{ "best_metric": null, "best_model_checkpoint": null, "epoch": 25.0, "eval_steps": 500, "global_step": 3875, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.06451612903225806, "grad_norm": 14.686903953552246, "learning_rate": 5.154639175257732e-06, "loss": 1.8113, "step": 10 }, { "epoch": 0.12903225806451613, "grad_norm": 9.743651390075684, "learning_rate": 1.0309278350515464e-05, "loss": 1.57, "step": 20 }, { "epoch": 0.1935483870967742, "grad_norm": 2.2041146755218506, "learning_rate": 1.5463917525773197e-05, "loss": 0.6814, "step": 30 }, { "epoch": 0.25806451612903225, "grad_norm": 3.186547040939331, "learning_rate": 2.0618556701030927e-05, "loss": 0.4245, "step": 40 }, { "epoch": 0.3225806451612903, "grad_norm": 1.2101554870605469, "learning_rate": 2.5773195876288658e-05, "loss": 0.2726, "step": 50 }, { "epoch": 0.3870967741935484, "grad_norm": 0.7474563121795654, "learning_rate": 3.0927835051546395e-05, "loss": 0.2088, "step": 60 }, { "epoch": 0.45161290322580644, "grad_norm": 0.9382892847061157, "learning_rate": 3.6082474226804125e-05, "loss": 0.2013, "step": 70 }, { "epoch": 0.5161290322580645, "grad_norm": 1.0286656618118286, "learning_rate": 4.1237113402061855e-05, "loss": 0.1683, "step": 80 }, { "epoch": 0.5806451612903226, "grad_norm": 0.867642343044281, "learning_rate": 4.639175257731959e-05, "loss": 0.1561, "step": 90 }, { "epoch": 0.6451612903225806, "grad_norm": 1.0434398651123047, "learning_rate": 5.1546391752577315e-05, "loss": 0.1509, "step": 100 }, { "epoch": 0.7096774193548387, "grad_norm": 1.3080216646194458, "learning_rate": 5.670103092783505e-05, "loss": 0.1291, "step": 110 }, { "epoch": 0.7741935483870968, "grad_norm": 0.38779398798942566, "learning_rate": 6.185567010309279e-05, "loss": 0.1209, "step": 120 }, { "epoch": 0.8387096774193549, "grad_norm": 0.49606090784072876, "learning_rate": 6.701030927835051e-05, "loss": 0.1202, "step": 130 }, { "epoch": 0.9032258064516129, "grad_norm": 0.7980489730834961, "learning_rate": 7.216494845360825e-05, "loss": 0.1153, "step": 140 }, { "epoch": 0.967741935483871, "grad_norm": 0.8646230697631836, "learning_rate": 7.731958762886599e-05, "loss": 0.1068, "step": 150 }, { "epoch": 1.032258064516129, "grad_norm": 0.6217169165611267, "learning_rate": 8.247422680412371e-05, "loss": 0.1049, "step": 160 }, { "epoch": 1.096774193548387, "grad_norm": 0.7635177373886108, "learning_rate": 8.762886597938145e-05, "loss": 0.0953, "step": 170 }, { "epoch": 1.1612903225806452, "grad_norm": 0.5218844413757324, "learning_rate": 9.278350515463918e-05, "loss": 0.098, "step": 180 }, { "epoch": 1.2258064516129032, "grad_norm": 0.5090600848197937, "learning_rate": 9.793814432989691e-05, "loss": 0.1049, "step": 190 }, { "epoch": 1.2903225806451613, "grad_norm": 0.5775400400161743, "learning_rate": 9.999934444418978e-05, "loss": 0.0995, "step": 200 }, { "epoch": 1.3548387096774195, "grad_norm": 0.739729106426239, "learning_rate": 9.999533833204613e-05, "loss": 0.0879, "step": 210 }, { "epoch": 1.4193548387096775, "grad_norm": 0.4800107479095459, "learning_rate": 9.99876905968797e-05, "loss": 0.0854, "step": 220 }, { "epoch": 1.4838709677419355, "grad_norm": 0.7302453517913818, "learning_rate": 9.997640179574574e-05, "loss": 0.0923, "step": 230 }, { "epoch": 1.5483870967741935, "grad_norm": 0.362665057182312, "learning_rate": 9.996147275091215e-05, "loss": 0.0873, "step": 240 }, { "epoch": 1.6129032258064515, "grad_norm": 0.502337634563446, "learning_rate": 9.994290454979931e-05, "loss": 0.0833, "step": 250 }, { "epoch": 1.6774193548387095, "grad_norm": 0.6111910939216614, "learning_rate": 9.992069854490114e-05, "loss": 0.0806, "step": 260 }, { "epoch": 1.7419354838709677, "grad_norm": 0.5922254920005798, "learning_rate": 9.989485635368641e-05, "loss": 0.0838, "step": 270 }, { "epoch": 1.8064516129032258, "grad_norm": 1.0255498886108398, "learning_rate": 9.986537985848097e-05, "loss": 0.0838, "step": 280 }, { "epoch": 1.870967741935484, "grad_norm": 0.8610944151878357, "learning_rate": 9.983227120633066e-05, "loss": 0.0831, "step": 290 }, { "epoch": 1.935483870967742, "grad_norm": 0.5282477736473083, "learning_rate": 9.979553280884494e-05, "loss": 0.0768, "step": 300 }, { "epoch": 2.0, "grad_norm": 0.7257622480392456, "learning_rate": 9.975516734202118e-05, "loss": 0.082, "step": 310 }, { "epoch": 2.064516129032258, "grad_norm": 0.5951042771339417, "learning_rate": 9.971117774604977e-05, "loss": 0.0785, "step": 320 }, { "epoch": 2.129032258064516, "grad_norm": 1.008421778678894, "learning_rate": 9.966356722509995e-05, "loss": 0.0756, "step": 330 }, { "epoch": 2.193548387096774, "grad_norm": 0.6158808469772339, "learning_rate": 9.961233924708644e-05, "loss": 0.0698, "step": 340 }, { "epoch": 2.258064516129032, "grad_norm": 0.5757787227630615, "learning_rate": 9.95574975434168e-05, "loss": 0.0703, "step": 350 }, { "epoch": 2.3225806451612905, "grad_norm": 0.6450510621070862, "learning_rate": 9.949904610871968e-05, "loss": 0.0685, "step": 360 }, { "epoch": 2.3870967741935485, "grad_norm": 0.5834465026855469, "learning_rate": 9.943698920055384e-05, "loss": 0.0676, "step": 370 }, { "epoch": 2.4516129032258065, "grad_norm": 0.634287416934967, "learning_rate": 9.937133133909801e-05, "loss": 0.0639, "step": 380 }, { "epoch": 2.5161290322580645, "grad_norm": 0.32366201281547546, "learning_rate": 9.93020773068217e-05, "loss": 0.0734, "step": 390 }, { "epoch": 2.5806451612903225, "grad_norm": 0.6290192008018494, "learning_rate": 9.922923214813675e-05, "loss": 0.0748, "step": 400 }, { "epoch": 2.6451612903225805, "grad_norm": 0.5695043206214905, "learning_rate": 9.915280116903003e-05, "loss": 0.066, "step": 410 }, { "epoch": 2.709677419354839, "grad_norm": 0.5956539511680603, "learning_rate": 9.907278993667681e-05, "loss": 0.065, "step": 420 }, { "epoch": 2.774193548387097, "grad_norm": 0.505896806716919, "learning_rate": 9.89892042790354e-05, "loss": 0.0637, "step": 430 }, { "epoch": 2.838709677419355, "grad_norm": 0.6462914943695068, "learning_rate": 9.890205028442252e-05, "loss": 0.0742, "step": 440 }, { "epoch": 2.903225806451613, "grad_norm": 0.6515319347381592, "learning_rate": 9.88113343010699e-05, "loss": 0.068, "step": 450 }, { "epoch": 2.967741935483871, "grad_norm": 0.5437873601913452, "learning_rate": 9.87170629366619e-05, "loss": 0.0734, "step": 460 }, { "epoch": 3.032258064516129, "grad_norm": 0.5710848569869995, "learning_rate": 9.861924305785409e-05, "loss": 0.072, "step": 470 }, { "epoch": 3.096774193548387, "grad_norm": 0.47411492466926575, "learning_rate": 9.851788178977327e-05, "loss": 0.0632, "step": 480 }, { "epoch": 3.161290322580645, "grad_norm": 