| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 78, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.223182126879692, |
| "epoch": 0.012903225806451613, |
| "grad_norm": 1.076210379600525, |
| "learning_rate": 0.0, |
| "loss": 2.058398723602295, |
| "mean_token_accuracy": 0.5399363860487938, |
| "num_tokens": 129656.0, |
| "step": 1 |
| }, |
| { |
| "entropy": 1.2220996767282486, |
| "epoch": 0.025806451612903226, |
| "grad_norm": 1.0572984218597412, |
| "learning_rate": 4.166666666666667e-06, |
| "loss": 2.070119857788086, |
| "mean_token_accuracy": 0.5379651635885239, |
| "num_tokens": 257608.0, |
| "step": 2 |
| }, |
| { |
| "entropy": 1.2825039327144623, |
| "epoch": 0.03870967741935484, |
| "grad_norm": 0.9764279723167419, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 2.0715339183807373, |
| "mean_token_accuracy": 0.5355395078659058, |
| "num_tokens": 385654.0, |
| "step": 3 |
| }, |
| { |
| "entropy": 1.3505332916975021, |
| "epoch": 0.05161290322580645, |
| "grad_norm": 0.8341186046600342, |
| "learning_rate": 1.25e-05, |
| "loss": 2.0449037551879883, |
| "mean_token_accuracy": 0.5337138548493385, |
| "num_tokens": 513578.0, |
| "step": 4 |
| }, |
| { |
| "entropy": 1.3795223087072372, |
| "epoch": 0.06451612903225806, |
| "grad_norm": 0.6299435496330261, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 1.9157527685165405, |
| "mean_token_accuracy": 0.5517139434814453, |
| "num_tokens": 643891.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 1.4815990179777145, |
| "epoch": 0.07741935483870968, |
| "grad_norm": 0.4900154173374176, |
| "learning_rate": 2.0833333333333336e-05, |
| "loss": 1.891552448272705, |
| "mean_token_accuracy": 0.5499126613140106, |
| "num_tokens": 773552.0, |
| "step": 6 |
| }, |
| { |
| "entropy": 1.5551793575286865, |
| "epoch": 0.09032258064516129, |
| "grad_norm": 0.36406657099723816, |
| "learning_rate": 2.5e-05, |
| "loss": 1.8251445293426514, |
| "mean_token_accuracy": 0.5556761026382446, |
| "num_tokens": 902676.0, |
| "step": 7 |
| }, |
| { |
| "entropy": 1.6218345165252686, |
| "epoch": 0.1032258064516129, |
| "grad_norm": 0.3063054084777832, |
| "learning_rate": 2.916666666666667e-05, |
| "loss": 1.7969532012939453, |
| "mean_token_accuracy": 0.5583265572786331, |
| "num_tokens": 1031062.0, |
| "step": 8 |
| }, |
| { |
| "entropy": 1.7121598720550537, |
| "epoch": 0.11612903225806452, |
| "grad_norm": 0.2525842785835266, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 1.7734124660491943, |
| "mean_token_accuracy": 0.563885897397995, |
| "num_tokens": 1159144.0, |
| "step": 9 |
| }, |
| { |
| "entropy": 1.7930369228124619, |
| "epoch": 0.12903225806451613, |
| "grad_norm": 0.24535015225410461, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 1.7790138721466064, |
| "mean_token_accuracy": 0.5602867007255554, |
| "num_tokens": 1286047.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.8023179322481155, |
| "epoch": 0.14193548387096774, |
| "grad_norm": 0.2574515640735626, |
| "learning_rate": 4.166666666666667e-05, |
| "loss": 1.7038989067077637, |
| "mean_token_accuracy": 0.5719256922602654, |
| "num_tokens": 1414956.0, |
| "step": 11 |
| }, |
| { |
| "entropy": 1.8494196385145187, |
