| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 2571, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.88671875, |
| "epoch": 0.011668611435239206, |
| "grad_norm": 3.6080080932769607, |
| "learning_rate": 6.976744186046511e-09, |
| "loss": 1.0903, |
| "mean_token_accuracy": 0.7742526292800903, |
| "num_tokens": 254566.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.894921875, |
| "epoch": 0.023337222870478413, |
| "grad_norm": 4.67772028725951, |
| "learning_rate": 1.4728682170542636e-08, |
| "loss": 1.0551, |
| "mean_token_accuracy": 0.7796614915132523, |
| "num_tokens": 494709.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.8853515625, |
| "epoch": 0.03500583430571762, |
| "grad_norm": 4.379003948551158, |
| "learning_rate": 2.248062015503876e-08, |
| "loss": 1.1326, |
| "mean_token_accuracy": 0.7659079849720001, |
| "num_tokens": 713648.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.9412109375, |
| "epoch": 0.046674445740956826, |
| "grad_norm": 4.993354359311264, |
| "learning_rate": 3.023255813953488e-08, |
| "loss": 1.1758, |
| "mean_token_accuracy": 0.7609771907329559, |
| "num_tokens": 918922.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.877734375, |
| "epoch": 0.058343057176196034, |
| "grad_norm": 3.3322654341290545, |
| "learning_rate": 3.7984496124031005e-08, |
| "loss": 1.1344, |
| "mean_token_accuracy": 0.7650036484003067, |
| "num_tokens": 1157460.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8763671875, |
| "epoch": 0.07001166861143523, |
| "grad_norm": 4.0137274659845135, |
| "learning_rate": 4.573643410852713e-08, |
| "loss": 1.102, |
| "mean_token_accuracy": 0.7667693287134171, |
| "num_tokens": 1391171.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.9154296875, |
| "epoch": 0.08168028004667445, |
| "grad_norm": 4.862708876936719, |
| "learning_rate": 5.348837209302325e-08, |
| "loss": 1.2146, |
| "mean_token_accuracy": 0.7537672340869903, |
| "num_tokens": 1587034.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.855859375, |
| "epoch": 0.09334889148191365, |
| "grad_norm": 2.6239840887365773, |
| "learning_rate": 6.124031007751938e-08, |
| "loss": 1.085, |
| "mean_token_accuracy": 0.7720604538917542, |
| "num_tokens": 1855100.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.891015625, |
| "epoch": 0.10501750291715285, |
| "grad_norm": 2.973282890785539, |
| "learning_rate": 6.89922480620155e-08, |
| "loss": 1.1208, |
| "mean_token_accuracy": 0.7673993885517121, |
| "num_tokens": 2085112.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.9234375, |
| "epoch": 0.11668611435239207, |
| "grad_norm": 3.2359060103195847, |
| "learning_rate": 7.674418604651163e-08, |
| "loss": 1.1592, |
| "mean_token_accuracy": 0.7620196729898453, |
| "num_tokens": 2311360.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.9013671875, |
| "epoch": 0.12835472578763127, |
| "grad_norm": 5.187549002171917, |
| "learning_rate": 8.449612403100774e-08, |
| "loss": 1.1213, |
| "mean_token_accuracy": 0.7649570018053055, |
| "num_tokens": 2532845.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.876953125, |
| "epoch": 0.14002333722287047, |
| "grad_norm": 3.8614501464636883, |
| "learning_rate": 9.224806201550387e-08, |
| "loss": 1.0691, |
| "mean_token_accuracy": 0.7747613906860351, |
| "num_tokens": 2757235.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.87421875, |
| "epoch": 0.1516919486581097, |
| "grad_norm": 3.9167215531492126, |
| "learning_rate": 1e-07, |
| "loss": 1.0801, |
| "mean_token_accuracy": 0.7733830362558365, |
| "num_tokens": 2988877.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.9126953125, |
| "epoch": 0.1633605600933489, |
| "grad_norm": 4.611622597457114, |
| "learning_rate": 1.077519379844961e-07, |
| "loss": 1.1657, |
| "mean_token_accuracy": 0.7569267064332962, |
| "num_tokens": 3205857.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.9212890625, |
| "epoch": 0.1750291715285881, |
| "grad_norm": 2.72070625776444, |
| "learning_rate": 1.1550387596899225e-07, |
| "loss": 1.0573, |
| "mean_token_accuracy": 0.7718144595623017, |
| "num_tokens": 3441131.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8599609375, |
| "epoch": 0.1866977829638273, |
| "grad_norm": 4.033564986530464, |
| "learning_rate": 1.2325581395348835e-07, |
| "loss": 0.997, |
| "mean_token_accuracy": 0.781052827835083, |
| "num_tokens": 3690901.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.9373046875, |
| "epoch": 0.1983663943990665, |
| "grad_norm": 2.8745592477568214, |
| "learning_rate": 1.3100775193798451e-07, |
| "loss": 1.0512, |
| "mean_token_accuracy": 0.7725604116916657, |
| "num_tokens": 3916710.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.91875, |
| "epoch": 0.2100350058343057, |
| "grad_norm": 3.6584187527962344, |
| "learning_rate": 1.3875968992248062e-07, |
| "loss": 1.0793, |
| "mean_token_accuracy": 0.765195780992508, |
| "num_tokens": 4137899.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8421875, |
| "epoch": 0.22170361726954493, |
| "grad_norm": 3.0859574408080266, |
| "learning_rate": 1.4651162790697673e-07, |
| "loss": 0.9577, |
| "mean_token_accuracy": 0.7878951787948608, |
| "num_tokens": 4396754.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.91484375, |
| "epoch": 0.23337222870478413, |
| "grad_norm": 4.0532827924941754, |
| "learning_rate": 1.5426356589147287e-07, |
| "loss": 1.0154, |
| "mean_token_accuracy": 0.7769698202610016, |
| "num_tokens": 4619096.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.884375, |
| "epoch": 0.24504084014002334, |
| "grad_norm": 3.5928931037629246, |
| "learning_rate": 1.6201550387596898e-07, |
| "loss": 0.9392, |
| "mean_token_accuracy": 0.7889974921941757, |
| "num_tokens": 4838862.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.9412109375, |
| "epoch": 0.25670945157526254, |
| "grad_norm": 3.2089579044686984, |
| "learning_rate": 1.697674418604651e-07, |
| "loss": 0.9215, |
| "mean_token_accuracy": 0.7862808167934418, |
| "num_tokens": 5057494.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.8984375, |
| "epoch": 0.26837806301050177, |
| "grad_norm": 3.0691116190169834, |
| "learning_rate": 1.7751937984496123e-07, |
| "loss": 0.8943, |
| "mean_token_accuracy": 0.7916853636503219, |
| "num_tokens": 5300840.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.8431640625, |
| "epoch": 0.28004667444574094, |
| "grad_norm": 3.737002727055304, |
| "learning_rate": 1.8527131782945736e-07, |
| "loss": 0.8154, |
| "mean_token_accuracy": 0.8023237615823746, |
| "num_tokens": 5554598.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.8529296875, |
| "epoch": 0.29171528588098017, |
| "grad_norm": 2.6923833826145405, |
| "learning_rate": 1.930232558139535e-07, |
| "loss": 0.8453, |
| "mean_token_accuracy": 0.7966215431690216, |
| "num_tokens": 5784536.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7712890625, |
| "epoch": 0.3033838973162194, |
| "grad_norm": 3.3474224741372502, |
| "learning_rate": 1.9991353220925205e-07, |
| "loss": 0.7672, |
| "mean_token_accuracy": 0.813620638847351, |
| "num_tokens": 6047008.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.8587890625, |
| "epoch": 0.31505250875145857, |
| "grad_norm": 2.786477330167304, |
| "learning_rate": 1.9904885430177258e-07, |
| "loss": 0.794, |
| "mean_token_accuracy": 0.8021291077136994, |
| "num_tokens": 6261330.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.79609375, |
| "epoch": 0.3267211201866978, |
| "grad_norm": 2.4679664525984326, |
| "learning_rate": 1.9818417639429312e-07, |
| "loss": 0.757, |
| "mean_token_accuracy": 0.8114175617694854, |
| "num_tokens": 6482724.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.7783203125, |
| "epoch": 0.338389731621937, |
| "grad_norm": 2.558203148388588, |
| "learning_rate": 1.9731949848681366e-07, |
| "loss": 0.7082, |
| "mean_token_accuracy": 0.8224216103553772, |
| "num_tokens": 6713489.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7888671875, |
| "epoch": 0.3500583430571762, |
| "grad_norm": 2.1772672136231104, |
| "learning_rate": 1.9645482057933417e-07, |
| "loss": 0.734, |
| "mean_token_accuracy": 0.8173721730709076, |
| "num_tokens": 6922995.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7560546875, |
| "epoch": 0.3617269544924154, |
| "grad_norm": 2.1435739756993164, |
| "learning_rate": 1.955901426718547e-07, |
| "loss": 0.7044, |
| "mean_token_accuracy": 0.8226464986801147, |
| "num_tokens": 7138620.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.7416015625, |
| "epoch": 0.3733955659276546, |
| "grad_norm": 2.0894737901361666, |
| "learning_rate": 1.9472546476437527e-07, |
| "loss": 0.6985, |
| "mean_token_accuracy": 0.8249520629644393, |
| "num_tokens": 7349920.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.7126953125, |