0.40361443161964417, "learning_rate": 9.841298651549831e-05, "loss": 0.058, "step": 490 }, { "epoch": 3.225806451612903, "grad_norm": 0.5025441646575928, "learning_rate": 9.83045648755225e-05, "loss": 0.0594, "step": 500 }, { "epoch": 3.2903225806451615, "grad_norm": 0.6928324103355408, "learning_rate": 9.819262476719692e-05, "loss": 0.061, "step": 510 }, { "epoch": 3.3548387096774195, "grad_norm": 0.45038920640945435, "learning_rate": 9.807717434415527e-05, "loss": 0.0597, "step": 520 }, { "epoch": 3.4193548387096775, "grad_norm": 0.35843929648399353, "learning_rate": 9.795822201571998e-05, "loss": 0.0637, "step": 530 }, { "epoch": 3.4838709677419355, "grad_norm": 0.727299690246582, "learning_rate": 9.78357764462896e-05, "loss": 0.062, "step": 540 }, { "epoch": 3.5483870967741935, "grad_norm": 0.5294666886329651, "learning_rate": 9.770984655470775e-05, "loss": 0.0616, "step": 550 }, { "epoch": 3.6129032258064515, "grad_norm": 0.24387721717357635, "learning_rate": 9.758044151361347e-05, "loss": 0.0599, "step": 560 }, { "epoch": 3.6774193548387095, "grad_norm": 0.42391934990882874, "learning_rate": 9.744757074877312e-05, "loss": 0.0572, "step": 570 }, { "epoch": 3.741935483870968, "grad_norm": 0.43955346941947937, "learning_rate": 9.731124393839375e-05, "loss": 0.0562, "step": 580 }, { "epoch": 3.806451612903226, "grad_norm": 0.6420799493789673, "learning_rate": 9.717147101241816e-05, "loss": 0.0568, "step": 590 }, { "epoch": 3.870967741935484, "grad_norm": 0.2555811405181885, "learning_rate": 9.70282621518017e-05, "loss": 0.0539, "step": 600 }, { "epoch": 3.935483870967742, "grad_norm": 0.408918559551239, "learning_rate": 9.688162778777056e-05, "loss": 0.0585, "step": 610 }, { "epoch": 4.0, "grad_norm": 0.422941118478775, "learning_rate": 9.673157860106206e-05, "loss": 0.0547, "step": 620 }, { "epoch": 4.064516129032258, "grad_norm": 0.4911976158618927, "learning_rate": 9.657812552114667e-05, "loss": 0.0559, "step": 630 }, { "epoch": 4.129032258064516, "grad_norm": 0.715604841709137, "learning_rate": 9.642127972543184e-05, "loss": 0.0536, "step": 640 }, { "epoch": 4.193548387096774, "grad_norm": 0.530028760433197, "learning_rate": 9.626105263844797e-05, "loss": 0.0538, "step": 650 }, { "epoch": 4.258064516129032, "grad_norm": 0.4460236728191376, "learning_rate": 9.609745593101611e-05, "loss": 0.0533, "step": 660 }, { "epoch": 4.32258064516129, "grad_norm": 0.422241747379303, "learning_rate": 9.5930501519398e-05, "loss": 0.0573, "step": 670 }, { "epoch": 4.387096774193548, "grad_norm": 0.5292857885360718, "learning_rate": 9.576020156442802e-05, "loss": 0.0558, "step": 680 }, { "epoch": 4.451612903225806, "grad_norm": 0.3689402937889099, "learning_rate": 9.55865684706274e-05, "loss": 0.0502, "step": 690 }, { "epoch": 4.516129032258064, "grad_norm": 0.490133672952652, "learning_rate": 9.540961488530074e-05, "loss": 0.0574, "step": 700 }, { "epoch": 4.580645161290323, "grad_norm": 0.40120774507522583, "learning_rate": 9.522935369761475e-05, "loss": 0.0557, "step": 710 }, { "epoch": 4.645161290322581, "grad_norm": 0.4087425172328949, "learning_rate": 9.504579803765943e-05, "loss": 0.0504, "step": 720 }, { "epoch": 4.709677419354839, "grad_norm": 0.6458566188812256, "learning_rate": 9.485896127549163e-05, "loss": 0.0564, "step": 730 }, { "epoch": 4.774193548387097, "grad_norm": 0.43900686502456665, "learning_rate": 9.466885702016127e-05, "loss": 0.0566, "step": 740 }, { "epoch": 4.838709677419355, "grad_norm": 0.4088176190853119, "learning_rate": 9.447549911872002e-05, "loss": 0.058, "step": 750 }, { "epoch": 4.903225806451613, "grad_norm": 0.40604203939437866, "learning_rate": 9.427890165521269e-05, "loss": 0.0499, "step": 760 }, { "epoch": 4.967741935483871, "grad_norm": 0.41275304555892944, "learning_rate": 9.407907894965137e-05, "loss": 0.0499, "step": 770 }, { "epoch": 5.032258064516129, "grad_norm": 0.2919212281703949, "learning_rate": 9.387604555697239e-05, "loss": 0.0555, "step": 780 }, { "epoch": 5.096774193548387, "grad_norm": 0.39005234837532043, "learning_rate": 9.366981626597608e-05, "loss": 0.0502, "step": 790 }, { "epoch": 5.161290322580645, "grad_norm": 0.4438292980194092, "learning_rate": 9.346040609824962e-05, "loss": 0.0568, "step": 800 }, { "epoch": 5.225806451612903, "grad_norm": 0.5161975622177124, "learning_rate": 9.324783030707292e-05, "loss": 0.0532, "step": 810 }, { "epoch": 5.290322580645161, "grad_norm": 0.40832287073135376, "learning_rate": 9.303210437630746e-05, "loss": 0.0552, "step": 820 }, { "epoch": 5.354838709677419, "grad_norm": 0.3923545777797699, "learning_rate": 9.28132440192686e-05, "loss": 0.0557, "step": 830 }, { "epoch": 5.419354838709677, "grad_norm": 0.33239755034446716, "learning_rate": 9.25912651775809e-05, "loss": 0.054, "step": 840 }, { "epoch": 5.483870967741936, "grad_norm": 0.40319114923477173, "learning_rate": 9.236618402001707e-05, "loss": 0.0501, "step": 850 }, { "epoch": 5.548387096774194, "grad_norm": 0.49930399656295776, "learning_rate": 9.213801694132012e-05, "loss": 0.0541, "step": 860 }, { "epoch": 5.612903225806452, "grad_norm": 0.3818790912628174, "learning_rate": 9.190678056100933e-05, "loss": 0.0501, "step": 870 }, { "epoch": 5.67741935483871, "grad_norm": 0.5576165914535522, "learning_rate": 9.16724917221695e-05, "loss": 0.0506, "step": 880 }, { "epoch": 5.741935483870968, "grad_norm": 0.5411040782928467, "learning_rate": 9.143516749022433e-05, "loss": 0.0461, "step": 890 }, { "epoch": 5.806451612903226, "grad_norm": 0.35116180777549744, "learning_rate": 9.119482515169324e-05, "loss": 0.0479, "step": 900 }, { "epoch": 5.870967741935484, "grad_norm": 0.30685099959373474, "learning_rate": 9.095148221293225e-05, "loss": 0.0459, "step": 910 }, { "epoch": 5.935483870967742, "grad_norm": 0.4129478633403778, "learning_rate": 9.070515639885888e-05, "loss": 0.0505, "step": 920 }, { "epoch": 6.0, "grad_norm": 0.5053979754447937, "learning_rate": 9.045586565166107e-05, "loss": 0.0525, "step": 930 }, { "epoch": 6.064516129032258, "grad_norm": 0.3175376057624817, "learning_rate": 9.020362812949024e-05, "loss": 0.0456, "step": 940 }, { "epoch": 6.129032258064516, "grad_norm": 0.4626881182193756, "learning_rate": 8.994846220513872e-05, "loss": 0.0441, "step": 950 }, { "epoch": 6.193548387096774, "grad_norm": 0.321857750415802, "learning_rate": 8.969038646470143e-05, "loss": 0.0468, "step": 960 }, { "epoch": 6.258064516129032, "grad_norm": 0.4286996126174927, "learning_rate": 8.942941970622216e-05, "loss": 0.0488, "step": 970 }, { "epoch": 6.32258064516129, "grad_norm": 0.5266990065574646, "learning_rate": 8.916558093832424e-05, "loss": 