| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9998149861239592, |
| "eval_steps": 100, |
| "global_step": 1351, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0037002775208140612, |
| "grad_norm": 3.239915386527218, |
| "learning_rate": 7.352941176470589e-07, |
| "loss": 1.2082, |
| "mean_token_accuracy": 0.6904086382244544, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.0074005550416281225, |
| "grad_norm": 2.8480570558845186, |
| "learning_rate": 1.4705882352941177e-06, |
| "loss": 1.2101, |
| "mean_token_accuracy": 0.688909531616815, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.011100832562442183, |
| "grad_norm": 2.425023099484193, |
| "learning_rate": 2.2058823529411767e-06, |
| "loss": 1.1302, |
| "mean_token_accuracy": 0.7074111736775649, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.014801110083256245, |
| "grad_norm": 1.8152419374117768, |
| "learning_rate": 2.9411764705882355e-06, |
| "loss": 1.1277, |
| "mean_token_accuracy": 0.704069345586691, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.018501387604070305, |
| "grad_norm": 1.5172611375815148, |
| "learning_rate": 3.6764705882352946e-06, |
| "loss": 1.0942, |
| "mean_token_accuracy": 0.7089358281360809, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.022201665124884366, |
| "grad_norm": 1.2356119543050552, |
| "learning_rate": 4.411764705882353e-06, |
| "loss": 1.04, |
| "mean_token_accuracy": 0.7169246015229918, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.025901942645698426, |
| "grad_norm": 1.2317231184779383, |
| "learning_rate": 5.147058823529411e-06, |
| "loss": 1.0126, |
| "mean_token_accuracy": 0.720928252798097, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.02960222016651249, |
| "grad_norm": 1.094000980605897, |
| "learning_rate": 5.882352941176471e-06, |
| "loss": 0.9576, |
| "mean_token_accuracy": 0.7326068845537983, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.03330249768732655, |
| "grad_norm": 0.8898258637631962, |
| "learning_rate": 6.61764705882353e-06, |
| "loss": 0.9722, |
| "mean_token_accuracy": 0.7278379936229317, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.03700277520814061, |
| "grad_norm": 0.9622645010733354, |
| "learning_rate": 7.352941176470589e-06, |
| "loss": 0.9281, |
| "mean_token_accuracy": 0.7367932115453535, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.04070305272895467, |
| "grad_norm": 0.9736757281340813, |
| "learning_rate": 8.088235294117648e-06, |
| "loss": 0.9054, |
| "mean_token_accuracy": 0.7442706350426345, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.04440333024976873, |
| "grad_norm": 1.042826803009636, |
| "learning_rate": 8.823529411764707e-06, |
| "loss": 0.8948, |
| "mean_token_accuracy": 0.7443939037516841, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.04810360777058279, |
| "grad_norm": 0.9490284166948695, |
| "learning_rate": 9.558823529411766e-06, |
| "loss": 0.9171, |
| "mean_token_accuracy": 0.7375623924285362, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.05180388529139685, |
| "grad_norm": 0.9728295714818891, |
| "learning_rate": 1.0294117647058823e-05, |
| "loss": 0.8826, |
| "mean_token_accuracy": 0.7468788718777591, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.05550416281221091, |
| "grad_norm": 0.8791162221139119, |
| "learning_rate": 1.1029411764705885e-05, |
| "loss": 0.9399, |
| "mean_token_accuracy": 0.7313582120538165, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.05920444033302498, |
| "grad_norm": 0.8984533054240379, |
| "learning_rate": 1.1764705882352942e-05, |
| "loss": 0.8946, |
| "mean_token_accuracy": 0.74295032091544, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.06290471785383904, |
| "grad_norm": 1.011416137474189, |
| "learning_rate": 1.25e-05, |
| "loss": 0.9146, |
| "mean_token_accuracy": 0.7366250851820709, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.0666049953746531, |
| "grad_norm": 1.0421339267946321, |
| "learning_rate": 1.323529411764706e-05, |
| "loss": 0.8674, |
| "mean_token_accuracy": 0.7498001647381672, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.07030527289546716, |
| "grad_norm": 0.9085816073653793, |
| "learning_rate": 1.3970588235294118e-05, |
| "loss": 0.8848, |
| "mean_token_accuracy": 0.7459213455908199, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.07400555041628122, |
| "grad_norm": 1.108924753563469, |
| "learning_rate": 1.4705882352941179e-05, |
| "loss": 0.8984, |
| "mean_token_accuracy": 0.7407395864040437, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07400555041628122, |
| "eval_loss": 0.8967365026473999, |
| "eval_mean_token_accuracy": 0.7362222154420599, |
| "eval_runtime": 14.9927, |
| "eval_samples_per_second": 17.142, |
| "eval_steps_per_second": 1.134, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07770582793709528, |
| "grad_norm": 1.0749851910600146, |
| "learning_rate": 1.5441176470588237e-05, |
| "loss": 0.9179, |
| "mean_token_accuracy": 0.7357399681666867, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.08140610545790934, |
| "grad_norm": 0.9047444784181967, |
| "learning_rate": 1.6176470588235296e-05, |
| "loss": 0.8645, |
| "mean_token_accuracy": 0.7504314694269566, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.0851063829787234, |
| "grad_norm": 0.9376349243794684, |
| "learning_rate": 1.6911764705882355e-05, |
| "loss": 0.8529, |
| "mean_token_accuracy": 0.7516550848858883, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.08880666049953746, |
| "grad_norm": 1.0917351629769148, |
| "learning_rate": 1.7647058823529414e-05, |
| "loss": 0.8017, |
| "mean_token_accuracy": 0.7668538841405976, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.09250693802035152, |
| "grad_norm": 1.001563932314443, |
| "learning_rate": 1.8382352941176472e-05, |
| "loss": 0.839, |
| "mean_token_accuracy": 0.7564128590771355, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.09620721554116558, |
| "grad_norm": 1.0798631661399483, |
| "learning_rate": 1.911764705882353e-05, |
| "loss": 0.8884, |
| "mean_token_accuracy": 0.7426265352132153, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.09990749306197964, |
| "grad_norm": 0.9968532290106067, |
| "learning_rate": 1.985294117647059e-05, |
| "loss": 0.8807, |
| "mean_token_accuracy": 0.7442322917876578, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.1036077705827937, |
| "grad_norm": 1.109473932122476, |
| "learning_rate": 1.9999465148392906e-05, |
| "loss": 0.8846, |
| "mean_token_accuracy": 0.74246846644443, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.10730804810360776, |
| "grad_norm": 0.9648937799207011, |
| "learning_rate": 1.999729241179462e-05, |
| "loss": 0.8574, |
| "mean_token_accuracy": 0.7496809848232011, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.11100832562442182, |
| "grad_norm": 1.1187334021697934, |
| "learning_rate": 1.999344872485215e-05, |
| "loss": 0.887, |
| "mean_token_accuracy": 0.7429769166906245, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.1147086031452359, |
| "grad_norm": 1.0683788241365297, |
| "learning_rate": 1.9987934730000457e-05, |
| "loss": 0.8419, |
| "mean_token_accuracy": 0.7538007970944414, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.11840888066604996, |
| "grad_norm": 1.1320511934956428, |
| "learning_rate": 1.998075134885022e-05, |
| "loss": 0.867, |
| "mean_token_accuracy": 0.7468512909716426, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.12210915818686402, |
| "grad_norm": 0.9103029614671497, |
| "learning_rate": 1.9971899782033853e-05, |
| "loss": 0.8576, |
| "mean_token_accuracy": 0.7499938720204854, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.12580943570767808, |
| "grad_norm": 0.981129476759838, |
| "learning_rate": 1.9961381509004785e-05, |
