuracy": 0.7255880139768124, |
| "num_tokens": 25230959.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 1.0651828214526176, |
| "epoch": 2.08, |
| "grad_norm": 3.25, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 1.0108, |
| "mean_token_accuracy": 0.727525070682168, |
| "num_tokens": 25422625.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 1.0677382831573485, |
| "eval_biology_loss": 1.290307641029358, |
| "eval_biology_mean_token_accuracy": 0.6951840238571168, |
| "eval_biology_num_tokens": 25422625.0, |
| "eval_biology_runtime": 38.6809, |
| "eval_biology_samples_per_second": 12.926, |
| "eval_biology_steps_per_second": 3.232, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.5783973336219788, |
| "eval_physics_loss": 0.6678071022033691, |
| "eval_physics_mean_token_accuracy": 0.8184197754859924, |
| "eval_physics_num_tokens": 25422625.0, |
| "eval_physics_runtime": 56.9183, |
| "eval_physics_samples_per_second": 8.785, |
| "eval_physics_steps_per_second": 2.196, |
| "step": 1300 |
| }, |
| { |
| "entropy": 1.0735113076865672, |
| "epoch": 2.096, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 1.0185, |
| "mean_token_accuracy": 0.7264322709292174, |
| "num_tokens": 25614286.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 1.0468817871063947, |
| "epoch": 2.112, |
| "grad_norm": 3.015625, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.9933, |
| "mean_token_accuracy": 0.7301209557801485, |
| "num_tokens": 25810384.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 1.057022250816226, |
| "epoch": 2.128, |
| "grad_norm": 3.5, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 1.0037, |
| "mean_token_accuracy": 0.7284754924476147, |
| "num_tokens": 26005047.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 1.0537616554647684, |
| "epoch": 2.144, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 1.0101, |
| "mean_token_accuracy": 0.7281042989343405, |
| "num_tokens": 26207169.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 1.0744788832962513, |
| "epoch": 2.16, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 1.0203, |
| "mean_token_accuracy": 0.7265679869800806, |
| "num_tokens": 26402150.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 1.0806033357977867, |
| "epoch": 2.176, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 1.0319, |
| "mean_token_accuracy": 0.7219661567360163, |
| "num_tokens": 26599464.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 1.0509729556739331, |
| "epoch": 2.192, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.918e-05, |
| "loss": 0.9872, |
| "mean_token_accuracy": 0.7306252598762513, |
| "num_tokens": 26790601.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 1.068820282816887, |
| "epoch": 2.208, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 1.0224, |
| "mean_token_accuracy": 0.7255501691251993, |
| "num_tokens": 26988598.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 1.0967998839914799, |
| "epoch": 2.224, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 1.0393, |
| "mean_token_accuracy": 0.7217595972120762, |
| "num_tokens": 27181933.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 1.0403927110135556, |
| "epoch": 2.24, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.982, |
| "mean_token_accuracy": 0.7317505519837141, |
| "num_tokens": 27378686.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 1.053062720298767, |
| "eval_biology_loss": 1.2769854068756104, |
| "eval_biology_mean_token_accuracy": 0.6977940616607666, |
| "eval_biology_num_tokens": 27378686.0, |
| "eval_biology_runtime": 38.6188, |
| "eval_biology_samples_per_second": 12.947, |
| "eval_biology_steps_per_second": 3.237, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.5554806351661682, |
| "eval_physics_loss": 0.6756051182746887, |
| "eval_physics_mean_token_accuracy": 0.8174587740898133, |
| "eval_physics_num_tokens": 27378686.0, |
| "eval_physics_runtime": 56.8627, |
| "eval_physics_samples_per_second": 8.793, |
| "eval_physics_steps_per_second": 2.198, |
| "step": 1400 |
| }, |
| { |
| "entropy": 1.0398972254246472, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.9991, |
| "mean_token_accuracy": 0.7316640116274357, |
| "num_tokens": 27571908.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 1.0447341743856668, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.9972, |
| "mean_token_accuracy": 0.728259452059865, |
| "num_tokens": 27770478.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 1.0823504395782948, |