0.0478, "step": 980 }, { "epoch": 6.387096774193548, "grad_norm": 0.2945270538330078, "learning_rate": 8.889888937882604e-05, "loss": 0.052, "step": 990 }, { "epoch": 6.451612903225806, "grad_norm": 0.32375168800354004, "learning_rate": 8.862936445334116e-05, "loss": 0.0484, "step": 1000 }, { "epoch": 6.516129032258064, "grad_norm": 0.3967803120613098, "learning_rate": 8.835702579386342e-05, "loss": 0.0549, "step": 1010 }, { "epoch": 6.580645161290323, "grad_norm": 0.35367152094841003, "learning_rate": 8.808189323733693e-05, "loss": 0.0476, "step": 1020 }, { "epoch": 6.645161290322581, "grad_norm": 0.46086496114730835, "learning_rate": 8.780398682421117e-05, "loss": 0.0463, "step": 1030 }, { "epoch": 6.709677419354839, "grad_norm": 0.5151063799858093, "learning_rate": 8.752332679698128e-05, "loss": 0.0468, "step": 1040 }, { "epoch": 6.774193548387097, "grad_norm": 0.5349656343460083, "learning_rate": 8.723993359871354e-05, "loss": 0.0472, "step": 1050 }, { "epoch": 6.838709677419355, "grad_norm": 0.4370066821575165, "learning_rate": 8.695382787155643e-05, "loss": 0.0414, "step": 1060 }, { "epoch": 6.903225806451613, "grad_norm": 0.39678850769996643, "learning_rate": 8.666503045523692e-05, "loss": 0.0399, "step": 1070 }, { "epoch": 6.967741935483871, "grad_norm": 0.5493939518928528, "learning_rate": 8.63735623855427e-05, "loss": 0.0545, "step": 1080 }, { "epoch": 7.032258064516129, "grad_norm": 0.2903650104999542, "learning_rate": 8.607944489278976e-05, "loss": 0.0423, "step": 1090 }, { "epoch": 7.096774193548387, "grad_norm": 0.5154908895492554, "learning_rate": 8.578269940027616e-05, "loss": 0.0441, "step": 1100 }, { "epoch": 7.161290322580645, "grad_norm": 0.3061012625694275, "learning_rate": 8.548334752272146e-05, "loss": 0.0455, "step": 1110 }, { "epoch": 7.225806451612903, "grad_norm": 0.34137821197509766, "learning_rate": 8.518141106469233e-05, "loss": 0.0426, "step": 1120 }, { "epoch": 7.290322580645161, "grad_norm": 0.2501906156539917, "learning_rate": 8.487691201901439e-05, "loss": 0.0471, "step": 1130 }, { "epoch": 7.354838709677419, "grad_norm": 0.5570168495178223, "learning_rate": 8.456987256517023e-05, "loss": 0.0415, "step": 1140 }, { "epoch": 7.419354838709677, "grad_norm": 0.36885467171669006, "learning_rate": 8.426031506768381e-05, "loss": 0.0488, "step": 1150 }, { "epoch": 7.483870967741936, "grad_norm": 0.2259949892759323, "learning_rate": 8.394826207449157e-05, "loss": 0.0443, "step": 1160 }, { "epoch": 7.548387096774194, "grad_norm": 0.31414613127708435, "learning_rate": 8.363373631529998e-05, "loss": 0.0462, "step": 1170 }, { "epoch": 7.612903225806452, "grad_norm": 0.43513432145118713, "learning_rate": 8.331676069992989e-05, "loss": 0.0417, "step": 1180 }, { "epoch": 7.67741935483871, "grad_norm": 0.2809264361858368, "learning_rate": 8.299735831664788e-05, "loss": 0.0448, "step": 1190 }, { "epoch": 7.741935483870968, "grad_norm": 0.3759617209434509, "learning_rate": 8.267555243048446e-05, "loss": 0.0432, "step": 1200 }, { "epoch": 7.806451612903226, "grad_norm": 0.33138594031333923, "learning_rate": 8.235136648153954e-05, "loss": 0.0434, "step": 1210 }, { "epoch": 7.870967741935484, "grad_norm": 0.47733554244041443, "learning_rate": 8.202482408327497e-05, "loss": 0.0406, "step": 1220 }, { "epoch": 7.935483870967742, "grad_norm": 0.3868086338043213, "learning_rate": 8.16959490207946e-05, "loss": 0.0394, "step": 1230 }, { "epoch": 8.0, "grad_norm": 0.44184964895248413, "learning_rate": 8.136476524911187e-05, "loss": 0.0393, "step": 1240 }, { "epoch": 8.064516129032258, "grad_norm": 0.3758850693702698, "learning_rate": 8.103129689140479e-05, "loss": 0.0454, "step": 1250 }, { "epoch": 8.129032258064516, "grad_norm": 0.5086485743522644, "learning_rate": 8.069556823725895e-05, "loss": 0.042, "step": 1260 }, { "epoch": 8.193548387096774, "grad_norm": 0.3741207420825958, "learning_rate": 8.03576037408982e-05, "loss": 0.0439, "step": 1270 }, { "epoch": 8.258064516129032, "grad_norm": 0.49306967854499817, "learning_rate": 8.001742801940349e-05, "loss": 0.04, "step": 1280 }, { "epoch": 8.32258064516129, "grad_norm": 0.3498378098011017, "learning_rate": 7.96750658509198e-05, "loss": 0.0448, "step": 1290 }, { "epoch": 8.387096774193548, "grad_norm": 0.3390824496746063, "learning_rate": 7.933054217285118e-05, "loss": 0.0472, "step": 1300 }, { "epoch": 8.451612903225806, "grad_norm": 0.27651500701904297, "learning_rate": 7.898388208004449e-05, "loss": 0.0368, "step": 1310 }, { "epoch": 8.516129032258064, "grad_norm": 0.33945196866989136, "learning_rate": 7.86351108229614e-05, "loss": 0.0376, "step": 1320 }, { "epoch": 8.580645161290322, "grad_norm": 0.2656388580799103, "learning_rate": 7.828425380583923e-05, "loss": 0.0415, "step": 1330 }, { "epoch": 8.64516129032258, "grad_norm": 0.2991732954978943, "learning_rate": 7.793133658484051e-05, "loss": 0.0434, "step": 1340 }, { "epoch": 8.709677419354838, "grad_norm": 0.5004262328147888, "learning_rate": 7.757638486619144e-05, "loss": 0.0418, "step": 1350 }, { "epoch": 8.774193548387096, "grad_norm": 0.3629301190376282, "learning_rate": 7.72194245043095e-05, "loss": 0.0368, "step": 1360 }, { "epoch": 8.838709677419354, "grad_norm": 0.4232747256755829, "learning_rate": 7.686048149992032e-05, "loss": 0.0393, "step": 1370 }, { "epoch": 8.903225806451612, "grad_norm": 0.3285137116909027, "learning_rate": 7.649958199816363e-05, "loss": 0.0373, "step": 1380 }, { "epoch": 8.967741935483872, "grad_norm": 0.3300577402114868, "learning_rate": 7.613675228668904e-05, "loss": 0.0376, "step": 1390 }, { "epoch": 9.03225806451613, "grad_norm": 0.27525898814201355, "learning_rate": 7.577201879374115e-05, "loss": 0.0401, "step": 1400 }, { "epoch": 9.096774193548388, "grad_norm": 0.2236848622560501, "learning_rate": 7.540540808623455e-05, "loss": 0.0407, "step": 1410 }, { "epoch": 9.161290322580646, "grad_norm": 0.25309398770332336, "learning_rate": 7.50369468678188e-05, "loss": 0.0373, "step": 1420 }, { "epoch": 9.225806451612904, "grad_norm": 0.46527889370918274, "learning_rate": 7.466666197693326e-05, "loss": 0.0397, "step": 1430 }, { "epoch": 9.290322580645162, "grad_norm": 0.5017492175102234, "learning_rate": 7.429458038485222e-05, "loss": 0.0429, "step": 1440 }, { "epoch": 9.35483870967742, "grad_norm": 0.3527701795101166, "learning_rate": 7.392072919372032e-05, "loss": 0.0356, "step": 1450 }, { "epoch": 9.419354838709678, "grad_norm": 0.37153077125549316, "learning_rate": 7.354513563457855e-05, "loss": 0.038, "step": 1460 }, { "epoch": 9.483870967741936, "grad_norm": 0.5790444016456604, "learning_rate": 7.316782706538056e-05, "loss": 0.0408, "step": 1470 }, { "epoch": 9.548387096774194, "grad_norm": 0.3680723309516907, "learning_rate": 7.278883096900017e-05, "loss": 0.0402, "step": 1480 }, { "epoch": 9.612903225806452, "grad_norm": 0.40202105045318604, "learning_rate": 7.240817495122935e-05, "loss": 0.0376, "step": 1490 }, { "epoch": 9.67741935483871, "grad_norm": 0.24831123650074005, "learning_rate": 7.202588673876751e-05, "loss": 0.0371, "step": 1500 }, { "epoch": 9.741935483870968, "grad_norm": 0.4128989279270172, "learning_rate": 7.16419941772019e-05, "loss": 0.0337, "step": 1510 }, { "epoch": 9.806451612903226, "grad_norm": 0.2867458164691925, "learning_rate": 7.125652522897936e-05, "loss": 0.0383, "step": 1520 }, { "epoch": 9.870967741935484, "grad_norm": 0.32444578409194946, "learning_rate": 7.086950797136959e-05, "loss": 0.0425, "step": 1530 }, { "epoch": 9.935483870967742, "grad_norm": 0.8138065338134766, "learning_rate": 7.04809705944199e-05, "loss": 0.0422, "step": 1540 }, { "epoch": 10.0, "grad_norm": 0.5340340733528137, "learning_rate": 7.009094139890206e-05, "loss": 0.0386, "step": 1550 }, { "epoch": 10.064516129032258, "grad_norm": 0.3925679922103882, "learning_rate": 6.969944879425069e-05, "loss": 0.0375, "step": 1560 }, { "epoch": 10.129032258064516, "grad_norm": 0.32105499505996704, "learning_rate": 6.930652129649411e-05, "loss": 0.0412, "step": 1570 }, { "epoch": 10.193548387096774, "grad_norm": 0.31760817766189575, "learning_rate": 6.891218752617716e-05, "loss": 0.034, "step": 1580 }, { "epoch": 10.258064516129032, "grad_norm": 0.36178886890411377, "learning_rate": 6.851647620627648e-05, "loss": 0.0338, "step": 1590 }, { "epoch": 10.32258064516129, "grad_norm": 0.29000788927078247, "learning_rate": 6.811941616010847e-05, "loss": 0.0379, "step": 1600 }, { "epoch": 10.387096774193548, "grad_norm": 0.40512529015541077, "learning_rate": 6.772103630922964e-05, "loss": 0.0399, "step": 1610 }, { "epoch": 10.451612903225806, "grad_norm": 0.3559243083000183, "learning_rate": 6.732136567133019e-05, "loss": 0.0349, "step": 1620 }, { "epoch": 10.516129032258064, "grad_norm": 0.5191022157669067, "learning_rate": 6.69204333581202e-05, "loss": 0.0342, "step": 1630 }, { "epoch": 10.580645161290322, "grad_norm": 0.21767917275428772, "learning_rate": 6.651826857320926e-05, "loss": 0.0386, "step": 1640 }, { "epoch": 10.64516129032258, "grad_norm": 0.39933088421821594, "learning_rate": 6.611490060997933e-05, "loss": 0.0375, "step": 1650 }, { "epoch": 10.709677419354838, "grad_norm": 0.32508501410484314, "learning_rate": 6.57103588494509e-05, "loss": 0.0348, "step": 1660 }, { "epoch": 10.774193548387096, "grad_norm": 0.6718910932540894, "learning_rate": 6.530467275814301e-05, "loss": 0.0405, "step": 1670 }, { "epoch": 10.838709677419354, "grad_norm": 0.46869444847106934, "learning_rate": 6.489787188592693e-05, "loss": 0.035, "step": 1680 }, { "epoch": 10.903225806451612, "grad_norm": 0.39346209168434143, "learning_rate": 6.448998586387365e-05, "loss": 0.0344, "step": 1690 }, { "epoch": 10.967741935483872, "grad_norm": 0.4559718370437622, "learning_rate": 6.408104440209574e-05, "loss": 0.038, "step": 1700 }, { "epoch": 11.03225806451613, "grad_norm": 0.7514871954917908, "learning_rate": 6.367107728758319e-05, "loss": 0.0339, "step": 1710 }, { "epoch": 11.096774193548388, "grad_norm": 0.2884572148323059, "learning_rate": 6.326011438203378e-05, "loss": 0.0362, "step": 1720 }, { "epoch": 11.161290322580646, "grad_norm": 0.3102618157863617, "learning_rate": 6.28481856196779e-05, "loss": 0.0326, "step": 1730 }, { "epoch": 11.225806451612904, "grad_norm": 0.25867921113967896, "learning_rate": 6.24353210050983e-05, "loss": 0.0398, "step": 1740 }, { "epoch": 11.290322580645162, "grad_norm": 0.28058162331581116, "learning_rate": 6.202155061104448e-05, "loss": 0.0362, "step": 1750 }, { "epoch": 11.35483870967742, "grad_norm": 0.360380619764328, "learning_rate": 6.160690457624223e-05, "loss": 0.0374, "step": 1760 }, { "epoch": 11.419354838709678, "grad_norm": 0.40499451756477356, "learning_rate": 6.119141310319838e-05, "loss": 0.0355, "step": 1770 }, { "epoch": 11.483870967741936, "grad_norm": 0.3203999996185303, "learning_rate": 6.077510645600081e-05, "loss": 0.0349, "step": 1780 }, { "epoch": 11.548387096774194, "grad_norm": 0.2518363296985626, "learning_rate": 6.035801495811411e-05, "loss": 0.0356, "step": 1790 }, { "epoch": 11.612903225806452, "grad_norm": 0.4911201000213623, "learning_rate": 5.99401689901708e-05, "loss": 0.0331, "step": 1800 }, { "epoch": 11.67741935483871, "grad_norm": 0.30413028597831726, "learning_rate": 5.9521598987758395e-05, "loss": 0.0333, "step": 1810 }, { "epoch": 11.741935483870968, "grad_norm": 0.37464162707328796, "learning_rate": 5.910233543920263e-05, "loss": 0.0336, "step": 1820 }, { "epoch": 11.806451612903226, "grad_norm": 0.352731853723526, "learning_rate": 5.868240888334653e-05, "loss": 0.0327, "step": 1830 }, { "epoch": 11.870967741935484, "grad_norm": 0.4417053461074829, "learning_rate": 5.826184990732609e-05, "loss": 0.0327, "step": 1840 }, { "epoch": 11.935483870967742, "grad_norm": 0.3826858401298523, "learning_rate": 5.784068914434238e-05, "loss": 0.0333, "step": 1850 }, { "epoch": 12.0, "grad_norm": 0.4200211763381958, "learning_rate": 5.741895727143013e-05, "loss": 0.0368, "step": 1860 }, { "epoch": 12.064516129032258, "grad_norm": 0.22449392080307007, "learning_rate": 5.699668500722328e-05, "loss": 0.0326, "step": 1870 }, { "epoch": 12.129032258064516, "grad_norm": 0.2556811571121216, "learning_rate": 5.65739031097175e-05, "loss": 0.0305, "step": 1880 }, { "epoch": 12.193548387096774, "grad_norm": 0.34918874502182007, "learning_rate": 5.615064237402974e-05, "loss": 0.0294, "step": 1890 }, { "epoch": 12.258064516129032, "grad_norm": 0.3033507764339447, "learning_rate": 5.5726933630155196e-05, "loss": 0.0311, "step": 1900 }, { "epoch": 12.32258064516129, "grad_norm": 0.2633187770843506, "learning_rate": 5.5302807740721606e-05, "loss": 0.0369, "step": 1910 }, { "epoch": 12.387096774193548, "grad_norm": 0.33521905541419983, "learning_rate": 5.4878295598741325e-05, "loss": 0.0299, "step": 1920 }, { "epoch": 12.451612903225806, "grad_norm": 0.28460752964019775, "learning_rate": 5.4453428125361e-05, "loss": 0.0297, "step": 1930 }, { "epoch": 12.516129032258064, "grad_norm": 0.24392494559288025, "learning_rate": 5.40282362676094e-05, "loss": 0.0338, "step": 1940 }, { "epoch": 12.580645161290322, "grad_norm": 0.24771785736083984, "learning_rate": 5.360275099614319e-05, "loss": 0.0327, "step": 1950 }, { "epoch": 12.64516129032258, "grad_norm": 0.41986414790153503, "learning_rate": 5.317700330299107e-05, "loss": 0.0273, "step": 