| "epoch": 0.15483870967741936, |
| "grad_norm": 0.28199130296707153, |
| "learning_rate": 4.5833333333333334e-05, |
| "loss": 1.6980609893798828, |
| "mean_token_accuracy": 0.5720911771059036, |
| "num_tokens": 1544083.0, |
| "step": 12 |
| }, |
| { |
| "entropy": 1.8534925431013107, |
| "epoch": 0.16774193548387098, |
| "grad_norm": 0.29593202471733093, |
| "learning_rate": 5e-05, |
| "loss": 1.6669628620147705, |
| "mean_token_accuracy": 0.5764533281326294, |
| "num_tokens": 1673817.0, |
| "step": 13 |
| }, |
| { |
| "entropy": 1.7742547988891602, |
| "epoch": 0.18064516129032257, |
| "grad_norm": 0.2752297818660736, |
| "learning_rate": 4.9997496794168726e-05, |
| "loss": 1.5782192945480347, |
| "mean_token_accuracy": 0.5955501049757004, |
| "num_tokens": 1801770.0, |
| "step": 14 |
| }, |
| { |
| "entropy": 1.8175705522298813, |
| "epoch": 0.1935483870967742, |
| "grad_norm": 0.2872096300125122, |
| "learning_rate": 4.998998767795805e-05, |
| "loss": 1.5999691486358643, |
| "mean_token_accuracy": 0.5888361930847168, |
| "num_tokens": 1931723.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 1.7427806854248047, |
| "epoch": 0.2064516129032258, |
| "grad_norm": 0.2521703839302063, |
| "learning_rate": 4.9977474155117045e-05, |
| "loss": 1.5546669960021973, |
| "mean_token_accuracy": 0.5967967882752419, |
| "num_tokens": 2061345.0, |
| "step": 16 |
| }, |
| { |
| "entropy": 1.6578464657068253, |
| "epoch": 0.21935483870967742, |
| "grad_norm": 0.22542445361614227, |
| "learning_rate": 4.995995873155958e-05, |
| "loss": 1.5100972652435303, |
| "mean_token_accuracy": 0.6021811291575432, |
| "num_tokens": 2190508.0, |
| "step": 17 |
| }, |
| { |
| "entropy": 1.598375290632248, |
| "epoch": 0.23225806451612904, |
| "grad_norm": 0.1917019784450531, |
| "learning_rate": 4.99374449148625e-05, |
| "loss": 1.490499496459961, |
| "mean_token_accuracy": 0.60594841837883, |
| "num_tokens": 2318941.0, |
| "step": 18 |
| }, |
| { |
| "entropy": 1.5606479495763779, |
| "epoch": 0.24516129032258063, |
| "grad_norm": 0.1699151247739792, |
| "learning_rate": 4.9909937213563165e-05, |
| "loss": 1.480535626411438, |
| "mean_token_accuracy": 0.607219472527504, |
| "num_tokens": 2448059.0, |
| "step": 19 |
| }, |
| { |
| "entropy": 1.5005332827568054, |
| "epoch": 0.25806451612903225, |
| "grad_norm": 0.15443331003189087, |
| "learning_rate": 4.987744113625665e-05, |
| "loss": 1.4707814455032349, |
| "mean_token_accuracy": 0.6087440326809883, |
| "num_tokens": 2577375.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.3930998891592026, |
| "epoch": 0.2709677419354839, |
| "grad_norm": 0.1559259444475174, |
| "learning_rate": 4.9839963190492576e-05, |
| "loss": 1.410855770111084, |
| "mean_token_accuracy": 0.6227314248681068, |
| "num_tokens": 2707055.0, |
| "step": 21 |
| }, |
| { |
| "entropy": 1.36322420835495, |
| "epoch": 0.2838709677419355, |
| "grad_norm": 0.20294561982154846, |
| "learning_rate": 4.979751088147192e-05, |
| "loss": 1.4284950494766235, |
| "mean_token_accuracy": 0.6158238276839256, |
| "num_tokens": 2836471.0, |
| "step": 22 |
| }, |
| { |
| "entropy": 1.3222410529851913, |
| "epoch": 0.2967741935483871, |
| "grad_norm": 0.22203199565410614, |
