| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9590544533558464, |
| "eval_steps": 500, |
| "global_step": 1136, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.004221190375685944, |
| "grad_norm": 14.098027229309082, |
| "learning_rate": 2.162496240601504e-06, |
| "loss": 0.3742, |
| "mean_token_accuracy": 0.8917215526103973, |
| "num_tokens": 1015112.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.008442380751371888, |
| "grad_norm": 2.7210590839385986, |
| "learning_rate": 4.865616541353383e-06, |
| "loss": 0.1792, |
| "mean_token_accuracy": 0.950411856174469, |
| "num_tokens": 2007804.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.01266357112705783, |
| "grad_norm": 1.2011040449142456, |
| "learning_rate": 7.568736842105265e-06, |
| "loss": 0.1621, |
| "mean_token_accuracy": 0.9531072795391082, |
| "num_tokens": 3001408.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.016884761502743775, |
| "grad_norm": 1.0074273347854614, |
| "learning_rate": 1.0271857142857143e-05, |
| "loss": 0.1263, |
| "mean_token_accuracy": 0.959876012802124, |
| "num_tokens": 4017789.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.021105951878429716, |
| "grad_norm": 0.9568006992340088, |
| "learning_rate": 1.2974977443609024e-05, |
| "loss": 0.1215, |
| "mean_token_accuracy": 0.9621701657772064, |
| "num_tokens": 5006152.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.02532714225411566, |
| "grad_norm": 0.784275233745575, |
| "learning_rate": 1.5678097744360906e-05, |
| "loss": 0.1171, |
| "mean_token_accuracy": 0.9627571046352387, |
| "num_tokens": 6034663.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.029548332629801603, |
| "grad_norm": 0.8781489133834839, |
| "learning_rate": 1.8381218045112784e-05, |
| "loss": 0.1067, |
| "mean_token_accuracy": 0.9636131703853608, |
| "num_tokens": 7039058.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.03376952300548755, |
| "grad_norm": 1.1658774614334106, |
| "learning_rate": 1.8921417739803738e-05, |
| "loss": 0.1309, |
| "mean_token_accuracy": 0.9636468350887298, |
| "num_tokens": 8040405.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.03799071338117349, |
| "grad_norm": 0.7406081557273865, |
| "learning_rate": 1.8919693842119104e-05, |
| "loss": 0.1004, |
| "mean_token_accuracy": 0.9690930128097535, |
| "num_tokens": 9058721.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.04221190375685943, |
| "grad_norm": 0.5902578234672546, |
| "learning_rate": 1.891664421130453e-05, |
| "loss": 0.1168, |
| "mean_token_accuracy": 0.9648330688476563, |
| "num_tokens": 10034955.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.04643309413254538, |
| "grad_norm": 0.5456152558326721, |
| "learning_rate": 1.8912269417314536e-05, |
| "loss": 0.0826, |
| "mean_token_accuracy": 0.9749838471412658, |
| "num_tokens": 11035568.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.05065428450823132, |
| "grad_norm": 0.484600692987442, |
| "learning_rate": 1.89065702777673e-05, |
| "loss": 0.0865, |
| "mean_token_accuracy": 0.9740271627902984, |
| "num_tokens": 12029633.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.054875474883917266, |
| "grad_norm": 0.7990138530731201, |
| "learning_rate": 1.8899547857791854e-05, |
| "loss": 0.0796, |
| "mean_token_accuracy": 0.9745240807533264, |
| "num_tokens": 13018372.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.05909666525960321, |
| "grad_norm": 0.8144208788871765, |
| "learning_rate": 1.8891203469829024e-05, |
| "loss": 0.0962, |
| "mean_token_accuracy": 0.9704646944999695, |
| "num_tokens": 14001672.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.06331785563528915, |
| "grad_norm": 0.5985013842582703, |
| "learning_rate": 1.8881538673386147e-05, |
| "loss": 0.083, |
| "mean_token_accuracy": 0.9728354692459107, |
| "num_tokens": 14988387.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.0675390460109751, |
| "grad_norm": 0.8302649855613708, |
| "learning_rate": 1.8870555274745598e-05, |
| "loss": 0.087, |
| "mean_token_accuracy": 0.9741718530654907, |
| "num_tokens": 15996580.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.07176023638666104, |
| "grad_norm": 0.6325950026512146, |
| "learning_rate": 1.8858255326627203e-05, |
| "loss": 0.0968, |
| "mean_token_accuracy": 0.9712941884994507, |
| "num_tokens": 16997865.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.07598142676234698, |
| "grad_norm": 0.6739248633384705, |
| "learning_rate": 1.8844641127804635e-05, |
| "loss": 0.0887, |
| "mean_token_accuracy": 0.9721704483032226, |
| "num_tokens": 17998653.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.08020261713803292, |
| "grad_norm": 0.708213210105896, |
| "learning_rate": 1.8829715222675744e-05, |
| "loss": 0.0907, |
| "mean_token_accuracy": 0.9724434196949006, |
| "num_tokens": 18984397.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.08442380751371886, |
| "grad_norm": 0.7431834936141968, |
| "learning_rate": 1.881348040078706e-05, |
| "loss": 0.0993, |
| "mean_token_accuracy": 0.9711518526077271, |
| "num_tokens": 19981414.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.08864499788940482, |
| "grad_norm": 0.49537602066993713, |
| "learning_rate": 1.8795939696312434e-05, |
| "loss": 0.1022, |
| "mean_token_accuracy": 0.9704329133033752, |
| "num_tokens": 20986629.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.09286618826509076, |
| "grad_norm": 0.6155444979667664, |
| "learning_rate": 1.8777096387485975e-05, |
| "loss": 0.0815, |
| "mean_token_accuracy": 0.9749746024608612, |
| "num_tokens": 21983717.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.0970873786407767, |
| "grad_norm": 0.7579423785209656, |
| "learning_rate": 1.8756953995989364e-05, |
| "loss": 0.0747, |
| "mean_token_accuracy": 0.9750826537609101, |
| "num_tokens": 22975708.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.10130856901646264, |
| "grad_norm": 0.6443991661071777, |
| "learning_rate": 1.873551628629368e-05, |
| "loss": 0.0738, |
| "mean_token_accuracy": 0.9769075989723206, |
| "num_tokens": 23992322.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.10552975939214859, |
| "grad_norm": 0.681242048740387, |
| "learning_rate": 1.8712787264955866e-05, |
| "loss": 0.1026, |
| "mean_token_accuracy": 0.9702336013317108, |
| "num_tokens": 24978489.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.10975094976783453, |
| "grad_norm": 0.6523096561431885, |
| "learning_rate": 1.868877117986989e-05, |
| "loss": 0.0974, |
| "mean_token_accuracy": 0.9709220051765441, |
| "num_tokens": 25977116.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.11397214014352047, |
| "grad_norm": 0.5705572962760925, |
| "learning_rate": 1.8663472519472888e-05, |
| "loss": 0.0737, |
| "mean_token_accuracy": 0.9760664999485016, |
| "num_tokens": 26983504.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.11819333051920641, |
| "grad_norm": 0.7862875461578369, |
| "learning_rate": 1.8636896011906278e-05, |
| "loss": 0.0823, |
| "mean_token_accuracy": 0.9751252889633178, |