1960 }, { "epoch": 12.709677419354838, "grad_norm": 0.3345301151275635, "learning_rate": 5.275102419929632e-05, "loss": 0.0329, "step": 1970 }, { "epoch": 12.774193548387096, "grad_norm": 0.35422471165657043, "learning_rate": 5.232484471305804e-05, "loss": 0.035, "step": 1980 }, { "epoch": 12.838709677419354, "grad_norm": 0.2817692160606384, "learning_rate": 5.189849588687099e-05, "loss": 0.0288, "step": 1990 }, { "epoch": 12.903225806451612, "grad_norm": 0.4547789394855499, "learning_rate": 5.147200877566456e-05, "loss": 0.0347, "step": 2000 }, { "epoch": 12.967741935483872, "grad_norm": 0.2539784908294678, "learning_rate": 5.104541444444071e-05, "loss": 0.0352, "step": 2010 }, { "epoch": 13.03225806451613, "grad_norm": 0.2622417211532593, "learning_rate": 5.061874396601121e-05, "loss": 0.0296, "step": 2020 }, { "epoch": 13.096774193548388, "grad_norm": 0.3102482259273529, "learning_rate": 5.0192028418734335e-05, "loss": 0.0313, "step": 2030 }, { "epoch": 13.161290322580646, "grad_norm": 0.25369301438331604, "learning_rate": 4.9765298884251186e-05, "loss": 0.0309, "step": 2040 }, { "epoch": 13.225806451612904, "grad_norm": 0.3013876974582672, "learning_rate": 4.933858644522165e-05, "loss": 0.0372, "step": 2050 }, { "epoch": 13.290322580645162, "grad_norm": 0.28694114089012146, "learning_rate": 4.8911922183060356e-05, "loss": 0.0295, "step": 2060 }, { "epoch": 13.35483870967742, "grad_norm": 0.4296466112136841, "learning_rate": 4.848533717567283e-05, "loss": 0.0305, "step": 2070 }, { "epoch": 13.419354838709678, "grad_norm": 0.2966064512729645, "learning_rate": 4.805886249519172e-05, "loss": 0.0354, "step": 2080 }, { "epoch": 13.483870967741936, "grad_norm": 0.25592073798179626, "learning_rate": 4.763252920571351e-05, "loss": 0.0332, "step": 2090 }, { "epoch": 13.548387096774194, "grad_norm": 0.42708101868629456, "learning_rate": 4.720636836103593e-05, "loss": 0.0362, "step": 2100 }, { "epoch": 13.612903225806452, "grad_norm": 0.44891414046287537, "learning_rate": 4.6780411002395904e-05, "loss": 0.0312, "step": 2110 }, { "epoch": 13.67741935483871, "grad_norm": 0.33705928921699524, "learning_rate": 4.6354688156208616e-05, "loss": 0.0322, "step": 2120 }, { "epoch": 13.741935483870968, "grad_norm": 0.34952977299690247, "learning_rate": 4.592923083180752e-05, "loss": 0.0297, "step": 2130 }, { "epoch": 13.806451612903226, "grad_norm": 0.2684139013290405, "learning_rate": 4.55040700191857e-05, "loss": 0.0334, "step": 2140 }, { "epoch": 13.870967741935484, "grad_norm": 0.3220863342285156, "learning_rate": 4.507923668673847e-05, "loss": 0.0312, "step": 2150 }, { "epoch": 13.935483870967742, "grad_norm": 0.31924524903297424, "learning_rate": 4.465476177900782e-05, "loss": 0.0303, "step": 2160 }, { "epoch": 14.0, "grad_norm": 0.5949335694313049, "learning_rate": 4.423067621442831e-05, "loss": 0.0275, "step": 2170 }, { "epoch": 14.064516129032258, "grad_norm": 0.28924089670181274, "learning_rate": 4.380701088307501e-05, "loss": 0.0272, "step": 2180 }, { "epoch": 14.129032258064516, "grad_norm": 0.3126032054424286, "learning_rate": 4.338379664441357e-05, "loss": 0.0305, "step": 2190 }, { "epoch": 14.193548387096774, "grad_norm": 0.2669059634208679, "learning_rate": 4.2961064325052335e-05, "loss": 0.0286, "step": 2200 }, { "epoch": 14.258064516129032, "grad_norm": 0.3986091911792755, "learning_rate": 4.253884471649707e-05, "loss": 0.0305, "step": 2210 }, { "epoch": 14.32258064516129, "grad_norm": 0.397923082113266, "learning_rate": 4.2117168572907996e-05, "loss": 0.0324, "step": 2220 }, { "epoch": 14.387096774193548, "grad_norm": 0.3409479856491089, "learning_rate": 4.169606660885982e-05, "loss": 0.0276, "step": 2230 }, { "epoch": 14.451612903225806, "grad_norm": 0.2611437737941742, "learning_rate": 4.127556949710437e-05, "loss": 0.0283, "step": 2240 }, { "epoch": 14.516129032258064, "grad_norm": 0.3530644178390503, "learning_rate": 4.085570786633656e-05, "loss": 0.0278, "step": 2250 }, { "epoch": 14.580645161290322, "grad_norm": 0.23989705741405487, "learning_rate": 4.0436512298963305e-05, "loss": 0.0318, "step": 2260 }, { "epoch": 14.64516129032258, "grad_norm": 0.2740343511104584, "learning_rate": 4.001801332887595e-05, "loss": 0.0297, "step": 2270 }, { "epoch": 14.709677419354838, "grad_norm": 0.39286455512046814, "learning_rate": 3.960024143922625e-05, "loss": 0.0291, "step": 2280 }, { "epoch": 14.774193548387096, "grad_norm": 0.7122088074684143, "learning_rate": 3.918322706020593e-05, "loss": 0.0368, "step": 2290 }, { "epoch": 14.838709677419354, "grad_norm": 0.22896356880664825, "learning_rate": 3.876700056683026e-05, "loss": 0.027, "step": 2300 }, { "epoch": 14.903225806451612, "grad_norm": 0.3285376727581024, "learning_rate": 3.8351592276725425e-05, "loss": 0.0248, "step": 2310 }, { "epoch": 14.967741935483872, "grad_norm": 0.23626801371574402, "learning_rate": 3.793703244792038e-05, "loss": 0.0295, "step": 2320 }, { "epoch": 15.03225806451613, "grad_norm": 0.39834141731262207, "learning_rate": 3.7523351276642736e-05, "loss": 0.0292, "step": 2330 }, { "epoch": 15.096774193548388, "grad_norm": 0.29628074169158936, "learning_rate": 3.71105788951194e-05, "loss": 0.0275, "step": 2340 }, { "epoch": 15.161290322580646, "grad_norm": 0.18978668749332428, "learning_rate": 3.669874536938164e-05, "loss": 0.0265, "step": 2350 }, { "epoch": 15.225806451612904, "grad_norm": 0.30174311995506287, "learning_rate": 3.628788069707524e-05, "loss": 0.0266, "step": 2360 }, { "epoch": 15.290322580645162, "grad_norm": 0.24498514831066132, "learning_rate": 3.587801480527542e-05, "loss": 0.0259, "step": 2370 }, { "epoch": 15.35483870967742, "grad_norm": 0.3696959316730499, "learning_rate": 3.546917754830691e-05, "loss": 0.032, "step": 2380 }, { "epoch": 15.419354838709678, "grad_norm": 0.31626754999160767, "learning_rate": 3.5061398705569545e-05, "loss": 0.0264, "step": 2390 }, { "epoch": 15.483870967741936, "grad_norm": 0.5776691436767578, "learning_rate": 3.465470797936896e-05, "loss": 0.0244, "step": 2400 }, { "epoch": 15.548387096774194, "grad_norm": 0.25500577688217163, "learning_rate": 3.424913499275328e-05, "loss": 0.0304, "step": 2410 }, { "epoch": 15.612903225806452, "grad_norm": 0.3002663552761078, "learning_rate": 3.384470928735527e-05, "loss": 0.0287, "step": 2420 }, { "epoch": 15.67741935483871, "grad_norm": 0.45792222023010254, "learning_rate": 3.3441460321240605e-05, "loss": 0.0316, "step": 2430 }, { "epoch": 15.741935483870968, "grad_norm": 0.20462553203105927, "learning_rate": 3.303941746676212e-05, "loss": 0.0255, "step": 2440 }, { "epoch": 15.806451612903226, "grad_norm": 0.29612648487091064, "learning_rate": 