| "epoch": 0.38506417736289383, |
| "grad_norm": 1.9392442613728555, |
| "learning_rate": 1.938607868568958e-07, |
| "loss": 0.6763, |
| "mean_token_accuracy": 0.8288130193948746, |
| "num_tokens": 7573495.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.7271484375, |
| "epoch": 0.396732788798133, |
| "grad_norm": 1.658755902363955, |
| "learning_rate": 1.9299610894941635e-07, |
| "loss": 0.6795, |
| "mean_token_accuracy": 0.8278997927904129, |
| "num_tokens": 7796467.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.6904296875, |
| "epoch": 0.40840140023337224, |
| "grad_norm": 1.383050584833463, |
| "learning_rate": 1.9213143104193688e-07, |
| "loss": 0.6568, |
| "mean_token_accuracy": 0.8330555349588394, |
| "num_tokens": 8027438.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.6525390625, |
| "epoch": 0.4200700116686114, |
| "grad_norm": 0.9650432762082423, |
| "learning_rate": 1.9126675313445742e-07, |
| "loss": 0.6174, |
| "mean_token_accuracy": 0.8411101251840591, |
| "num_tokens": 8286417.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.6767578125, |
| "epoch": 0.43173862310385064, |
| "grad_norm": 1.0003620196260086, |
| "learning_rate": 1.9040207522697793e-07, |
| "loss": 0.6392, |
| "mean_token_accuracy": 0.8370698988437653, |
| "num_tokens": 8497089.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.6603515625, |
| "epoch": 0.44340723453908987, |
| "grad_norm": 1.1924960866089696, |
| "learning_rate": 1.8953739731949847e-07, |
| "loss": 0.6386, |
| "mean_token_accuracy": 0.8361361563205719, |
| "num_tokens": 8711822.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.6767578125, |
| "epoch": 0.45507584597432904, |
| "grad_norm": 0.8863022029681805, |
| "learning_rate": 1.88672719412019e-07, |
| "loss": 0.652, |
| "mean_token_accuracy": 0.8311549305915833, |
| "num_tokens": 8938553.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.633984375, |
| "epoch": 0.46674445740956827, |
| "grad_norm": 0.8435417013428951, |
| "learning_rate": 1.8780804150453954e-07, |
| "loss": 0.6011, |
| "mean_token_accuracy": 0.8425446510314941, |
| "num_tokens": 9210424.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.6443359375, |
| "epoch": 0.47841306884480744, |
| "grad_norm": 0.8580123634338545, |
| "learning_rate": 1.8694336359706008e-07, |
| "loss": 0.6224, |
| "mean_token_accuracy": 0.838255450129509, |
| "num_tokens": 9450850.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.6400390625, |
| "epoch": 0.49008168028004667, |
| "grad_norm": 1.0637818778036898, |
| "learning_rate": 1.8607868568958062e-07, |
| "loss": 0.6193, |
| "mean_token_accuracy": 0.8387202978134155, |
| "num_tokens": 9689411.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.644921875, |
| "epoch": 0.5017502917152858, |
| "grad_norm": 1.0063166867599815, |
| "learning_rate": 1.8521400778210115e-07, |
| "loss": 0.608, |
| "mean_token_accuracy": 0.8414727926254273, |
| "num_tokens": 9943210.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.62421875, |
| "epoch": 0.5134189031505251, |
| "grad_norm": 0.8184605065948258, |
| "learning_rate": 1.843493298746217e-07, |
| "loss": 0.5912, |
| "mean_token_accuracy": 0.8459764152765274, |
| "num_tokens": 10185134.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.657421875, |
| "epoch": 0.5250875145857643, |
| "grad_norm": 0.8381842906003937, |
| "learning_rate": 1.8348465196714223e-07, |
| "loss": 0.6372, |
| "mean_token_accuracy": 0.8356127142906189, |
| "num_tokens": 10430571.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.6337890625, |
| "epoch": 0.5367561260210035, |
| "grad_norm": 1.1661960687869792, |
| "learning_rate": 1.8261997405966277e-07, |
| "loss": 0.6002, |
| "mean_token_accuracy": 0.845330348610878, |
| "num_tokens": 10642769.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.59951171875, |
| "epoch": 0.5484247374562428, |
| "grad_norm": 0.9271960502602317, |
| "learning_rate": 1.817552961521833e-07, |
| "loss": 0.5816, |
| "mean_token_accuracy": 0.8478394538164139, |
| "num_tokens": 10871702.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.6298828125, |
| "epoch": 0.5600933488914819, |
| "grad_norm": 0.8601480377641353, |
| "learning_rate": 1.8089061824470384e-07, |
| "loss": 0.5983, |
| "mean_token_accuracy": 0.8441305816173553, |
| "num_tokens": 11086221.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.65234375, |
| "epoch": 0.5717619603267211, |
| "grad_norm": 1.133103034000805, |
| "learning_rate": 1.8002594033722438e-07, |
| "loss": 0.6194, |
| "mean_token_accuracy": 0.8399488240480423, |
| "num_tokens": 11297451.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.6349609375, |
| "epoch": 0.5834305717619603, |
| "grad_norm": 1.2150447819710926, |
| "learning_rate": 1.7916126242974491e-07, |
| "loss": 0.6089, |
| "mean_token_accuracy": 0.8407108932733536, |
| "num_tokens": 11528722.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.6544921875, |
| "epoch": 0.5950991831971996, |
| "grad_norm": 1.145061927973259, |
| "learning_rate": 1.7829658452226545e-07, |
| "loss": 0.6289, |
| "mean_token_accuracy": 0.8377750784158706, |
| "num_tokens": 11729846.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.62265625, |
| "epoch": 0.6067677946324388, |
| "grad_norm": 1.2209891545018745, |
| "learning_rate": 1.7743190661478596e-07, |
| "loss": 0.5899, |
| "mean_token_accuracy": 0.8470842450857162, |
| "num_tokens": 11949454.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.60390625, |
| "epoch": 0.6184364060676779, |
| "grad_norm": 0.829259857793348, |
| "learning_rate": 1.7656722870730653e-07, |
| "loss": 0.5779, |
| "mean_token_accuracy": 0.8504338830709457, |
| "num_tokens": 12184931.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.5990234375, |
| "epoch": 0.6301050175029171, |
| "grad_norm": 0.9497377077030208, |
| "learning_rate": 1.7570255079982706e-07, |
| "loss": 0.5565, |
| "mean_token_accuracy": 0.8523817807435989, |
| "num_tokens": 12408293.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.61015625, |
| "epoch": 0.6417736289381564, |
| "grad_norm": 1.059510938316246, |
| "learning_rate": 1.748378728923476e-07, |
| "loss": 0.5826, |
| "mean_token_accuracy": 0.8484628945589066, |
| "num_tokens": 12622069.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.610546875, |
| "epoch": 0.6534422403733956, |
| "grad_norm": 1.167113128375012, |
| "learning_rate": 1.7397319498486814e-07, |
| "loss": 0.5887, |
| "mean_token_accuracy": 0.8465761065483093, |
| "num_tokens": 12843568.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.5943359375, |
| "epoch": 0.6651108518086347, |
| "grad_norm": 0.7745163180830756, |
| "learning_rate": 1.7310851707738868e-07, |
| "loss": 0.5754, |
| "mean_token_accuracy": 0.8500663459300994, |
| "num_tokens": 13123081.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.60810546875, |
| "epoch": 0.676779463243874, |
| "grad_norm": 1.1596670328373013, |
| "learning_rate": 1.722438391699092e-07, |
| "loss": 0.5817, |
| "mean_token_accuracy": 0.8473448395729065, |
| "num_tokens": 13365614.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.610546875, |
| "epoch": 0.6884480746791132, |
| "grad_norm": 1.154287886290439, |
| "learning_rate": 1.7137916126242972e-07, |
| "loss": 0.5933, |
| "mean_token_accuracy": 0.8445927232503891, |
| "num_tokens": 13598488.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6158203125, |
| "epoch": 0.7001166861143524, |
| "grad_norm": 1.3178950290555602, |
| "learning_rate": 1.7051448335495026e-07, |
| "loss": 0.5806, |
| "mean_token_accuracy": 0.8462684005498886, |
| "num_tokens": 13804279.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.599609375, |
| "epoch": 0.7117852975495916, |
| "grad_norm": 0.9624061606614607, |
| "learning_rate": 1.696498054474708e-07, |
| "loss": 0.5672, |
| "mean_token_accuracy": 0.8530819177627563, |
| "num_tokens": 14047828.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.607421875, |
| "epoch": 0.7234539089848308, |
| "grad_norm": 1.1818863446599894, |
| "learning_rate": 1.6878512753999134e-07, |
| "loss": 0.5755, |
| "mean_token_accuracy": 0.8491891711950302, |
| "num_tokens": 14273878.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.5873046875, |
| "epoch": 0.73512252042007, |
| "grad_norm": 0.9598520593156886, |
| "learning_rate": 1.6792044963251187e-07, |
| "loss": 0.552, |
| "mean_token_accuracy": 0.8552615612745285, |
| "num_tokens": 14500012.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.573828125, |
| "epoch": 0.7467911318553092, |
| "grad_norm": 0.8750249032476805, |
| "learning_rate": 1.6705577172503244e-07, |
| "loss": 0.5501, |
| "mean_token_accuracy": 0.8538237124681473, |
| "num_tokens": 14736031.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.57607421875, |
| "epoch": 0.7584597432905484, |
| "grad_norm": 0.8675819982547629, |
