{ "best_metric": 1.601110577583313, "best_model_checkpoint": "outputs/checkpoint-13300", "epoch": 2.999832822513235, "eval_steps": 100, "global_step": 13458, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.022290331568682084, "grad_norm": 820153.75, "learning_rate": 6e-06, "loss": 3.4337, "step": 100 }, { "epoch": 0.022290331568682084, "eval_loss": 2.219454050064087, "eval_runtime": 63.1076, "eval_samples_per_second": 211.829, "eval_steps_per_second": 13.247, "step": 100 }, { "epoch": 0.04458066313736417, "grad_norm": 595487.125, "learning_rate": 1.2e-05, "loss": 2.4245, "step": 200 }, { "epoch": 0.04458066313736417, "eval_loss": 1.938194990158081, "eval_runtime": 63.043, "eval_samples_per_second": 212.046, "eval_steps_per_second": 13.261, "step": 200 }, { "epoch": 0.06687099470604625, "grad_norm": 613520.625, "learning_rate": 1.8e-05, "loss": 2.2722, "step": 300 }, { "epoch": 0.06687099470604625, "eval_loss": 1.8768234252929688, "eval_runtime": 63.5969, "eval_samples_per_second": 210.199, "eval_steps_per_second": 13.145, "step": 300 }, { "epoch": 0.08916132627472834, "grad_norm": 566655.875, "learning_rate": 2.4e-05, "loss": 2.1739, "step": 400 }, { "epoch": 0.08916132627472834, "eval_loss": 1.8509334325790405, "eval_runtime": 63.6581, "eval_samples_per_second": 209.997, "eval_steps_per_second": 13.133, "step": 400 }, { "epoch": 0.11145165784341042, "grad_norm": 618861.25, "learning_rate": 3e-05, "loss": 2.1388, "step": 500 }, { "epoch": 0.11145165784341042, "eval_loss": 1.8238554000854492, "eval_runtime": 63.0746, "eval_samples_per_second": 211.939, "eval_steps_per_second": 13.254, "step": 500 }, { "epoch": 0.1337419894120925, "grad_norm": 594465.5, "learning_rate": 2.9768482790554097e-05, "loss": 2.1116, "step": 600 }, { "epoch": 0.1337419894120925, "eval_loss": 1.813140869140625, "eval_runtime": 63.2719, "eval_samples_per_second": 211.279, "eval_steps_per_second": 13.213, "step": 600 }, { "epoch": 0.1560323209807746, "grad_norm": 528193.5625, "learning_rate": 2.9536965581108196e-05, "loss": 2.0772, "step": 700 }, { "epoch": 0.1560323209807746, "eval_loss": 1.7901383638381958, "eval_runtime": 63.0937, "eval_samples_per_second": 211.876, "eval_steps_per_second": 13.25, "step": 700 }, { "epoch": 0.17832265254945667, "grad_norm": 500452.375, "learning_rate": 2.9305448371662296e-05, "loss": 2.0526, "step": 800 }, { "epoch": 0.17832265254945667, "eval_loss": 1.7764878273010254, "eval_runtime": 63.8628, "eval_samples_per_second": 209.324, "eval_steps_per_second": 13.091, "step": 800 }, { "epoch": 0.20061298411813877, "grad_norm": 492339.875, "learning_rate": 2.9073931162216392e-05, "loss": 2.0419, "step": 900 }, { "epoch": 0.20061298411813877, "eval_loss": 1.7795604467391968, "eval_runtime": 63.1552, "eval_samples_per_second": 211.669, "eval_steps_per_second": 13.237, "step": 900 }, { "epoch": 0.22290331568682084, "grad_norm": 549590.1875, "learning_rate": 2.884241395277049e-05, "loss": 2.0305, "step": 1000 }, { "epoch": 0.22290331568682084, "eval_loss": 1.7810730934143066, "eval_runtime": 62.8059, "eval_samples_per_second": 212.846, "eval_steps_per_second": 13.311, "step": 1000 }, { "epoch": 0.24519364725550294, "grad_norm": 463470.46875, "learning_rate": 2.8610896743324588e-05, "loss": 2.0123, "step": 1100 }, { "epoch": 0.24519364725550294, "eval_loss": 1.739020586013794, "eval_runtime": 63.3157, "eval_samples_per_second": 211.133, "eval_steps_per_second": 13.204, "step": 1100 }, { "epoch": 0.267483978824185, "grad_norm": 483440.3125, "learning_rate": 2.8379379533878684e-05, "loss": 2.0051, "step": 1200 }, { "epoch": 0.267483978824185, "eval_loss": 1.754070520401001, "eval_runtime": 63.6661, "eval_samples_per_second": 209.97, "eval_steps_per_second": 13.131, "step": 1200 }, { "epoch": 0.2897743103928671, "grad_norm": 515932.6875, "learning_rate": 