| "num_tokens": 27984528.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.12241452089489235, |
| "grad_norm": 0.6491385102272034, |
| "learning_rate": 1.8609046624132114e-05, |
| "loss": 0.0969, |
| "mean_token_accuracy": 0.9717353045940399, |
| "num_tokens": 28985575.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.1266357112705783, |
| "grad_norm": 0.4432844817638397, |
| "learning_rate": 1.8579929561004796e-05, |
| "loss": 0.0763, |
| "mean_token_accuracy": 0.9764184832572937, |
| "num_tokens": 29992777.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.13085690164626423, |
| "grad_norm": 0.6792006492614746, |
| "learning_rate": 1.8549550264298318e-05, |
| "loss": 0.0725, |
| "mean_token_accuracy": 0.9753028631210328, |
| "num_tokens": 30988761.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.1350780920219502, |
| "grad_norm": 0.5308874249458313, |
| "learning_rate": 1.851791441168923e-05, |
| "loss": 0.0854, |
| "mean_token_accuracy": 0.975176191329956, |
| "num_tokens": 31984456.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.13929928239763614, |
| "grad_norm": 0.5485326051712036, |
| "learning_rate": 1.8485027915695546e-05, |
| "loss": 0.0735, |
| "mean_token_accuracy": 0.9766063809394836, |
| "num_tokens": 32977478.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.14352047277332208, |
| "grad_norm": 0.586266040802002, |
| "learning_rate": 1.8450896922571697e-05, |
| "loss": 0.0748, |
| "mean_token_accuracy": 0.9772798240184783, |
| "num_tokens": 33974101.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.14774166314900802, |
| "grad_norm": 0.6779407858848572, |
| "learning_rate": 1.841552781115988e-05, |
| "loss": 0.0862, |
| "mean_token_accuracy": 0.9731907069683075, |
| "num_tokens": 34969453.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.15196285352469396, |
| "grad_norm": 0.5682159066200256, |
| "learning_rate": 1.8378927191697873e-05, |
| "loss": 0.0741, |
| "mean_token_accuracy": 0.9774076879024506, |
| "num_tokens": 35970233.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.1561840439003799, |
| "grad_norm": 0.5332077741622925, |
| "learning_rate": 1.8341101904583627e-05, |
| "loss": 0.0634, |
| "mean_token_accuracy": 0.9798624992370606, |
| "num_tokens": 36974750.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.16040523427606584, |
| "grad_norm": 0.629296064376831, |
| "learning_rate": 1.830205901909686e-05, |
| "loss": 0.0814, |
| "mean_token_accuracy": 0.9750245630741119, |
| "num_tokens": 37976351.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.16462642465175178, |
| "grad_norm": 0.5438576340675354, |
| "learning_rate": 1.8261805832077837e-05, |
| "loss": 0.0846, |
| "mean_token_accuracy": 0.9750823318958283, |
| "num_tokens": 38986485.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.16884761502743773, |
| "grad_norm": 0.7107402086257935, |
| "learning_rate": 1.822034986656367e-05, |
| "loss": 0.0767, |
| "mean_token_accuracy": 0.9756914496421814, |
| "num_tokens": 39983288.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1730688054031237, |
| "grad_norm": 0.44018906354904175, |
| "learning_rate": 1.8177698870382283e-05, |
| "loss": 0.0612, |
| "mean_token_accuracy": 0.9797536730766296, |
| "num_tokens": 40979214.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.17728999577880963, |
| "grad_norm": 0.47697558999061584, |
| "learning_rate": 1.8133860814704424e-05, |
| "loss": 0.0768, |
| "mean_token_accuracy": 0.9753372669219971, |
| "num_tokens": 41976378.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.18151118615449557, |
| "grad_norm": 0.4467805027961731, |
| "learning_rate": 1.8088843892553893e-05, |
| "loss": 0.0892, |
| "mean_token_accuracy": 0.9738144636154175, |
| "num_tokens": 42966433.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.18573237653018151, |
| "grad_norm": 0.49840766191482544, |
| "learning_rate": 1.804265651727634e-05, |
| "loss": 0.0605, |
| "mean_token_accuracy": 0.9797145128250122, |
| "num_tokens": 43965812.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.18995356690586745, |
| "grad_norm": 0.6677643060684204, |
| "learning_rate": 1.799530732096684e-05, |
| "loss": 0.068, |
| "mean_token_accuracy": 0.9776868879795074, |
| "num_tokens": 44962121.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.1941747572815534, |
| "grad_norm": 0.5379244089126587, |
| "learning_rate": 1.7946805152856656e-05, |
| "loss": 0.0774, |
| "mean_token_accuracy": 0.9764459311962128, |
| "num_tokens": 45952901.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.19839594765723934, |
| "grad_norm": 0.7632982730865479, |
| "learning_rate": 1.7897159077659345e-05, |
| "loss": 0.0769, |
| "mean_token_accuracy": 0.9761867642402648, |
| "num_tokens": 46964495.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.20261713803292528, |
| "grad_norm": 0.6408265829086304, |
| "learning_rate": 1.784637837387665e-05, |
| "loss": 0.0692, |
| "mean_token_accuracy": 0.9783361613750458, |
| "num_tokens": 47955983.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.20683832840861122, |
| "grad_norm": 0.3746410608291626, |
| "learning_rate": 1.779447253206439e-05, |
| "loss": 0.0609, |
| "mean_token_accuracy": 0.9804293215274811, |
| "num_tokens": 48949400.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.21105951878429718, |
| "grad_norm": 0.36423608660697937, |
| "learning_rate": 1.774145125305878e-05, |
| "loss": 0.0791, |
| "mean_token_accuracy": 0.9767191052436829, |
| "num_tokens": 49956318.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.21528070915998312, |
| "grad_norm": 0.6634694337844849, |
| "learning_rate": 1.768732444616336e-05, |
| "loss": 0.0724, |
| "mean_token_accuracy": 0.978820675611496, |
| "num_tokens": 50969232.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.21950189953566906, |
| "grad_norm": 0.5039066076278687, |
| "learning_rate": 1.7632102227297065e-05, |
| "loss": 0.0855, |
| "mean_token_accuracy": 0.9745490431785584, |
| "num_tokens": 51981247.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.223723089911355, |
| "grad_norm": 0.6930281519889832, |
| "learning_rate": 1.757579491710361e-05, |
| "loss": 0.0825, |
| "mean_token_accuracy": 0.9763229966163636, |
| "num_tokens": 52987145.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.22794428028704095, |
| "grad_norm": 0.3987239599227905, |
| "learning_rate": 1.7518413039022622e-05, |
| "loss": 0.0637, |
| "mean_token_accuracy": 0.978488165140152, |
| "num_tokens": 53996958.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.23216547066272689, |
| "grad_norm": 0.35757359862327576, |
| "learning_rate": 1.7459967317322915e-05, |
| "loss": 0.0611, |
| "mean_token_accuracy": 0.9797202467918396, |
| "num_tokens": 55009068.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.23638666103841283, |
| "grad_norm": 0.5069255828857422, |
| "learning_rate": 1.7400468675098167e-05, |
| "loss": 0.0853, |
| "mean_token_accuracy": 0.9749383151531219, |
| "num_tokens": 56019129.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.24060785141409877, |