| "learning_rate": 1.6619109381755297e-07, |
| "loss": 0.5419, |
| "mean_token_accuracy": 0.8552770107984543, |
| "num_tokens": 14992429.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.5853515625, |
| "epoch": 0.7701283547257877, |
| "grad_norm": 0.9565395364172973, |
| "learning_rate": 1.6532641591007348e-07, |
| "loss": 0.5607, |
| "mean_token_accuracy": 0.8527290880680084, |
| "num_tokens": 15227620.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.591015625, |
| "epoch": 0.7817969661610268, |
| "grad_norm": 1.1578458622367023, |
| "learning_rate": 1.6446173800259402e-07, |
| "loss": 0.5728, |
| "mean_token_accuracy": 0.8488179862499237, |
| "num_tokens": 15463966.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.584375, |
| "epoch": 0.793465577596266, |
| "grad_norm": 0.8737876936181449, |
| "learning_rate": 1.6359706009511456e-07, |
| "loss": 0.527, |
| "mean_token_accuracy": 0.8584627479314804, |
| "num_tokens": 15705712.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.5869140625, |
| "epoch": 0.8051341890315052, |
| "grad_norm": 0.9812167818608712, |
| "learning_rate": 1.627323821876351e-07, |
| "loss": 0.5586, |
| "mean_token_accuracy": 0.8537031769752502, |
| "num_tokens": 15953256.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.5861328125, |
| "epoch": 0.8168028004667445, |
| "grad_norm": 0.9216693008774387, |
| "learning_rate": 1.6186770428015563e-07, |
| "loss": 0.5464, |
| "mean_token_accuracy": 0.8546293020248413, |
| "num_tokens": 16191397.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.55634765625, |
| "epoch": 0.8284714119019837, |
| "grad_norm": 1.1641527912702896, |
| "learning_rate": 1.6100302637267617e-07, |
| "loss": 0.5272, |
| "mean_token_accuracy": 0.861052593588829, |
| "num_tokens": 16454768.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.594921875, |
| "epoch": 0.8401400233372228, |
| "grad_norm": 1.2491634105697131, |
| "learning_rate": 1.601383484651967e-07, |
| "loss": 0.5737, |
| "mean_token_accuracy": 0.8475441306829452, |
| "num_tokens": 16669613.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.572265625, |
| "epoch": 0.851808634772462, |
| "grad_norm": 0.9911570250266466, |
| "learning_rate": 1.5927367055771725e-07, |
| "loss": 0.5487, |
| "mean_token_accuracy": 0.8555569291114807, |
| "num_tokens": 16895746.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.5921875, |
| "epoch": 0.8634772462077013, |
| "grad_norm": 0.8961119033273289, |
| "learning_rate": 1.5840899265023778e-07, |
| "loss": 0.5858, |
| "mean_token_accuracy": 0.8470549911260605, |
| "num_tokens": 17129338.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.55166015625, |
| "epoch": 0.8751458576429405, |
| "grad_norm": 0.7460003163568061, |
| "learning_rate": 1.5754431474275832e-07, |
| "loss": 0.5254, |
| "mean_token_accuracy": 0.8598881781101226, |
| "num_tokens": 17394693.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.55302734375, |
| "epoch": 0.8868144690781797, |
| "grad_norm": 0.8690986716224312, |
| "learning_rate": 1.5667963683527886e-07, |
| "loss": 0.5264, |
| "mean_token_accuracy": 0.8602248638868332, |
| "num_tokens": 17649014.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.544921875, |
| "epoch": 0.8984830805134189, |
| "grad_norm": 0.9977977925950072, |
| "learning_rate": 1.558149589277994e-07, |
| "loss": 0.5273, |
| "mean_token_accuracy": 0.8606024384498596, |
| "num_tokens": 17898184.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.53291015625, |
| "epoch": 0.9101516919486581, |
| "grad_norm": 0.8785408102804774, |
| "learning_rate": 1.5495028102031993e-07, |
| "loss": 0.5179, |
| "mean_token_accuracy": 0.8633781671524048, |
| "num_tokens": 18159169.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.55830078125, |
| "epoch": 0.9218203033838973, |
| "grad_norm": 1.1086430672891179, |
| "learning_rate": 1.5408560311284047e-07, |
| "loss": 0.5362, |
| "mean_token_accuracy": 0.8577522933483124, |
| "num_tokens": 18398687.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.583203125, |
| "epoch": 0.9334889148191365, |
| "grad_norm": 1.0620067323343811, |
| "learning_rate": 1.53220925205361e-07, |
| "loss": 0.5629, |
| "mean_token_accuracy": 0.8510100096464157, |
| "num_tokens": 18628906.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.5697265625, |
| "epoch": 0.9451575262543758, |
| "grad_norm": 1.1638187060960867, |
| "learning_rate": 1.5235624729788152e-07, |
| "loss": 0.5459, |
| "mean_token_accuracy": 0.8565227925777436, |
| "num_tokens": 18829223.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.575, |
| "epoch": 0.9568261376896149, |
| "grad_norm": 0.9289998794570573, |
| "learning_rate": 1.5149156939040205e-07, |
| "loss": 0.5503, |
| "mean_token_accuracy": 0.8538018435239791, |
| "num_tokens": 19056784.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.56669921875, |
| "epoch": 0.9684947491248541, |
| "grad_norm": 1.0151711709172289, |
| "learning_rate": 1.506268914829226e-07, |
| "loss": 0.5341, |
| "mean_token_accuracy": 0.8585916042327881, |
| "num_tokens": 19297326.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.5625, |
| "epoch": 0.9801633605600933, |
| "grad_norm": 0.9429753468924236, |
| "learning_rate": 1.4976221357544313e-07, |
| "loss": 0.5385, |
| "mean_token_accuracy": 0.8597111642360687, |
| "num_tokens": 19551128.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.56767578125, |
| "epoch": 0.9918319719953326, |
| "grad_norm": 0.9520274009237335, |
| "learning_rate": 1.488975356679637e-07, |
| "loss": 0.5382, |
| "mean_token_accuracy": 0.8579822689294815, |
| "num_tokens": 19782079.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.5681640625, |
| "epoch": 1.0035005834305717, |
| "grad_norm": 1.2950489880418383, |
| "learning_rate": 1.4803285776048423e-07, |
| "loss": 0.5417, |
| "mean_token_accuracy": 0.8574281573295593, |
| "num_tokens": 20024595.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.555859375, |
| "epoch": 1.015169194865811, |
| "grad_norm": 0.9783943269073984, |
| "learning_rate": 1.4716817985300477e-07, |
| "loss": 0.5318, |
| "mean_token_accuracy": 0.859162762761116, |
| "num_tokens": 20262937.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.5609375, |
| "epoch": 1.0268378063010501, |
| "grad_norm": 0.8982951974845514, |
| "learning_rate": 1.4630350194552528e-07, |
| "loss": 0.5338, |
| "mean_token_accuracy": 0.8586933076381683, |
| "num_tokens": 20483406.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.5478515625, |
| "epoch": 1.0385064177362895, |
| "grad_norm": 0.8548825563485252, |
| "learning_rate": 1.4543882403804581e-07, |
| "loss": 0.522, |
| "mean_token_accuracy": 0.8596341460943222, |
| "num_tokens": 20724364.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.53681640625, |
| "epoch": 1.0501750291715286, |
| "grad_norm": 1.1036103111651139, |
| "learning_rate": 1.4457414613056635e-07, |
| "loss": 0.5075, |
| "mean_token_accuracy": 0.8639997214078903, |
| "num_tokens": 20968277.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.5716796875, |
| "epoch": 1.0618436406067677, |
| "grad_norm": 1.110568703897574, |
| "learning_rate": 1.437094682230869e-07, |
| "loss": 0.5411, |
| "mean_token_accuracy": 0.8587539702653885, |
| "num_tokens": 21200267.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.56201171875, |
| "epoch": 1.073512252042007, |
| "grad_norm": 1.2849682784296736, |
| "learning_rate": 1.4284479031560743e-07, |
| "loss": 0.539, |
| "mean_token_accuracy": 0.8571189701557159, |
| "num_tokens": 21419056.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.53701171875, |
| "epoch": 1.0851808634772462, |
| "grad_norm": 0.9350766598947755, |
| "learning_rate": 1.4198011240812796e-07, |
| "loss": 0.5187, |
| "mean_token_accuracy": 0.8637947797775268, |
| "num_tokens": 21680057.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.562109375, |
| "epoch": 1.0968494749124855, |
| "grad_norm": 0.9865952755042443, |
| "learning_rate": 1.411154345006485e-07, |
| "loss": 0.5348, |
| "mean_token_accuracy": 0.8581738114356995, |
| "num_tokens": 21916091.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.546484375, |
| "epoch": 1.1085180863477246, |
| "grad_norm": 1.2877808351471172, |
| "learning_rate": 1.4025075659316904e-07, |
| "loss": 0.5278, |
| "mean_token_accuracy": 0.8611021101474762, |
| "num_tokens": 22160097.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.544140625, |
| "epoch": 1.1201866977829638, |
| "grad_norm": 1.2400877531187657, |
| "learning_rate": 1.3938607868568958e-07, |
| "loss": 0.5095, |
| "mean_token_accuracy": 0.8626355141401291, |
| "num_tokens": 22410240.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.5384765625, |
| "epoch": 1.131855309218203, |
| "grad_norm": 1.142269477097539, |
| "learning_rate": 1.385214007782101e-07, |
| "loss": 0.5095, |