| "loss": 0.8662, |
| "mean_token_accuracy": 0.7473178017703339, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.12950971322849214, |
| "grad_norm": 1.0506640779724106, |
| "learning_rate": 1.9949198287790215e-05, |
| "loss": 0.8448, |
| "mean_token_accuracy": 0.7532863939278541, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.1332099907493062, |
| "grad_norm": 1.026757070568888, |
| "learning_rate": 1.9935352154697257e-05, |
| "loss": 0.8289, |
| "mean_token_accuracy": 0.7571774470525083, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.13691026827012026, |
| "grad_norm": 1.024410882605988, |
| "learning_rate": 1.9919845423972603e-05, |
| "loss": 0.8708, |
| "mean_token_accuracy": 0.7462624656471927, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.14061054579093432, |
| "grad_norm": 0.9976772238865053, |
| "learning_rate": 1.9902680687415704e-05, |
| "loss": 0.8516, |
| "mean_token_accuracy": 0.7509371514896652, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.14431082331174838, |
| "grad_norm": 0.9523302539052055, |
| "learning_rate": 1.9883860813945596e-05, |
| "loss": 0.866, |
| "mean_token_accuracy": 0.7468851074446029, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.14801110083256244, |
| "grad_norm": 0.9850889047221103, |
| "learning_rate": 1.986338894912137e-05, |
| "loss": 0.8455, |
| "mean_token_accuracy": 0.7527866267647105, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.14801110083256244, |
| "eval_loss": 0.865836501121521, |
| "eval_mean_token_accuracy": 0.7421542283263433, |
| "eval_runtime": 14.4283, |
| "eval_samples_per_second": 17.812, |
| "eval_steps_per_second": 1.178, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1517113783533765, |
| "grad_norm": 1.0229873534868676, |
| "learning_rate": 1.9841268514616434e-05, |
| "loss": 0.8252, |
| "mean_token_accuracy": 0.7574686598573471, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.15541165587419056, |
| "grad_norm": 0.9622467160372926, |
| "learning_rate": 1.9817503207646606e-05, |
| "loss": 0.8554, |
| "mean_token_accuracy": 0.7478246659284545, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.15911193339500462, |
| "grad_norm": 0.9131276938848428, |
| "learning_rate": 1.979209700035216e-05, |
| "loss": 0.8244, |
| "mean_token_accuracy": 0.7584659661867365, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.16281221091581868, |
| "grad_norm": 0.9051238256817666, |
| "learning_rate": 1.976505413913393e-05, |
| "loss": 0.8335, |
| "mean_token_accuracy": 0.7552826906170925, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.16651248843663274, |
| "grad_norm": 0.8323380442965079, |
| "learning_rate": 1.9736379143943565e-05, |
| "loss": 0.8206, |
| "mean_token_accuracy": 0.7582460617181218, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.1702127659574468, |
| "grad_norm": 0.9191267666766435, |
| "learning_rate": 1.9706076807528044e-05, |
| "loss": 0.8497, |
| "mean_token_accuracy": 0.7505949453078798, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.17391304347826086, |
| "grad_norm": 0.8750491803678914, |
| "learning_rate": 1.967415219462864e-05, |
| "loss": 0.8676, |
| "mean_token_accuracy": 0.7436590583163561, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.17761332099907493, |
| "grad_norm": 0.9239017879909354, |
| "learning_rate": 1.9640610641134383e-05, |
| "loss": 0.8559, |
| "mean_token_accuracy": 0.7475432119302973, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.18131359851988899, |
| "grad_norm": 0.9375160912449586, |
| "learning_rate": 1.9605457753190224e-05, |
| "loss": 0.835, |
| "mean_token_accuracy": 0.7547475756381888, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.18501387604070305, |
| "grad_norm": 0.9492613957957171, |
| "learning_rate": 1.9568699406260016e-05, |
| "loss": 0.8051, |
| "mean_token_accuracy": 0.7635365375511847, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.1887141535615171, |
| "grad_norm": 0.8639060336030094, |
| "learning_rate": 1.953034174414449e-05, |
| "loss": 0.8424, |
| "mean_token_accuracy": 0.7528160265907698, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.19241443108233117, |
| "grad_norm": 0.8984792755320175, |
| "learning_rate": 1.9490391177954383e-05, |
| "loss": 0.8386, |
| "mean_token_accuracy": 0.7550477950939954, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.19611470860314523, |
| "grad_norm": 1.122121252111232, |
| "learning_rate": 1.944885438503888e-05, |
| "loss": 0.8063, |
| "mean_token_accuracy": 0.7605705069134607, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.1998149861239593, |
| "grad_norm": 0.9283272086566748, |
| "learning_rate": 1.9405738307869565e-05, |
| "loss": 0.8276, |
| "mean_token_accuracy": 0.7565203062763346, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.20351526364477335, |
| "grad_norm": 0.8413884737177622, |
| "learning_rate": 1.936105015288003e-05, |
| "loss": 0.8282, |
| "mean_token_accuracy": 0.7568203022016651, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.2072155411655874, |
| "grad_norm": 0.9409390882210004, |
| "learning_rate": 1.9314797389261426e-05, |
| "loss": 0.8123, |
| "mean_token_accuracy": 0.7606654417233719, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.21091581868640147, |
| "grad_norm": 0.9393490463149954, |
| "learning_rate": 1.9266987747714036e-05, |
| "loss": 0.8573, |
| "mean_token_accuracy": 0.747553958645714, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.21461609620721553, |
| "grad_norm": 0.8638492423841437, |
| "learning_rate": 1.9217629219155172e-05, |
| "loss": 0.8123, |
| "mean_token_accuracy": 0.7590153618433214, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.2183163737280296, |
| "grad_norm": 0.846309176173066, |
| "learning_rate": 1.916673005338357e-05, |
| "loss": 0.8056, |
| "mean_token_accuracy": 0.7618184813128182, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.22201665124884365, |
| "grad_norm": 0.872871552134233, |
| "learning_rate": 1.9114298757700508e-05, |
| "loss": 0.8481, |
| "mean_token_accuracy": 0.7489313630762335, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.22201665124884365, |
| "eval_loss": 0.8483777642250061, |
| "eval_mean_token_accuracy": 0.7457049964976622, |
| "eval_runtime": 14.427, |
| "eval_samples_per_second": 17.814, |
| "eval_steps_per_second": 1.178, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.22571692876965774, |
| "grad_norm": 0.999826315813207, |
| "learning_rate": 1.9060344095487916e-05, |
| "loss": 0.8166, |
| "mean_token_accuracy": 0.758320414906213, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.2294172062904718, |
| "grad_norm": 0.8969904896726869, |
| "learning_rate": 1.9004875084743624e-05, |
| "loss": 0.8315, |
| "mean_token_accuracy": 0.7541319645798021, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.23311748381128586, |
| "grad_norm": 0.8663093018774404, |
| "learning_rate": 1.8947900996574133e-05, |
| "loss": 0.8213, |
| "mean_token_accuracy": 0.7579362435854416, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.23681776133209992, |
| "grad_norm": 0.9755265816894861, |
| "learning_rate": 1.8889431353645004e-05, |
| "loss": 0.8488, |
| "mean_token_accuracy": 0.7504882625355576, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.24051803885291398, |
| "grad_norm": 0.9335300912517395, |
| "learning_rate": 1.8829475928589272e-05, |
| "loss": 0.827, |
| "mean_token_accuracy": 0.7555998423977681, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.24421831637372804, |
| "grad_norm": 0.8900082915560992, |
| "learning_rate": 1.8768044742374008e-05, |
| "loss": 0.8152, |
| "mean_token_accuracy": 0.7585311452224557, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.2479185938945421, |
| "grad_norm": 0.9294234165727436, |
| "learning_rate": 1.870514806262544e-05, |