| "grad_norm": 0.40278518199920654, |
| "learning_rate": 1.7339928232225487e-05, |
| "loss": 0.0668, |
| "mean_token_accuracy": 0.9784162878990174, |
| "num_tokens": 57017311.0, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.2448290417897847, |
| "grad_norm": 0.512820303440094, |
| "learning_rate": 1.7278357303287197e-05, |
| "loss": 0.0765, |
| "mean_token_accuracy": 0.9770870089530945, |
| "num_tokens": 58009057.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.24905023216547068, |
| "grad_norm": 0.32328370213508606, |
| "learning_rate": 1.7215767395456195e-05, |
| "loss": 0.0602, |
| "mean_token_accuracy": 0.9798195004463196, |
| "num_tokens": 59016847.0, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.2532714225411566, |
| "grad_norm": 0.3779190480709076, |
| "learning_rate": 1.715217020634536e-05, |
| "loss": 0.0826, |
| "mean_token_accuracy": 0.9751770973205567, |
| "num_tokens": 60009790.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.25749261291684256, |
| "grad_norm": 0.5218421220779419, |
| "learning_rate": 1.7087577621821343e-05, |
| "loss": 0.0598, |
| "mean_token_accuracy": 0.980607008934021, |
| "num_tokens": 61010880.0, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.26171380329252847, |
| "grad_norm": 0.35313546657562256, |
| "learning_rate": 1.70220017137832e-05, |
| "loss": 0.0686, |
| "mean_token_accuracy": 0.9785439670085907, |
| "num_tokens": 62016215.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.26593499366821444, |
| "grad_norm": 0.38041552901268005, |
| "learning_rate": 1.69554547379062e-05, |
| "loss": 0.0585, |
| "mean_token_accuracy": 0.9798943281173706, |
| "num_tokens": 63015841.0, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.2701561840439004, |
| "grad_norm": 0.467557430267334, |
| "learning_rate": 1.6887949131351377e-05, |
| "loss": 0.0548, |
| "mean_token_accuracy": 0.9805089175701142, |
| "num_tokens": 64019133.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.2743773744195863, |
| "grad_norm": 0.46841275691986084, |
| "learning_rate": 1.6819497510441067e-05, |
| "loss": 0.0777, |
| "mean_token_accuracy": 0.9744618594646454, |
| "num_tokens": 65027125.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.2785985647952723, |
| "grad_norm": 0.414924293756485, |
| "learning_rate": 1.6750112668301037e-05, |
| "loss": 0.0636, |
| "mean_token_accuracy": 0.9790864646434784, |
| "num_tokens": 66014932.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.2828197551709582, |
| "grad_norm": 0.4480653703212738, |
| "learning_rate": 1.6679807572469516e-05, |
| "loss": 0.0754, |
| "mean_token_accuracy": 0.9772217273712158, |
| "num_tokens": 67000562.0, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.28704094554664417, |
| "grad_norm": 0.5654832124710083, |
| "learning_rate": 1.6608595362473684e-05, |
| "loss": 0.0776, |
| "mean_token_accuracy": 0.9737156569957733, |
| "num_tokens": 68005427.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.2912621359223301, |
| "grad_norm": 0.437466561794281, |
| "learning_rate": 1.6536489347373948e-05, |
| "loss": 0.0627, |
| "mean_token_accuracy": 0.9805977463722229, |
| "num_tokens": 68998626.0, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.29548332629801605, |
| "grad_norm": 0.5042473077774048, |
| "learning_rate": 1.6463503003276605e-05, |
| "loss": 0.0633, |
| "mean_token_accuracy": 0.9787137031555175, |
| "num_tokens": 70020109.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.29970451667370196, |
| "grad_norm": 0.31632962822914124, |
| "learning_rate": 1.6389649970815236e-05, |
| "loss": 0.0574, |
| "mean_token_accuracy": 0.9804539918899536, |
| "num_tokens": 71028280.0, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.3039257070493879, |
| "grad_norm": 0.5246463418006897, |
| "learning_rate": 1.6314944052601375e-05, |
| "loss": 0.067, |
| "mean_token_accuracy": 0.9787373781204224, |
| "num_tokens": 72043894.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.3081468974250739, |
| "grad_norm": 0.36841967701911926, |
| "learning_rate": 1.6239399210644896e-05, |
| "loss": 0.0571, |
| "mean_token_accuracy": 0.9811227321624756, |
| "num_tokens": 73048135.0, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.3123680878007598, |
| "grad_norm": 0.49402135610580444, |
| "learning_rate": 1.6163029563744605e-05, |
| "loss": 0.0757, |
| "mean_token_accuracy": 0.9763616681098938, |
| "num_tokens": 74033780.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.3165892781764458, |
| "grad_norm": 0.5389291644096375, |
| "learning_rate": 1.6085849384849547e-05, |
| "loss": 0.0648, |
| "mean_token_accuracy": 0.9801723957061768, |
| "num_tokens": 75036067.0, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.3208104685521317, |
| "grad_norm": 0.7108619213104248, |
| "learning_rate": 1.600787309839149e-05, |
| "loss": 0.0772, |
| "mean_token_accuracy": 0.9763815343379975, |
| "num_tokens": 76037294.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.32503165892781766, |
| "grad_norm": 0.5434122085571289, |
| "learning_rate": 1.592911527758908e-05, |
| "loss": 0.0735, |
| "mean_token_accuracy": 0.9775534689426422, |
| "num_tokens": 77026856.0, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.32925284930350357, |
| "grad_norm": 0.5309452414512634, |
| "learning_rate": 1.5849590641724254e-05, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9794178307056427, |
| "num_tokens": 78032516.0, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.33347403967918954, |
| "grad_norm": 0.5379695892333984, |
| "learning_rate": 1.5769314053391268e-05, |
| "loss": 0.0739, |
| "mean_token_accuracy": 0.9780141413211823, |
| "num_tokens": 79038536.0, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.33769523005487545, |
| "grad_norm": 0.5310766696929932, |
| "learning_rate": 1.5688300515718992e-05, |
| "loss": 0.0745, |
| "mean_token_accuracy": 0.976209282875061, |
| "num_tokens": 80033980.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.3419164204305614, |
| "grad_norm": 0.33774712681770325, |
| "learning_rate": 1.5606565169566948e-05, |
| "loss": 0.0624, |
| "mean_token_accuracy": 0.9799854457378387, |
| "num_tokens": 81025858.0, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.3461376108062474, |
| "grad_norm": 0.3490382134914398, |
| "learning_rate": 1.5524123290695567e-05, |
| "loss": 0.0526, |
| "mean_token_accuracy": 0.9817829906940461, |
| "num_tokens": 82020883.0, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.3503588011819333, |
| "grad_norm": 0.43765324354171753, |
| "learning_rate": 1.5440990286911282e-05, |
| "loss": 0.085, |
| "mean_token_accuracy": 0.9749090909957886, |
| "num_tokens": 83016121.0, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.35457999155761927, |
| "grad_norm": 0.3650844991207123, |
| "learning_rate": 1.5357181695186886e-05, |
| "loss": 0.072, |
| "mean_token_accuracy": 0.9782340586185455, |
| "num_tokens": 84013389.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.3588011819333052, |
| "grad_norm": 0.38040757179260254, |