| "mean_token_accuracy": 0.8640122473239898, |
| "num_tokens": 22671161.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.54169921875, |
| "epoch": 1.1435239206534422, |
| "grad_norm": 1.2044942584540295, |
| "learning_rate": 1.3765672287073065e-07, |
| "loss": 0.511, |
| "mean_token_accuracy": 0.8647431403398513, |
| "num_tokens": 22902585.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.53408203125, |
| "epoch": 1.1551925320886816, |
| "grad_norm": 1.1142593964061263, |
| "learning_rate": 1.367920449632512e-07, |
| "loss": 0.5017, |
| "mean_token_accuracy": 0.8659891307353973, |
| "num_tokens": 23146577.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.53232421875, |
| "epoch": 1.1668611435239207, |
| "grad_norm": 1.2124771694087062, |
| "learning_rate": 1.3592736705577172e-07, |
| "loss": 0.5041, |
| "mean_token_accuracy": 0.8665498048067093, |
| "num_tokens": 23377464.0, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.538671875, |
| "epoch": 1.1785297549591598, |
| "grad_norm": 1.1052892167059623, |
| "learning_rate": 1.3506268914829226e-07, |
| "loss": 0.5132, |
| "mean_token_accuracy": 0.8649571925401688, |
| "num_tokens": 23593172.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.552734375, |
| "epoch": 1.1901983663943991, |
| "grad_norm": 0.9208155400056344, |
| "learning_rate": 1.3419801124081277e-07, |
| "loss": 0.5267, |
| "mean_token_accuracy": 0.8597115635871887, |
| "num_tokens": 23831856.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.53095703125, |
| "epoch": 1.2018669778296382, |
| "grad_norm": 0.8245068210183225, |
| "learning_rate": 1.333333333333333e-07, |
| "loss": 0.5084, |
| "mean_token_accuracy": 0.8654226243495942, |
| "num_tokens": 24086745.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.559765625, |
| "epoch": 1.2135355892648776, |
| "grad_norm": 1.0374246128657951, |
| "learning_rate": 1.3246865542585385e-07, |
| "loss": 0.5445, |
| "mean_token_accuracy": 0.8561158120632172, |
| "num_tokens": 24330531.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.525390625, |
| "epoch": 1.2252042007001167, |
| "grad_norm": 0.9236290967908911, |
| "learning_rate": 1.316039775183744e-07, |
| "loss": 0.4926, |
| "mean_token_accuracy": 0.8676816105842591, |
| "num_tokens": 24566554.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.5494140625, |
| "epoch": 1.2368728121353558, |
| "grad_norm": 1.1596172578224395, |
| "learning_rate": 1.3073929961089495e-07, |
| "loss": 0.5226, |
| "mean_token_accuracy": 0.8613315314054489, |
| "num_tokens": 24772892.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.53896484375, |
| "epoch": 1.2485414235705952, |
| "grad_norm": 1.0896322702812646, |
| "learning_rate": 1.2987462170341549e-07, |
| "loss": 0.5202, |
| "mean_token_accuracy": 0.863583680987358, |
| "num_tokens": 25020314.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.53212890625, |
| "epoch": 1.2602100350058343, |
| "grad_norm": 0.9840834886328633, |
| "learning_rate": 1.2900994379593602e-07, |
| "loss": 0.5082, |
| "mean_token_accuracy": 0.865423783659935, |
| "num_tokens": 25273465.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.5767578125, |
| "epoch": 1.2718786464410736, |
| "grad_norm": 0.9327547225739615, |
| "learning_rate": 1.2814526588845656e-07, |
| "loss": 0.549, |
| "mean_token_accuracy": 0.8544924229383468, |
| "num_tokens": 25503539.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.52109375, |
| "epoch": 1.2835472578763127, |
| "grad_norm": 1.0702437453482427, |
| "learning_rate": 1.2728058798097707e-07, |
| "loss": 0.4864, |
| "mean_token_accuracy": 0.870351767539978, |
| "num_tokens": 25737984.0, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.532421875, |
| "epoch": 1.2952158693115519, |
| "grad_norm": 0.8429024339136488, |
| "learning_rate": 1.264159100734976e-07, |
| "loss": 0.5084, |
| "mean_token_accuracy": 0.8641570597887039, |
| "num_tokens": 25967737.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.5439453125, |
| "epoch": 1.3068844807467912, |
| "grad_norm": 0.9601648724027674, |
| "learning_rate": 1.2555123216601814e-07, |
| "loss": 0.513, |
| "mean_token_accuracy": 0.8645294100046158, |
| "num_tokens": 26196017.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.54296875, |
| "epoch": 1.3185530921820303, |
| "grad_norm": 1.111976817654636, |
| "learning_rate": 1.2468655425853868e-07, |
| "loss": 0.5169, |
| "mean_token_accuracy": 0.8617329627275467, |
| "num_tokens": 26418098.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.540234375, |
| "epoch": 1.3302217036172697, |
| "grad_norm": 1.1363027352157586, |
| "learning_rate": 1.2382187635105922e-07, |
| "loss": 0.5234, |
| "mean_token_accuracy": 0.8613226413726807, |
| "num_tokens": 26635406.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.55029296875, |
| "epoch": 1.3418903150525088, |
| "grad_norm": 1.0513544602880793, |
| "learning_rate": 1.2295719844357976e-07, |
| "loss": 0.5227, |
| "mean_token_accuracy": 0.8615224540233613, |
| "num_tokens": 26866702.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.534375, |
| "epoch": 1.353558926487748, |
| "grad_norm": 0.9582742247172211, |
| "learning_rate": 1.2209252053610032e-07, |
| "loss": 0.5116, |
| "mean_token_accuracy": 0.8651837110519409, |
| "num_tokens": 27120190.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.5330078125, |
| "epoch": 1.3652275379229872, |
| "grad_norm": 1.2049169350194822, |
| "learning_rate": 1.2122784262862083e-07, |
| "loss": 0.5102, |
| "mean_token_accuracy": 0.8653781920671463, |
| "num_tokens": 27350715.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.5373046875, |
| "epoch": 1.3768961493582264, |
| "grad_norm": 0.8831191465174705, |
| "learning_rate": 1.2036316472114137e-07, |
| "loss": 0.511, |
| "mean_token_accuracy": 0.8633464246988296, |
| "num_tokens": 27585706.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.54462890625, |
| "epoch": 1.3885647607934657, |
| "grad_norm": 1.360740792990045, |
| "learning_rate": 1.194984868136619e-07, |
| "loss": 0.5169, |
| "mean_token_accuracy": 0.8637441724538804, |
| "num_tokens": 27814345.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.55009765625, |
| "epoch": 1.4002333722287048, |
| "grad_norm": 0.933816810313179, |
| "learning_rate": 1.1863380890618244e-07, |
| "loss": 0.5213, |
| "mean_token_accuracy": 0.861369925737381, |
| "num_tokens": 28042274.0, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.5306640625, |
| "epoch": 1.411901983663944, |
| "grad_norm": 1.0013169013500245, |
| "learning_rate": 1.1776913099870298e-07, |
| "loss": 0.5032, |
| "mean_token_accuracy": 0.8648687392473221, |
| "num_tokens": 28253075.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.54541015625, |
| "epoch": 1.4235705950991833, |
| "grad_norm": 1.3146657829224968, |
| "learning_rate": 1.1690445309122352e-07, |
| "loss": 0.5314, |
| "mean_token_accuracy": 0.8584078788757324, |
| "num_tokens": 28458030.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.52578125, |
| "epoch": 1.4352392065344224, |
| "grad_norm": 0.8652684653136519, |
| "learning_rate": 1.1603977518374405e-07, |
| "loss": 0.4952, |
| "mean_token_accuracy": 0.8684214323759079, |
| "num_tokens": 28689552.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.528515625, |
| "epoch": 1.4469078179696617, |
| "grad_norm": 1.169898829036548, |
| "learning_rate": 1.1517509727626458e-07, |
| "loss": 0.5055, |
| "mean_token_accuracy": 0.8671382158994675, |
| "num_tokens": 28930976.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.53388671875, |
| "epoch": 1.4585764294049008, |
| "grad_norm": 1.136236349171374, |
| "learning_rate": 1.1431041936878512e-07, |
| "loss": 0.5056, |
| "mean_token_accuracy": 0.8670242369174957, |
| "num_tokens": 29193803.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.54560546875, |
| "epoch": 1.47024504084014, |
| "grad_norm": 1.0437895552590635, |
| "learning_rate": 1.1344574146130565e-07, |
| "loss": 0.5144, |
| "mean_token_accuracy": 0.8618934273719787, |
| "num_tokens": 29419925.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.54111328125, |
| "epoch": 1.4819136522753793, |
| "grad_norm": 1.3308285992741864, |
| "learning_rate": 1.1258106355382619e-07, |
| "loss": 0.5256, |
| "mean_token_accuracy": 0.8580889046192169, |
| "num_tokens": 29667384.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.52724609375, |
| "epoch": 1.4935822637106184, |
| "grad_norm": 0.8734484568658543, |
| "learning_rate": 1.1171638564634673e-07, |
| "loss": 0.5007, |
| "mean_token_accuracy": 0.8688851237297058, |
| "num_tokens": 29887992.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.533203125, |
| "epoch": 1.5052508751458578, |
| "grad_norm": 0.7306388444788622, |
| "learning_rate": 1.1085170773886728e-07, |
| "loss": 0.5065, |
| "mean_token_accuracy": 0.8635075688362122, |