| "epoch": 2.288, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 1.0252, |
| "mean_token_accuracy": 0.7220408860594034, |
| "num_tokens": 27965558.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 1.0753419052809476, |
| "epoch": 2.304, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 1.0203, |
| "mean_token_accuracy": 0.7257058396935463, |
| "num_tokens": 28165624.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 1.0463432766497136, |
| "epoch": 2.32, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.9989, |
| "mean_token_accuracy": 0.731269496306777, |
| "num_tokens": 28363594.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 1.0562681667506695, |
| "epoch": 2.336, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.898e-05, |
| "loss": 1.0089, |
| "mean_token_accuracy": 0.7288151986896991, |
| "num_tokens": 28558104.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 1.07792302146554, |
| "epoch": 2.352, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 1.0219, |
| "mean_token_accuracy": 0.7233119513839483, |
| "num_tokens": 28747265.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 1.06618184261024, |
| "epoch": 2.368, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 1.0027, |
| "mean_token_accuracy": 0.7261252999305725, |
| "num_tokens": 28944694.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 1.082535107806325, |
| "epoch": 2.384, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 1.0256, |
| "mean_token_accuracy": 0.7226470891386271, |
| "num_tokens": 29136617.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 1.0529342222958804, |
| "epoch": 2.4, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.9972, |
| "mean_token_accuracy": 0.7272268772125244, |
| "num_tokens": 29333526.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 1.046908447265625, |
| "eval_biology_loss": 1.2635581493377686, |
| "eval_biology_mean_token_accuracy": 0.7004145407676696, |
| "eval_biology_num_tokens": 29333526.0, |
| "eval_biology_runtime": 38.7807, |
| "eval_biology_samples_per_second": 12.893, |
| "eval_biology_steps_per_second": 3.223, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.5629296214580536, |
| "eval_physics_loss": 0.6746402978897095, |
| "eval_physics_mean_token_accuracy": 0.8170881199836731, |
| "eval_physics_num_tokens": 29333526.0, |
| "eval_physics_runtime": 56.9496, |
| "eval_physics_samples_per_second": 8.78, |
| "eval_physics_steps_per_second": 2.195, |
| "step": 1500 |
| }, |
| { |
| "entropy": 1.0541506320238114, |
| "epoch": 2.416, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 1.0065, |
| "mean_token_accuracy": 0.7273991428315639, |
| "num_tokens": 29534153.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 1.053679431974888, |
| "epoch": 2.432, |
| "grad_norm": 3.328125, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.9968, |
| "mean_token_accuracy": 0.7284378662705422, |
| "num_tokens": 29736571.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 1.0887313183397054, |
| "epoch": 2.448, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 1.0429, |
| "mean_token_accuracy": 0.7227712966501713, |
| "num_tokens": 29932423.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 1.0347874112427236, |
| "epoch": 2.464, |
| "grad_norm": 3.640625, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.9796, |
| "mean_token_accuracy": 0.7323532313108444, |
| "num_tokens": 30137125.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 1.048006160184741, |
| "epoch": 2.48, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.878e-05, |
| "loss": 1.001, |
| "mean_token_accuracy": 0.7286384627223015, |
| "num_tokens": 30341656.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 1.052099209651351, |
| "epoch": 2.496, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.9888, |
| "mean_token_accuracy": 0.7288977019488811, |
| "num_tokens": 30533732.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 1.0728581339120864, |
| "epoch": 2.512, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 1.0181, |
| "mean_token_accuracy": 0.7265560120344162, |
| "num_tokens": 30730599.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 1.0553664829581977, |
| "epoch": 2.528, |
| "grad_norm": 3.640625, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.9993, |
| "mean_token_accuracy": 0.7286464177072048, |
| "num_tokens": 30927950.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 1.0592417079955339, |