2.8147862324432783e-05, "loss": 2.0018, "step": 1300 }, { "epoch": 0.2897743103928671, "eval_loss": 1.7487035989761353, "eval_runtime": 63.3358, "eval_samples_per_second": 211.065, "eval_steps_per_second": 13.199, "step": 1300 }, { "epoch": 0.3120646419615492, "grad_norm": 510042.875, "learning_rate": 2.7916345114986883e-05, "loss": 1.9878, "step": 1400 }, { "epoch": 0.3120646419615492, "eval_loss": 1.7291030883789062, "eval_runtime": 63.1739, "eval_samples_per_second": 211.607, "eval_steps_per_second": 13.233, "step": 1400 }, { "epoch": 0.33435497353023125, "grad_norm": 508744.25, "learning_rate": 2.768482790554098e-05, "loss": 1.9885, "step": 1500 }, { "epoch": 0.33435497353023125, "eval_loss": 1.7247428894042969, "eval_runtime": 63.3733, "eval_samples_per_second": 210.941, "eval_steps_per_second": 13.192, "step": 1500 }, { "epoch": 0.35664530509891335, "grad_norm": 512775.4375, "learning_rate": 2.7453310696095075e-05, "loss": 1.9864, "step": 1600 }, { "epoch": 0.35664530509891335, "eval_loss": 1.7243812084197998, "eval_runtime": 63.6899, "eval_samples_per_second": 209.892, "eval_steps_per_second": 13.126, "step": 1600 }, { "epoch": 0.37893563666759544, "grad_norm": 480830.6875, "learning_rate": 2.7221793486649175e-05, "loss": 1.9735, "step": 1700 }, { "epoch": 0.37893563666759544, "eval_loss": 1.7244142293930054, "eval_runtime": 63.3501, "eval_samples_per_second": 211.018, "eval_steps_per_second": 13.196, "step": 1700 }, { "epoch": 0.40122596823627754, "grad_norm": 554714.125, "learning_rate": 2.6990276277203274e-05, "loss": 1.9756, "step": 1800 }, { "epoch": 0.40122596823627754, "eval_loss": 1.7170045375823975, "eval_runtime": 62.9859, "eval_samples_per_second": 212.238, "eval_steps_per_second": 13.273, "step": 1800 }, { "epoch": 0.4235162998049596, "grad_norm": 464572.84375, "learning_rate": 2.675875906775737e-05, "loss": 1.9538, "step": 1900 }, { "epoch": 0.4235162998049596, "eval_loss": 1.7204582691192627, "eval_runtime": 63.2678, "eval_samples_per_second": 211.292, "eval_steps_per_second": 13.214, "step": 1900 }, { "epoch": 0.4458066313736417, "grad_norm": 461049.8125, "learning_rate": 2.652724185831147e-05, "loss": 1.9519, "step": 2000 }, { "epoch": 0.4458066313736417, "eval_loss": 1.7141714096069336, "eval_runtime": 63.4578, "eval_samples_per_second": 210.66, "eval_steps_per_second": 13.174, "step": 2000 }, { "epoch": 0.4680969629423238, "grad_norm": 470016.21875, "learning_rate": 2.6295724648865566e-05, "loss": 1.9391, "step": 2100 }, { "epoch": 0.4680969629423238, "eval_loss": 1.7201862335205078, "eval_runtime": 63.3953, "eval_samples_per_second": 210.867, "eval_steps_per_second": 13.187, "step": 2100 }, { "epoch": 0.4903872945110059, "grad_norm": 485373.15625, "learning_rate": 2.6064207439419662e-05, "loss": 1.9352, "step": 2200 }, { "epoch": 0.4903872945110059, "eval_loss": 1.698974370956421, "eval_runtime": 63.411, "eval_samples_per_second": 210.815, "eval_steps_per_second": 13.184, "step": 2200 }, { "epoch": 0.5126776260796879, "grad_norm": 472483.78125, "learning_rate": 2.5832690229973762e-05, "loss": 1.9412, "step": 2300 }, { "epoch": 0.5126776260796879, "eval_loss": 1.7102808952331543, "eval_runtime": 63.3414, "eval_samples_per_second": 211.047, "eval_steps_per_second": 13.198, "step": 2300 }, { "epoch": 0.53496795764837, "grad_norm": 444579.25, "learning_rate": 2.560117302052786e-05, "loss": 1.9331, "step": 2400 }, { "epoch": 0.53496795764837, "eval_loss": 1.7026287317276, "eval_runtime": 63.7826, "eval_samples_per_second": 209.587, "eval_steps_per_second": 13.107, "step": 2400 }, { "epoch": 0.5572582892170521, "grad_norm": 544911.375, "learning_rate": 2.5369655811081957e-05, "loss": 1.933, "step": 2500 }, { "epoch": 0.5572582892170521, "eval_loss": 1.6917425394058228, "eval_runtime": 63.6992, "eval_samples_per_second": 