| "learning_rate": 1.5272713178757815e-05, |
| "loss": 0.0644, |
| "mean_token_accuracy": 0.9801673710346221, |
| "num_tokens": 85029073.0, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.36302237230899115, |
| "grad_norm": 0.4018385112285614, |
| "learning_rate": 1.5187600524194783e-05, |
| "loss": 0.0704, |
| "mean_token_accuracy": 0.97780482172966, |
| "num_tokens": 86022619.0, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.36724356268467706, |
| "grad_norm": 0.38161927461624146, |
| "learning_rate": 1.5101859638453367e-05, |
| "loss": 0.059, |
| "mean_token_accuracy": 0.981672877073288, |
| "num_tokens": 87008632.0, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.37146475306036303, |
| "grad_norm": 0.5053279995918274, |
| "learning_rate": 1.5015506545901146e-05, |
| "loss": 0.068, |
| "mean_token_accuracy": 0.9787163019180298, |
| "num_tokens": 88012179.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.37568594343604894, |
| "grad_norm": 0.6230582594871521, |
| "learning_rate": 1.4928557385322827e-05, |
| "loss": 0.0646, |
| "mean_token_accuracy": 0.9803561389446258, |
| "num_tokens": 88999968.0, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.3799071338117349, |
| "grad_norm": 0.3518192172050476, |
| "learning_rate": 1.4841028406904036e-05, |
| "loss": 0.0799, |
| "mean_token_accuracy": 0.9754970967769623, |
| "num_tokens": 90004569.0, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.3841283241874209, |
| "grad_norm": 0.5927233695983887, |
| "learning_rate": 1.475293596919427e-05, |
| "loss": 0.0576, |
| "mean_token_accuracy": 0.9809002757072449, |
| "num_tokens": 91001483.0, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.3883495145631068, |
| "grad_norm": 0.39552298188209534, |
| "learning_rate": 1.4664296536049597e-05, |
| "loss": 0.0573, |
| "mean_token_accuracy": 0.9819892585277558, |
| "num_tokens": 91998968.0, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.39257070493879276, |
| "grad_norm": 0.6548199653625488, |
| "learning_rate": 1.4575126673555682e-05, |
| "loss": 0.0797, |
| "mean_token_accuracy": 0.9763227999210358, |
| "num_tokens": 92987326.0, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.39679189531447867, |
| "grad_norm": 0.44273531436920166, |
| "learning_rate": 1.4485443046931691e-05, |
| "loss": 0.0765, |
| "mean_token_accuracy": 0.9767032504081726, |
| "num_tokens": 93986087.0, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.40101308569016464, |
| "grad_norm": 0.3193528950214386, |
| "learning_rate": 1.4395262417415695e-05, |
| "loss": 0.0682, |
| "mean_token_accuracy": 0.978713858127594, |
| "num_tokens": 94964202.0, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.40523427606585055, |
| "grad_norm": 0.523923933506012, |
| "learning_rate": 1.4304601639132101e-05, |
| "loss": 0.0806, |
| "mean_token_accuracy": 0.9752468287944793, |
| "num_tokens": 95969328.0, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.4094554664415365, |
| "grad_norm": 0.5847504138946533, |
| "learning_rate": 1.421347765594174e-05, |
| "loss": 0.0744, |
| "mean_token_accuracy": 0.9762375295162201, |
| "num_tokens": 96961433.0, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.41367665681722243, |
| "grad_norm": 0.44322511553764343, |
| "learning_rate": 1.412190749827517e-05, |
| "loss": 0.0633, |
| "mean_token_accuracy": 0.979754900932312, |
| "num_tokens": 97964016.0, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.4178978471929084, |
| "grad_norm": 0.6941162943840027, |
| "learning_rate": 1.4029908279949832e-05, |
| "loss": 0.0744, |
| "mean_token_accuracy": 0.9761178731918335, |
| "num_tokens": 98967481.0, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.42211903756859437, |
| "grad_norm": 0.5297139286994934, |
| "learning_rate": 1.3937497194971562e-05, |
| "loss": 0.0764, |
| "mean_token_accuracy": 0.97714062333107, |
| "num_tokens": 99962064.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4263402279442803, |
| "grad_norm": 0.4501569867134094, |
| "learning_rate": 1.3844691514321187e-05, |
| "loss": 0.0559, |
| "mean_token_accuracy": 0.9815985262393951, |
| "num_tokens": 100966846.0, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.43056141831996625, |
| "grad_norm": 0.30753740668296814, |
| "learning_rate": 1.3751508582726678e-05, |
| "loss": 0.0628, |
| "mean_token_accuracy": 0.9812352299690247, |
| "num_tokens": 101981290.0, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.43478260869565216, |
| "grad_norm": 0.498076856136322, |
| "learning_rate": 1.3657965815421552e-05, |
| "loss": 0.0648, |
| "mean_token_accuracy": 0.9800419569015503, |
| "num_tokens": 102984061.0, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.43900379907133813, |
| "grad_norm": 0.29834893345832825, |
| "learning_rate": 1.3564080694890087e-05, |
| "loss": 0.0693, |
| "mean_token_accuracy": 0.9796207189559937, |
| "num_tokens": 103987394.0, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.44322498944702404, |
| "grad_norm": 0.3692452013492584, |
| "learning_rate": 1.3469870767599967e-05, |
| "loss": 0.0707, |
| "mean_token_accuracy": 0.9781329989433288, |
| "num_tokens": 104978867.0, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.44744617982271, |
| "grad_norm": 0.5009628534317017, |
| "learning_rate": 1.3375353640722964e-05, |
| "loss": 0.0658, |
| "mean_token_accuracy": 0.9787944197654724, |
| "num_tokens": 105980380.0, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.4516673701983959, |
| "grad_norm": 0.3739317059516907, |
| "learning_rate": 1.3280546978844306e-05, |
| "loss": 0.061, |
| "mean_token_accuracy": 0.9798948109149933, |
| "num_tokens": 106983460.0, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.4558885605740819, |
| "grad_norm": 0.35601434111595154, |
| "learning_rate": 1.3185468500661253e-05, |
| "loss": 0.058, |
| "mean_token_accuracy": 0.9817588865756989, |
| "num_tokens": 107982873.0, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.46010975094976786, |
| "grad_norm": 0.4605794847011566, |
| "learning_rate": 1.3090135975671632e-05, |
| "loss": 0.0663, |
| "mean_token_accuracy": 0.9795182943344116, |
| "num_tokens": 108978610.0, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.46433094132545377, |
| "grad_norm": 0.4092525243759155, |
| "learning_rate": 1.2994567220852818e-05, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9795619785785675, |
| "num_tokens": 110008286.0, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.46855213170113974, |
| "grad_norm": 0.4141243100166321, |
| "learning_rate": 1.289878009733189e-05, |
| "loss": 0.0623, |
| "mean_token_accuracy": 0.9800220608711243, |
| "num_tokens": 110991293.0, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.47277332207682565, |
| "grad_norm": 0.30998215079307556, |
| "learning_rate": 1.2802792507047485e-05, |
| "loss": 0.056, |
| "mean_token_accuracy": 0.9825344681739807, |
| "num_tokens": 111994073.0, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.4769945124525116, |
| "grad_norm": 0.41842731833457947, |