| "loss": 0.8334, |
| "mean_token_accuracy": 0.7540371079296321, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.25161887141535616, |
| "grad_norm": 0.8974570795671991, |
| "learning_rate": 1.8640796401912805e-05, |
| "loss": 0.8387, |
| "mean_token_accuracy": 0.7533950740061741, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.2553191489361702, |
| "grad_norm": 0.9304337164615034, |
| "learning_rate": 1.8575000515991283e-05, |
| "loss": 0.8373, |
| "mean_token_accuracy": 0.7522675141398661, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.2590194264569843, |
| "grad_norm": 0.9033200669469308, |
| "learning_rate": 1.850777140200427e-05, |
| "loss": 0.8244, |
| "mean_token_accuracy": 0.7568203171895187, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.2627197039777983, |
| "grad_norm": 0.8758137392511858, |
| "learning_rate": 1.843912029664531e-05, |
| "loss": 0.7854, |
| "mean_token_accuracy": 0.7673373370996341, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.2664199814986124, |
| "grad_norm": 0.8264158075854122, |
| "learning_rate": 1.8369058674280004e-05, |
| "loss": 0.8315, |
| "mean_token_accuracy": 0.7537363388382118, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.27012025901942643, |
| "grad_norm": 0.9670401033713641, |
| "learning_rate": 1.8297598245028173e-05, |
| "loss": 0.8038, |
| "mean_token_accuracy": 0.7621257946600963, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.2738205365402405, |
| "grad_norm": 0.80837274332651, |
| "learning_rate": 1.8224750952806626e-05, |
| "loss": 0.8099, |
| "mean_token_accuracy": 0.7602013870948292, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.27752081406105455, |
| "grad_norm": 0.9804858790315341, |
| "learning_rate": 1.815052897333284e-05, |
| "loss": 0.8053, |
| "mean_token_accuracy": 0.7617221638816465, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.28122109158186864, |
| "grad_norm": 0.9022919090042508, |
| "learning_rate": 1.8074944712089925e-05, |
| "loss": 0.8125, |
| "mean_token_accuracy": 0.7580943305793408, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.2849213691026827, |
| "grad_norm": 0.8906726263514316, |
| "learning_rate": 1.799801080225316e-05, |
| "loss": 0.7982, |
| "mean_token_accuracy": 0.7643851506636017, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.28862164662349676, |
| "grad_norm": 0.9254743961939594, |
| "learning_rate": 1.7919740102578482e-05, |
| "loss": 0.8303, |
| "mean_token_accuracy": 0.7550160319572066, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.2923219241443108, |
| "grad_norm": 0.8866451111188923, |
| "learning_rate": 1.7840145695253258e-05, |
| "loss": 0.8289, |
| "mean_token_accuracy": 0.7549396953188148, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.2960222016651249, |
| "grad_norm": 0.9181449537488579, |
| "learning_rate": 1.7759240883709745e-05, |
| "loss": 0.8081, |
| "mean_token_accuracy": 0.7601195554230622, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2960222016651249, |
| "eval_loss": 0.8367598056793213, |
| "eval_mean_token_accuracy": 0.7479890842555498, |
| "eval_runtime": 14.4256, |
| "eval_samples_per_second": 17.816, |
| "eval_steps_per_second": 1.178, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.299722479185939, |
| "grad_norm": 1.020831268230977, |
| "learning_rate": 1.7677039190401538e-05, |
| "loss": 0.8015, |
| "mean_token_accuracy": 0.762128841557159, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.303422756706753, |
| "grad_norm": 0.9003231285941122, |
| "learning_rate": 1.759355435454342e-05, |
| "loss": 0.8058, |
| "mean_token_accuracy": 0.7601393580378268, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.3071230342275671, |
| "grad_norm": 0.8236287029577125, |
| "learning_rate": 1.7508800329814993e-05, |
| "loss": 0.8184, |
| "mean_token_accuracy": 0.7579837604334689, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.3108233117483811, |
| "grad_norm": 0.878406946421321, |
| "learning_rate": 1.7422791282028457e-05, |
| "loss": 0.8244, |
| "mean_token_accuracy": 0.756998624229836, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.3145235892691952, |
| "grad_norm": 0.8740565941244159, |
| "learning_rate": 1.7335541586760928e-05, |
| "loss": 0.8093, |
| "mean_token_accuracy": 0.758314911393715, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.31822386679000925, |
| "grad_norm": 0.8246101567782078, |
| "learning_rate": 1.7247065826951694e-05, |
| "loss": 0.841, |
| "mean_token_accuracy": 0.7503554458795133, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.32192414431082333, |
| "grad_norm": 0.9006537182294639, |
| "learning_rate": 1.715737879046483e-05, |
| "loss": 0.8274, |
| "mean_token_accuracy": 0.7544573421624281, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.32562442183163737, |
| "grad_norm": 0.8655847169317894, |
| "learning_rate": 1.7066495467617552e-05, |
| "loss": 0.7768, |
| "mean_token_accuracy": 0.768584957125417, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.32932469935245146, |
| "grad_norm": 0.8873250103741515, |
| "learning_rate": 1.6974431048674714e-05, |
| "loss": 0.8102, |
| "mean_token_accuracy": 0.7594749343726344, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.3330249768732655, |
| "grad_norm": 0.9413246521565958, |
| "learning_rate": 1.6881200921309914e-05, |
| "loss": 0.8249, |
| "mean_token_accuracy": 0.7560174557216164, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.3367252543940796, |
| "grad_norm": 0.824311990821373, |
| "learning_rate": 1.6786820668033596e-05, |
| "loss": 0.783, |
| "mean_token_accuracy": 0.7668897253243784, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.3404255319148936, |
| "grad_norm": 0.8487771854424017, |
| "learning_rate": 1.6691306063588583e-05, |
| "loss": 0.8029, |
| "mean_token_accuracy": 0.7621946857360491, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.3441258094357077, |
| "grad_norm": 0.8528899682952513, |
| "learning_rate": 1.6594673072313478e-05, |
| "loss": 0.8288, |
| "mean_token_accuracy": 0.7542124144532096, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.34782608695652173, |
| "grad_norm": 0.8630488475790786, |
| "learning_rate": 1.6496937845474375e-05, |
| "loss": 0.8226, |
| "mean_token_accuracy": 0.7555700287541637, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.3515263644773358, |
| "grad_norm": 0.8605658074115097, |
| "learning_rate": 1.639811671856535e-05, |
| "loss": 0.8236, |
| "mean_token_accuracy": 0.7557499752346134, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.35522664199814985, |
| "grad_norm": 0.7797518506572444, |
| "learning_rate": 1.6298226208578127e-05, |
| "loss": 0.7813, |
| "mean_token_accuracy": 0.7673481576245396, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.35892691951896394, |
| "grad_norm": 0.8968525161832575, |
| "learning_rate": 1.6197283011241423e-05, |
| "loss": 0.8144, |
| "mean_token_accuracy": 0.7588428362993352, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.36262719703977797, |
| "grad_norm": 0.8655098146119614, |
| "learning_rate": 1.6095303998230432e-05, |
| "loss": 0.7944, |
| "mean_token_accuracy": 0.7637976691011525, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.36632747456059206, |
| "grad_norm": 0.8535889063927236, |
| "learning_rate": 1.599230621434687e-05, |
| "loss": 0.812, |
| "mean_token_accuracy": 0.7584837868241149, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.3700277520814061, |
| "grad_norm": 0.869018976065693, |
| "learning_rate": 1.5888306874670112e-05, |
| "loss": 0.7997, |
| "mean_token_accuracy": 0.7605666985241383, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.3700277520814061, |
| "eval_loss": 0.8272697925567627, |
| "eval_mean_token_accuracy": 0.7506664118007147, |
| "eval_runtime": 14.4281, |