| "learning_rate": 4.975009271054409e-05, |
| "loss": 1.4026174545288086, |
| "mean_token_accuracy": 0.6221327260136604, |
| "num_tokens": 2967186.0, |
| "step": 23 |
| }, |
| { |
| "entropy": 1.3454352766275406, |
| "epoch": 0.3096774193548387, |
| "grad_norm": 0.2088484764099121, |
| "learning_rate": 4.969771817350445e-05, |
| "loss": 1.4195332527160645, |
| "mean_token_accuracy": 0.6193315461277962, |
| "num_tokens": 3091283.0, |
| "step": 24 |
| }, |
| { |
| "entropy": 1.3203742057085037, |
| "epoch": 0.3225806451612903, |
| "grad_norm": 0.1915750652551651, |
| "learning_rate": 4.9640397758692715e-05, |
| "loss": 1.3804551362991333, |
| "mean_token_accuracy": 0.626660019159317, |
| "num_tokens": 3221704.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.3405092060565948, |
| "epoch": 0.33548387096774196, |
| "grad_norm": 0.16226191818714142, |
| "learning_rate": 4.957814294489261e-05, |
| "loss": 1.3678839206695557, |
| "mean_token_accuracy": 0.6271623075008392, |
| "num_tokens": 3351359.0, |
| "step": 26 |
| }, |
| { |
| "entropy": 1.3681351393461227, |
| "epoch": 0.34838709677419355, |
| "grad_norm": 0.15001052618026733, |
| "learning_rate": 4.9510966199033174e-05, |
| "loss": 1.3961801528930664, |
| "mean_token_accuracy": 0.6210252344608307, |
| "num_tokens": 3480058.0, |
| "step": 27 |
| }, |
| { |
| "entropy": 1.360452800989151, |
| "epoch": 0.36129032258064514, |
| "grad_norm": 0.13115695118904114, |
| "learning_rate": 4.943888097369216e-05, |
| "loss": 1.3473964929580688, |
| "mean_token_accuracy": 0.6314521729946136, |
| "num_tokens": 3610897.0, |
| "step": 28 |
| }, |
| { |
| "entropy": 1.3818826526403427, |
| "epoch": 0.3741935483870968, |
| "grad_norm": 0.12812234461307526, |
| "learning_rate": 4.936190170440208e-05, |
| "loss": 1.3602492809295654, |
| "mean_token_accuracy": 0.6298193782567978, |
| "num_tokens": 3740627.0, |
| "step": 29 |
| }, |
| { |
| "entropy": 1.3282616436481476, |
| "epoch": 0.3870967741935484, |
| "grad_norm": 0.12813468277454376, |
| "learning_rate": 4.928004380675941e-05, |
| "loss": 1.2889750003814697, |
| "mean_token_accuracy": 0.6441097930073738, |
| "num_tokens": 3869695.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.3403072953224182, |
| "epoch": 0.4, |
| "grad_norm": 0.1254233419895172, |
| "learning_rate": 4.9193323673337476e-05, |
| "loss": 1.3034392595291138, |
| "mean_token_accuracy": 0.6405649855732918, |
| "num_tokens": 3998221.0, |
| "step": 31 |
| }, |
| { |
| "entropy": 1.3977190256118774, |
| "epoch": 0.4129032258064516, |
| "grad_norm": 0.13067315518856049, |
| "learning_rate": 4.910175867040377e-05, |
| "loss": 1.354722261428833, |
| "mean_token_accuracy": 0.6276756450533867, |
| "num_tokens": 4127108.0, |
| "step": 32 |
| }, |
| { |
| "entropy": 1.348634958267212, |
| "epoch": 0.4258064516129032, |
| "grad_norm": 0.12189042568206787, |
| "learning_rate": 4.9005367134442235e-05, |
| "loss": 1.3136909008026123, |
| "mean_token_accuracy": 0.6381376534700394, |
| "num_tokens": 4256994.0, |
| "step": 33 |
| }, |
| { |
| "entropy": 1.3185934275388718, |
| "epoch": 0.43870967741935485, |
| "grad_norm": 0.1224159300327301, |
| "learning_rate": 4.890416836848127e-05, |
| "loss": 1.2872580289840698, |