209.861, "eval_steps_per_second": 13.124, "step": 2500 }, { "epoch": 0.5795486207857342, "grad_norm": 442399.9375, "learning_rate": 2.5138138601636054e-05, "loss": 1.9209, "step": 2600 }, { "epoch": 0.5795486207857342, "eval_loss": 1.7024257183074951, "eval_runtime": 63.6754, "eval_samples_per_second": 209.94, "eval_steps_per_second": 13.129, "step": 2600 }, { "epoch": 0.6018389523544163, "grad_norm": 478748.46875, "learning_rate": 2.4906621392190153e-05, "loss": 1.9193, "step": 2700 }, { "epoch": 0.6018389523544163, "eval_loss": 1.7053234577178955, "eval_runtime": 63.8191, "eval_samples_per_second": 209.467, "eval_steps_per_second": 13.1, "step": 2700 }, { "epoch": 0.6241292839230984, "grad_norm": 468540.28125, "learning_rate": 2.4675104182744253e-05, "loss": 1.8972, "step": 2800 }, { "epoch": 0.6241292839230984, "eval_loss": 1.6971023082733154, "eval_runtime": 63.3867, "eval_samples_per_second": 210.896, "eval_steps_per_second": 13.189, "step": 2800 }, { "epoch": 0.6464196154917804, "grad_norm": 447092.15625, "learning_rate": 2.444358697329835e-05, "loss": 1.9116, "step": 2900 }, { "epoch": 0.6464196154917804, "eval_loss": 1.6840194463729858, "eval_runtime": 62.9634, "eval_samples_per_second": 212.314, "eval_steps_per_second": 13.278, "step": 2900 }, { "epoch": 0.6687099470604625, "grad_norm": 498904.8125, "learning_rate": 2.421206976385245e-05, "loss": 1.908, "step": 3000 }, { "epoch": 0.6687099470604625, "eval_loss": 1.69094979763031, "eval_runtime": 63.2731, "eval_samples_per_second": 211.275, "eval_steps_per_second": 13.213, "step": 3000 }, { "epoch": 0.6910002786291446, "grad_norm": 462770.21875, "learning_rate": 2.3980552554406545e-05, "loss": 1.9074, "step": 3100 }, { "epoch": 0.6910002786291446, "eval_loss": 1.6828372478485107, "eval_runtime": 63.5905, "eval_samples_per_second": 210.22, "eval_steps_per_second": 13.147, "step": 3100 }, { "epoch": 0.7132906101978267, "grad_norm": 510012.96875, "learning_rate": 2.374903534496064e-05, "loss": 1.8977, "step": 3200 }, { "epoch": 0.7132906101978267, "eval_loss": 1.6863964796066284, "eval_runtime": 63.6806, "eval_samples_per_second": 209.923, "eval_steps_per_second": 13.128, "step": 3200 }, { "epoch": 0.7355809417665088, "grad_norm": 471075.21875, "learning_rate": 2.351751813551474e-05, "loss": 1.9154, "step": 3300 }, { "epoch": 0.7355809417665088, "eval_loss": 1.6765246391296387, "eval_runtime": 63.5227, "eval_samples_per_second": 210.444, "eval_steps_per_second": 13.161, "step": 3300 }, { "epoch": 0.7578712733351909, "grad_norm": 459823.4375, "learning_rate": 2.328600092606884e-05, "loss": 1.8869, "step": 3400 }, { "epoch": 0.7578712733351909, "eval_loss": 1.6767551898956299, "eval_runtime": 62.9808, "eval_samples_per_second": 212.255, "eval_steps_per_second": 13.274, "step": 3400 }, { "epoch": 0.780161604903873, "grad_norm": 431094.84375, "learning_rate": 2.3054483716622936e-05, "loss": 1.8935, "step": 3500 }, { "epoch": 0.780161604903873, "eval_loss": 1.6749335527420044, "eval_runtime": 62.9334, "eval_samples_per_second": 212.415, "eval_steps_per_second": 13.284, "step": 3500 }, { "epoch": 0.8024519364725551, "grad_norm": 444868.96875, "learning_rate": 2.2822966507177032e-05, "loss": 1.8913, "step": 3600 }, { "epoch": 0.8024519364725551, "eval_loss": 1.6784425973892212, "eval_runtime": 63.2429, "eval_samples_per_second": 211.376, "eval_steps_per_second": 13.219, "step": 3600 }, { "epoch": 0.8247422680412371, "grad_norm": 451189.40625, "learning_rate": 2.259144929773113e-05, "loss": 1.8878, "step": 3700 }, { "epoch": 0.8247422680412371, "eval_loss": 1.6754982471466064, "eval_runtime": 63.7373, "eval_samples_per_second": 209.736, "eval_steps_per_second": 13.116, "step": 3700 }, { "epoch": 0.8470325996099192, "grad_norm": 441200.9375, "learning_rate": 2.235993208828523e-05, "loss": 