| "learning_rate": 1.2706622389404078e-05, |
| "loss": 0.058, |
| "mean_token_accuracy": 0.9809507966041565, |
| "num_tokens": 112998715.0, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.48121570282819753, |
| "grad_norm": 0.46353113651275635, |
| "learning_rate": 1.2610287717919215e-05, |
| "loss": 0.0579, |
| "mean_token_accuracy": 0.9815491259098053, |
| "num_tokens": 114007676.0, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.4854368932038835, |
| "grad_norm": 0.36348381638526917, |
| "learning_rate": 1.2513806496864394e-05, |
| "loss": 0.0605, |
| "mean_token_accuracy": 0.9798177361488343, |
| "num_tokens": 115007878.0, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.4896580835795694, |
| "grad_norm": 0.43649837374687195, |
| "learning_rate": 1.2417196757900202e-05, |
| "loss": 0.0598, |
| "mean_token_accuracy": 0.9803382277488708, |
| "num_tokens": 116015639.0, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.4938792739552554, |
| "grad_norm": 0.5160219669342041, |
| "learning_rate": 1.2320476556706314e-05, |
| "loss": 0.0731, |
| "mean_token_accuracy": 0.9775352954864502, |
| "num_tokens": 116991747.0, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.49810046433094135, |
| "grad_norm": 0.527754008769989, |
| "learning_rate": 1.2223663969607017e-05, |
| "loss": 0.063, |
| "mean_token_accuracy": 0.9807395756244659, |
| "num_tokens": 117983025.0, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.5023216547066273, |
| "grad_norm": 0.3984428346157074, |
| "learning_rate": 1.2126777090192877e-05, |
| "loss": 0.0581, |
| "mean_token_accuracy": 0.9815522134304047, |
| "num_tokens": 118974311.0, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.5065428450823132, |
| "grad_norm": 0.29826706647872925, |
| "learning_rate": 1.2029834025939158e-05, |
| "loss": 0.064, |
| "mean_token_accuracy": 0.9798329710960388, |
| "num_tokens": 119955478.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.5107640354579992, |
| "grad_norm": 0.43030846118927, |
| "learning_rate": 1.1932852894821697e-05, |
| "loss": 0.0542, |
| "mean_token_accuracy": 0.9823975384235382, |
| "num_tokens": 120961826.0, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.5149852258336851, |
| "grad_norm": 0.5718479156494141, |
| "learning_rate": 1.1835851821930737e-05, |
| "loss": 0.0854, |
| "mean_token_accuracy": 0.9766719162464141, |
| "num_tokens": 121946291.0, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.519206416209371, |
| "grad_norm": 0.27885958552360535, |
| "learning_rate": 1.1738848936083515e-05, |
| "loss": 0.0549, |
| "mean_token_accuracy": 0.9819578707218171, |
| "num_tokens": 122946869.0, |
| "step": 615 |
| }, |
| { |
| "epoch": 0.5234276065850569, |
| "grad_norm": 0.4546900987625122, |
| "learning_rate": 1.1641862366436083e-05, |
| "loss": 0.0628, |
| "mean_token_accuracy": 0.980167806148529, |
| "num_tokens": 123934725.0, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.527648796960743, |
| "grad_norm": 0.36613330245018005, |
| "learning_rate": 1.1544910239095123e-05, |
| "loss": 0.0665, |
| "mean_token_accuracy": 0.9814972579479218, |
| "num_tokens": 124924573.0, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.5318699873364289, |
| "grad_norm": 0.6353659629821777, |
| "learning_rate": 1.1448010673730277e-05, |
| "loss": 0.0701, |
| "mean_token_accuracy": 0.9788794994354248, |
| "num_tokens": 125923716.0, |
| "step": 630 |
| }, |
| { |
| "epoch": 0.5360911777121148, |
| "grad_norm": 0.38121145963668823, |
| "learning_rate": 1.1351181780187745e-05, |
| "loss": 0.0638, |
| "mean_token_accuracy": 0.978546005487442, |
| "num_tokens": 126922247.0, |
| "step": 635 |
| }, |
| { |
| "epoch": 0.5403123680878008, |
| "grad_norm": 0.3332112729549408, |
| "learning_rate": 1.1254441655105638e-05, |
| "loss": 0.0685, |
| "mean_token_accuracy": 0.9787221193313599, |
| "num_tokens": 127926789.0, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.5445335584634867, |
| "grad_norm": 0.3463106155395508, |
| "learning_rate": 1.1157808378531889e-05, |
| "loss": 0.0612, |
| "mean_token_accuracy": 0.9802833020687103, |
| "num_tokens": 128939279.0, |
| "step": 645 |
| }, |
| { |
| "epoch": 0.5487547488391726, |
| "grad_norm": 0.40639668703079224, |
| "learning_rate": 1.1061300010545174e-05, |
| "loss": 0.0486, |
| "mean_token_accuracy": 0.9833217918872833, |
| "num_tokens": 129935683.0, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.5529759392148585, |
| "grad_norm": 0.5526772737503052, |
| "learning_rate": 1.0964934587879663e-05, |
| "loss": 0.0616, |
| "mean_token_accuracy": 0.9801701128482818, |
| "num_tokens": 130919686.0, |
| "step": 655 |
| }, |
| { |
| "epoch": 0.5571971295905446, |
| "grad_norm": 0.4757324755191803, |
| "learning_rate": 1.0868730120554022e-05, |
| "loss": 0.0754, |
| "mean_token_accuracy": 0.9766432046890259, |
| "num_tokens": 131922785.0, |
| "step": 660 |
| }, |
| { |
| "epoch": 0.5614183199662305, |
| "grad_norm": 0.44919857382774353, |
| "learning_rate": 1.077270458850552e-05, |
| "loss": 0.0686, |
| "mean_token_accuracy": 0.9786693751811981, |
| "num_tokens": 132925457.0, |
| "step": 665 |
| }, |
| { |
| "epoch": 0.5656395103419164, |
| "grad_norm": 0.4338498115539551, |
| "learning_rate": 1.0676875938229672e-05, |
| "loss": 0.0828, |
| "mean_token_accuracy": 0.9770305335521698, |
| "num_tokens": 133920212.0, |
| "step": 670 |
| }, |
| { |
| "epoch": 0.5698607007176023, |
| "grad_norm": 0.4849966764450073, |
| "learning_rate": 1.0581262079426185e-05, |
| "loss": 0.0684, |
| "mean_token_accuracy": 0.9776960253715515, |
| "num_tokens": 134928292.0, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.5740818910932883, |
| "grad_norm": 0.4540240168571472, |
| "learning_rate": 1.0485880881651739e-05, |
| "loss": 0.0666, |
| "mean_token_accuracy": 0.9798371970653534, |
| "num_tokens": 135932867.0, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.5783030814689742, |
| "grad_norm": 0.39765167236328125, |
| "learning_rate": 1.039075017098032e-05, |
| "loss": 0.0489, |
| "mean_token_accuracy": 0.9826711773872375, |
| "num_tokens": 136936844.0, |
| "step": 685 |
| }, |
| { |
| "epoch": 0.5825242718446602, |
| "grad_norm": 0.3820544183254242, |
| "learning_rate": 1.0295887726671634e-05, |
| "loss": 0.0654, |
| "mean_token_accuracy": 0.9792472183704376, |
| "num_tokens": 137935792.0, |
| "step": 690 |
| }, |
| { |
| "epoch": 0.5867454622203462, |
| "grad_norm": 0.7194415330886841, |
| "learning_rate": 1.0201311277848296e-05, |
| "loss": 0.0764, |
| "mean_token_accuracy": 0.9776616096496582, |
| "num_tokens": 138946053.0, |
| "step": 695 |
| }, |
| { |
| "epoch": 0.5909666525960321, |
| "grad_norm": 0.48795053362846375, |
| "learning_rate": 1.010703850018237e-05, |
| "loss": 0.0533, |
| "mean_token_accuracy": 0.9802048861980438, |
| "num_tokens": 139953587.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.595187842971718, |
| "grad_norm": 0.7714407444000244, |
| "learning_rate": 1.0013087012591931e-05, |
| "loss": 0.0563, |
| "mean_token_accuracy": 0.9834475040435791, |