3.263861000842043e-05, "loss": 0.0292, "step": 2450 }, { "epoch": 15.870967741935484, "grad_norm": 0.3627132475376129, "learning_rate": 3.223906714073083e-05, "loss": 0.0277, "step": 2460 }, { "epoch": 15.935483870967742, "grad_norm": 0.4241921901702881, "learning_rate": 3.1840817966096684e-05, "loss": 0.0323, "step": 2470 }, { "epoch": 16.0, "grad_norm": 0.4466724693775177, "learning_rate": 3.1443891492689834e-05, "loss": 0.0257, "step": 2480 }, { "epoch": 16.06451612903226, "grad_norm": 0.33882275223731995, "learning_rate": 3.104831663233746e-05, "loss": 0.0288, "step": 2490 }, { "epoch": 16.129032258064516, "grad_norm": 0.28266456723213196, "learning_rate": 3.065412219841635e-05, "loss": 0.0266, "step": 2500 }, { "epoch": 16.193548387096776, "grad_norm": 0.2942975163459778, "learning_rate": 3.026133690375398e-05, "loss": 0.026, "step": 2510 }, { "epoch": 16.258064516129032, "grad_norm": 0.28727394342422485, "learning_rate": 2.9869989358537253e-05, "loss": 0.0282, "step": 2520 }, { "epoch": 16.322580645161292, "grad_norm": 0.28704243898391724, "learning_rate": 2.9480108068228397e-05, "loss": 0.0285, "step": 2530 }, { "epoch": 16.387096774193548, "grad_norm": 0.3424932360649109, "learning_rate": 2.9091721431488784e-05, "loss": 0.0298, "step": 2540 }, { "epoch": 16.451612903225808, "grad_norm": 0.4286734461784363, "learning_rate": 2.8704857738110368e-05, "loss": 0.0244, "step": 2550 }, { "epoch": 16.516129032258064, "grad_norm": 0.2728700339794159, "learning_rate": 2.8319545166954963e-05, "loss": 0.0265, "step": 2560 }, { "epoch": 16.580645161290324, "grad_norm": 0.5243811011314392, "learning_rate": 2.793581178390188e-05, "loss": 0.0242, "step": 2570 }, { "epoch": 16.64516129032258, "grad_norm": 0.3646429479122162, "learning_rate": 2.7553685539803488e-05, "loss": 0.027, "step": 2580 }, { "epoch": 16.70967741935484, "grad_norm": 0.24419990181922913, "learning_rate": 2.717319426844942e-05, "loss": 0.0274, "step": 2590 }, { "epoch": 16.774193548387096, "grad_norm": 0.3144078254699707, "learning_rate": 2.679436568453904e-05, "loss": 0.0266, "step": 2600 }, { "epoch": 16.838709677419356, "grad_norm": 0.3204503655433655, "learning_rate": 2.64172273816629e-05, "loss": 0.027, "step": 2610 }, { "epoch": 16.903225806451612, "grad_norm": 0.2552044093608856, "learning_rate": 2.6041806830292647e-05, "loss": 0.0269, "step": 2620 }, { "epoch": 16.967741935483872, "grad_norm": 0.25380566716194153, "learning_rate": 2.5668131375780276e-05, "loss": 0.0231, "step": 2630 }, { "epoch": 17.032258064516128, "grad_norm": 0.3057281970977783, "learning_rate": 2.5296228236366194e-05, "loss": 0.0266, "step": 2640 }, { "epoch": 17.096774193548388, "grad_norm": 0.26944369077682495, "learning_rate": 2.492612450119669e-05, "loss": 0.0264, "step": 2650 }, { "epoch": 17.161290322580644, "grad_norm": 0.31917497515678406, "learning_rate": 2.455784712835084e-05, "loss": 0.0265, "step": 2660 }, { "epoch": 17.225806451612904, "grad_norm": 0.2606102228164673, "learning_rate": 2.4191422942876803e-05, "loss": 0.0238, "step": 2670 }, { "epoch": 17.29032258064516, "grad_norm": 0.25582873821258545, "learning_rate": 2.3826878634838e-05, "loss": 0.024, "step": 2680 }, { "epoch": 17.35483870967742, "grad_norm": 0.25627753138542175, "learning_rate": 2.3464240757368944e-05, "loss": 0.0261, "step": 2690 }, { "epoch": 17.419354838709676, "grad_norm": 0.2993682622909546, "learning_rate": 2.3103535724741233e-05, "loss": 0.0267, "step": 2700 }, { "epoch": 17.483870967741936, "grad_norm": 0.24699701368808746, "learning_rate": 2.2744789810439422e-05, "loss": 0.0246, "step": 2710 }, { "epoch": 17.548387096774192, "grad_norm": 0.24654793739318848, "learning_rate": 2.2388029145247396e-05, "loss": 0.0216, "step": 2720 }, { "epoch": 17.612903225806452, "grad_norm": 0.3724614083766937, "learning_rate": 2.2033279715345006e-05, "loss": 0.0264, "step": 2730 }, { "epoch": 17.677419354838708, "grad_norm": 0.2768930196762085, "learning_rate": 2.1680567360415155e-05, "loss": 0.0264, "step": 2740 }, { "epoch": 17.741935483870968, "grad_norm": 0.1919165402650833, "learning_rate": 2.1329917771761805e-05, "loss": 0.0218, "step": 2750 }, { "epoch": 17.806451612903224, "grad_norm": 0.15743491053581238, "learning_rate": 2.0981356490438502e-05, "loss": 0.0213, "step": 2760 }, { "epoch": 17.870967741935484, "grad_norm": 0.3243708312511444, "learning_rate": 2.0634908905388123e-05, "loss": 0.0221, "step": 2770 }, { "epoch": 17.93548387096774, "grad_norm": 1.23497474193573, "learning_rate": 2.02906002515934e-05, "loss": 0.0252, "step": 2780 }, { "epoch": 18.0, "grad_norm": 0.3177509903907776, "learning_rate": 1.9948455608238985e-05, "loss": 0.0272, "step": 2790 }, { "epoch": 18.06451612903226, "grad_norm": 0.29627230763435364, "learning_rate": 1.960849989688463e-05, "loss": 0.0238, "step": 2800 }, { "epoch": 18.129032258064516, "grad_norm": 0.47126612067222595, "learning_rate": 1.927075787964985e-05, "loss": 0.0226, "step": 2810 }, { "epoch": 18.193548387096776, "grad_norm": 0.22074201703071594, "learning_rate": 1.8935254157410403e-05, "loss": 0.0227, "step": 2820 }, { "epoch": 18.258064516129032, "grad_norm": 0.23865483701229095, "learning_rate": 1.8602013168006293e-05, "loss": 0.0232, "step": 2830 }, { "epoch": 18.322580645161292, "grad_norm": 0.24160803854465485, "learning_rate": 1.827105918446178e-05, "loss": 0.0226, "step": 2840 }, { "epoch": 18.387096774193548, "grad_norm": 0.29074233770370483, "learning_rate": 1.794241631321727e-05, "loss": 0.0215, "step": 2850 }, { "epoch": 18.451612903225808, "grad_norm": 0.2850009799003601, "learning_rate": 1.7616108492373545e-05, "loss": 0.0209, "step": 2860 }, { "epoch": 18.516129032258064, "grad_norm": 0.39563989639282227, "learning_rate": 1.7292159489948012e-05, "loss": 0.0234, "step": 2870 }, { "epoch": 18.580645161290324, "grad_norm": 0.2690832316875458, "learning_rate": 1.697059290214356e-05, "loss": 0.0251, "step": 2880 }, { "epoch": 18.64516129032258, "grad_norm": 0.21451742947101593, "learning_rate": 1.6651432151629725e-05, "loss": 0.0251, "step": 2890 }, { "epoch": 18.70967741935484, "grad_norm": 0.5447711944580078, "learning_rate": 1.6334700485836714e-05, "loss": 0.0241, "step": 2900 }, { "epoch": 18.774193548387096, "grad_norm": 0.26067161560058594, "learning_rate": 1.6020420975261956e-05, "loss": 0.0279, "step": 2910 }, { "epoch": 18.838709677419356, "grad_norm": 0.3259269595146179, "learning_rate": 1.570861651178978e-05, "loss": 0.0219, "step": 2920 }, { "epoch": 18.903225806451612, "grad_norm": 0.2738400399684906, "learning_rate": 1.5399309807023944e-05, "loss": 0.0205, "step": 2930 }, { "epoch": 