| "mean_token_accuracy": 0.6426139622926712, |
| "num_tokens": 4385244.0, |
| "step": 34 |
| }, |
| { |
| "entropy": 1.2890230417251587, |
| "epoch": 0.45161290322580644, |
| "grad_norm": 0.11730781197547913, |
| "learning_rate": 4.8798182638228166e-05, |
| "loss": 1.2566399574279785, |
| "mean_token_accuracy": 0.649329274892807, |
| "num_tokens": 4515175.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 1.3161695450544357, |
| "epoch": 0.4645161290322581, |
| "grad_norm": 0.11564701795578003, |
| "learning_rate": 4.868743116801074e-05, |
| "loss": 1.2894011735916138, |
| "mean_token_accuracy": 0.6447330787777901, |
| "num_tokens": 4644464.0, |
| "step": 36 |
| }, |
| { |
| "entropy": 1.263451337814331, |
| "epoch": 0.4774193548387097, |
| "grad_norm": 0.11025005578994751, |
| "learning_rate": 4.857193613652711e-05, |
| "loss": 1.250872254371643, |
| "mean_token_accuracy": 0.6496427357196808, |
| "num_tokens": 4774388.0, |
| "step": 37 |
| }, |
| { |
| "entropy": 1.2831027507781982, |
| "epoch": 0.49032258064516127, |
| "grad_norm": 0.11225498467683792, |
| "learning_rate": 4.845172067240415e-05, |
| "loss": 1.2833701372146606, |
| "mean_token_accuracy": 0.642292931675911, |
| "num_tokens": 4903171.0, |
| "step": 38 |
| }, |
| { |
| "entropy": 1.28334479033947, |
| "epoch": 0.5032258064516129, |
| "grad_norm": 0.11390651762485504, |
| "learning_rate": 4.8326808849565936e-05, |
| "loss": 1.2927157878875732, |
| "mean_token_accuracy": 0.6402992084622383, |
| "num_tokens": 5033677.0, |
| "step": 39 |
| }, |
| { |
| "entropy": 1.2627993077039719, |
| "epoch": 0.5161290322580645, |
| "grad_norm": 0.1086907684803009, |
| "learning_rate": 4.819722568241274e-05, |
| "loss": 1.2574979066848755, |
| "mean_token_accuracy": 0.6465372666716576, |
| "num_tokens": 5163487.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.2626460939645767, |
| "epoch": 0.5290322580645161, |
| "grad_norm": 0.10910395532846451, |
| "learning_rate": 4.806299712081172e-05, |
| "loss": 1.266538381576538, |
| "mean_token_accuracy": 0.6473062336444855, |
| "num_tokens": 5292899.0, |
| "step": 41 |
| }, |
| { |
| "entropy": 1.2527801245450974, |
| "epoch": 0.5419354838709678, |
| "grad_norm": 0.10939069092273712, |
| "learning_rate": 4.792415004490034e-05, |
| "loss": 1.2544794082641602, |
| "mean_token_accuracy": 0.6493285819888115, |
| "num_tokens": 5422025.0, |
| "step": 42 |
| }, |
| { |
| "entropy": 1.2684112191200256, |
| "epoch": 0.5548387096774193, |
| "grad_norm": 0.10955163091421127, |
| "learning_rate": 4.77807122597034e-05, |
| "loss": 1.2640540599822998, |
| "mean_token_accuracy": 0.6442403197288513, |
| "num_tokens": 5551149.0, |
| "step": 43 |
| }, |
| { |
| "entropy": 1.202114075422287, |
| "epoch": 0.567741935483871, |
| "grad_norm": 0.10252036899328232, |
| "learning_rate": 4.7632712489564926e-05, |
| "loss": 1.2032322883605957, |
| "mean_token_accuracy": 0.6615511402487755, |
| "num_tokens": 5680701.0, |
| "step": 44 |
| }, |
| { |
| "entropy": 1.2185258865356445, |
| "epoch": 0.5806451612903226, |
| "grad_norm": 0.10858958959579468, |
| "learning_rate": 4.748018037239592e-05, |
| "loss": 1.2210724353790283, |
| "mean_token_accuracy": 0.6543648615479469, |