| "eval_samples_per_second": 17.813, |
| "eval_steps_per_second": 1.178, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.3737280296022202, |
| "grad_norm": 0.8518524507404158, |
| "learning_rate": 1.5783323361679865e-05, |
| "loss": 0.8168, |
| "mean_token_accuracy": 0.757339361928324, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.3774283071230342, |
| "grad_norm": 0.826955116427965, |
| "learning_rate": 1.567737322235084e-05, |
| "loss": 0.7977, |
| "mean_token_accuracy": 0.7626890473452755, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.3811285846438483, |
| "grad_norm": 0.81749324197106, |
| "learning_rate": 1.557047416521996e-05, |
| "loss": 0.8004, |
| "mean_token_accuracy": 0.7616827532540068, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.38482886216466233, |
| "grad_norm": 0.8922348675427028, |
| "learning_rate": 1.546264405742654e-05, |
| "loss": 0.8158, |
| "mean_token_accuracy": 0.7574806745169147, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.3885291396854764, |
| "grad_norm": 0.8314989282626728, |
| "learning_rate": 1.535390092172597e-05, |
| "loss": 0.8343, |
| "mean_token_accuracy": 0.7518461075825386, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.39222941720629045, |
| "grad_norm": 0.9803006136249208, |
| "learning_rate": 1.5244262933477401e-05, |
| "loss": 0.8125, |
| "mean_token_accuracy": 0.7575044146062361, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.39592969472710454, |
| "grad_norm": 0.8240339887315228, |
| "learning_rate": 1.5133748417605878e-05, |
| "loss": 0.8046, |
| "mean_token_accuracy": 0.7602933891525141, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.3996299722479186, |
| "grad_norm": 1.0034971319001893, |
| "learning_rate": 1.5022375845539537e-05, |
| "loss": 0.8428, |
| "mean_token_accuracy": 0.7486714904011393, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.40333024976873266, |
| "grad_norm": 0.8668996786648001, |
| "learning_rate": 1.4910163832122278e-05, |
| "loss": 0.8212, |
| "mean_token_accuracy": 0.7548812709692821, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.4070305272895467, |
| "grad_norm": 0.9141308972741845, |
| "learning_rate": 1.4797131132502464e-05, |
| "loss": 0.7719, |
| "mean_token_accuracy": 0.770355880311961, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.4107308048103608, |
| "grad_norm": 0.8537164385394441, |
| "learning_rate": 1.4683296638998192e-05, |
| "loss": 0.7992, |
| "mean_token_accuracy": 0.7629726221010971, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.4144310823311748, |
| "grad_norm": 0.8326818981936996, |
| "learning_rate": 1.4568679377939619e-05, |
| "loss": 0.7802, |
| "mean_token_accuracy": 0.7676385973158337, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.4181313598519889, |
| "grad_norm": 0.832442552040938, |
| "learning_rate": 1.4453298506488896e-05, |
| "loss": 0.7913, |
| "mean_token_accuracy": 0.7640136926495742, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.42183163737280294, |
| "grad_norm": 0.8484820193679721, |
| "learning_rate": 1.4337173309438236e-05, |
| "loss": 0.7775, |
| "mean_token_accuracy": 0.767823896982353, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.425531914893617, |
| "grad_norm": 0.8808404728533856, |
| "learning_rate": 1.4220323195986649e-05, |
| "loss": 0.8008, |
| "mean_token_accuracy": 0.7603237102804046, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.42923219241443106, |
| "grad_norm": 0.8447018318022254, |
| "learning_rate": 1.4102767696495885e-05, |
| "loss": 0.8307, |
| "mean_token_accuracy": 0.7518857396509806, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.43293246993524515, |
| "grad_norm": 0.8085246144584103, |
| "learning_rate": 1.398452645922611e-05, |
| "loss": 0.7829, |
| "mean_token_accuracy": 0.7674445881359748, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.4366327474560592, |
| "grad_norm": 0.8736484311681693, |
| "learning_rate": 1.3865619247051916e-05, |
| "loss": 0.7944, |
| "mean_token_accuracy": 0.7624168954027166, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.44033302497687327, |
| "grad_norm": 0.7989835433698877, |
| "learning_rate": 1.3746065934159123e-05, |
| "loss": 0.7934, |
| "mean_token_accuracy": 0.7634149574109332, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.4440333024976873, |
| "grad_norm": 0.8032554489053177, |
| "learning_rate": 1.3625886502723008e-05, |
| "loss": 0.7576, |
| "mean_token_accuracy": 0.7739669137787928, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.4440333024976873, |
| "eval_loss": 0.8197986483573914, |
| "eval_mean_token_accuracy": 0.7513783085827505, |
| "eval_runtime": 14.4284, |
| "eval_samples_per_second": 17.812, |
| "eval_steps_per_second": 1.178, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.4477335800185014, |
| "grad_norm": 0.9180541639485074, |
| "learning_rate": 1.3505101039568494e-05, |
| "loss": 0.8295, |
| "mean_token_accuracy": 0.752840844934445, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.4514338575393155, |
| "grad_norm": 0.7752356138164175, |
| "learning_rate": 1.3383729732812814e-05, |
| "loss": 0.7879, |
| "mean_token_accuracy": 0.764613230505008, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.4551341350601295, |
| "grad_norm": 0.83145675114526, |
| "learning_rate": 1.3261792868491267e-05, |
| "loss": 0.7914, |
| "mean_token_accuracy": 0.7635481399350937, |
| "step": 615 |
| }, |
| { |
| "epoch": 0.4588344125809436, |
| "grad_norm": 0.9055291804893332, |
| "learning_rate": 1.3139310827166613e-05, |
| "loss": 0.7701, |
| "mean_token_accuracy": 0.7685765923501461, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.46253469010175763, |
| "grad_norm": 0.7854598351220646, |
| "learning_rate": 1.3016304080522657e-05, |
| "loss": 0.7704, |
| "mean_token_accuracy": 0.7688856993881648, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.4662349676225717, |
| "grad_norm": 0.8441815485432693, |
| "learning_rate": 1.2892793187942588e-05, |
| "loss": 0.7626, |
| "mean_token_accuracy": 0.7723868739695786, |
| "step": 630 |
| }, |
| { |
| "epoch": 0.46993524514338575, |
| "grad_norm": 0.8667439139705619, |
| "learning_rate": 1.2768798793072708e-05, |
| "loss": 0.7833, |
| "mean_token_accuracy": 0.7663687223264415, |
| "step": 635 |
| }, |
| { |
| "epoch": 0.47363552266419984, |
| "grad_norm": 0.8321053356387224, |
| "learning_rate": 1.2644341620372025e-05, |
| "loss": 0.797, |
| "mean_token_accuracy": 0.7617867821279096, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.47733580018501387, |
| "grad_norm": 0.7876516967194911, |
| "learning_rate": 1.2519442471648364e-05, |
| "loss": 0.7935, |
| "mean_token_accuracy": 0.7634835072725441, |
| "step": 645 |
| }, |
| { |
| "epoch": 0.48103607770582796, |
| "grad_norm": 0.7906296446679114, |
| "learning_rate": 1.2394122222581557e-05, |
| "loss": 0.7903, |
| "mean_token_accuracy": 0.7648497977323175, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.484736355226642, |
| "grad_norm": 0.8411500456259221, |
| "learning_rate": 1.226840181923427e-05, |
| "loss": 0.7672, |
| "mean_token_accuracy": 0.7703495787160083, |
| "step": 655 |
| }, |
| { |
| "epoch": 0.4884366327474561, |
| "grad_norm": 0.8625525401373217, |
| "learning_rate": 1.214230227455106e-05, |
| "loss": 0.7905, |
| "mean_token_accuracy": 0.7643257328337282, |
| "step": 660 |
| }, |
| { |
| "epoch": 0.4921369102682701, |
| "grad_norm": 0.8769883186756258, |
| "learning_rate": 1.201584466484629e-05, |
| "loss": 0.8025, |
| "mean_token_accuracy": 0.7592148887938102, |
| "step": 665 |
| }, |
| { |
| "epoch": 0.4958371877890842, |
| "grad_norm": 0.7838112843020255, |
| "learning_rate": 1.1889050126281405e-05, |
| "loss": 0.7909, |
| "mean_token_accuracy": 0.7630134287220484, |
| "step": 670 |
| }, |
| { |
| "epoch": 0.49953746530989823, |