18.967741935483872, "grad_norm": 0.27828454971313477, "learning_rate": 1.5092523390633267e-05, "loss": 0.0226, "step": 2940 }, { "epoch": 19.032258064516128, "grad_norm": 0.23453515768051147, "learning_rate": 1.4788279608710732e-05, "loss": 0.025, "step": 2950 }, { "epoch": 19.096774193548388, "grad_norm": 0.3187068998813629, "learning_rate": 1.4486600622145652e-05, "loss": 0.0242, "step": 2960 }, { "epoch": 19.161290322580644, "grad_norm": 0.29299116134643555, "learning_rate": 1.4187508405009637e-05, "loss": 0.0216, "step": 2970 }, { "epoch": 19.225806451612904, "grad_norm": 0.25096479058265686, "learning_rate": 1.3891024742955877e-05, "loss": 0.0217, "step": 2980 }, { "epoch": 19.29032258064516, "grad_norm": 0.34088948369026184, "learning_rate": 1.3597171231632433e-05, "loss": 0.0233, "step": 2990 }, { "epoch": 19.35483870967742, "grad_norm": 0.255126416683197, "learning_rate": 1.3305969275109098e-05, "loss": 0.023, "step": 3000 }, { "epoch": 19.419354838709676, "grad_norm": 0.2508387863636017, "learning_rate": 1.3017440084318421e-05, "loss": 0.0205, "step": 3010 }, { "epoch": 19.483870967741936, "grad_norm": 0.21891236305236816, "learning_rate": 1.2731604675510728e-05, "loss": 0.0221, "step": 3020 }, { "epoch": 19.548387096774192, "grad_norm": 0.3564584255218506, "learning_rate": 1.2448483868723216e-05, "loss": 0.0243, "step": 3030 }, { "epoch": 19.612903225806452, "grad_norm": 0.2741142809391022, "learning_rate": 1.2168098286263557e-05, "loss": 0.022, "step": 3040 }, { "epoch": 19.677419354838708, "grad_norm": 0.40683725476264954, "learning_rate": 1.189046835120769e-05, "loss": 0.0265, "step": 3050 }, { "epoch": 19.741935483870968, "grad_norm": 0.3363529145717621, "learning_rate": 1.1615614285912301e-05, "loss": 0.0255, "step": 3060 }, { "epoch": 19.806451612903224, "grad_norm": 0.17003341019153595, "learning_rate": 1.134355611054176e-05, "loss": 0.0214, "step": 3070 }, { "epoch": 19.870967741935484, "grad_norm": 0.27854353189468384, "learning_rate": 1.1074313641609935e-05, "loss": 0.023, "step": 3080 }, { "epoch": 19.93548387096774, "grad_norm": 0.3035428524017334, "learning_rate": 1.0807906490536751e-05, "loss": 0.0233, "step": 3090 }, { "epoch": 20.0, "grad_norm": 0.3595190644264221, "learning_rate": 1.0544354062219658e-05, "loss": 0.0228, "step": 3100 }, { "epoch": 20.06451612903226, "grad_norm": 0.2843579649925232, "learning_rate": 1.0283675553620281e-05, "loss": 0.0192, "step": 3110 }, { "epoch": 20.129032258064516, "grad_norm": 0.4682447910308838, "learning_rate": 1.0025889952366085e-05, "loss": 0.0243, "step": 3120 }, { "epoch": 20.193548387096776, "grad_norm": 0.1721201092004776, "learning_rate": 9.77101603536728e-06, "loss": 0.0234, "step": 3130 }, { "epoch": 20.258064516129032, "grad_norm": 0.2416217029094696, "learning_rate": 9.519072367449239e-06, "loss": 0.0189, "step": 3140 }, { "epoch": 20.322580645161292, "grad_norm": 0.28073132038116455, "learning_rate": 9.270077300000145e-06, "loss": 0.021, "step": 3150 }, { "epoch": 20.387096774193548, "grad_norm": 0.2682770788669586, "learning_rate": 9.02404896963433e-06, "loss": 0.0201, "step": 3160 }, { "epoch": 20.451612903225808, "grad_norm": 0.29497283697128296, "learning_rate": 8.781005296871248e-06, "loss": 0.0305, "step": 3170 }, { "epoch": 20.516129032258064, "grad_norm": 0.2409420758485794, "learning_rate": 8.540963984830075e-06, "loss": 0.0206, "step": 3180 }, { "epoch": 20.580645161290324, "grad_norm": 0.23492980003356934, "learning_rate": 8.303942517940327e-06, "loss": 0.023, "step": 3190 }, { "epoch": 20.64516129032258, "grad_norm": 0.18785379827022552, "learning_rate": 8.069958160668256e-06, "loss": 0.0217, "step": 3200 }, { "epoch": 20.70967741935484, "grad_norm": 0.2183133363723755, "learning_rate": 7.839027956259282e-06, "loss": 0.0213, "step": 3210 }, { "epoch": 20.774193548387096, "grad_norm": 0.20956051349639893, "learning_rate": 7.6111687254966705e-06, "loss": 0.0188, "step": 3220 }, { "epoch": 20.838709677419356, "grad_norm": 0.2115253210067749, "learning_rate": 7.386397065476209e-06, "loss": 0.0219, "step": 3230 }, { "epoch": 20.903225806451612, "grad_norm": 0.27826544642448425, "learning_rate": 7.164729348397409e-06, "loss": 0.021, "step": 3240 }, { "epoch": 20.967741935483872, "grad_norm": 0.8272599577903748, "learning_rate": 6.946181720370825e-06, "loss": 0.0197, "step": 3250 }, { "epoch": 21.032258064516128, "grad_norm": 0.31838661432266235, "learning_rate": 6.73077010024214e-06, "loss": 0.025, "step": 3260 }, { "epoch": 21.096774193548388, "grad_norm": 0.32917433977127075, "learning_rate": 6.518510178432491e-06, "loss": 0.0209, "step": 3270 }, { "epoch": 21.161290322580644, "grad_norm": 0.21181564033031464, "learning_rate": 6.309417415795732e-06, "loss": 0.0203, "step": 3280 }, { "epoch": 21.225806451612904, "grad_norm": 0.13702601194381714, "learning_rate": 6.1035070424921905e-06, "loss": 0.0238, "step": 3290 }, { "epoch": 21.29032258064516, "grad_norm": 0.5193454027175903, "learning_rate": 5.900794056879317e-06, "loss": 0.021, "step": 3300 }, { "epoch": 21.35483870967742, "grad_norm": 0.2874934673309326, "learning_rate": 5.7012932244192745e-06, "loss": 0.0202, "step": 3310 }, { "epoch": 21.419354838709676, "grad_norm": 0.20375904440879822, "learning_rate": 5.505019076603352e-06, "loss": 0.0206, "step": 3320 }, { "epoch": 21.483870967741936, "grad_norm": 0.25239813327789307, "learning_rate": 5.311985909893591e-06, "loss": 0.0208, "step": 3330 }, { "epoch": 21.548387096774192, "grad_norm": 0.1695331186056137, "learning_rate": 5.122207784681337e-06, "loss": 0.0241, "step": 3340 }, { "epoch": 21.612903225806452, "grad_norm": 0.2161603420972824, "learning_rate": 4.935698524263216e-06, "loss": 0.0214, "step": 3350 }, { "epoch": 21.677419354838708, "grad_norm": 0.3184122145175934, "learning_rate": 4.7524717138341335e-06, "loss": 0.0205, "step": 3360 }, { "epoch": 21.741935483870968, "grad_norm": 0.24667605757713318, "learning_rate": 4.5725406994978235e-06, "loss": 0.0182, "step": 3370 }, { "epoch": 21.806451612903224, "grad_norm": 0.21050481498241425, "learning_rate": 4.395918587294701e-06, "loss": 0.0223, "step": 3380 }, { "epoch": 21.870967741935484, "grad_norm": 0.31748640537261963, "learning_rate": 4.222618242247195e-06, "loss": 0.0164, "step": 3390 }, { "epoch": 21.93548387096774, "grad_norm": 0.23009532690048218, "learning_rate": 4.052652287422754e-06, "loss": 0.0223, "step": 3400 }, { "epoch": 22.0, "grad_norm": 0.46855324506759644, "learning_rate": 3.886033103014286e-06, "loss": 0.0228, "step": 3410 }, { "epoch": 22.06451612903226, "grad_norm": 0.3878401219844818, "learning_rate": 3.7227728254384752e-06, "loss": 