| "num_tokens": 5808493.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 1.2259098291397095, |
| "epoch": 0.5935483870967742, |
| "grad_norm": 0.10566595196723938, |
| "learning_rate": 4.732314645373921e-05, |
| "loss": 1.2279314994812012, |
| "mean_token_accuracy": 0.6541318669915199, |
| "num_tokens": 5938296.0, |
| "step": 46 |
| }, |
| { |
| "entropy": 1.217906042933464, |
| "epoch": 0.6064516129032258, |
| "grad_norm": 0.10399080067873001, |
| "learning_rate": 4.7161642180652464e-05, |
| "loss": 1.2170512676239014, |
| "mean_token_accuracy": 0.6552224531769753, |
| "num_tokens": 6067439.0, |
| "step": 47 |
| }, |
| { |
| "entropy": 1.243887796998024, |
| "epoch": 0.6193548387096774, |
| "grad_norm": 0.10374640673398972, |
| "learning_rate": 4.699569989541074e-05, |
| "loss": 1.2446861267089844, |
| "mean_token_accuracy": 0.6509256213903427, |
| "num_tokens": 6197510.0, |
| "step": 48 |
| }, |
| { |
| "entropy": 1.2262021899223328, |
| "epoch": 0.632258064516129, |
| "grad_norm": 0.10401046276092529, |
| "learning_rate": 4.6825352829029705e-05, |
| "loss": 1.2263641357421875, |
| "mean_token_accuracy": 0.6529856100678444, |
| "num_tokens": 6327768.0, |
| "step": 49 |
| }, |
| { |
| "entropy": 1.1782560497522354, |
| "epoch": 0.6451612903225806, |
| "grad_norm": 0.10297931730747223, |
| "learning_rate": 4.665063509461097e-05, |
| "loss": 1.1814297437667847, |
| "mean_token_accuracy": 0.6635062471032143, |
| "num_tokens": 6455846.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.2240826934576035, |
| "epoch": 0.6580645161290323, |
| "grad_norm": 0.10559607297182083, |
| "learning_rate": 4.647158168051066e-05, |
| "loss": 1.2128801345825195, |
| "mean_token_accuracy": 0.659131720662117, |
| "num_tokens": 6585457.0, |
| "step": 51 |
| }, |
| { |
| "entropy": 1.2306331396102905, |
| "epoch": 0.6709677419354839, |
| "grad_norm": 0.10558925569057465, |
| "learning_rate": 4.628822844333278e-05, |
| "loss": 1.2250795364379883, |
| "mean_token_accuracy": 0.6541951075196266, |
| "num_tokens": 6716096.0, |
| "step": 52 |
| }, |
| { |
| "entropy": 1.2307018637657166, |
| "epoch": 0.6838709677419355, |
| "grad_norm": 0.106507308781147, |
| "learning_rate": 4.6100612100748765e-05, |
| "loss": 1.216687798500061, |
| "mean_token_accuracy": 0.657420352101326, |
| "num_tokens": 6844341.0, |
| "step": 53 |
| }, |
| { |
| "entropy": 1.1905661970376968, |
| "epoch": 0.6967741935483871, |
| "grad_norm": 0.1018037423491478, |
| "learning_rate": 4.59087702241444e-05, |
| "loss": 1.1824817657470703, |
| "mean_token_accuracy": 0.6638368740677834, |
| "num_tokens": 6971794.0, |
| "step": 54 |
| }, |
| { |
| "entropy": 1.2321506142616272, |
| "epoch": 0.7096774193548387, |
| "grad_norm": 0.1053023412823677, |
| "learning_rate": 4.571274123109606e-05, |
| "loss": 1.2273911237716675, |
| "mean_token_accuracy": 0.6531050652265549, |
| "num_tokens": 7100353.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 1.1877244710922241, |
| "epoch": 0.7225806451612903, |
| "grad_norm": 0.10472644120454788, |
| "learning_rate": 4.551256437767719e-05, |
| "loss": 1.1830646991729736, |
| "mean_token_accuracy": 0.662447988986969, |
| "num_tokens": 7230177.0, |
| "step": 56 |
| }, |
| { |