| "num_tokens": 140953652.0, |
| "step": 705 |
| }, |
| { |
| "epoch": 0.5994090333474039, |
| "grad_norm": 0.3944249451160431, |
| "learning_rate": 9.919474373948201e-06, |
| "loss": 0.0527, |
| "mean_token_accuracy": 0.9824005424976349, |
| "num_tokens": 141959115.0, |
| "step": 710 |
| }, |
| { |
| "epoch": 0.6036302237230899, |
| "grad_norm": 0.39308813214302063, |
| "learning_rate": 9.826218079793912e-06, |
| "loss": 0.0713, |
| "mean_token_accuracy": 0.9796823501586914, |
| "num_tokens": 142955339.0, |
| "step": 715 |
| }, |
| { |
| "epoch": 0.6078514140987759, |
| "grad_norm": 0.536405622959137, |
| "learning_rate": 9.733335559073532e-06, |
| "loss": 0.065, |
| "mean_token_accuracy": 0.9791229665279388, |
| "num_tokens": 143941329.0, |
| "step": 720 |
| }, |
| { |
| "epoch": 0.6120726044744618, |
| "grad_norm": 0.4547133147716522, |
| "learning_rate": 9.640844170875895e-06, |
| "loss": 0.0562, |
| "mean_token_accuracy": 0.9817556202411651, |
| "num_tokens": 144949773.0, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.6162937948501478, |
| "grad_norm": 0.3856240212917328, |
| "learning_rate": 9.548761201189938e-06, |
| "loss": 0.0632, |
| "mean_token_accuracy": 0.9790012300014496, |
| "num_tokens": 145965019.0, |
| "step": 730 |
| }, |
| { |
| "epoch": 0.6205149852258337, |
| "grad_norm": 0.3784791827201843, |
| "learning_rate": 9.457103859674036e-06, |
| "loss": 0.0573, |
| "mean_token_accuracy": 0.9810520112514496, |
| "num_tokens": 146966217.0, |
| "step": 735 |
| }, |
| { |
| "epoch": 0.6247361756015196, |
| "grad_norm": 0.694983720779419, |
| "learning_rate": 9.365889276439675e-06, |
| "loss": 0.0717, |
| "mean_token_accuracy": 0.9780777871608735, |
| "num_tokens": 147977344.0, |
| "step": 740 |
| }, |
| { |
| "epoch": 0.6289573659772055, |
| "grad_norm": 0.5198516845703125, |
| "learning_rate": 9.275134498849924e-06, |
| "loss": 0.0723, |
| "mean_token_accuracy": 0.9758940279483795, |
| "num_tokens": 148972986.0, |
| "step": 745 |
| }, |
| { |
| "epoch": 0.6331785563528916, |
| "grad_norm": 0.5585047006607056, |
| "learning_rate": 9.184856488333424e-06, |
| "loss": 0.0637, |
| "mean_token_accuracy": 0.9806169748306275, |
| "num_tokens": 149972677.0, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.6373997467285775, |
| "grad_norm": 0.3646729588508606, |
| "learning_rate": 9.095072117214402e-06, |
| "loss": 0.0528, |
| "mean_token_accuracy": 0.9825437963008881, |
| "num_tokens": 150969252.0, |
| "step": 755 |
| }, |
| { |
| "epoch": 0.6416209371042634, |
| "grad_norm": 0.3320753872394562, |
| "learning_rate": 9.00579816555937e-06, |
| "loss": 0.0576, |
| "mean_token_accuracy": 0.980225658416748, |
| "num_tokens": 151969368.0, |
| "step": 760 |
| }, |
| { |
| "epoch": 0.6458421274799493, |
| "grad_norm": 0.4518062174320221, |
| "learning_rate": 8.91705131804102e-06, |
| "loss": 0.0588, |
| "mean_token_accuracy": 0.9812731683254242, |
| "num_tokens": 152981304.0, |
| "step": 765 |
| }, |
| { |
| "epoch": 0.6500633178556353, |
| "grad_norm": 0.4851647615432739, |
| "learning_rate": 8.828848160820007e-06, |
| "loss": 0.0719, |
| "mean_token_accuracy": 0.9766692161560059, |
| "num_tokens": 153989333.0, |
| "step": 770 |
| }, |
| { |
| "epoch": 0.6542845082313212, |
| "grad_norm": 0.3385196924209595, |
| "learning_rate": 8.7412051784451e-06, |
| "loss": 0.0547, |
| "mean_token_accuracy": 0.9820794582366943, |
| "num_tokens": 154992382.0, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.6585056986070071, |
| "grad_norm": 0.6140791773796082, |
| "learning_rate": 8.65413875077233e-06, |
| "loss": 0.0625, |
| "mean_token_accuracy": 0.9814089715480805, |
| "num_tokens": 155982888.0, |
| "step": 780 |
| }, |
| { |
| "epoch": 0.6627268889826932, |
| "grad_norm": 0.3975711166858673, |
| "learning_rate": 8.567665149903717e-06, |
| "loss": 0.0743, |
| "mean_token_accuracy": 0.9773818850517273, |
| "num_tokens": 156977379.0, |
| "step": 785 |
| }, |
| { |
| "epoch": 0.6669480793583791, |
| "grad_norm": 0.35105836391448975, |
| "learning_rate": 8.481800537146122e-06, |
| "loss": 0.0559, |
| "mean_token_accuracy": 0.9833659827709198, |
| "num_tokens": 157977104.0, |
| "step": 790 |
| }, |
| { |
| "epoch": 0.671169269734065, |
| "grad_norm": 0.4171176552772522, |
| "learning_rate": 8.39656095999083e-06, |
| "loss": 0.0619, |
| "mean_token_accuracy": 0.9801605641841888, |
| "num_tokens": 158962353.0, |
| "step": 795 |
| }, |
| { |
| "epoch": 0.6753904601097509, |
| "grad_norm": 0.40441545844078064, |
| "learning_rate": 8.311962349114364e-06, |
| "loss": 0.0675, |
| "mean_token_accuracy": 0.9791892468929291, |
| "num_tokens": 159964238.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.6796116504854369, |
| "grad_norm": 0.2904331386089325, |
| "learning_rate": 8.228020515401172e-06, |
| "loss": 0.0499, |
| "mean_token_accuracy": 0.9829580664634705, |
| "num_tokens": 160962411.0, |
| "step": 805 |
| }, |
| { |
| "epoch": 0.6838328408611228, |
| "grad_norm": 0.38575753569602966, |
| "learning_rate": 8.144751146988668e-06, |
| "loss": 0.0546, |
| "mean_token_accuracy": 0.9813836991786957, |
| "num_tokens": 161968430.0, |
| "step": 810 |
| }, |
| { |
| "epoch": 0.6880540312368087, |
| "grad_norm": 0.43409463763237, |
| "learning_rate": 8.062169806335251e-06, |
| "loss": 0.0697, |
| "mean_token_accuracy": 0.9776336848735809, |
| "num_tokens": 162980798.0, |
| "step": 815 |
| }, |
| { |
| "epoch": 0.6922752216124948, |
| "grad_norm": 0.4342169463634491, |
| "learning_rate": 7.980291927311761e-06, |
| "loss": 0.0532, |
| "mean_token_accuracy": 0.9819177865982056, |
| "num_tokens": 163983922.0, |
| "step": 820 |
| }, |
| { |
| "epoch": 0.6964964119881807, |
| "grad_norm": 0.3181346356868744, |
| "learning_rate": 7.899132812317034e-06, |
| "loss": 0.0605, |
| "mean_token_accuracy": 0.98127081990242, |
| "num_tokens": 164979475.0, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.7007176023638666, |
| "grad_norm": 0.3032984435558319, |
| "learning_rate": 7.818707629417953e-06, |
| "loss": 0.0586, |
| "mean_token_accuracy": 0.9816713869571686, |
| "num_tokens": 165986361.0, |
| "step": 830 |
| }, |
| { |
| "epoch": 0.7049387927395525, |
| "grad_norm": 0.2787056863307953, |
| "learning_rate": 7.739031409514682e-06, |
| "loss": 0.0541, |
| "mean_token_accuracy": 0.9827538132667542, |
| "num_tokens": 166981954.0, |
| "step": 835 |
| }, |
| { |
| "epoch": 0.7091599831152385, |
| "grad_norm": 0.3149995505809784, |
| "learning_rate": 7.66011904353146e-06, |
| "loss": 0.0547, |
| "mean_token_accuracy": 0.9833314538002014, |
| "num_tokens": 167973037.0, |
| "step": 840 |
| }, |
| { |
| "epoch": 0.7133811734909244, |
| "grad_norm": 0.5108799338340759, |
| "learning_rate": 7.581985279633619e-06, |
| "loss": 0.0759, |
| "mean_token_accuracy": 0.9779969990253449, |
| "num_tokens": 168968687.0, |
| "step": 845 |
| }, |
| { |
| "epoch": 0.7176023638666104, |
| "grad_norm": 0.2869400382041931, |
| "learning_rate": 7.504644720471242e-06, |
| "loss": 0.0503, |
| "mean_token_accuracy": 0.9832251667976379, |