| "epoch": 2.544, |
| "grad_norm": 3.46875, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 1.0135, |
| "mean_token_accuracy": 0.7272635698318481, |
| "num_tokens": 31125667.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 1.0688998382538557, |
| "epoch": 2.56, |
| "grad_norm": 3.578125, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 1.0172, |
| "mean_token_accuracy": 0.7267541728913784, |
| "num_tokens": 31315742.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 1.0306266536712647, |
| "eval_biology_loss": 1.2540794610977173, |
| "eval_biology_mean_token_accuracy": 0.7014414758682251, |
| "eval_biology_num_tokens": 31315742.0, |
| "eval_biology_runtime": 38.994, |
| "eval_biology_samples_per_second": 12.822, |
| "eval_biology_steps_per_second": 3.206, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.5480240859985351, |
| "eval_physics_loss": 0.6785013675689697, |
| "eval_physics_mean_token_accuracy": 0.81651726770401, |
| "eval_physics_num_tokens": 31315742.0, |
| "eval_physics_runtime": 56.9063, |
| "eval_physics_samples_per_second": 8.786, |
| "eval_physics_steps_per_second": 2.197, |
| "step": 1600 |
| }, |
| { |
| "entropy": 1.072851437330246, |
| "epoch": 2.576, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 1.0177, |
| "mean_token_accuracy": 0.7259860239923001, |
| "num_tokens": 31505374.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 1.0911020763218402, |
| "epoch": 2.592, |
| "grad_norm": 3.375, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 1.0375, |
| "mean_token_accuracy": 0.7193623151630163, |
| "num_tokens": 31701680.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 1.0759818445891143, |
| "epoch": 2.608, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 1.024, |
| "mean_token_accuracy": 0.7245077144354581, |
| "num_tokens": 31893221.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 1.0434412751346827, |
| "epoch": 2.624, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.858e-05, |
| "loss": 0.9884, |
| "mean_token_accuracy": 0.7292033292353153, |
| "num_tokens": 32087176.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 1.077140349522233, |
| "epoch": 2.64, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 1.0218, |
| "mean_token_accuracy": 0.7251086611300707, |
| "num_tokens": 32280689.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 1.0449730832129718, |
| "epoch": 2.656, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.9897, |
| "mean_token_accuracy": 0.7329251658171415, |
| "num_tokens": 32483392.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 1.0476958800107241, |
| "epoch": 2.672, |
| "grad_norm": 3.328125, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.9996, |
| "mean_token_accuracy": 0.7290864728391171, |
| "num_tokens": 32683246.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 1.0343516409397124, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.9794, |
| "mean_token_accuracy": 0.7327864274382592, |
| "num_tokens": 32884993.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 1.059150441735983, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 3.25, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 1.004, |
| "mean_token_accuracy": 0.7282994613051414, |
| "num_tokens": 33088388.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 1.0319584112614393, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 3.53125, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.9858, |
| "mean_token_accuracy": 0.7318238351494074, |
| "num_tokens": 33287208.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 1.0519089736938476, |
| "eval_biology_loss": 1.2423198223114014, |
| "eval_biology_mean_token_accuracy": 0.7037161078453064, |
| "eval_biology_num_tokens": 33287208.0, |
| "eval_biology_runtime": 38.8171, |
| "eval_biology_samples_per_second": 12.881, |
| "eval_biology_steps_per_second": 3.22, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.5594763560295105, |
| "eval_physics_loss": 0.677318811416626, |
| "eval_physics_mean_token_accuracy": 0.8168268208503723, |
| "eval_physics_num_tokens": 33287208.0, |
| "eval_physics_runtime": 57.185, |
| "eval_physics_samples_per_second": 8.744, |
| "eval_physics_steps_per_second": 2.186, |
| "step": 1700 |
| }, |
| { |
| "entropy": 1.0736609652638436, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 3.15625, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 1.017, |
| "mean_token_accuracy": 0.7253531437367201, |