| "num_tokens": 30110668.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.52392578125, |
| "epoch": 1.5169194865810969, |
| "grad_norm": 1.0234505032150005, |
| "learning_rate": 1.0998702983138782e-07, |
| "loss": 0.507, |
| "mean_token_accuracy": 0.8656572282314301, |
| "num_tokens": 30356416.0, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.52646484375, |
| "epoch": 1.528588098016336, |
| "grad_norm": 0.8996970492635795, |
| "learning_rate": 1.0912235192390833e-07, |
| "loss": 0.5071, |
| "mean_token_accuracy": 0.8644882917404175, |
| "num_tokens": 30606131.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5259765625, |
| "epoch": 1.5402567094515751, |
| "grad_norm": 1.2061738135326794, |
| "learning_rate": 1.0825767401642886e-07, |
| "loss": 0.4884, |
| "mean_token_accuracy": 0.8696462869644165, |
| "num_tokens": 30803972.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.53349609375, |
| "epoch": 1.5519253208868145, |
| "grad_norm": 1.0923502318206546, |
| "learning_rate": 1.0739299610894941e-07, |
| "loss": 0.5123, |
| "mean_token_accuracy": 0.8642762959003448, |
| "num_tokens": 31022510.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5263671875, |
| "epoch": 1.5635939323220538, |
| "grad_norm": 1.012567521115995, |
| "learning_rate": 1.0652831820146995e-07, |
| "loss": 0.4991, |
| "mean_token_accuracy": 0.8673476189374923, |
| "num_tokens": 31274558.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.5474609375, |
| "epoch": 1.575262543757293, |
| "grad_norm": 0.9663872736444783, |
| "learning_rate": 1.0566364029399049e-07, |
| "loss": 0.527, |
| "mean_token_accuracy": 0.8608141630887985, |
| "num_tokens": 31497581.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.5341796875, |
| "epoch": 1.586931155192532, |
| "grad_norm": 1.2324821824221774, |
| "learning_rate": 1.0479896238651103e-07, |
| "loss": 0.4989, |
| "mean_token_accuracy": 0.8665684014558792, |
| "num_tokens": 31726544.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.54189453125, |
| "epoch": 1.5985997666277711, |
| "grad_norm": 0.9634311905999515, |
| "learning_rate": 1.0393428447903156e-07, |
| "loss": 0.5174, |
| "mean_token_accuracy": 0.8613916575908661, |
| "num_tokens": 31933350.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.540625, |
| "epoch": 1.6102683780630105, |
| "grad_norm": 0.8452580326701845, |
| "learning_rate": 1.0306960657155209e-07, |
| "loss": 0.5201, |
| "mean_token_accuracy": 0.8623594820499421, |
| "num_tokens": 32187176.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.53349609375, |
| "epoch": 1.6219369894982498, |
| "grad_norm": 1.0364707435030027, |
| "learning_rate": 1.0220492866407262e-07, |
| "loss": 0.5096, |
| "mean_token_accuracy": 0.8641870081424713, |
| "num_tokens": 32436326.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.53359375, |
| "epoch": 1.633605600933489, |
| "grad_norm": 1.2855344419133643, |
| "learning_rate": 1.0134025075659316e-07, |
| "loss": 0.4929, |
| "mean_token_accuracy": 0.8663494795560837, |
| "num_tokens": 32665220.0, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.54033203125, |
| "epoch": 1.645274212368728, |
| "grad_norm": 0.9249491071469745, |
| "learning_rate": 1.004755728491137e-07, |
| "loss": 0.5157, |
| "mean_token_accuracy": 0.8621318072080613, |
| "num_tokens": 32891165.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.52421875, |
| "epoch": 1.6569428238039672, |
| "grad_norm": 0.9686580912510933, |
| "learning_rate": 9.961089494163424e-08, |
| "loss": 0.5025, |
| "mean_token_accuracy": 0.8648477911949157, |
| "num_tokens": 33144037.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.5203125, |
| "epoch": 1.6686114352392065, |
| "grad_norm": 1.3402296685930952, |
| "learning_rate": 9.874621703415477e-08, |
| "loss": 0.4765, |
| "mean_token_accuracy": 0.8707535088062286, |
| "num_tokens": 33385702.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.521484375, |
| "epoch": 1.6802800466744459, |
| "grad_norm": 1.0705276983138934, |
| "learning_rate": 9.788153912667531e-08, |
| "loss": 0.4955, |
| "mean_token_accuracy": 0.8672397702932357, |
| "num_tokens": 33592547.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.53505859375, |
| "epoch": 1.691948658109685, |
| "grad_norm": 1.1082302086540656, |
| "learning_rate": 9.701686121919585e-08, |
| "loss": 0.5112, |
| "mean_token_accuracy": 0.8654333740472794, |
| "num_tokens": 33805248.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.50341796875, |
| "epoch": 1.703617269544924, |
| "grad_norm": 0.9023577167954476, |
| "learning_rate": 9.615218331171638e-08, |
| "loss": 0.473, |
| "mean_token_accuracy": 0.8713460355997086, |
| "num_tokens": 34060142.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5111328125, |
| "epoch": 1.7152858809801632, |
| "grad_norm": 1.4017038317486485, |
| "learning_rate": 9.528750540423691e-08, |
| "loss": 0.4933, |
| "mean_token_accuracy": 0.8676174700260162, |
| "num_tokens": 34336364.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.49755859375, |
| "epoch": 1.7269544924154026, |
| "grad_norm": 1.1670507371305696, |
| "learning_rate": 9.442282749675745e-08, |
| "loss": 0.4811, |
| "mean_token_accuracy": 0.8718594759702682, |
| "num_tokens": 34574499.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.528515625, |
| "epoch": 1.738623103850642, |
| "grad_norm": 1.0619074065136254, |
| "learning_rate": 9.3558149589278e-08, |
| "loss": 0.4874, |
| "mean_token_accuracy": 0.8688185155391693, |
| "num_tokens": 34814847.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.51416015625, |
| "epoch": 1.750291715285881, |
| "grad_norm": 1.1682300773753995, |
| "learning_rate": 9.269347168179853e-08, |
| "loss": 0.4924, |
| "mean_token_accuracy": 0.8689872920513153, |
| "num_tokens": 35029943.0, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.50341796875, |
| "epoch": 1.7619603267211201, |
| "grad_norm": 0.8324080494080839, |
| "learning_rate": 9.182879377431906e-08, |
| "loss": 0.4776, |
| "mean_token_accuracy": 0.871701642870903, |
| "num_tokens": 35263797.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.53818359375, |
| "epoch": 1.7736289381563592, |
| "grad_norm": 1.01722355964336, |
| "learning_rate": 9.09641158668396e-08, |
| "loss": 0.5078, |
| "mean_token_accuracy": 0.8638970285654068, |
| "num_tokens": 35480324.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.5095703125, |
| "epoch": 1.7852975495915986, |
| "grad_norm": 1.3774042656503047, |
| "learning_rate": 9.009943795936013e-08, |
| "loss": 0.4946, |
| "mean_token_accuracy": 0.8686927855014801, |
| "num_tokens": 35729838.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.54619140625, |
| "epoch": 1.796966161026838, |
| "grad_norm": 1.3757668345279621, |
| "learning_rate": 8.923476005188067e-08, |
| "loss": 0.5147, |
| "mean_token_accuracy": 0.8646462053060532, |
| "num_tokens": 35960576.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.51875, |
| "epoch": 1.808634772462077, |
| "grad_norm": 0.9655632525935715, |
| "learning_rate": 8.837008214440121e-08, |
| "loss": 0.492, |
| "mean_token_accuracy": 0.8697095483541488, |
| "num_tokens": 36196125.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.50703125, |
| "epoch": 1.8203033838973162, |
| "grad_norm": 1.1426049591921774, |
| "learning_rate": 8.750540423692174e-08, |
| "loss": 0.4857, |
| "mean_token_accuracy": 0.8702061146497726, |
| "num_tokens": 36433907.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5359375, |
| "epoch": 1.8319719953325553, |
| "grad_norm": 0.8911223314495179, |
| "learning_rate": 8.664072632944228e-08, |
| "loss": 0.5153, |
| "mean_token_accuracy": 0.86376294195652, |
| "num_tokens": 36658969.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.53076171875, |
| "epoch": 1.8436406067677946, |
| "grad_norm": 1.146919379630077, |
| "learning_rate": 8.57760484219628e-08, |
| "loss": 0.4982, |
| "mean_token_accuracy": 0.8675561279058457, |
| "num_tokens": 36904610.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.536328125, |
| "epoch": 1.855309218203034, |
| "grad_norm": 1.3098678650609548, |
| "learning_rate": 8.491137051448334e-08, |
| "loss": 0.5016, |
| "mean_token_accuracy": 0.8657185763120652, |
| "num_tokens": 37118754.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.51201171875, |
| "epoch": 1.866977829638273, |
| "grad_norm": 0.9507752101814033, |
| "learning_rate": 8.404669260700389e-08, |
| "loss": 0.4953, |
| "mean_token_accuracy": 0.8687389105558395, |
| "num_tokens": 37364878.0, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.5212890625, |
| "epoch": 1.8786464410735122, |
| "grad_norm": 1.3099025347849076, |
| "learning_rate": 8.318201469952443e-08, |
| "loss": 0.5015, |
| "mean_token_accuracy": 0.8662092357873916, |
| "num_tokens": 37591802.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.51689453125, |