0.0214, "step": 3420 }, { "epoch": 22.129032258064516, "grad_norm": 0.26421085000038147, "learning_rate": 3.5628833464517398e-06, "loss": 0.0214, "step": 3430 }, { "epoch": 22.193548387096776, "grad_norm": 0.2137315571308136, "learning_rate": 3.4063763122840654e-06, "loss": 0.0185, "step": 3440 }, { "epoch": 22.258064516129032, "grad_norm": 0.20358504354953766, "learning_rate": 3.2532631227906562e-06, "loss": 0.0205, "step": 3450 }, { "epoch": 22.322580645161292, "grad_norm": 0.2786470055580139, "learning_rate": 3.103554930621644e-06, "loss": 0.0225, "step": 3460 }, { "epoch": 22.387096774193548, "grad_norm": 0.25646957755088806, "learning_rate": 2.9572626404096917e-06, "loss": 0.0191, "step": 3470 }, { "epoch": 22.451612903225808, "grad_norm": 0.15664377808570862, "learning_rate": 2.8143969079757083e-06, "loss": 0.02, "step": 3480 }, { "epoch": 22.516129032258064, "grad_norm": 0.21855230629444122, "learning_rate": 2.6749681395527194e-06, "loss": 0.0187, "step": 3490 }, { "epoch": 22.580645161290324, "grad_norm": 0.15705305337905884, "learning_rate": 2.538986491027845e-06, "loss": 0.0198, "step": 3500 }, { "epoch": 22.64516129032258, "grad_norm": 0.3838616609573364, "learning_rate": 2.4064618672025917e-06, "loss": 0.0203, "step": 3510 }, { "epoch": 22.70967741935484, "grad_norm": 0.1712443083524704, "learning_rate": 2.2774039210713584e-06, "loss": 0.0205, "step": 3520 }, { "epoch": 22.774193548387096, "grad_norm": 0.3191426396369934, "learning_rate": 2.151822053118352e-06, "loss": 0.0204, "step": 3530 }, { "epoch": 22.838709677419356, "grad_norm": 0.21879324316978455, "learning_rate": 2.029725410632821e-06, "loss": 0.019, "step": 3540 }, { "epoch": 22.903225806451612, "grad_norm": 0.25989222526550293, "learning_rate": 1.9111228870428267e-06, "loss": 0.019, "step": 3550 }, { "epoch": 22.967741935483872, "grad_norm": 0.22728225588798523, "learning_rate": 1.7960231212674095e-06, "loss": 0.019, "step": 3560 }, { "epoch": 23.032258064516128, "grad_norm": 0.3381948173046112, "learning_rate": 1.684434497087356e-06, "loss": 0.0204, "step": 3570 }, { "epoch": 23.096774193548388, "grad_norm": 0.3168867230415344, "learning_rate": 1.5763651425345306e-06, "loss": 0.0173, "step": 3580 }, { "epoch": 23.161290322580644, "grad_norm": 0.4736108183860779, "learning_rate": 1.4718229292998265e-06, "loss": 0.0193, "step": 3590 }, { "epoch": 23.225806451612904, "grad_norm": 0.1927083134651184, "learning_rate": 1.3708154721598142e-06, "loss": 0.0162, "step": 3600 }, { "epoch": 23.29032258064516, "grad_norm": 0.26042094826698303, "learning_rate": 1.2733501284220573e-06, "loss": 0.019, "step": 3610 }, { "epoch": 23.35483870967742, "grad_norm": 0.3572435975074768, "learning_rate": 1.1794339973892509e-06, "loss": 0.0187, "step": 3620 }, { "epoch": 23.419354838709676, "grad_norm": 0.31472429633140564, "learning_rate": 1.0890739198420763e-06, "loss": 0.0223, "step": 3630 }, { "epoch": 23.483870967741936, "grad_norm": 0.28377604484558105, "learning_rate": 1.0022764775409532e-06, "loss": 0.0187, "step": 3640 }, { "epoch": 23.548387096774192, "grad_norm": 0.2872464656829834, "learning_rate": 9.190479927466023e-07, "loss": 0.0195, "step": 3650 }, { "epoch": 23.612903225806452, "grad_norm": 0.2037927806377411, "learning_rate": 8.393945277595682e-07, "loss": 0.0209, "step": 3660 }, { "epoch": 23.677419354838708, "grad_norm": 0.21536409854888916, "learning_rate": 7.633218844786238e-07, "loss": 0.0209, "step": 3670 }, { "epoch": 23.741935483870968, "grad_norm": 0.19614161550998688, "learning_rate": 6.908356039781516e-07, "loss": 0.0188, "step": 3680 }, { "epoch": 23.806451612903224, "grad_norm": 0.2418230175971985, "learning_rate": 6.219409661045839e-07, "loss": 0.0209, "step": 3690 }, { "epoch": 23.870967741935484, "grad_norm": 0.41142016649246216, "learning_rate": 5.566429890917601e-07, "loss": 0.0201, "step": 3700 }, { "epoch": 23.93548387096774, "grad_norm": 0.22285668551921844, "learning_rate": 4.949464291954642e-07, "loss": 0.0183, "step": 3710 }, { "epoch": 24.0, "grad_norm": 0.27075517177581787, "learning_rate": 4.368557803469342e-07, "loss": 0.0185, "step": 3720 }, { "epoch": 24.06451612903226, "grad_norm": 0.31757959723472595, "learning_rate": 3.823752738255471e-07, "loss": 0.022, "step": 3730 }, { "epoch": 24.129032258064516, "grad_norm": 0.18830479681491852, "learning_rate": 3.315088779506259e-07, "loss": 0.0198, "step": 3740 }, { "epoch": 24.193548387096776, "grad_norm": 0.5040279626846313, "learning_rate": 2.8426029779236586e-07, "loss": 0.0241, "step": 3750 }, { "epoch": 24.258064516129032, "grad_norm": 0.7920020222663879, "learning_rate": 2.406329749019942e-07, "loss": 0.0213, "step": 3760 }, { "epoch": 24.322580645161292, "grad_norm": 0.18160688877105713, "learning_rate": 2.0063008706103782e-07, "loss": 0.0186, "step": 3770 }, { "epoch": 24.387096774193548, "grad_norm": 0.29395878314971924, "learning_rate": 1.6425454804991913e-07, "loss": 0.0198, "step": 3780 }, { "epoch": 24.451612903225808, "grad_norm": 0.2948833107948303, "learning_rate": 1.3150900743566509e-07, "loss": 0.02, "step": 3790 }, { "epoch": 24.516129032258064, "grad_norm": 0.40248140692710876, "learning_rate": 1.0239585037894462e-07, "loss": 0.0179, "step": 3800 }, { "epoch": 24.580645161290324, "grad_norm": 0.2159591168165207, "learning_rate": 7.691719746033e-08, "loss": 0.0151, "step": 3810 }, { "epoch": 24.64516129032258, "grad_norm": 0.3262782692909241, "learning_rate": 5.507490452582587e-08, "loss": 0.0214, "step": 3820 }, { "epoch": 24.70967741935484, "grad_norm": 0.47080403566360474, "learning_rate": 3.687056255169963e-08, "loss": 0.0234, "step": 3830 }, { "epoch": 24.774193548387096, "grad_norm": 0.5033960342407227, "learning_rate": 2.2305497528601892e-08, "loss": 0.023, "step": 3840 }, { "epoch": 24.838709677419356, "grad_norm": 0.3430034816265106, "learning_rate": 1.1380770364960436e-08, "loss": 0.0234, "step": 3850 }, { "epoch": 24.903225806451612, "grad_norm": 0.23608365654945374, "learning_rate": 4.0971768097253224e-09, "loss": 0.0199, "step": 3860 }, { "epoch": 24.967741935483872, "grad_norm": 0.6302427649497986, "learning_rate": 4.55247394404168e-10, "loss": 0.0207, "step": 3870 }, { "epoch": 25.0, "step": 3875, "total_flos": 4.172300356853376e+17, "train_loss": 0.051802249112436846, "train_runtime": 4196.5277, "train_samples_per_second": 45.103, "train_steps_per_second": 0.923 } ], "logging_steps": 10, "max_steps": 3875, "num_input_tokens_seen": 0, "num_train_epochs": 25, "save_steps": 10000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 4.172300356853376e+17, "train_batch_size": 49, "trial_name": null, "trial_params": null }