1.8948, "step": 3800 }, { "epoch": 0.8470325996099192, "eval_loss": 1.6711094379425049, "eval_runtime": 62.7885, "eval_samples_per_second": 212.905, "eval_steps_per_second": 13.315, "step": 3800 }, { "epoch": 0.8693229311786013, "grad_norm": 467896.84375, "learning_rate": 2.2128414878839327e-05, "loss": 1.8816, "step": 3900 }, { "epoch": 0.8693229311786013, "eval_loss": 1.671681523323059, "eval_runtime": 63.1597, "eval_samples_per_second": 211.654, "eval_steps_per_second": 13.236, "step": 3900 }, { "epoch": 0.8916132627472834, "grad_norm": 456236.3125, "learning_rate": 2.1896897669393427e-05, "loss": 1.8864, "step": 4000 }, { "epoch": 0.8916132627472834, "eval_loss": 1.673299789428711, "eval_runtime": 63.6594, "eval_samples_per_second": 209.992, "eval_steps_per_second": 13.132, "step": 4000 }, { "epoch": 0.9139035943159655, "grad_norm": 470346.40625, "learning_rate": 2.1665380459947523e-05, "loss": 1.879, "step": 4100 }, { "epoch": 0.9139035943159655, "eval_loss": 1.6596086025238037, "eval_runtime": 63.3568, "eval_samples_per_second": 210.995, "eval_steps_per_second": 13.195, "step": 4100 }, { "epoch": 0.9361939258846476, "grad_norm": 468418.625, "learning_rate": 2.143386325050162e-05, "loss": 1.8862, "step": 4200 }, { "epoch": 0.9361939258846476, "eval_loss": 1.6646808385849, "eval_runtime": 63.5493, "eval_samples_per_second": 210.356, "eval_steps_per_second": 13.155, "step": 4200 }, { "epoch": 0.9584842574533297, "grad_norm": 454157.0, "learning_rate": 2.120234604105572e-05, "loss": 1.8722, "step": 4300 }, { "epoch": 0.9584842574533297, "eval_loss": 1.667905330657959, "eval_runtime": 62.7216, "eval_samples_per_second": 213.132, "eval_steps_per_second": 13.329, "step": 4300 }, { "epoch": 0.9807745890220118, "grad_norm": 462536.9375, "learning_rate": 2.0970828831609818e-05, "loss": 1.8669, "step": 4400 }, { "epoch": 0.9807745890220118, "eval_loss": 1.65095853805542, "eval_runtime": 62.9484, "eval_samples_per_second": 212.364, "eval_steps_per_second": 13.281, "step": 4400 }, { "epoch": 1.0030649205906939, "grad_norm": 478586.0, "learning_rate": 2.0739311622163914e-05, "loss": 1.8636, "step": 4500 }, { "epoch": 1.0030649205906939, "eval_loss": 1.6621143817901611, "eval_runtime": 63.2068, "eval_samples_per_second": 211.496, "eval_steps_per_second": 13.226, "step": 4500 }, { "epoch": 1.0253552521593758, "grad_norm": 539991.8125, "learning_rate": 2.050779441271801e-05, "loss": 1.8262, "step": 4600 }, { "epoch": 1.0253552521593758, "eval_loss": 1.6551061868667603, "eval_runtime": 63.124, "eval_samples_per_second": 211.774, "eval_steps_per_second": 13.244, "step": 4600 }, { "epoch": 1.047645583728058, "grad_norm": 428922.53125, "learning_rate": 2.027627720327211e-05, "loss": 1.8142, "step": 4700 }, { "epoch": 1.047645583728058, "eval_loss": 1.6545677185058594, "eval_runtime": 63.0786, "eval_samples_per_second": 211.926, "eval_steps_per_second": 13.253, "step": 4700 }, { "epoch": 1.06993591529674, "grad_norm": 438794.46875, "learning_rate": 2.004475999382621e-05, "loss": 1.8097, "step": 4800 }, { "epoch": 1.06993591529674, "eval_loss": 1.6556122303009033, "eval_runtime": 63.6203, "eval_samples_per_second": 210.122, "eval_steps_per_second": 13.14, "step": 4800 }, { "epoch": 1.092226246865422, "grad_norm": 468837.34375, "learning_rate": 1.9813242784380306e-05, "loss": 1.8242, "step": 4900 }, { "epoch": 1.092226246865422, "eval_loss": 1.657508134841919, "eval_runtime": 63.2746, "eval_samples_per_second": 211.27, "eval_steps_per_second": 13.212, "step": 4900 }, { "epoch": 1.1145165784341042, "grad_norm": 449878.40625, "learning_rate": 1.9581725574934405e-05, "loss": 1.8143, "step": 5000 }, { "epoch": 1.1145165784341042, "eval_loss": 1.6671580076217651, "eval_runtime": 63.9114, "eval_samples_per_second": 209.165, "eval_steps_per_second": 13.081, "step": 