| "num_tokens": 169961683.0, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.7218235542422963, |
| "grad_norm": 0.3674737215042114, |
| "learning_rate": 7.428111820450033e-06, |
| "loss": 0.0555, |
| "mean_token_accuracy": 0.9819011390209198, |
| "num_tokens": 170956044.0, |
| "step": 855 |
| }, |
| { |
| "epoch": 0.7260447446179823, |
| "grad_norm": 0.3464158773422241, |
| "learning_rate": 7.352400883029895e-06, |
| "loss": 0.0578, |
| "mean_token_accuracy": 0.981655764579773, |
| "num_tokens": 171953288.0, |
| "step": 860 |
| }, |
| { |
| "epoch": 0.7302659349936682, |
| "grad_norm": 0.5077041983604431, |
| "learning_rate": 7.277526058051712e-06, |
| "loss": 0.0713, |
| "mean_token_accuracy": 0.9765253722667694, |
| "num_tokens": 172950908.0, |
| "step": 865 |
| }, |
| { |
| "epoch": 0.7344871253693541, |
| "grad_norm": 0.3088940382003784, |
| "learning_rate": 7.203501339092841e-06, |
| "loss": 0.0589, |
| "mean_token_accuracy": 0.9817701578140259, |
| "num_tokens": 173950644.0, |
| "step": 870 |
| }, |
| { |
| "epoch": 0.7387083157450401, |
| "grad_norm": 0.41986432671546936, |
| "learning_rate": 7.1303405608518076e-06, |
| "loss": 0.0683, |
| "mean_token_accuracy": 0.9800974488258362, |
| "num_tokens": 174962849.0, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.7429295061207261, |
| "grad_norm": 0.405791312456131, |
| "learning_rate": 7.0580573965627105e-06, |
| "loss": 0.0581, |
| "mean_token_accuracy": 0.9800466537475586, |
| "num_tokens": 175961404.0, |
| "step": 880 |
| }, |
| { |
| "epoch": 0.747150696496412, |
| "grad_norm": 0.2580551505088806, |
| "learning_rate": 6.986665355439776e-06, |
| "loss": 0.045, |
| "mean_token_accuracy": 0.9854241788387299, |
| "num_tokens": 176971909.0, |
| "step": 885 |
| }, |
| { |
| "epoch": 0.7513718868720979, |
| "grad_norm": 0.4034690260887146, |
| "learning_rate": 6.916177780152595e-06, |
| "loss": 0.0666, |
| "mean_token_accuracy": 0.9804459571838379, |
| "num_tokens": 177979657.0, |
| "step": 890 |
| }, |
| { |
| "epoch": 0.7555930772477839, |
| "grad_norm": 0.3841191828250885, |
| "learning_rate": 6.846607844332456e-06, |
| "loss": 0.0429, |
| "mean_token_accuracy": 0.9856389760971069, |
| "num_tokens": 178998823.0, |
| "step": 895 |
| }, |
| { |
| "epoch": 0.7598142676234698, |
| "grad_norm": 0.34408411383628845, |
| "learning_rate": 6.777968550110304e-06, |
| "loss": 0.0596, |
| "mean_token_accuracy": 0.981190150976181, |
| "num_tokens": 180000875.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.7640354579991557, |
| "grad_norm": 0.4975771903991699, |
| "learning_rate": 6.710272725686721e-06, |
| "loss": 0.062, |
| "mean_token_accuracy": 0.9801162958145142, |
| "num_tokens": 180980656.0, |
| "step": 905 |
| }, |
| { |
| "epoch": 0.7682566483748418, |
| "grad_norm": 0.3257213830947876, |
| "learning_rate": 6.6435330229344295e-06, |
| "loss": 0.0556, |
| "mean_token_accuracy": 0.9824772357940674, |
| "num_tokens": 181987680.0, |
| "step": 910 |
| }, |
| { |
| "epoch": 0.7724778387505277, |
| "grad_norm": 0.27520373463630676, |
| "learning_rate": 6.577761915033747e-06, |
| "loss": 0.0554, |
| "mean_token_accuracy": 0.9823514938354492, |
| "num_tokens": 182984683.0, |
| "step": 915 |
| }, |
| { |
| "epoch": 0.7766990291262136, |
| "grad_norm": 0.41581806540489197, |
| "learning_rate": 6.512971694141437e-06, |
| "loss": 0.0627, |
| "mean_token_accuracy": 0.9802701056003571, |
| "num_tokens": 183971189.0, |
| "step": 920 |
| }, |
| { |
| "epoch": 0.7809202195018995, |
| "grad_norm": 0.4702225923538208, |
| "learning_rate": 6.449174469093386e-06, |
| "loss": 0.0632, |
| "mean_token_accuracy": 0.9811832010746002, |
| "num_tokens": 184969949.0, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.7851414098775855, |
| "grad_norm": 0.49398288130760193, |
| "learning_rate": 6.38638216314156e-06, |
| "loss": 0.0516, |
| "mean_token_accuracy": 0.9828024804592133, |
| "num_tokens": 185960010.0, |
| "step": 930 |
| }, |
| { |
| "epoch": 0.7893626002532714, |
| "grad_norm": 0.40433740615844727, |
| "learning_rate": 6.324606511725622e-06, |
| "loss": 0.0494, |
| "mean_token_accuracy": 0.9838074147701263, |
| "num_tokens": 186963900.0, |
| "step": 935 |
| }, |
| { |
| "epoch": 0.7935837906289573, |
| "grad_norm": 0.389503538608551, |
| "learning_rate": 6.263859060279658e-06, |
| "loss": 0.0573, |
| "mean_token_accuracy": 0.9814595222473145, |
| "num_tokens": 187980618.0, |
| "step": 940 |
| }, |
| { |
| "epoch": 0.7978049810046433, |
| "grad_norm": 0.4310465455055237, |
| "learning_rate": 6.20415116207443e-06, |
| "loss": 0.0661, |
| "mean_token_accuracy": 0.9793638586997986, |
| "num_tokens": 188984212.0, |
| "step": 945 |
| }, |
| { |
| "epoch": 0.8020261713803293, |
| "grad_norm": 0.6339879035949707, |
| "learning_rate": 6.1454939760955165e-06, |
| "loss": 0.0634, |
| "mean_token_accuracy": 0.9799651265144348, |
| "num_tokens": 190001533.0, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.8062473617560152, |
| "grad_norm": 0.5607146620750427, |
| "learning_rate": 6.087898464957788e-06, |
| "loss": 0.0574, |
| "mean_token_accuracy": 0.9805450439453125, |
| "num_tokens": 191003443.0, |
| "step": 955 |
| }, |
| { |
| "epoch": 0.8104685521317011, |
| "grad_norm": 0.3527127802371979, |
| "learning_rate": 6.031375392856559e-06, |
| "loss": 0.05, |
| "mean_token_accuracy": 0.9833211183547974, |
| "num_tokens": 191999134.0, |
| "step": 960 |
| }, |
| { |
| "epoch": 0.8146897425073871, |
| "grad_norm": 0.34288936853408813, |
| "learning_rate": 5.9759353235558494e-06, |
| "loss": 0.0578, |
| "mean_token_accuracy": 0.9805012285709381, |
| "num_tokens": 193001376.0, |
| "step": 965 |
| }, |
| { |
| "epoch": 0.818910932883073, |
| "grad_norm": 0.3201969265937805, |
| "learning_rate": 5.92158861841408e-06, |
| "loss": 0.0551, |
| "mean_token_accuracy": 0.9816329479217529, |
| "num_tokens": 193990536.0, |
| "step": 970 |
| }, |
| { |
| "epoch": 0.823132123258759, |
| "grad_norm": 0.45922183990478516, |
| "learning_rate": 5.868345434447615e-06, |
| "loss": 0.0493, |
| "mean_token_accuracy": 0.9836022198200226, |
| "num_tokens": 194992672.0, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.8273533136344449, |
| "grad_norm": 0.38956931233406067, |
| "learning_rate": 5.816215722432479e-06, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9808073222637177, |
| "num_tokens": 195992361.0, |
| "step": 980 |
| }, |
| { |
| "epoch": 0.8315745040101309, |
| "grad_norm": 0.4778905212879181, |
| "learning_rate": 5.765209225044643e-06, |
| "loss": 0.0624, |
| "mean_token_accuracy": 0.9804998755455017, |
| "num_tokens": 196987807.0, |
| "step": 985 |
| }, |
| { |
| "epoch": 0.8357956943858168, |
| "grad_norm": 0.5508166551589966, |
| "learning_rate": 5.715335475039162e-06, |
| "loss": 0.0557, |
| "mean_token_accuracy": 0.9825825035572052, |
| "num_tokens": 197973788.0, |
| "step": 990 |
| }, |
| { |
| "epoch": 0.8400168847615027, |
| "grad_norm": 0.36449918150901794, |
| "learning_rate": 5.666603793468598e-06, |
| "loss": 0.0711, |
| "mean_token_accuracy": 0.9793222069740295, |