| "epoch": 1.8903150525087513, |
| "grad_norm": 0.8713486719264236, |
| "learning_rate": 8.231733679204495e-08, |
| "loss": 0.4888, |
| "mean_token_accuracy": 0.867431354522705, |
| "num_tokens": 37835361.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.52783203125, |
| "epoch": 1.9019836639439907, |
| "grad_norm": 0.9636335029109001, |
| "learning_rate": 8.145265888456549e-08, |
| "loss": 0.5073, |
| "mean_token_accuracy": 0.8635091751813888, |
| "num_tokens": 38054909.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.50966796875, |
| "epoch": 1.91365227537923, |
| "grad_norm": 0.9361509351276102, |
| "learning_rate": 8.058798097708603e-08, |
| "loss": 0.4868, |
| "mean_token_accuracy": 0.8701409250497818, |
| "num_tokens": 38297597.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.5109375, |
| "epoch": 1.9253208868144691, |
| "grad_norm": 1.280585376243716, |
| "learning_rate": 7.972330306960658e-08, |
| "loss": 0.5032, |
| "mean_token_accuracy": 0.8657286465167999, |
| "num_tokens": 38551858.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.52392578125, |
| "epoch": 1.9369894982497082, |
| "grad_norm": 1.0469321053275862, |
| "learning_rate": 7.88586251621271e-08, |
| "loss": 0.4926, |
| "mean_token_accuracy": 0.8677780538797378, |
| "num_tokens": 38756948.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5095703125, |
| "epoch": 1.9486581096849473, |
| "grad_norm": 1.1453317081467052, |
| "learning_rate": 7.799394725464764e-08, |
| "loss": 0.4776, |
| "mean_token_accuracy": 0.8704618036746978, |
| "num_tokens": 38981572.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.51162109375, |
| "epoch": 1.9603267211201867, |
| "grad_norm": 0.9348030322118813, |
| "learning_rate": 7.712926934716818e-08, |
| "loss": 0.471, |
| "mean_token_accuracy": 0.8728257834911346, |
| "num_tokens": 39192455.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.50322265625, |
| "epoch": 1.971995332555426, |
| "grad_norm": 0.8532803514765905, |
| "learning_rate": 7.62645914396887e-08, |
| "loss": 0.4604, |
| "mean_token_accuracy": 0.874236273765564, |
| "num_tokens": 39438128.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.517578125, |
| "epoch": 1.9836639439906651, |
| "grad_norm": 1.1018964727470024, |
| "learning_rate": 7.539991353220925e-08, |
| "loss": 0.4864, |
| "mean_token_accuracy": 0.8695633262395859, |
| "num_tokens": 39657071.0, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.5267578125, |
| "epoch": 1.9953325554259043, |
| "grad_norm": 1.0301557153596899, |
| "learning_rate": 7.453523562472979e-08, |
| "loss": 0.5069, |
| "mean_token_accuracy": 0.8655430823564529, |
| "num_tokens": 39877709.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.5185546875, |
| "epoch": 2.0070011668611434, |
| "grad_norm": 1.0795255550550795, |
| "learning_rate": 7.367055771725033e-08, |
| "loss": 0.4816, |
| "mean_token_accuracy": 0.8720827460289001, |
| "num_tokens": 40118111.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.50224609375, |
| "epoch": 2.018669778296383, |
| "grad_norm": 0.8523953167706487, |
| "learning_rate": 7.280587980977085e-08, |
| "loss": 0.479, |
| "mean_token_accuracy": 0.8715966522693634, |
| "num_tokens": 40367267.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.51640625, |
| "epoch": 2.030338389731622, |
| "grad_norm": 0.9567843234490496, |
| "learning_rate": 7.194120190229139e-08, |
| "loss": 0.4816, |
| "mean_token_accuracy": 0.8685865789651871, |
| "num_tokens": 40588742.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5134765625, |
| "epoch": 2.042007001166861, |
| "grad_norm": 1.0804255756017007, |
| "learning_rate": 7.107652399481194e-08, |
| "loss": 0.48, |
| "mean_token_accuracy": 0.8702541202306747, |
| "num_tokens": 40818124.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.51884765625, |
| "epoch": 2.0536756126021003, |
| "grad_norm": 1.1700694135825165, |
| "learning_rate": 7.021184608733246e-08, |
| "loss": 0.4822, |
| "mean_token_accuracy": 0.8714286834001541, |
| "num_tokens": 41041609.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.50693359375, |
| "epoch": 2.0653442240373394, |
| "grad_norm": 1.3758144488531516, |
| "learning_rate": 6.9347168179853e-08, |
| "loss": 0.4898, |
| "mean_token_accuracy": 0.8695096075534821, |
| "num_tokens": 41267461.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.48779296875, |
| "epoch": 2.077012835472579, |
| "grad_norm": 1.0242700126773803, |
| "learning_rate": 6.848249027237354e-08, |
| "loss": 0.4658, |
| "mean_token_accuracy": 0.874448499083519, |
| "num_tokens": 41507251.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.487890625, |
| "epoch": 2.088681446907818, |
| "grad_norm": 0.9059026565393827, |
| "learning_rate": 6.761781236489407e-08, |
| "loss": 0.4397, |
| "mean_token_accuracy": 0.8805317610502243, |
| "num_tokens": 41764587.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.49189453125, |
| "epoch": 2.100350058343057, |
| "grad_norm": 1.0141996916061835, |
| "learning_rate": 6.675313445741461e-08, |
| "loss": 0.4666, |
| "mean_token_accuracy": 0.8730397671461105, |
| "num_tokens": 42031122.0, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.50771484375, |
| "epoch": 2.1120186697782963, |
| "grad_norm": 0.963318398388354, |
| "learning_rate": 6.588845654993515e-08, |
| "loss": 0.4821, |
| "mean_token_accuracy": 0.8706251591444015, |
| "num_tokens": 42275162.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.516796875, |
| "epoch": 2.1236872812135354, |
| "grad_norm": 0.9659882939489175, |
| "learning_rate": 6.502377864245569e-08, |
| "loss": 0.4908, |
| "mean_token_accuracy": 0.8684900850057602, |
| "num_tokens": 42499755.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.5041015625, |
| "epoch": 2.135355892648775, |
| "grad_norm": 1.0713407563312622, |
| "learning_rate": 6.415910073497622e-08, |
| "loss": 0.4775, |
| "mean_token_accuracy": 0.8699021428823471, |
| "num_tokens": 42731832.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.48896484375, |
| "epoch": 2.147024504084014, |
| "grad_norm": 1.0897153935245856, |
| "learning_rate": 6.329442282749675e-08, |
| "loss": 0.4541, |
| "mean_token_accuracy": 0.8789973556995392, |
| "num_tokens": 43006203.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.51318359375, |
| "epoch": 2.1586931155192532, |
| "grad_norm": 1.018958247344349, |
| "learning_rate": 6.242974492001728e-08, |
| "loss": 0.4934, |
| "mean_token_accuracy": 0.8682343244552613, |
| "num_tokens": 43229874.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.520703125, |
| "epoch": 2.1703617269544924, |
| "grad_norm": 1.2364812697957024, |
| "learning_rate": 6.156506701253784e-08, |
| "loss": 0.4972, |
| "mean_token_accuracy": 0.8680180430412292, |
| "num_tokens": 43456429.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.4978515625, |
| "epoch": 2.1820303383897315, |
| "grad_norm": 0.8907920281335129, |
| "learning_rate": 6.070038910505836e-08, |
| "loss": 0.4686, |
| "mean_token_accuracy": 0.8744249463081359, |
| "num_tokens": 43710091.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.51806640625, |
| "epoch": 2.193698949824971, |
| "grad_norm": 1.0573271571016871, |
| "learning_rate": 5.98357111975789e-08, |
| "loss": 0.4901, |
| "mean_token_accuracy": 0.868583670258522, |
| "num_tokens": 43913791.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.51298828125, |
| "epoch": 2.20536756126021, |
| "grad_norm": 1.2221913686769739, |
| "learning_rate": 5.8971033290099434e-08, |
| "loss": 0.4803, |
| "mean_token_accuracy": 0.8703306466341019, |
| "num_tokens": 44122915.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.510546875, |
| "epoch": 2.2170361726954493, |
| "grad_norm": 1.038917339571604, |
| "learning_rate": 5.810635538261998e-08, |
| "loss": 0.4735, |
| "mean_token_accuracy": 0.870867383480072, |
| "num_tokens": 44366169.0, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.4953125, |
| "epoch": 2.2287047841306884, |
| "grad_norm": 0.8301493675525253, |
| "learning_rate": 5.72416774751405e-08, |
| "loss": 0.4689, |
| "mean_token_accuracy": 0.8745264947414398, |
| "num_tokens": 44636039.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.5044921875, |
| "epoch": 2.2403733955659275, |
| "grad_norm": 1.2406960812943308, |
| "learning_rate": 5.6376999567661045e-08, |
| "loss": 0.4803, |
| "mean_token_accuracy": 0.8719497144222259, |
| "num_tokens": 44852034.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.516796875, |
| "epoch": 2.252042007001167, |
| "grad_norm": 1.2001739100351527, |
| "learning_rate": 5.551232166018158e-08, |
| "loss": 0.4916, |
| "mean_token_accuracy": 0.8681713461875915, |
| "num_tokens": 45086831.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.5060546875, |
| "epoch": 2.263710618436406, |
| "grad_norm": 1.2201939696883253, |