5000 }, { "epoch": 1.1368069100027862, "grad_norm": 427682.84375, "learning_rate": 1.93502083654885e-05, "loss": 1.8242, "step": 5100 }, { "epoch": 1.1368069100027862, "eval_loss": 1.6521689891815186, "eval_runtime": 62.5274, "eval_samples_per_second": 213.794, "eval_steps_per_second": 13.37, "step": 5100 }, { "epoch": 1.1590972415714684, "grad_norm": 441866.34375, "learning_rate": 1.9118691156042597e-05, "loss": 1.8105, "step": 5200 }, { "epoch": 1.1590972415714684, "eval_loss": 1.6530238389968872, "eval_runtime": 62.6567, "eval_samples_per_second": 213.353, "eval_steps_per_second": 13.343, "step": 5200 }, { "epoch": 1.1813875731401504, "grad_norm": 455879.25, "learning_rate": 1.8887173946596697e-05, "loss": 1.813, "step": 5300 }, { "epoch": 1.1813875731401504, "eval_loss": 1.654367208480835, "eval_runtime": 63.6338, "eval_samples_per_second": 210.077, "eval_steps_per_second": 13.138, "step": 5300 }, { "epoch": 1.2036779047088326, "grad_norm": 410671.53125, "learning_rate": 1.8655656737150796e-05, "loss": 1.8201, "step": 5400 }, { "epoch": 1.2036779047088326, "eval_loss": 1.6491619348526, "eval_runtime": 63.5953, "eval_samples_per_second": 210.204, "eval_steps_per_second": 13.146, "step": 5400 }, { "epoch": 1.2259682362775146, "grad_norm": 439893.84375, "learning_rate": 1.8424139527704893e-05, "loss": 1.8078, "step": 5500 }, { "epoch": 1.2259682362775146, "eval_loss": 1.6452112197875977, "eval_runtime": 63.1345, "eval_samples_per_second": 211.738, "eval_steps_per_second": 13.242, "step": 5500 }, { "epoch": 1.2482585678461966, "grad_norm": 436707.84375, "learning_rate": 1.819262231825899e-05, "loss": 1.8081, "step": 5600 }, { "epoch": 1.2482585678461966, "eval_loss": 1.63877272605896, "eval_runtime": 63.3151, "eval_samples_per_second": 211.134, "eval_steps_per_second": 13.204, "step": 5600 }, { "epoch": 1.2705488994148788, "grad_norm": 453170.34375, "learning_rate": 1.7961105108813088e-05, "loss": 1.809, "step": 5700 }, { "epoch": 1.2705488994148788, "eval_loss": 1.6391615867614746, "eval_runtime": 63.0066, "eval_samples_per_second": 212.168, "eval_steps_per_second": 13.268, "step": 5700 }, { "epoch": 1.292839230983561, "grad_norm": 419482.625, "learning_rate": 1.7729587899367188e-05, "loss": 1.7993, "step": 5800 }, { "epoch": 1.292839230983561, "eval_loss": 1.6412807703018188, "eval_runtime": 62.9872, "eval_samples_per_second": 212.234, "eval_steps_per_second": 13.273, "step": 5800 }, { "epoch": 1.315129562552243, "grad_norm": 439712.59375, "learning_rate": 1.7498070689921284e-05, "loss": 1.806, "step": 5900 }, { "epoch": 1.315129562552243, "eval_loss": 1.6369620561599731, "eval_runtime": 62.9462, "eval_samples_per_second": 212.372, "eval_steps_per_second": 13.281, "step": 5900 }, { "epoch": 1.337419894120925, "grad_norm": 433225.875, "learning_rate": 1.7266553480475383e-05, "loss": 1.8054, "step": 6000 }, { "epoch": 1.337419894120925, "eval_loss": 1.6376512050628662, "eval_runtime": 63.3967, "eval_samples_per_second": 210.863, "eval_steps_per_second": 13.187, "step": 6000 }, { "epoch": 1.3597102256896072, "grad_norm": 442755.03125, "learning_rate": 1.703503627102948e-05, "loss": 1.7996, "step": 6100 }, { "epoch": 1.3597102256896072, "eval_loss": 1.6386842727661133, "eval_runtime": 64.1806, "eval_samples_per_second": 208.287, "eval_steps_per_second": 13.026, "step": 6100 }, { "epoch": 1.3820005572582892, "grad_norm": 444388.1875, "learning_rate": 1.6803519061583576e-05, "loss": 1.7973, "step": 6200 }, { "epoch": 1.3820005572582892, "eval_loss": 1.637714147567749, "eval_runtime": 62.6809, "eval_samples_per_second": 213.271, "eval_steps_per_second": 13.337, "step": 6200 }, { "epoch": 1.4042908888269712, "grad_norm": 457163.5625, "learning_rate": 1.657200185213768e-05, "loss": 1.8103, "step": 6300 }, { "epoch": 1.4042908888269712, "eval_loss": 