| "num_tokens": 33474750.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 1.0549831412732602, |
| "epoch": 2.752, |
| "grad_norm": 3.625, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 1.0054, |
| "mean_token_accuracy": 0.7258293054997921, |
| "num_tokens": 33664302.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 1.0876980781555177, |
| "epoch": 2.768, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.0317, |
| "mean_token_accuracy": 0.7200849443674088, |
| "num_tokens": 33854081.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 1.046105058863759, |
| "epoch": 2.784, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.9956, |
| "mean_token_accuracy": 0.7304057106375694, |
| "num_tokens": 34053149.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 1.0452948212623596, |
| "epoch": 2.8, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.9964, |
| "mean_token_accuracy": 0.7318151947110891, |
| "num_tokens": 34254105.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 1.0454461514949798, |
| "epoch": 2.816, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.9864, |
| "mean_token_accuracy": 0.7298544138669968, |
| "num_tokens": 34453342.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 1.0891260366886855, |
| "epoch": 2.832, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 1.0274, |
| "mean_token_accuracy": 0.7241302687674761, |
| "num_tokens": 34644113.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 1.0674222145229577, |
| "epoch": 2.848, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 1.0191, |
| "mean_token_accuracy": 0.7263091344386339, |
| "num_tokens": 34835062.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 1.047906568646431, |
| "epoch": 2.864, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 1.0015, |
| "mean_token_accuracy": 0.7286620266735554, |
| "num_tokens": 35032361.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 1.0795419167727232, |
| "epoch": 2.88, |
| "grad_norm": 3.421875, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 1.0231, |
| "mean_token_accuracy": 0.7250057961791754, |
| "num_tokens": 35226653.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 1.0592120084762573, |
| "eval_biology_loss": 1.2307487726211548, |
| "eval_biology_mean_token_accuracy": 0.7039423842430115, |
| "eval_biology_num_tokens": 35226653.0, |
| "eval_biology_runtime": 38.811, |
| "eval_biology_samples_per_second": 12.883, |
| "eval_biology_steps_per_second": 3.221, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.5607087211608887, |
| "eval_physics_loss": 0.6777584552764893, |
| "eval_physics_mean_token_accuracy": 0.816848436832428, |
| "eval_physics_num_tokens": 35226653.0, |
| "eval_physics_runtime": 56.9878, |
| "eval_physics_samples_per_second": 8.774, |
| "eval_physics_steps_per_second": 2.193, |
| "step": 1800 |
| }, |
| { |
| "entropy": 1.0618976611644029, |
| "epoch": 2.896, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 1.0068, |
| "mean_token_accuracy": 0.7253085561096668, |
| "num_tokens": 35421873.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 1.0810394674539565, |
| "epoch": 2.912, |
| "grad_norm": 3.6875, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.0287, |
| "mean_token_accuracy": 0.7242106914520263, |
| "num_tokens": 35614590.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 1.0641514841467141, |
| "epoch": 2.928, |
| "grad_norm": 3.578125, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 1.0163, |
| "mean_token_accuracy": 0.7251328159123659, |
| "num_tokens": 35806494.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 1.0547591600567103, |
| "epoch": 2.944, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 1.0087, |
| "mean_token_accuracy": 0.7291445326060056, |
| "num_tokens": 36001867.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 1.0459509972482919, |
| "epoch": 2.96, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.9846, |
| "mean_token_accuracy": 0.7305242072790861, |
| "num_tokens": 36198806.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 1.0438212916254996, |
| "epoch": 2.976, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.9897, |
| "mean_token_accuracy": 0.7307346038520336, |
| "num_tokens": 36393098.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 1.0572807922959329, |
| "epoch": 2.992, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 1.005, |
| "mean_token_accuracy": 0.728681518137455, |
| "num_tokens": 36589074.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.9687618412077427, |