| "learning_rate": 5.4647643752702113e-08, |
| "loss": 0.4791, |
| "mean_token_accuracy": 0.8719442188739777, |
| "num_tokens": 45320636.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.49951171875, |
| "epoch": 2.2753792298716453, |
| "grad_norm": 1.0182529704846792, |
| "learning_rate": 5.378296584522265e-08, |
| "loss": 0.4707, |
| "mean_token_accuracy": 0.8711633592844009, |
| "num_tokens": 45573294.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.4943359375, |
| "epoch": 2.2870478413068844, |
| "grad_norm": 0.807485477152529, |
| "learning_rate": 5.291828793774319e-08, |
| "loss": 0.4653, |
| "mean_token_accuracy": 0.8742965906858444, |
| "num_tokens": 45846897.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.51396484375, |
| "epoch": 2.2987164527421236, |
| "grad_norm": 1.083937844553969, |
| "learning_rate": 5.2053610030263725e-08, |
| "loss": 0.4929, |
| "mean_token_accuracy": 0.8691924005746842, |
| "num_tokens": 46077019.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.5154296875, |
| "epoch": 2.310385064177363, |
| "grad_norm": 0.900664769665513, |
| "learning_rate": 5.1188932122784256e-08, |
| "loss": 0.4802, |
| "mean_token_accuracy": 0.8698993623256683, |
| "num_tokens": 46330180.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.518359375, |
| "epoch": 2.3220536756126022, |
| "grad_norm": 1.0749851110661846, |
| "learning_rate": 5.032425421530479e-08, |
| "loss": 0.499, |
| "mean_token_accuracy": 0.8690392822027206, |
| "num_tokens": 46536646.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.503515625, |
| "epoch": 2.3337222870478413, |
| "grad_norm": 1.0939645475896693, |
| "learning_rate": 4.945957630782534e-08, |
| "loss": 0.4746, |
| "mean_token_accuracy": 0.8731703281402587, |
| "num_tokens": 46797880.0, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.51689453125, |
| "epoch": 2.3453908984830805, |
| "grad_norm": 1.2272100669192603, |
| "learning_rate": 4.859489840034587e-08, |
| "loss": 0.4886, |
| "mean_token_accuracy": 0.8702142089605331, |
| "num_tokens": 47001629.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.51318359375, |
| "epoch": 2.3570595099183196, |
| "grad_norm": 1.6466101040962522, |
| "learning_rate": 4.7730220492866405e-08, |
| "loss": 0.4876, |
| "mean_token_accuracy": 0.8689578950405121, |
| "num_tokens": 47226383.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.50537109375, |
| "epoch": 2.368728121353559, |
| "grad_norm": 1.0332091804494559, |
| "learning_rate": 4.686554258538694e-08, |
| "loss": 0.4858, |
| "mean_token_accuracy": 0.8702533215284347, |
| "num_tokens": 47481540.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.512890625, |
| "epoch": 2.3803967327887983, |
| "grad_norm": 1.1717077993524783, |
| "learning_rate": 4.600086467790747e-08, |
| "loss": 0.4854, |
| "mean_token_accuracy": 0.8703208565711975, |
| "num_tokens": 47693664.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.518359375, |
| "epoch": 2.3920653442240374, |
| "grad_norm": 1.1614508515952189, |
| "learning_rate": 4.5136186770428016e-08, |
| "loss": 0.4973, |
| "mean_token_accuracy": 0.866780811548233, |
| "num_tokens": 47940930.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.49873046875, |
| "epoch": 2.4037339556592765, |
| "grad_norm": 1.0911950675475588, |
| "learning_rate": 4.427150886294855e-08, |
| "loss": 0.4619, |
| "mean_token_accuracy": 0.8770538687705993, |
| "num_tokens": 48180508.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.51025390625, |
| "epoch": 2.4154025670945156, |
| "grad_norm": 1.1799726485957962, |
| "learning_rate": 4.3406830955469084e-08, |
| "loss": 0.488, |
| "mean_token_accuracy": 0.8678788334131241, |
| "num_tokens": 48392986.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.50390625, |
| "epoch": 2.427071178529755, |
| "grad_norm": 1.0676083287695486, |
| "learning_rate": 4.254215304798962e-08, |
| "loss": 0.4827, |
| "mean_token_accuracy": 0.8717045158147811, |
| "num_tokens": 48641322.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.50712890625, |
| "epoch": 2.4387397899649943, |
| "grad_norm": 1.132323092412435, |
| "learning_rate": 4.167747514051016e-08, |
| "loss": 0.4888, |
| "mean_token_accuracy": 0.8690821349620819, |
| "num_tokens": 48881292.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.50546875, |
| "epoch": 2.4504084014002334, |
| "grad_norm": 0.9627850217175202, |
| "learning_rate": 4.0812797233030696e-08, |
| "loss": 0.4907, |
| "mean_token_accuracy": 0.8702411115169525, |
| "num_tokens": 49101943.0, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.53798828125, |
| "epoch": 2.4620770128354725, |
| "grad_norm": 1.2150324403662869, |
| "learning_rate": 3.9948119325551233e-08, |
| "loss": 0.5041, |
| "mean_token_accuracy": 0.8649662494659424, |
| "num_tokens": 49334455.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.49423828125, |
| "epoch": 2.4737456242707117, |
| "grad_norm": 0.8930680639667525, |
| "learning_rate": 3.9083441418071764e-08, |
| "loss": 0.48, |
| "mean_token_accuracy": 0.8709317803382873, |
| "num_tokens": 49596945.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.48525390625, |
| "epoch": 2.485414235705951, |
| "grad_norm": 1.0764627011799328, |
| "learning_rate": 3.82187635105923e-08, |
| "loss": 0.4574, |
| "mean_token_accuracy": 0.8758306086063385, |
| "num_tokens": 49827975.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.50498046875, |
| "epoch": 2.4970828471411903, |
| "grad_norm": 1.0702848047299671, |
| "learning_rate": 3.735408560311284e-08, |
| "loss": 0.4715, |
| "mean_token_accuracy": 0.8726317018270493, |
| "num_tokens": 50067801.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.5103515625, |
| "epoch": 2.5087514585764294, |
| "grad_norm": 0.794873638706198, |
| "learning_rate": 3.6489407695633376e-08, |
| "loss": 0.484, |
| "mean_token_accuracy": 0.8697792887687683, |
| "num_tokens": 50302736.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.50029296875, |
| "epoch": 2.5204200700116686, |
| "grad_norm": 0.9787422558055034, |
| "learning_rate": 3.562472978815391e-08, |
| "loss": 0.4924, |
| "mean_token_accuracy": 0.8689286947250366, |
| "num_tokens": 50579280.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.5119140625, |
| "epoch": 2.5320886814469077, |
| "grad_norm": 1.1083646420008544, |
| "learning_rate": 3.4760051880674444e-08, |
| "loss": 0.4938, |
| "mean_token_accuracy": 0.8684186100959778, |
| "num_tokens": 50803957.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.4935546875, |
| "epoch": 2.5437572928821472, |
| "grad_norm": 0.6894786121123152, |
| "learning_rate": 3.389537397319499e-08, |
| "loss": 0.4658, |
| "mean_token_accuracy": 0.8758569538593293, |
| "num_tokens": 51060311.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.50986328125, |
| "epoch": 2.5554259043173864, |
| "grad_norm": 0.8343809154897581, |
| "learning_rate": 3.303069606571552e-08, |
| "loss": 0.4845, |
| "mean_token_accuracy": 0.8703210353851318, |
| "num_tokens": 51300412.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.50576171875, |
| "epoch": 2.5670945157526255, |
| "grad_norm": 1.178383407277546, |
| "learning_rate": 3.2166018158236055e-08, |
| "loss": 0.4909, |
| "mean_token_accuracy": 0.8689768821001053, |
| "num_tokens": 51537247.0, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.49921875, |
| "epoch": 2.5787631271878646, |
| "grad_norm": 1.0175757280100703, |
| "learning_rate": 3.130134025075659e-08, |
| "loss": 0.4707, |
| "mean_token_accuracy": 0.8722994387149811, |
| "num_tokens": 51758422.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.48623046875, |
| "epoch": 2.5904317386231037, |
| "grad_norm": 1.1683941191425213, |
| "learning_rate": 3.043666234327712e-08, |
| "loss": 0.4615, |
| "mean_token_accuracy": 0.8765349298715591, |
| "num_tokens": 52010173.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.4931640625, |
| "epoch": 2.6021003500583433, |
| "grad_norm": 1.0548518983119062, |
| "learning_rate": 2.9571984435797664e-08, |
| "loss": 0.4636, |
| "mean_token_accuracy": 0.8731530159711838, |
| "num_tokens": 52237298.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.5025390625, |
| "epoch": 2.6137689614935824, |
| "grad_norm": 1.1795120615781054, |
| "learning_rate": 2.8707306528318198e-08, |
| "loss": 0.4857, |
| "mean_token_accuracy": 0.8704844772815704, |
| "num_tokens": 52492216.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.5251953125, |
| "epoch": 2.6254375729288215, |
| "grad_norm": 1.0613512308250002, |
| "learning_rate": 2.7842628620838738e-08, |
| "loss": 0.4993, |
| "mean_token_accuracy": 0.8662945479154587, |
| "num_tokens": 52691121.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.50361328125, |
| "epoch": 2.6371061843640606, |
| "grad_norm": 0.803924606819153, |
| "learning_rate": 2.6977950713359272e-08, |
| "loss": 0.4736, |