1.6353328227996826, "eval_runtime": 63.7218, "eval_samples_per_second": 209.787, "eval_steps_per_second": 13.12, "step": 6300 }, { "epoch": 1.4265812203956534, "grad_norm": 423063.71875, "learning_rate": 1.6340484642691775e-05, "loss": 1.8193, "step": 6400 }, { "epoch": 1.4265812203956534, "eval_loss": 1.632241129875183, "eval_runtime": 62.6795, "eval_samples_per_second": 213.276, "eval_steps_per_second": 13.338, "step": 6400 }, { "epoch": 1.4488715519643356, "grad_norm": 450858.5625, "learning_rate": 1.610896743324587e-05, "loss": 1.8133, "step": 6500 }, { "epoch": 1.4488715519643356, "eval_loss": 1.642052412033081, "eval_runtime": 64.1592, "eval_samples_per_second": 208.357, "eval_steps_per_second": 13.03, "step": 6500 }, { "epoch": 1.4711618835330176, "grad_norm": 424338.96875, "learning_rate": 1.5877450223799967e-05, "loss": 1.8016, "step": 6600 }, { "epoch": 1.4711618835330176, "eval_loss": 1.6317675113677979, "eval_runtime": 63.0019, "eval_samples_per_second": 212.184, "eval_steps_per_second": 13.269, "step": 6600 }, { "epoch": 1.4934522151016996, "grad_norm": 441573.03125, "learning_rate": 1.5645933014354067e-05, "loss": 1.7923, "step": 6700 }, { "epoch": 1.4934522151016996, "eval_loss": 1.6302847862243652, "eval_runtime": 63.6702, "eval_samples_per_second": 209.957, "eval_steps_per_second": 13.13, "step": 6700 }, { "epoch": 1.5157425466703818, "grad_norm": 460539.0625, "learning_rate": 1.5414415804908166e-05, "loss": 1.7969, "step": 6800 }, { "epoch": 1.5157425466703818, "eval_loss": 1.6323357820510864, "eval_runtime": 62.8247, "eval_samples_per_second": 212.783, "eval_steps_per_second": 13.307, "step": 6800 }, { "epoch": 1.5380328782390638, "grad_norm": 432685.96875, "learning_rate": 1.5182898595462264e-05, "loss": 1.7955, "step": 6900 }, { "epoch": 1.5380328782390638, "eval_loss": 1.6306127309799194, "eval_runtime": 63.1442, "eval_samples_per_second": 211.706, "eval_steps_per_second": 13.24, "step": 6900 }, { "epoch": 1.5603232098077457, "grad_norm": 414684.125, "learning_rate": 1.495138138601636e-05, "loss": 1.7933, "step": 7000 }, { "epoch": 1.5603232098077457, "eval_loss": 1.6255934238433838, "eval_runtime": 63.4943, "eval_samples_per_second": 210.539, "eval_steps_per_second": 13.167, "step": 7000 }, { "epoch": 1.582613541376428, "grad_norm": 423529.65625, "learning_rate": 1.471986417657046e-05, "loss": 1.7951, "step": 7100 }, { "epoch": 1.582613541376428, "eval_loss": 1.6387838125228882, "eval_runtime": 63.598, "eval_samples_per_second": 210.195, "eval_steps_per_second": 13.145, "step": 7100 }, { "epoch": 1.6049038729451102, "grad_norm": 421001.21875, "learning_rate": 1.4488346967124556e-05, "loss": 1.8165, "step": 7200 }, { "epoch": 1.6049038729451102, "eval_loss": 1.6351381540298462, "eval_runtime": 63.005, "eval_samples_per_second": 212.174, "eval_steps_per_second": 13.269, "step": 7200 }, { "epoch": 1.6271942045137922, "grad_norm": 401336.375, "learning_rate": 1.4256829757678654e-05, "loss": 1.7934, "step": 7300 }, { "epoch": 1.6271942045137922, "eval_loss": 1.630112886428833, "eval_runtime": 63.0665, "eval_samples_per_second": 211.967, "eval_steps_per_second": 13.256, "step": 7300 }, { "epoch": 1.6494845360824741, "grad_norm": 439442.84375, "learning_rate": 1.4025312548232753e-05, "loss": 1.7809, "step": 7400 }, { "epoch": 1.6494845360824741, "eval_loss": 1.6299716234207153, "eval_runtime": 63.6859, "eval_samples_per_second": 209.905, "eval_steps_per_second": 13.127, "step": 7400 }, { "epoch": 1.6717748676511563, "grad_norm": 407261.71875, "learning_rate": 1.379379533878685e-05, "loss": 1.7899, "step": 7500 }, { "epoch": 1.6717748676511563, "eval_loss": 1.6285357475280762, "eval_runtime": 63.1642, "eval_samples_per_second": 211.639, "eval_steps_per_second": 13.235, "step": 7500 }, { "epoch": 1.6940651992198386, "grad_norm": 