| "entropy": 1.1354379504919052, |
| "epoch": 0.7354838709677419, |
| "grad_norm": 0.10116837918758392, |
| "learning_rate": 4.530827975059715e-05, |
| "loss": 1.1367874145507812, |
| "mean_token_accuracy": 0.6742443069815636, |
| "num_tokens": 7359042.0, |
| "step": 57 |
| }, |
| { |
| "entropy": 1.1323172450065613, |
| "epoch": 0.7483870967741936, |
| "grad_norm": 0.10149563103914261, |
| "learning_rate": 4.5099928259173516e-05, |
| "loss": 1.1368333101272583, |
| "mean_token_accuracy": 0.6757072061300278, |
| "num_tokens": 7488137.0, |
| "step": 58 |
| }, |
| { |
| "entropy": 1.1689679026603699, |
| "epoch": 0.7612903225806451, |
| "grad_norm": 0.10289032757282257, |
| "learning_rate": 4.488755162713975e-05, |
| "loss": 1.170515775680542, |
| "mean_token_accuracy": 0.6659915670752525, |
| "num_tokens": 7618282.0, |
| "step": 59 |
| }, |
| { |
| "entropy": 1.1511125713586807, |
| "epoch": 0.7741935483870968, |
| "grad_norm": 0.10344893485307693, |
| "learning_rate": 4.467119238428975e-05, |
| "loss": 1.1578235626220703, |
| "mean_token_accuracy": 0.6681751757860184, |
| "num_tokens": 7748017.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.15065398812294, |
| "epoch": 0.7870967741935484, |
| "grad_norm": 0.10410141944885254, |
| "learning_rate": 4.445089385796099e-05, |
| "loss": 1.1545982360839844, |
| "mean_token_accuracy": 0.6693510711193085, |
| "num_tokens": 7876986.0, |
| "step": 61 |
| }, |
| { |
| "entropy": 1.1854369789361954, |
| "epoch": 0.8, |
| "grad_norm": 0.10658932477235794, |
| "learning_rate": 4.422670016435792e-05, |
| "loss": 1.1886390447616577, |
| "mean_token_accuracy": 0.6610049903392792, |
| "num_tokens": 8006794.0, |
| "step": 62 |
| }, |
| { |
| "entropy": 1.1734878718852997, |
| "epoch": 0.8129032258064516, |
| "grad_norm": 0.0999564528465271, |
| "learning_rate": 4.3998656199717435e-05, |
| "loss": 1.1779112815856934, |
| "mean_token_accuracy": 0.6633528992533684, |
| "num_tokens": 8135408.0, |
| "step": 63 |
| }, |
| { |
| "entropy": 1.1888367980718613, |
| "epoch": 0.8258064516129032, |
| "grad_norm": 0.1022685319185257, |
| "learning_rate": 4.3766807631318106e-05, |
| "loss": 1.1841375827789307, |
| "mean_token_accuracy": 0.6636051461100578, |
| "num_tokens": 8265182.0, |
| "step": 64 |
| }, |
| { |
| "entropy": 1.1773615032434464, |
| "epoch": 0.8387096774193549, |
| "grad_norm": 0.10359368473291397, |
| "learning_rate": 4.353120088833501e-05, |
| "loss": 1.1596578359603882, |
| "mean_token_accuracy": 0.6664805710315704, |
| "num_tokens": 8395710.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 1.1735037863254547, |
| "epoch": 0.8516129032258064, |
| "grad_norm": 0.10424409806728363, |
| "learning_rate": 4.329188315254196e-05, |
| "loss": 1.1645269393920898, |
| "mean_token_accuracy": 0.6646556481719017, |
| "num_tokens": 8525026.0, |
| "step": 66 |
| }, |
| { |
| "entropy": 1.1600711345672607, |
| "epoch": 0.864516129032258, |
| "grad_norm": 0.10298816114664078, |
| "learning_rate": 4.3048902348863116e-05, |
| "loss": 1.146191120147705, |
| "mean_token_accuracy": 0.6732870414853096, |
| "num_tokens": 8655176.0, |
| "step": 67 |
| }, |
| { |
| "entropy": 1.1675972044467926, |
| "epoch": 0.8774193548387097, |
| "grad_norm": 0.09964212030172348, |