| "epoch": 3.008, |
| "grad_norm": 4.75, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.8593, |
| "mean_token_accuracy": 0.7585306584835052, |
| "num_tokens": 36778710.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.7038324676454067, |
| "epoch": 3.024, |
| "grad_norm": 3.671875, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.666, |
| "mean_token_accuracy": 0.79880424849689, |
| "num_tokens": 36966789.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.6823407102376222, |
| "epoch": 3.04, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.621, |
| "mean_token_accuracy": 0.8089036166667938, |
| "num_tokens": 37156529.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 0.7511428904533386, |
| "eval_biology_loss": 1.4531614780426025, |
| "eval_biology_mean_token_accuracy": 0.6934510517120361, |
| "eval_biology_num_tokens": 37156529.0, |
| "eval_biology_runtime": 38.62, |
| "eval_biology_samples_per_second": 12.947, |
| "eval_biology_steps_per_second": 3.237, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.4265694625377655, |
| "eval_physics_loss": 0.781083881855011, |
| "eval_physics_mean_token_accuracy": 0.8090727143287658, |
| "eval_physics_num_tokens": 37156529.0, |
| "eval_physics_runtime": 56.8784, |
| "eval_physics_samples_per_second": 8.791, |
| "eval_physics_steps_per_second": 2.198, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.6869338469579815, |
| "epoch": 3.056, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.798e-05, |
| "loss": 0.6247, |
| "mean_token_accuracy": 0.8083337739109993, |
| "num_tokens": 37351325.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.6717990661039949, |
| "epoch": 3.072, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.6161, |
| "mean_token_accuracy": 0.8123877976089716, |
| "num_tokens": 37541069.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.6788237554952502, |
| "epoch": 3.088, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.6233, |
| "mean_token_accuracy": 0.8102868665009737, |
| "num_tokens": 37733440.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.6710379313677549, |
| "epoch": 3.104, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.6095, |
| "mean_token_accuracy": 0.811876380816102, |
| "num_tokens": 37933661.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.6801185900345444, |
| "epoch": 3.12, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.6255, |
| "mean_token_accuracy": 0.8087910477072, |
| "num_tokens": 38126905.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.6823285862803459, |
| "epoch": 3.136, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.619, |
| "mean_token_accuracy": 0.8105869103223086, |
| "num_tokens": 38316042.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.6686201065778732, |
| "epoch": 3.152, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.6139, |
| "mean_token_accuracy": 0.811134846508503, |
| "num_tokens": 38514212.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.696059150248766, |
| "epoch": 3.168, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.6377, |
| "mean_token_accuracy": 0.8045505676418543, |
| "num_tokens": 38708822.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.6871348209679127, |
| "epoch": 3.184, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.632, |
| "mean_token_accuracy": 0.8063865445554257, |
| "num_tokens": 38900708.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.6824637034907937, |
| "epoch": 3.2, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.626, |
| "mean_token_accuracy": 0.8084224767982959, |
| "num_tokens": 39102325.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 0.7477709612846375, |
| "eval_biology_loss": 1.473409652709961, |
| "eval_biology_mean_token_accuracy": 0.6936884145736695, |
| "eval_biology_num_tokens": 39102325.0, |
| "eval_biology_runtime": 38.6411, |
| "eval_biology_samples_per_second": 12.94, |
| "eval_biology_steps_per_second": 3.235, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.4129572913646698, |
| "eval_physics_loss": 0.8035010695457458, |
| "eval_physics_mean_token_accuracy": 0.8071833181381226, |
| "eval_physics_num_tokens": 39102325.0, |
| "eval_physics_runtime": 56.8955, |
| "eval_physics_samples_per_second": 8.788, |
| "eval_physics_steps_per_second": 2.197, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 16, |
| "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": 3.279038413131436e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