403575.21875, "learning_rate": 1.3562278129340949e-05, "loss": 1.7887, "step": 7600 }, { "epoch": 1.6940651992198386, "eval_loss": 1.6282232999801636, "eval_runtime": 62.5905, "eval_samples_per_second": 213.579, "eval_steps_per_second": 13.357, "step": 7600 }, { "epoch": 1.7163555307885203, "grad_norm": 440494.6875, "learning_rate": 1.3330760919895045e-05, "loss": 1.7859, "step": 7700 }, { "epoch": 1.7163555307885203, "eval_loss": 1.622281551361084, "eval_runtime": 63.4527, "eval_samples_per_second": 210.677, "eval_steps_per_second": 13.175, "step": 7700 }, { "epoch": 1.7386458623572025, "grad_norm": 414236.25, "learning_rate": 1.3099243710449143e-05, "loss": 1.8, "step": 7800 }, { "epoch": 1.7386458623572025, "eval_loss": 1.623448371887207, "eval_runtime": 62.7927, "eval_samples_per_second": 212.891, "eval_steps_per_second": 13.314, "step": 7800 }, { "epoch": 1.7609361939258847, "grad_norm": 438500.71875, "learning_rate": 1.2867726501003242e-05, "loss": 1.7907, "step": 7900 }, { "epoch": 1.7609361939258847, "eval_loss": 1.6255710124969482, "eval_runtime": 63.6158, "eval_samples_per_second": 210.137, "eval_steps_per_second": 13.141, "step": 7900 }, { "epoch": 1.7832265254945667, "grad_norm": 413099.34375, "learning_rate": 1.2636209291557339e-05, "loss": 1.8052, "step": 8000 }, { "epoch": 1.7832265254945667, "eval_loss": 1.6219525337219238, "eval_runtime": 63.4015, "eval_samples_per_second": 210.847, "eval_steps_per_second": 13.186, "step": 8000 }, { "epoch": 1.8055168570632487, "grad_norm": 442402.25, "learning_rate": 1.2404692082111438e-05, "loss": 1.7881, "step": 8100 }, { "epoch": 1.8055168570632487, "eval_loss": 1.6209757328033447, "eval_runtime": 62.8346, "eval_samples_per_second": 212.749, "eval_steps_per_second": 13.305, "step": 8100 }, { "epoch": 1.827807188631931, "grad_norm": 482508.3125, "learning_rate": 1.2173174872665536e-05, "loss": 1.7967, "step": 8200 }, { "epoch": 1.827807188631931, "eval_loss": 1.6282365322113037, "eval_runtime": 63.2359, "eval_samples_per_second": 211.399, "eval_steps_per_second": 13.22, "step": 8200 }, { "epoch": 1.8500975202006131, "grad_norm": 443377.25, "learning_rate": 1.1941657663219632e-05, "loss": 1.7921, "step": 8300 }, { "epoch": 1.8500975202006131, "eval_loss": 1.624315619468689, "eval_runtime": 63.0489, "eval_samples_per_second": 212.026, "eval_steps_per_second": 13.26, "step": 8300 }, { "epoch": 1.8723878517692951, "grad_norm": 397804.8125, "learning_rate": 1.1710140453773732e-05, "loss": 1.7898, "step": 8400 }, { "epoch": 1.8723878517692951, "eval_loss": 1.6195310354232788, "eval_runtime": 63.4127, "eval_samples_per_second": 210.81, "eval_steps_per_second": 13.183, "step": 8400 }, { "epoch": 1.894678183337977, "grad_norm": 422386.96875, "learning_rate": 1.1478623244327828e-05, "loss": 1.779, "step": 8500 }, { "epoch": 1.894678183337977, "eval_loss": 1.6192779541015625, "eval_runtime": 63.476, "eval_samples_per_second": 210.599, "eval_steps_per_second": 13.17, "step": 8500 }, { "epoch": 1.9169685149066593, "grad_norm": 411348.03125, "learning_rate": 1.1247106034881927e-05, "loss": 1.7852, "step": 8600 }, { "epoch": 1.9169685149066593, "eval_loss": 1.6169335842132568, "eval_runtime": 63.0679, "eval_samples_per_second": 211.962, "eval_steps_per_second": 13.256, "step": 8600 }, { "epoch": 1.9392588464753413, "grad_norm": 444615.65625, "learning_rate": 1.1015588825436025e-05, "loss": 1.7813, "step": 8700 }, { "epoch": 1.9392588464753413, "eval_loss": 1.6124924421310425, "eval_runtime": 63.9864, "eval_samples_per_second": 208.919, "eval_steps_per_second": 13.065, "step": 8700 }, { "epoch": 1.9615491780440233, "grad_norm": 414404.125, "learning_rate": 1.0784071615990121e-05, "loss": 1.7893, "step": 8800 }, { "epoch": 1.9615491780440233, "eval_loss": 1.6191661357879639, "eval_runtime": 