| "grad_norm": 0.874298908965393, |
| "learning_rate": 1.1761939851332241e-05, |
| "loss": 0.7691, |
| "mean_token_accuracy": 0.7678435427558977, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.5032377428307123, |
| "grad_norm": 0.806343521404904, |
| "learning_rate": 1.1634535085246903e-05, |
| "loss": 0.7892, |
| "mean_token_accuracy": 0.7629181254463294, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.5069380203515264, |
| "grad_norm": 0.8419778500876326, |
| "learning_rate": 1.1506857122494832e-05, |
| "loss": 0.8051, |
| "mean_token_accuracy": 0.7606549427357923, |
| "step": 685 |
| }, |
| { |
| "epoch": 0.5106382978723404, |
| "grad_norm": 0.7877466333281109, |
| "learning_rate": 1.1378927303207637e-05, |
| "loss": 0.7768, |
| "mean_token_accuracy": 0.7675751056593663, |
| "step": 690 |
| }, |
| { |
| "epoch": 0.5143385753931545, |
| "grad_norm": 0.9351990005133819, |
| "learning_rate": 1.12507670096123e-05, |
| "loss": 0.786, |
| "mean_token_accuracy": 0.762621351731412, |
| "step": 695 |
| }, |
| { |
| "epoch": 0.5180388529139686, |
| "grad_norm": 0.854162080782847, |
| "learning_rate": 1.1122397662457352e-05, |
| "loss": 0.7829, |
| "mean_token_accuracy": 0.7657675443087595, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.5180388529139686, |
| "eval_loss": 0.8126487135887146, |
| "eval_mean_token_accuracy": 0.7547239760153944, |
| "eval_runtime": 14.4295, |
| "eval_samples_per_second": 17.811, |
| "eval_steps_per_second": 1.178, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.5217391304347826, |
| "grad_norm": 0.8645040092134876, |
| "learning_rate": 1.0993840717432582e-05, |
| "loss": 0.7931, |
| "mean_token_accuracy": 0.7640120292612093, |
| "step": 705 |
| }, |
| { |
| "epoch": 0.5254394079555966, |
| "grad_norm": 0.8306086563251239, |
| "learning_rate": 1.0865117661582958e-05, |
| "loss": 0.7908, |
| "mean_token_accuracy": 0.7626317172664291, |
| "step": 710 |
| }, |
| { |
| "epoch": 0.5291396854764108, |
| "grad_norm": 0.8020466355732573, |
| "learning_rate": 1.0736250009717249e-05, |
| "loss": 0.7725, |
| "mean_token_accuracy": 0.7694951870682798, |
| "step": 715 |
| }, |
| { |
| "epoch": 0.5328399629972248, |
| "grad_norm": 0.8902488653913394, |
| "learning_rate": 1.0607259300812047e-05, |
| "loss": 0.7956, |
| "mean_token_accuracy": 0.7613192935638047, |
| "step": 720 |
| }, |
| { |
| "epoch": 0.5365402405180388, |
| "grad_norm": 0.8214216473694839, |
| "learning_rate": 1.0478167094411733e-05, |
| "loss": 0.7813, |
| "mean_token_accuracy": 0.7666191635211226, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.5402405180388529, |
| "grad_norm": 0.7826795279308211, |
| "learning_rate": 1.0348994967025012e-05, |
| "loss": 0.7904, |
| "mean_token_accuracy": 0.7615527950542014, |
| "step": 730 |
| }, |
| { |
| "epoch": 0.543940795559667, |
| "grad_norm": 0.7400827688733695, |
| "learning_rate": 1.0219764508518595e-05, |
| "loss": 0.813, |
| "mean_token_accuracy": 0.7559395006050915, |
| "step": 735 |
| }, |
| { |
| "epoch": 0.547641073080481, |
| "grad_norm": 0.8998834617524183, |
| "learning_rate": 1.0090497318508687e-05, |
| "loss": 0.8011, |
| "mean_token_accuracy": 0.7619206875092797, |
| "step": 740 |
| }, |
| { |
| "epoch": 0.5513413506012951, |
| "grad_norm": 0.7948918646757527, |
| "learning_rate": 9.961215002750799e-06, |
| "loss": 0.7761, |
| "mean_token_accuracy": 0.7669139188304434, |
| "step": 745 |
| }, |
| { |
| "epoch": 0.5550416281221091, |
| "grad_norm": 0.8243901741332256, |
| "learning_rate": 9.831939169528565e-06, |
| "loss": 0.7877, |
| "mean_token_accuracy": 0.7651489953053691, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.5587419056429233, |
| "grad_norm": 0.8525260298716054, |
| "learning_rate": 9.702691426042124e-06, |
| "loss": 0.8082, |
| "mean_token_accuracy": 0.7574811962767495, |
| "step": 755 |
| }, |
| { |
| "epoch": 0.5624421831637373, |
| "grad_norm": 0.8537478736826423, |
| "learning_rate": 9.573493374796694e-06, |
| "loss": 0.7921, |
| "mean_token_accuracy": 0.7631372878452614, |
| "step": 760 |
| }, |
| { |
| "epoch": 0.5661424606845513, |
| "grad_norm": 0.8430252803933753, |
| "learning_rate": 9.444366609991916e-06, |
| "loss": 0.7816, |
| "mean_token_accuracy": 0.7655811713537104, |
| "step": 765 |
| }, |
| { |
| "epoch": 0.5698427382053654, |
| "grad_norm": 0.7819283016691769, |
| "learning_rate": 9.315332713912593e-06, |
| "loss": 0.7766, |
| "mean_token_accuracy": 0.7672953120755375, |
| "step": 770 |
| }, |
| { |
| "epoch": 0.5735430157261795, |
| "grad_norm": 0.7854217248693685, |
| "learning_rate": 9.18641325332142e-06, |
| "loss": 0.7791, |
| "mean_token_accuracy": 0.7654781090301902, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.5772432932469935, |
| "grad_norm": 0.7797472528150745, |
| "learning_rate": 9.057629775854314e-06, |
| "loss": 0.786, |
| "mean_token_accuracy": 0.7640452228560917, |
| "step": 780 |
| }, |
| { |
| "epoch": 0.5809435707678076, |
| "grad_norm": 0.7467612879682155, |
| "learning_rate": 8.929003806418934e-06, |
| "loss": 0.7896, |
| "mean_token_accuracy": 0.7636666656288214, |
| "step": 785 |
| }, |
| { |
| "epoch": 0.5846438482886216, |
| "grad_norm": 0.8108378163356028, |
| "learning_rate": 8.800556843597002e-06, |
| "loss": 0.7947, |
| "mean_token_accuracy": 0.7620531323988468, |
| "step": 790 |
| }, |
| { |
| "epoch": 0.5883441258094357, |
| "grad_norm": 0.8711772849690956, |
| "learning_rate": 8.672310356051023e-06, |
| "loss": 0.7978, |
| "mean_token_accuracy": 0.7624323429145154, |
| "step": 795 |
| }, |
| { |
| "epoch": 0.5920444033302498, |
| "grad_norm": 0.8137827849359022, |
| "learning_rate": 8.544285778936004e-06, |
| "loss": 0.8179, |
| "mean_token_accuracy": 0.7553667683649044, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5920444033302498, |
| "eval_loss": 0.807556688785553, |
| "eval_mean_token_accuracy": 0.7549181587193091, |
| "eval_runtime": 14.4357, |
| "eval_samples_per_second": 17.803, |
| "eval_steps_per_second": 1.178, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5957446808510638, |
| "grad_norm": 0.9030717079561702, |
| "learning_rate": 8.416504510316774e-06, |
| "loss": 0.7273, |
| "mean_token_accuracy": 0.7819614746996253, |
| "step": 805 |
| }, |
| { |
| "epoch": 0.599444958371878, |
| "grad_norm": 0.8188514357400608, |
| "learning_rate": 8.28898790759152e-06, |
| "loss": 0.7799, |
| "mean_token_accuracy": 0.7673006684255306, |
| "step": 810 |
| }, |
| { |
| "epoch": 0.603145235892692, |
| "grad_norm": 0.8057726363169286, |
| "learning_rate": 8.161757283922084e-06, |
| "loss": 0.7727, |
| "mean_token_accuracy": 0.7686252774437621, |
| "step": 815 |
| }, |
| { |
| "epoch": 0.606845513413506, |
| "grad_norm": 0.8421037028882805, |
| "learning_rate": 8.034833904671698e-06, |
| "loss": 0.786, |
| "mean_token_accuracy": 0.7641876212394639, |
| "step": 820 |
| }, |
| { |
| "epoch": 0.61054579093432, |
| "grad_norm": 0.8334908295515484, |
| "learning_rate": 7.908238983850666e-06, |
| "loss": 0.7932, |
| "mean_token_accuracy": 0.7616740275010933, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.6142460684551342, |
| "grad_norm": 0.7732913249201244, |
| "learning_rate": 7.781993680570656e-06, |
| "loss": 0.8014, |
| "mean_token_accuracy": 0.7606977785932187, |
| "step": 830 |
| }, |
| { |
| "epoch": 0.6179463459759482, |
| "grad_norm": 0.8529863935049032, |
| "learning_rate": 7.656119095508155e-06, |
| "loss": 0.761, |
| "mean_token_accuracy": 0.7720907621215842, |
| "step": 835 |
| }, |
| { |
| "epoch": 0.6216466234967623, |
| "grad_norm": 0.7790046842614121, |
| "learning_rate": 7.530636267377706e-06, |
| "loss": 0.7634, |
| "mean_token_accuracy": 0.7701395597098513, |
| "step": 840 |
| }, |
| { |
| "epoch": 0.6253469010175763, |
| "grad_norm": 0.8225616793183522, |
