| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 857, |
| "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 |
| } |
| ], |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 66207481462784.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