| "mean_token_accuracy": 0.872618991136551, |
| "num_tokens": 52926598.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.5169921875, |
| "epoch": 2.6487747957992998, |
| "grad_norm": 1.161833032024034, |
| "learning_rate": 2.611327280587981e-08, |
| "loss": 0.48, |
| "mean_token_accuracy": 0.8722775608301163, |
| "num_tokens": 53131790.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.51552734375, |
| "epoch": 2.6604434072345393, |
| "grad_norm": 1.1227006450354817, |
| "learning_rate": 2.5248594898400343e-08, |
| "loss": 0.5036, |
| "mean_token_accuracy": 0.8652754247188568, |
| "num_tokens": 53372586.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.489453125, |
| "epoch": 2.6721120186697784, |
| "grad_norm": 1.2251762237024462, |
| "learning_rate": 2.438391699092088e-08, |
| "loss": 0.4491, |
| "mean_token_accuracy": 0.8776472002267838, |
| "num_tokens": 53606048.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.51103515625, |
| "epoch": 2.6837806301050176, |
| "grad_norm": 1.3510266856978659, |
| "learning_rate": 2.3519239083441418e-08, |
| "loss": 0.4857, |
| "mean_token_accuracy": 0.870513316988945, |
| "num_tokens": 53845042.0, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.51318359375, |
| "epoch": 2.6954492415402567, |
| "grad_norm": 1.0353462330817174, |
| "learning_rate": 2.2654561175961952e-08, |
| "loss": 0.4874, |
| "mean_token_accuracy": 0.8694655030965805, |
| "num_tokens": 54088107.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.52353515625, |
| "epoch": 2.707117852975496, |
| "grad_norm": 0.887494267056635, |
| "learning_rate": 2.178988326848249e-08, |
| "loss": 0.4931, |
| "mean_token_accuracy": 0.8688248664140701, |
| "num_tokens": 54294782.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.52275390625, |
| "epoch": 2.7187864644107353, |
| "grad_norm": 1.0323821297654974, |
| "learning_rate": 2.0925205361003026e-08, |
| "loss": 0.4957, |
| "mean_token_accuracy": 0.8669693738222122, |
| "num_tokens": 54514357.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.5142578125, |
| "epoch": 2.7304550758459745, |
| "grad_norm": 1.0459665505256175, |
| "learning_rate": 2.0060527453523564e-08, |
| "loss": 0.4886, |
| "mean_token_accuracy": 0.8690153062343597, |
| "num_tokens": 54727816.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.5115234375, |
| "epoch": 2.7421236872812136, |
| "grad_norm": 1.0569922738520776, |
| "learning_rate": 1.9195849546044098e-08, |
| "loss": 0.486, |
| "mean_token_accuracy": 0.8685569494962693, |
| "num_tokens": 54963984.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.48935546875, |
| "epoch": 2.7537922987164527, |
| "grad_norm": 1.0045090803577597, |
| "learning_rate": 1.833117163856463e-08, |
| "loss": 0.4662, |
| "mean_token_accuracy": 0.8752841472625732, |
| "num_tokens": 55220830.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.4923828125, |
| "epoch": 2.765460910151692, |
| "grad_norm": 0.8439517288934741, |
| "learning_rate": 1.746649373108517e-08, |
| "loss": 0.4626, |
| "mean_token_accuracy": 0.8767544478178024, |
| "num_tokens": 55462755.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.49501953125, |
| "epoch": 2.7771295215869314, |
| "grad_norm": 1.1583920816335884, |
| "learning_rate": 1.6601815823605706e-08, |
| "loss": 0.4645, |
| "mean_token_accuracy": 0.8747540920972824, |
| "num_tokens": 55705248.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.49912109375, |
| "epoch": 2.7887981330221705, |
| "grad_norm": 0.8368158721184437, |
| "learning_rate": 1.5737137916126243e-08, |
| "loss": 0.4817, |
| "mean_token_accuracy": 0.8708766758441925, |
| "num_tokens": 55958763.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.49990234375, |
| "epoch": 2.8004667444574096, |
| "grad_norm": 1.1758528324203712, |
| "learning_rate": 1.4872460008646779e-08, |
| "loss": 0.4712, |
| "mean_token_accuracy": 0.872442701458931, |
| "num_tokens": 56179313.0, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.4869140625, |
| "epoch": 2.8121353558926487, |
| "grad_norm": 1.2177742662753448, |
| "learning_rate": 1.4007782101167314e-08, |
| "loss": 0.4612, |
| "mean_token_accuracy": 0.8746420115232467, |
| "num_tokens": 56415877.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.50009765625, |
| "epoch": 2.823803967327888, |
| "grad_norm": 1.0170691477685734, |
| "learning_rate": 1.3143104193687852e-08, |
| "loss": 0.4674, |
| "mean_token_accuracy": 0.8730188280344009, |
| "num_tokens": 56689071.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.51162109375, |
| "epoch": 2.8354725787631274, |
| "grad_norm": 0.9060289658767671, |
| "learning_rate": 1.2278426286208386e-08, |
| "loss": 0.4997, |
| "mean_token_accuracy": 0.8675457507371902, |
| "num_tokens": 56920613.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.49921875, |
| "epoch": 2.8471411901983665, |
| "grad_norm": 1.142218279415812, |
| "learning_rate": 1.1413748378728923e-08, |
| "loss": 0.4705, |
| "mean_token_accuracy": 0.874994769692421, |
| "num_tokens": 57157309.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.4919921875, |
| "epoch": 2.8588098016336057, |
| "grad_norm": 1.2783487187746796, |
| "learning_rate": 1.0549070471249458e-08, |
| "loss": 0.4693, |
| "mean_token_accuracy": 0.8738763213157654, |
| "num_tokens": 57384007.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.511328125, |
| "epoch": 2.8704784130688448, |
| "grad_norm": 0.9680171059677461, |
| "learning_rate": 9.684392563769996e-09, |
| "loss": 0.4859, |
| "mean_token_accuracy": 0.8674903780221939, |
| "num_tokens": 57625809.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.487109375, |
| "epoch": 2.882147024504084, |
| "grad_norm": 1.1956425518353855, |
| "learning_rate": 8.819714656290531e-09, |
| "loss": 0.4657, |
| "mean_token_accuracy": 0.8743886262178421, |
| "num_tokens": 57852230.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.50927734375, |
| "epoch": 2.8938156359393234, |
| "grad_norm": 1.053684403779092, |
| "learning_rate": 7.955036748811067e-09, |
| "loss": 0.4816, |
| "mean_token_accuracy": 0.8711407810449601, |
| "num_tokens": 58091216.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.50478515625, |
| "epoch": 2.9054842473745626, |
| "grad_norm": 0.9440599133851945, |
| "learning_rate": 7.090358841331603e-09, |
| "loss": 0.4762, |
| "mean_token_accuracy": 0.8718186289072036, |
| "num_tokens": 58332703.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.5123046875, |
| "epoch": 2.9171528588098017, |
| "grad_norm": 1.0951256474359146, |
| "learning_rate": 6.22568093385214e-09, |
| "loss": 0.4943, |
| "mean_token_accuracy": 0.8665006667375564, |
| "num_tokens": 58572805.0, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.5029296875, |
| "epoch": 2.928821470245041, |
| "grad_norm": 1.0383538654622884, |
| "learning_rate": 5.361003026372675e-09, |
| "loss": 0.4774, |
| "mean_token_accuracy": 0.8724213868379593, |
| "num_tokens": 58789696.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.49296875, |
| "epoch": 2.94049008168028, |
| "grad_norm": 1.0615564642672966, |
| "learning_rate": 4.496325118893212e-09, |
| "loss": 0.4621, |
| "mean_token_accuracy": 0.8735111504793167, |
| "num_tokens": 59041021.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.51220703125, |
| "epoch": 2.9521586931155195, |
| "grad_norm": 1.0144786467220435, |
| "learning_rate": 3.6316472114137486e-09, |
| "loss": 0.4931, |
| "mean_token_accuracy": 0.8699846416711807, |
| "num_tokens": 59268730.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.5169921875, |
| "epoch": 2.9638273045507586, |
| "grad_norm": 1.0453853578762702, |
| "learning_rate": 2.766969303934284e-09, |
| "loss": 0.4862, |
| "mean_token_accuracy": 0.8693469822406769, |
| "num_tokens": 59496852.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.5158203125, |
| "epoch": 2.9754959159859977, |
| "grad_norm": 1.0607397353947798, |
| "learning_rate": 1.90229139645482e-09, |
| "loss": 0.4877, |
| "mean_token_accuracy": 0.8695639640092849, |
| "num_tokens": 59736589.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.4892578125, |
| "epoch": 2.987164527421237, |
| "grad_norm": 1.098774912647474, |
| "learning_rate": 1.0376134889753566e-09, |
| "loss": 0.4584, |
| "mean_token_accuracy": 0.8759648203849792, |
| "num_tokens": 59972332.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.51708984375, |
| "epoch": 2.998833138856476, |
| "grad_norm": 1.1733382416108404, |
| "learning_rate": 1.7293558149589276e-10, |
| "loss": 0.5003, |
| "mean_token_accuracy": 0.8674053460359573, |
| "num_tokens": 60193812.0, |
| "step": 2570 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 2571, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 1000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 199613630971904.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