| "learning_rate": 7.405566169415481e-06, |
| "loss": 0.7736, |
| "mean_token_accuracy": 0.7693085459512584, |
| "step": 845 |
| }, |
| { |
| "epoch": 0.6290471785383904, |
| "grad_norm": 0.7995615588703834, |
| "learning_rate": 7.280929705873818e-06, |
| "loss": 0.8007, |
| "mean_token_accuracy": 0.7614276357940275, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.6327474560592045, |
| "grad_norm": 0.7568937242521389, |
| "learning_rate": 7.15674770852727e-06, |
| "loss": 0.7358, |
| "mean_token_accuracy": 0.7783971531791127, |
| "step": 855 |
| }, |
| { |
| "epoch": 0.6364477335800185, |
| "grad_norm": 0.7756432966734992, |
| "learning_rate": 7.033040933190776e-06, |
| "loss": 0.803, |
| "mean_token_accuracy": 0.7595857140786891, |
| "step": 860 |
| }, |
| { |
| "epoch": 0.6401480111008325, |
| "grad_norm": 0.8218979611135344, |
| "learning_rate": 6.909830056250527e-06, |
| "loss": 0.7839, |
| "mean_token_accuracy": 0.7639222007393546, |
| "step": 865 |
| }, |
| { |
| "epoch": 0.6438482886216467, |
| "grad_norm": 0.7688613456202562, |
| "learning_rate": 6.787135671208126e-06, |
| "loss": 0.7572, |
| "mean_token_accuracy": 0.7727190284700474, |
| "step": 870 |
| }, |
| { |
| "epoch": 0.6475485661424607, |
| "grad_norm": 0.7591858458864597, |
| "learning_rate": 6.6649782852385554e-06, |
| "loss": 0.7714, |
| "mean_token_accuracy": 0.7696598108468179, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.6512488436632747, |
| "grad_norm": 0.7777825640170363, |
| "learning_rate": 6.543378315762634e-06, |
| "loss": 0.7571, |
| "mean_token_accuracy": 0.7740491454057086, |
| "step": 880 |
| }, |
| { |
| "epoch": 0.6549491211840888, |
| "grad_norm": 0.7917932009302419, |
| "learning_rate": 6.42235608703441e-06, |
| "loss": 0.7971, |
| "mean_token_accuracy": 0.7617020404301962, |
| "step": 885 |
| }, |
| { |
| "epoch": 0.6586493987049029, |
| "grad_norm": 0.7691064861967819, |
| "learning_rate": 6.301931826744189e-06, |
| "loss": 0.7739, |
| "mean_token_accuracy": 0.7681618066719192, |
| "step": 890 |
| }, |
| { |
| "epoch": 0.6623496762257169, |
| "grad_norm": 0.7996344447354707, |
| "learning_rate": 6.18212566263765e-06, |
| "loss": 0.7856, |
| "mean_token_accuracy": 0.7637340466330773, |
| "step": 895 |
| }, |
| { |
| "epoch": 0.666049953746531, |
| "grad_norm": 0.8226708174097512, |
| "learning_rate": 6.0629576191517035e-06, |
| "loss": 0.7633, |
| "mean_token_accuracy": 0.7714390934857177, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.666049953746531, |
| "eval_loss": 0.8014763593673706, |
| "eval_mean_token_accuracy": 0.7567653930021836, |
| "eval_runtime": 14.4293, |
| "eval_samples_per_second": 17.811, |
| "eval_steps_per_second": 1.178, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.669750231267345, |
| "grad_norm": 0.8005238604002097, |
| "learning_rate": 5.944447614067588e-06, |
| "loss": 0.7575, |
| "mean_token_accuracy": 0.771954640275627, |
| "step": 905 |
| }, |
| { |
| "epoch": 0.6734505087881592, |
| "grad_norm": 0.8084142017934829, |
| "learning_rate": 5.8266154551818225e-06, |
| "loss": 0.7384, |
| "mean_token_accuracy": 0.777882152510502, |
| "step": 910 |
| }, |
| { |
| "epoch": 0.6771507863089732, |
| "grad_norm": 0.7778625881879635, |
| "learning_rate": 5.709480836995509e-06, |
| "loss": 0.7701, |
| "mean_token_accuracy": 0.7698965224005008, |
| "step": 915 |
| }, |
| { |
| "epoch": 0.6808510638297872, |
| "grad_norm": 0.7577955469774116, |
| "learning_rate": 5.593063337422595e-06, |
| "loss": 0.7805, |
| "mean_token_accuracy": 0.765536335028292, |
| "step": 920 |
| }, |
| { |
| "epoch": 0.6845513413506013, |
| "grad_norm": 0.9227266710988027, |
| "learning_rate": 5.477382414517625e-06, |
| "loss": 0.798, |
| "mean_token_accuracy": 0.7598567457034476, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.6882516188714154, |
| "grad_norm": 0.8508290504613905, |
| "learning_rate": 5.362457403223495e-06, |
| "loss": 0.7884, |
| "mean_token_accuracy": 0.7633147515282703, |
| "step": 930 |
| }, |
| { |
| "epoch": 0.6919518963922294, |
| "grad_norm": 0.789919298542449, |
| "learning_rate": 5.248307512139818e-06, |
| "loss": 0.7862, |
| "mean_token_accuracy": 0.7651956223578331, |
| "step": 935 |
| }, |
| { |
| "epoch": 0.6956521739130435, |
| "grad_norm": 0.775028897496873, |
| "learning_rate": 5.134951820312402e-06, |
| "loss": 0.7608, |
| "mean_token_accuracy": 0.7710794013752429, |
| "step": 940 |
| }, |
| { |
| "epoch": 0.6993524514338575, |
| "grad_norm": 0.8430703919544058, |
| "learning_rate": 5.022409274044346e-06, |
| "loss": 0.7915, |
| "mean_token_accuracy": 0.7635476913138959, |
| "step": 945 |
| }, |
| { |
| "epoch": 0.7030527289546716, |
| "grad_norm": 0.8044082255650751, |
| "learning_rate": 4.910698683729371e-06, |
| "loss": 0.7638, |
| "mean_token_accuracy": 0.7705783943727836, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.7067530064754857, |
| "grad_norm": 0.7439402505547626, |
| "learning_rate": 4.799838720707847e-06, |
| "loss": 0.7835, |
| "mean_token_accuracy": 0.76646702805682, |
| "step": 955 |
| }, |
| { |
| "epoch": 0.7104532839962997, |
| "grad_norm": 0.8141310220719153, |
| "learning_rate": 4.6898479141460415e-06, |
| "loss": 0.7583, |
| "mean_token_accuracy": 0.7721620830045813, |
| "step": 960 |
| }, |
| { |
| "epoch": 0.7141535615171137, |
| "grad_norm": 0.7863013724531347, |
| "learning_rate": 4.580744647939163e-06, |
| "loss": 0.7648, |
| "mean_token_accuracy": 0.7700705090647101, |
| "step": 965 |
| }, |
| { |
| "epoch": 0.7178538390379279, |
| "grad_norm": 0.7919610069204563, |
| "learning_rate": 4.472547157638674e-06, |
| "loss": 0.7756, |
| "mean_token_accuracy": 0.7670807847229706, |
| "step": 970 |
| }, |
| { |
| "epoch": 0.7215541165587419, |
| "grad_norm": 0.8241671907963101, |
| "learning_rate": 4.365273527404384e-06, |
| "loss": 0.7595, |
| "mean_token_accuracy": 0.7707993119384532, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.7252543940795559, |
| "grad_norm": 0.7638723992566342, |
| "learning_rate": 4.258941686981864e-06, |
| "loss": 0.7529, |
| "mean_token_accuracy": 0.7727072913006545, |
| "step": 980 |
| }, |
| { |
| "epoch": 0.72895467160037, |
| "grad_norm": 0.7934309413562728, |
| "learning_rate": 4.15356940870567e-06, |
| "loss": 0.7721, |
| "mean_token_accuracy": 0.7670910625057229, |
| "step": 985 |
| }, |
| { |
| "epoch": 0.7326549491211841, |
| "grad_norm": 0.8485940316250707, |
| "learning_rate": 4.049174304528857e-06, |
| "loss": 0.802, |
| "mean_token_accuracy": 0.7588791241905861, |
| "step": 990 |
| }, |
| { |
| "epoch": 0.7363552266419982, |
| "grad_norm": 0.8810207610120062, |
| "learning_rate": 3.945773823079315e-06, |
| "loss": 0.7595, |
| "mean_token_accuracy": 0.7711046119500352, |
| "step": 995 |
| }, |
| { |
| "epoch": 0.7400555041628122, |
| "grad_norm": 0.833110497686039, |
| "learning_rate": 3.8433852467434175e-06, |
| "loss": 0.7722, |
| "mean_token_accuracy": 0.7675894720358956, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7400555041628122, |
| "eval_loss": 0.7973992824554443, |
| "eval_mean_token_accuracy": 0.757112345312609, |
| "eval_runtime": 14.427, |
| "eval_samples_per_second": 17.814, |
| "eval_steps_per_second": 1.178, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7437557816836263, |
| "grad_norm": 0.8237006050954356, |
| "learning_rate": 3.742025688777413e-06, |
| "loss": 0.8037, |
| "mean_token_accuracy": 0.7586444848081989, |
| "step": 1005 |
| }, |
| { |
| "epoch": 0.7474560592044404, |
| "grad_norm": 0.7526823694442858, |
| "learning_rate": 3.641712090447125e-06, |
| "loss": 0.7716, |
| "mean_token_accuracy": 0.7684351614558499, |
| "step": 1010 |
| }, |
| { |
| "epoch": 0.7511563367252544, |
| "grad_norm": 0.7943312700833844, |
| "learning_rate": 3.542461218196379e-06, |
| "loss": 0.731, |
| "mean_token_accuracy": 0.7802367124302477, |