| "num_tokens": 198974891.0, |
| "step": 995 |
| }, |
| { |
| "epoch": 0.8442380751371887, |
| "grad_norm": 0.49898725748062134, |
| "learning_rate": 5.619023287940956e-06, |
| "loss": 0.0669, |
| "mean_token_accuracy": 0.9799689531326294, |
| "num_tokens": 199964089.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8484592655128746, |
| "grad_norm": 0.40632086992263794, |
| "learning_rate": 5.572602850917558e-06, |
| "loss": 0.0588, |
| "mean_token_accuracy": 0.9810280919075012, |
| "num_tokens": 200961905.0, |
| "step": 1005 |
| }, |
| { |
| "epoch": 0.8526804558885606, |
| "grad_norm": 0.48627394437789917, |
| "learning_rate": 5.527351158051086e-06, |
| "loss": 0.0553, |
| "mean_token_accuracy": 0.982756781578064, |
| "num_tokens": 201952631.0, |
| "step": 1010 |
| }, |
| { |
| "epoch": 0.8569016462642465, |
| "grad_norm": 0.3169231712818146, |
| "learning_rate": 5.48327666656418e-06, |
| "loss": 0.0443, |
| "mean_token_accuracy": 0.984506243467331, |
| "num_tokens": 202949197.0, |
| "step": 1015 |
| }, |
| { |
| "epoch": 0.8611228366399325, |
| "grad_norm": 0.5935258865356445, |
| "learning_rate": 5.440387613668826e-06, |
| "loss": 0.0572, |
| "mean_token_accuracy": 0.9830881118774414, |
| "num_tokens": 203953300.0, |
| "step": 1020 |
| }, |
| { |
| "epoch": 0.8653440270156184, |
| "grad_norm": 0.563925564289093, |
| "learning_rate": 5.3986920150268925e-06, |
| "loss": 0.0567, |
| "mean_token_accuracy": 0.982083934545517, |
| "num_tokens": 204946698.0, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.8695652173913043, |
| "grad_norm": 0.29804691672325134, |
| "learning_rate": 5.358197663252053e-06, |
| "loss": 0.0479, |
| "mean_token_accuracy": 0.9836555004119873, |
| "num_tokens": 205950230.0, |
| "step": 1030 |
| }, |
| { |
| "epoch": 0.8737864077669902, |
| "grad_norm": 0.43488752841949463, |
| "learning_rate": 5.318912126453407e-06, |
| "loss": 0.0677, |
| "mean_token_accuracy": 0.9799270749092102, |
| "num_tokens": 206955427.0, |
| "step": 1035 |
| }, |
| { |
| "epoch": 0.8780075981426763, |
| "grad_norm": 0.39133355021476746, |
| "learning_rate": 5.280842746821047e-06, |
| "loss": 0.068, |
| "mean_token_accuracy": 0.9783591270446778, |
| "num_tokens": 207941668.0, |
| "step": 1040 |
| }, |
| { |
| "epoch": 0.8822287885183622, |
| "grad_norm": 0.39996597170829773, |
| "learning_rate": 5.243996639253857e-06, |
| "loss": 0.0596, |
| "mean_token_accuracy": 0.9805743098258972, |
| "num_tokens": 208944975.0, |
| "step": 1045 |
| }, |
| { |
| "epoch": 0.8864499788940481, |
| "grad_norm": 0.3374752104282379, |
| "learning_rate": 5.208380690029793e-06, |
| "loss": 0.0555, |
| "mean_token_accuracy": 0.9818054139614105, |
| "num_tokens": 209928961.0, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.8906711692697341, |
| "grad_norm": 0.4433048665523529, |
| "learning_rate": 5.174001555518876e-06, |
| "loss": 0.0534, |
| "mean_token_accuracy": 0.9823460936546325, |
| "num_tokens": 210924119.0, |
| "step": 1055 |
| }, |
| { |
| "epoch": 0.89489235964542, |
| "grad_norm": 0.23922964930534363, |
| "learning_rate": 5.140865660939177e-06, |
| "loss": 0.0496, |
| "mean_token_accuracy": 0.983320826292038, |
| "num_tokens": 211917903.0, |
| "step": 1060 |
| }, |
| { |
| "epoch": 0.8991135500211059, |
| "grad_norm": 0.2743581235408783, |
| "learning_rate": 5.1089791991559745e-06, |
| "loss": 0.0511, |
| "mean_token_accuracy": 0.9829956293106079, |
| "num_tokens": 212901425.0, |
| "step": 1065 |
| }, |
| { |
| "epoch": 0.9033347403967918, |
| "grad_norm": 0.5250707864761353, |
| "learning_rate": 5.078348129524362e-06, |
| "loss": 0.0604, |
| "mean_token_accuracy": 0.9811276137828827, |
| "num_tokens": 213890839.0, |
| "step": 1070 |
| }, |
| { |
| "epoch": 0.9075559307724779, |
| "grad_norm": 0.3646811544895172, |
| "learning_rate": 5.048978176775484e-06, |
| "loss": 0.0492, |
| "mean_token_accuracy": 0.9845901250839233, |
| "num_tokens": 214887090.0, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.9117771211481638, |
| "grad_norm": 0.32301196455955505, |
| "learning_rate": 5.020874829946623e-06, |
| "loss": 0.0643, |
| "mean_token_accuracy": 0.9794431924819946, |
| "num_tokens": 215879299.0, |
| "step": 1080 |
| }, |
| { |
| "epoch": 0.9159983115238497, |
| "grad_norm": 0.3308781683444977, |
| "learning_rate": 4.994043341355334e-06, |
| "loss": 0.0624, |
| "mean_token_accuracy": 0.9805513024330139, |
| "num_tokens": 216876360.0, |
| "step": 1085 |
| }, |
| { |
| "epoch": 0.9202195018995357, |
| "grad_norm": 0.5843566656112671, |
| "learning_rate": 4.968488725617828e-06, |
| "loss": 0.0809, |
| "mean_token_accuracy": 0.9752067506313324, |
| "num_tokens": 217860939.0, |
| "step": 1090 |
| }, |
| { |
| "epoch": 0.9244406922752216, |
| "grad_norm": 0.6650079488754272, |
| "learning_rate": 4.944215758711777e-06, |
| "loss": 0.0638, |
| "mean_token_accuracy": 0.9792717456817627, |
| "num_tokens": 218850584.0, |
| "step": 1095 |
| }, |
| { |
| "epoch": 0.9286618826509075, |
| "grad_norm": 0.3985125422477722, |
| "learning_rate": 4.9212289770837105e-06, |
| "loss": 0.05, |
| "mean_token_accuracy": 0.983612322807312, |
| "num_tokens": 219848686.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9328830730265935, |
| "grad_norm": 0.4458722174167633, |
| "learning_rate": 4.899532676801194e-06, |
| "loss": 0.0575, |
| "mean_token_accuracy": 0.9825719773769379, |
| "num_tokens": 220857976.0, |
| "step": 1105 |
| }, |
| { |
| "epoch": 0.9371042634022795, |
| "grad_norm": 0.5457350015640259, |
| "learning_rate": 4.879130912749917e-06, |
| "loss": 0.0695, |
| "mean_token_accuracy": 0.9788504898548126, |
| "num_tokens": 221861159.0, |
| "step": 1110 |
| }, |
| { |
| "epoch": 0.9413254537779654, |
| "grad_norm": 0.3502093553543091, |
| "learning_rate": 4.8600274978758636e-06, |
| "loss": 0.0505, |
| "mean_token_accuracy": 0.9830187797546387, |
| "num_tokens": 222871892.0, |
| "step": 1115 |
| }, |
| { |
| "epoch": 0.9455466441536513, |
| "grad_norm": 0.32649070024490356, |
| "learning_rate": 4.84222600247271e-06, |
| "loss": 0.0542, |
| "mean_token_accuracy": 0.9818561136722564, |
| "num_tokens": 223874296.0, |
| "step": 1120 |
| }, |
| { |
| "epoch": 0.9497678345293372, |
| "grad_norm": 0.4820583462715149, |
| "learning_rate": 4.825729753514547e-06, |
| "loss": 0.0532, |
| "mean_token_accuracy": 0.9828004717826844, |
| "num_tokens": 224882069.0, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.9539890249050232, |
| "grad_norm": 0.5502006411552429, |
| "learning_rate": 4.810541834034105e-06, |
| "loss": 0.0583, |
| "mean_token_accuracy": 0.9812820851802826, |
| "num_tokens": 225881271.0, |
| "step": 1130 |
| }, |
| { |
| "epoch": 0.9582102152807092, |
| "grad_norm": 0.3897382616996765, |
| "learning_rate": 4.796665082546546e-06, |
| "loss": 0.0492, |
| "mean_token_accuracy": 0.9832486152648926, |
| "num_tokens": 226876358.0, |
| "step": 1135 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1184, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.4496849821958144e+19, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