| "learning_rate": 4.280230713577564e-05, |
| "loss": 1.1639752388000488, |
| "mean_token_accuracy": 0.6678973436355591, |
| "num_tokens": 8784950.0, |
| "step": 68 |
| }, |
| { |
| "entropy": 1.1486363634467125, |
| "epoch": 0.8903225806451613, |
| "grad_norm": 0.10526347905397415, |
| "learning_rate": 4.255214689556557e-05, |
| "loss": 1.1550880670547485, |
| "mean_token_accuracy": 0.669487789273262, |
| "num_tokens": 8915424.0, |
| "step": 69 |
| }, |
| { |
| "entropy": 1.1960512697696686, |
| "epoch": 0.9032258064516129, |
| "grad_norm": 0.10434431582689285, |
| "learning_rate": 4.229847172443866e-05, |
| "loss": 1.195779800415039, |
| "mean_token_accuracy": 0.6591227054595947, |
| "num_tokens": 9043958.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.1383194029331207, |
| "epoch": 0.9161290322580645, |
| "grad_norm": 0.10281933099031448, |
| "learning_rate": 4.204133242248832e-05, |
| "loss": 1.1457536220550537, |
| "mean_token_accuracy": 0.670663595199585, |
| "num_tokens": 9173238.0, |
| "step": 71 |
| }, |
| { |
| "entropy": 1.1284168064594269, |
| "epoch": 0.9290322580645162, |
| "grad_norm": 0.10491844266653061, |
| "learning_rate": 4.1780780483522575e-05, |
| "loss": 1.1363515853881836, |
| "mean_token_accuracy": 0.6729972660541534, |
| "num_tokens": 9303178.0, |
| "step": 72 |
| }, |
| { |
| "entropy": 1.155638337135315, |
| "epoch": 0.9419354838709677, |
| "grad_norm": 0.10739096254110336, |
| "learning_rate": 4.151686808475204e-05, |
| "loss": 1.1668245792388916, |
| "mean_token_accuracy": 0.6647603213787079, |
| "num_tokens": 9432086.0, |
| "step": 73 |
| }, |
| { |
| "entropy": 1.1737090498209, |
| "epoch": 0.9548387096774194, |
| "grad_norm": 0.10476558655500412, |
| "learning_rate": 4.1249648076341165e-05, |
| "loss": 1.1709017753601074, |
| "mean_token_accuracy": 0.6655773147940636, |
| "num_tokens": 9560333.0, |
| "step": 74 |
| }, |
| { |
| "entropy": 1.1429373025894165, |
| "epoch": 0.967741935483871, |
| "grad_norm": 0.09973709285259247, |
| "learning_rate": 4.0979173970824626e-05, |
| "loss": 1.1408506631851196, |
| "mean_token_accuracy": 0.6709853634238243, |
| "num_tokens": 9689775.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 1.131405919790268, |
| "epoch": 0.9806451612903225, |
| "grad_norm": 0.10272659361362457, |
| "learning_rate": 4.070549993239106e-05, |
| "loss": 1.1342134475708008, |
| "mean_token_accuracy": 0.6737439185380936, |
| "num_tokens": 9818901.0, |
| "step": 76 |
| }, |
| { |
| "entropy": 1.1382378786802292, |
| "epoch": 0.9935483870967742, |
| "grad_norm": 0.10625957697629929, |
| "learning_rate": 4.0428680766036384e-05, |
| "loss": 1.1314501762390137, |
| "mean_token_accuracy": 0.6726028546690941, |
| "num_tokens": 9948460.0, |
| "step": 77 |
| }, |
| { |
| "entropy": 1.1366769075393677, |
| "epoch": 1.0, |
| "grad_norm": 0.14248473942279816, |
| "learning_rate": 4.0148771906588706e-05, |
| "loss": 1.120863676071167, |
| "mean_token_accuracy": 0.6762505769729614, |
| "num_tokens": 9997067.0, |
| "step": 78 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 234, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.884823710923489e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