| "step": 1015 |
| }, |
| { |
| "epoch": 0.7548566142460684, |
| "grad_norm": 0.7798186842393364, |
| "learning_rate": 3.444289660844665e-06, |
| "loss": 0.8063, |
| "mean_token_accuracy": 0.7586571868727487, |
| "step": 1020 |
| }, |
| { |
| "epoch": 0.7585568917668826, |
| "grad_norm": 0.765497355738645, |
| "learning_rate": 3.347213826814456e-06, |
| "loss": 0.7583, |
| "mean_token_accuracy": 0.7724547572887597, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.7622571692876966, |
| "grad_norm": 0.7857443566387864, |
| "learning_rate": 3.2512499413887255e-06, |
| "loss": 0.7543, |
| "mean_token_accuracy": 0.7736901779455366, |
| "step": 1030 |
| }, |
| { |
| "epoch": 0.7659574468085106, |
| "grad_norm": 0.802530120535559, |
| "learning_rate": 3.1564140439990256e-06, |
| "loss": 0.7567, |
| "mean_token_accuracy": 0.7718877631758259, |
| "step": 1035 |
| }, |
| { |
| "epoch": 0.7696577243293247, |
| "grad_norm": 0.7989942251381315, |
| "learning_rate": 3.0627219855446667e-06, |
| "loss": 0.7661, |
| "mean_token_accuracy": 0.7682660057488334, |
| "step": 1040 |
| }, |
| { |
| "epoch": 0.7733580018501388, |
| "grad_norm": 0.840504807173006, |
| "learning_rate": 2.970189425743383e-06, |
| "loss": 0.7551, |
| "mean_token_accuracy": 0.7720390981661918, |
| "step": 1045 |
| }, |
| { |
| "epoch": 0.7770582793709528, |
| "grad_norm": 0.7779487830464692, |
| "learning_rate": 2.8788318305139808e-06, |
| "loss": 0.7655, |
| "mean_token_accuracy": 0.7693840268148309, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.7807585568917669, |
| "grad_norm": 0.7927243683319699, |
| "learning_rate": 2.7886644693913333e-06, |
| "loss": 0.7743, |
| "mean_token_accuracy": 0.7681611530624838, |
| "step": 1055 |
| }, |
| { |
| "epoch": 0.7844588344125809, |
| "grad_norm": 0.8373582713281018, |
| "learning_rate": 2.6997024129742544e-06, |
| "loss": 0.7785, |
| "mean_token_accuracy": 0.766058403299143, |
| "step": 1060 |
| }, |
| { |
| "epoch": 0.788159111933395, |
| "grad_norm": 0.8441970949667804, |
| "learning_rate": 2.611960530406572e-06, |
| "loss": 0.7648, |
| "mean_token_accuracy": 0.7704607514620688, |
| "step": 1065 |
| }, |
| { |
| "epoch": 0.7918593894542091, |
| "grad_norm": 0.7819472527987796, |
| "learning_rate": 2.5254534868919077e-06, |
| "loss": 0.745, |
| "mean_token_accuracy": 0.7760465989840439, |
| "step": 1070 |
| }, |
| { |
| "epoch": 0.7955596669750231, |
| "grad_norm": 0.7781088934347913, |
| "learning_rate": 2.4401957412425213e-06, |
| "loss": 0.7672, |
| "mean_token_accuracy": 0.7700818907749126, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.7992599444958371, |
| "grad_norm": 0.8191690736963815, |
| "learning_rate": 2.3562015434626784e-06, |
| "loss": 0.7427, |
| "mean_token_accuracy": 0.7759877209976991, |
| "step": 1080 |
| }, |
| { |
| "epoch": 0.8029602220166513, |
| "grad_norm": 0.7935070459901263, |
| "learning_rate": 2.273484932366874e-06, |
| "loss": 0.8047, |
| "mean_token_accuracy": 0.7583013868934, |
| "step": 1085 |
| }, |
| { |
| "epoch": 0.8066604995374653, |
| "grad_norm": 0.7331859606400628, |
| "learning_rate": 2.192059733233408e-06, |
| "loss": 0.8008, |
| "mean_token_accuracy": 0.7600684615036566, |
| "step": 1090 |
| }, |
| { |
| "epoch": 0.8103607770582794, |
| "grad_norm": 0.8074434856892805, |
| "learning_rate": 2.111939555493603e-06, |
| "loss": 0.7794, |
| "mean_token_accuracy": 0.7657683649247553, |
| "step": 1095 |
| }, |
| { |
| "epoch": 0.8140610545790934, |
| "grad_norm": 0.8086682573035457, |
| "learning_rate": 2.0331377904571303e-06, |
| "loss": 0.7586, |
| "mean_token_accuracy": 0.7709054565375336, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.8140610545790934, |
| "eval_loss": 0.7944868803024292, |
| "eval_mean_token_accuracy": 0.7578580899604019, |
| "eval_runtime": 14.41, |
| "eval_samples_per_second": 17.835, |
| "eval_steps_per_second": 1.18, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.8177613320999075, |
| "grad_norm": 0.8343820210266487, |
| "learning_rate": 1.9556676090737803e-06, |
| "loss": 0.7709, |
| "mean_token_accuracy": 0.7677138846863649, |
| "step": 1105 |
| }, |
| { |
| "epoch": 0.8214616096207216, |
| "grad_norm": 0.8274108362888261, |
| "learning_rate": 1.879541959732072e-06, |
| "loss": 0.772, |
| "mean_token_accuracy": 0.767794811632847, |
| "step": 1110 |
| }, |
| { |
| "epoch": 0.8251618871415356, |
| "grad_norm": 0.7986915851539247, |
| "learning_rate": 1.8047735660950427e-06, |
| "loss": 0.7824, |
| "mean_token_accuracy": 0.7641448767963437, |
| "step": 1115 |
| }, |
| { |
| "epoch": 0.8288621646623496, |
| "grad_norm": 0.7912930392931041, |
| "learning_rate": 1.7313749249736266e-06, |
| "loss": 0.7722, |
| "mean_token_accuracy": 0.7688348884744852, |
| "step": 1120 |
| }, |
| { |
| "epoch": 0.8325624421831638, |
| "grad_norm": 0.7998408807770482, |
| "learning_rate": 1.6593583042379192e-06, |
| "loss": 0.7777, |
| "mean_token_accuracy": 0.7664428074095984, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.8362627197039778, |
| "grad_norm": 0.7582195922034395, |
| "learning_rate": 1.5887357407667314e-06, |
| "loss": 0.7635, |
| "mean_token_accuracy": 0.7685496955885053, |
| "step": 1130 |
| }, |
| { |
| "epoch": 0.8399629972247918, |
| "grad_norm": 0.7846551733743673, |
| "learning_rate": 1.5195190384357405e-06, |
| "loss": 0.7623, |
| "mean_token_accuracy": 0.7700296736101973, |
| "step": 1135 |
| }, |
| { |
| "epoch": 0.8436632747456059, |
| "grad_norm": 0.8102544575623821, |
| "learning_rate": 1.4517197661445893e-06, |
| "loss": 0.8151, |
| "mean_token_accuracy": 0.7554465880686968, |
| "step": 1140 |
| }, |
| { |
| "epoch": 0.84736355226642, |
| "grad_norm": 0.7915517548834473, |
| "learning_rate": 1.3853492558832472e-06, |
| "loss": 0.8012, |
| "mean_token_accuracy": 0.759899401391139, |
| "step": 1145 |
| }, |
| { |
| "epoch": 0.851063829787234, |
| "grad_norm": 0.771064167787231, |
| "learning_rate": 1.3204186008379926e-06, |
| "loss": 0.7466, |
| "mean_token_accuracy": 0.7737188447666707, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.8547641073080481, |
| "grad_norm": 0.7634680191740536, |
| "learning_rate": 1.2569386535372807e-06, |
| "loss": 0.7571, |
| "mean_token_accuracy": 0.7721920358524959, |
| "step": 1155 |
| }, |
| { |
| "epoch": 0.8584643848288621, |
| "grad_norm": 0.7702642560131528, |
| "learning_rate": 1.1949200240378577e-06, |
| "loss": 0.7681, |
| "mean_token_accuracy": 0.7690720719541746, |
| "step": 1160 |
| }, |
| { |
| "epoch": 0.8621646623496763, |
| "grad_norm": 0.7798114837698836, |
| "learning_rate": 1.1343730781513896e-06, |
| "loss": 0.747, |
| "mean_token_accuracy": 0.774875145015143, |
| "step": 1165 |
| }, |
| { |
| "epoch": 0.8658649398704903, |
| "grad_norm": 0.7949938733034089, |
| "learning_rate": 1.0753079357119134e-06, |
| "loss": 0.7849, |
| "mean_token_accuracy": 0.765570759792907, |
| "step": 1170 |
| }, |
| { |
| "epoch": 0.8695652173913043, |
| "grad_norm": 0.7698843089224456, |
| "learning_rate": 1.017734468884417e-06, |
| "loss": 0.7242, |
| "mean_token_accuracy": 0.7807185988785081, |
| "step": 1175 |
| }, |
| { |
| "epoch": 0.8732654949121184, |
| "grad_norm": 0.7280858038390325, |
| "learning_rate": 9.616623005147952e-07, |
| "loss": 0.7586, |
| "mean_token_accuracy": 0.7718187615798253, |
| "step": 1180 |
| }, |
| { |
| "epoch": 0.8769657724329325, |
| "grad_norm": 0.839720555473333, |
| "learning_rate": 9.071008025214767e-07, |
| "loss": 0.7946, |
| "mean_token_accuracy": 0.7638036068173906, |
| "step": 1185 |
| }, |
| { |
| "epoch": 0.8806660499537465, |
| "grad_norm": 0.7555588990753439, |
| "learning_rate": 8.540590943290128e-07, |
| "loss": 0.7571, |
| "mean_token_accuracy": 0.7721729084932626, |
| "step": 1190 |
| }, |
| { |
| "epoch": 0.8843663274745606, |
| "grad_norm": 0.7969942371823111, |
| "learning_rate": 8.025460413438457e-07, |
| "loss": 0.8112, |
| "mean_token_accuracy": 0.75678643764634, |
| "step": 1195 |
| }, |
| { |
| "epoch": 0.8880666049953746, |
| "grad_norm": 0.7364947816684886, |
| "learning_rate": 7.525702534725443e-07, |
| "loss": 0.7528, |
| "mean_token_accuracy": 0.7727457500345997, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.8880666049953746, |
| "eval_loss": 0.7930368185043335, |
| "eval_mean_token_accuracy": 0.7581294874020488, |
| "eval_runtime": 23.741, |
| "eval_samples_per_second": 10.825, |
| "eval_steps_per_second": 0.716, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.8917668825161887, |
| "grad_norm": 0.8056054421008482, |
| "learning_rate": 7.041400836827439e-07, |
| "loss": 0.7526, |
| "mean_token_accuracy": 0.7748112663916402, |
| "step": 1205 |
| }, |
| { |
| "epoch": 0.8954671600370028, |
| "grad_norm": 0.7430784621258909, |
| "learning_rate": 6.572636266070265e-07, |
| "loss": 0.7695, |
| "mean_token_accuracy": 0.7691522840115936, |
| "step": 1210 |
| }, |
| { |
| "epoch": 0.8991674375578168, |
| "grad_norm": 0.7294844027936921, |
| "learning_rate": 6.119487171899807e-07, |
| "loss": 0.7605, |
| "mean_token_accuracy": 0.7704982028684197, |
| "step": 1215 |
| }, |
| { |
| "epoch": 0.902867715078631, |
| "grad_norm": 0.8002640649815743, |
| "learning_rate": 5.682029293786673e-07, |
| "loss": 0.7926, |
| "mean_token_accuracy": 0.7625801633473456, |
| "step": 1220 |
| }, |
| { |
| "epoch": 0.906567992599445, |
| "grad_norm": 0.7836990755620872, |
| "learning_rate": 5.26033574856708e-07, |
| "loss": 0.7672, |
| "mean_token_accuracy": 0.7700011696772932, |
| "step": 1225 |
| }, |
| { |
| "epoch": 0.910268270120259, |
| "grad_norm": 0.8159687086578604, |
| "learning_rate": 4.854477018222103e-07, |
| "loss": 0.8008, |
| "mean_token_accuracy": 0.7607755917340568, |
| "step": 1230 |
| }, |
| { |
| "epoch": 0.913968547641073, |
| "grad_norm": 0.7645217245522713, |
| "learning_rate": 4.464520938097294e-07, |
| "loss": 0.7788, |
| "mean_token_accuracy": 0.7658377209811824, |
| "step": 1235 |
| }, |
| { |
| "epoch": 0.9176688251618872, |
| "grad_norm": 0.7673840812542029, |
| "learning_rate": 4.0905326855646186e-07, |
| "loss": 0.7573, |
| "mean_token_accuracy": 0.7721266403991972, |
| "step": 1240 |
| }, |
| { |
| "epoch": 0.9213691026827012, |
| "grad_norm": 0.7454299232676801, |
| "learning_rate": 3.732574769128738e-07, |
| "loss": 0.7659, |
| "mean_token_accuracy": 0.7694314927154986, |
| "step": 1245 |
| }, |
| { |
| "epoch": 0.9250693802035153, |
| "grad_norm": 0.7654595178316245, |
| "learning_rate": 3.390707017979311e-07, |
| "loss": 0.745, |
| "mean_token_accuracy": 0.7750965684798319, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.9287696577243293, |
| "grad_norm": 0.8067427671781613, |
| "learning_rate": 3.06498657199108e-07, |
| "loss": 0.7732, |
| "mean_token_accuracy": 0.768182908414515, |
| "step": 1255 |
| }, |
| { |
| "epoch": 0.9324699352451434, |
| "grad_norm": 0.7562039729125304, |
| "learning_rate": 2.7554678721735675e-07, |
| "loss": 0.7787, |
| "mean_token_accuracy": 0.7642553045991999, |
| "step": 1260 |
| }, |
| { |
| "epoch": 0.9361702127659575, |
| "grad_norm": 0.7430603594511384, |
| "learning_rate": 2.4622026515717654e-07, |
| "loss": 0.8077, |
| "mean_token_accuracy": 0.7573527261389765, |
| "step": 1265 |
| }, |
| { |
| "epoch": 0.9398704902867715, |
| "grad_norm": 0.7593630984334607, |
| "learning_rate": 2.1852399266194312e-07, |
| "loss": 0.7793, |
| "mean_token_accuracy": 0.764669397435321, |
| "step": 1270 |
| }, |
| { |
| "epoch": 0.9435707678075855, |
| "grad_norm": 0.7357935865065981, |
| "learning_rate": 1.9246259889464935e-07, |
| "loss": 0.7781, |
| "mean_token_accuracy": 0.7665922229483206, |
| "step": 1275 |
| }, |
| { |
| "epoch": 0.9472710453283997, |
| "grad_norm": 0.7288331937335505, |
| "learning_rate": 1.6804043976418438e-07, |
| "loss": 0.7704, |
| "mean_token_accuracy": 0.7690436128045272, |
| "step": 1280 |
| }, |
| { |
| "epoch": 0.9509713228492137, |
| "grad_norm": 0.7843857219453947, |
| "learning_rate": 1.4526159719728595e-07, |
| "loss": 0.7768, |
| "mean_token_accuracy": 0.7667279502666673, |
| "step": 1285 |
| }, |
| { |
| "epoch": 0.9546716003700277, |
| "grad_norm": 0.8205041527973095, |
| "learning_rate": 1.24129878456285e-07, |
| "loss": 0.808, |
| "mean_token_accuracy": 0.759965865777703, |
| "step": 1290 |
| }, |
| { |
| "epoch": 0.9583718778908418, |
| "grad_norm": 0.7786180946366419, |
| "learning_rate": 1.0464881550276362e-07, |
| "loss": 0.7656, |
| "mean_token_accuracy": 0.7689572567793539, |
| "step": 1295 |
| }, |
| { |
| "epoch": 0.9620721554116559, |
| "grad_norm": 0.8324732784427329, |
| "learning_rate": 8.682166440721729e-08, |
| "loss": 0.7554, |
| "mean_token_accuracy": 0.7726877638582714, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.9620721554116559, |
| "eval_loss": 0.7924448251724243, |
| "eval_mean_token_accuracy": 0.7585140244194415, |
| "eval_runtime": 14.4043, |
| "eval_samples_per_second": 17.842, |
| "eval_steps_per_second": 1.18, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.96577243293247, |
| "grad_norm": 0.8003294805368201, |
| "learning_rate": 7.065140480483235e-08, |
| "loss": 0.7953, |
| "mean_token_accuracy": 0.7616640102372079, |
| "step": 1305 |
| }, |
| { |
| "epoch": 0.969472710453284, |
| "grad_norm": 0.8056079696464618, |
| "learning_rate": 5.6140739397474445e-08, |
| "loss": 0.7998, |
| "mean_token_accuracy": 0.7604724691954434, |
| "step": 1310 |
| }, |
| { |
| "epoch": 0.973172987974098, |
| "grad_norm": 0.7398525398002953, |
| "learning_rate": 4.329209350195651e-08, |
| "loss": 0.7687, |
| "mean_token_accuracy": 0.7694833489373637, |
| "step": 1315 |
| }, |
| { |
| "epoch": 0.9768732654949122, |
| "grad_norm": 0.767475456178718, |
| "learning_rate": 3.210761464466639e-08, |
| "loss": 0.7683, |
| "mean_token_accuracy": 0.7690996438734377, |
| "step": 1320 |
| }, |
| { |
| "epoch": 0.9805735430157262, |
| "grad_norm": 0.804536573614064, |
| "learning_rate": 2.2589172202635014e-08, |
| "loss": 0.7773, |
| "mean_token_accuracy": 0.7664188170808871, |
| "step": 1325 |
| }, |
| { |
| "epoch": 0.9842738205365402, |
| "grad_norm": 0.8454088605208244, |
| "learning_rate": 1.4738357091084177e-08, |
| "loss": 0.768, |
| "mean_token_accuracy": 0.7688064486234464, |
| "step": 1330 |
| }, |
| { |
| "epoch": 0.9879740980573543, |
| "grad_norm": 0.8321469884127546, |
| "learning_rate": 8.556481497521418e-09, |
| "loss": 0.764, |
| "mean_token_accuracy": 0.7714298871126578, |
| "step": 1335 |
| }, |
| { |
| "epoch": 0.9916743755781684, |
| "grad_norm": 0.8508534520033296, |
| "learning_rate": 4.044578662419918e-09, |
| "loss": 0.7853, |
| "mean_token_accuracy": 0.7642099197772693, |
| "step": 1340 |
| }, |
| { |
| "epoch": 0.9953746530989824, |
| "grad_norm": 0.7643692450259899, |
| "learning_rate": 1.203402706525525e-09, |
| "loss": 0.7728, |
| "mean_token_accuracy": 0.7677490371223821, |
| "step": 1345 |
| }, |
| { |
| "epoch": 0.9990749306197965, |
| "grad_norm": 0.7766881531885754, |
| "learning_rate": 3.342850480869686e-11, |
| "loss": 0.7842, |
| "mean_token_accuracy": 0.7655295798263551, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.9998149861239592, |
| "mean_token_accuracy": 0.7588118329384822, |
| "step": 1351, |
| "total_flos": 76959556042752.0, |
| "train_loss": 0.8100585834261755, |
| "train_runtime": 5700.9233, |
| "train_samples_per_second": 7.585, |
| "train_steps_per_second": 0.237 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1351, |
| "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": false, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 76959556042752.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