63.6221, "eval_samples_per_second": 210.116, "eval_steps_per_second": 13.14, "step": 8800 }, { "epoch": 1.9838395096127055, "grad_norm": 445457.1875, "learning_rate": 1.055255440654422e-05, "loss": 1.7905, "step": 8900 }, { "epoch": 1.9838395096127055, "eval_loss": 1.6134953498840332, "eval_runtime": 63.3933, "eval_samples_per_second": 210.874, "eval_steps_per_second": 13.188, "step": 8900 }, { "epoch": 2.0061298411813877, "grad_norm": 439959.25, "learning_rate": 1.0321037197098317e-05, "loss": 1.7873, "step": 9000 }, { "epoch": 2.0061298411813877, "eval_loss": 1.6150258779525757, "eval_runtime": 62.9808, "eval_samples_per_second": 212.255, "eval_steps_per_second": 13.274, "step": 9000 }, { "epoch": 2.0284201727500695, "grad_norm": 399886.46875, "learning_rate": 1.0089519987652416e-05, "loss": 1.7304, "step": 9100 }, { "epoch": 2.0284201727500695, "eval_loss": 1.615242600440979, "eval_runtime": 63.4005, "eval_samples_per_second": 210.85, "eval_steps_per_second": 13.186, "step": 9100 }, { "epoch": 2.0507105043187517, "grad_norm": 432171.59375, "learning_rate": 9.858002778206514e-06, "loss": 1.748, "step": 9200 }, { "epoch": 2.0507105043187517, "eval_loss": 1.6178884506225586, "eval_runtime": 63.1372, "eval_samples_per_second": 211.729, "eval_steps_per_second": 13.241, "step": 9200 }, { "epoch": 2.073000835887434, "grad_norm": 438172.875, "learning_rate": 9.62648556876061e-06, "loss": 1.7488, "step": 9300 }, { "epoch": 2.073000835887434, "eval_loss": 1.615029215812683, "eval_runtime": 63.7948, "eval_samples_per_second": 209.547, "eval_steps_per_second": 13.105, "step": 9300 }, { "epoch": 2.095291167456116, "grad_norm": 422492.28125, "learning_rate": 9.39496835931471e-06, "loss": 1.7482, "step": 9400 }, { "epoch": 2.095291167456116, "eval_loss": 1.6207051277160645, "eval_runtime": 63.458, "eval_samples_per_second": 210.659, "eval_steps_per_second": 13.174, "step": 9400 }, { "epoch": 2.117581499024798, "grad_norm": 438706.75, "learning_rate": 9.163451149868806e-06, "loss": 1.739, "step": 9500 }, { "epoch": 2.117581499024798, "eval_loss": 1.6097335815429688, "eval_runtime": 63.1577, "eval_samples_per_second": 211.66, "eval_steps_per_second": 13.237, "step": 9500 }, { "epoch": 2.13987183059348, "grad_norm": 418132.5625, "learning_rate": 8.931933940422906e-06, "loss": 1.7599, "step": 9600 }, { "epoch": 2.13987183059348, "eval_loss": 1.6131938695907593, "eval_runtime": 63.4093, "eval_samples_per_second": 210.821, "eval_steps_per_second": 13.184, "step": 9600 }, { "epoch": 2.1621621621621623, "grad_norm": 430660.1875, "learning_rate": 8.700416730977003e-06, "loss": 1.7473, "step": 9700 }, { "epoch": 2.1621621621621623, "eval_loss": 1.6116857528686523, "eval_runtime": 63.5141, "eval_samples_per_second": 210.473, "eval_steps_per_second": 13.162, "step": 9700 }, { "epoch": 2.184452493730844, "grad_norm": 410735.0625, "learning_rate": 8.4688995215311e-06, "loss": 1.75, "step": 9800 }, { "epoch": 2.184452493730844, "eval_loss": 1.611021876335144, "eval_runtime": 63.3479, "eval_samples_per_second": 211.025, "eval_steps_per_second": 13.197, "step": 9800 }, { "epoch": 2.2067428252995263, "grad_norm": 437123.8125, "learning_rate": 8.237382312085199e-06, "loss": 1.7432, "step": 9900 }, { "epoch": 2.2067428252995263, "eval_loss": 1.614385962486267, "eval_runtime": 63.4099, "eval_samples_per_second": 210.819, "eval_steps_per_second": 13.184, "step": 9900 }, { "epoch": 2.2290331568682085, "grad_norm": 434528.1875, "learning_rate": 8.005865102639295e-06, "loss": 1.7491, "step": 10000 }, { "epoch": 2.2290331568682085, "eval_loss": 1.6123193502426147, "eval_runtime": 63.2705, "eval_samples_per_second": 211.283, "eval_steps_per_second": 13.213, "step": 10000 }, { "epoch": 2.2513234884368907, "grad_norm": 394168.875, "learning_rate": 7.774347893193395e-06, "loss": 1.7377, "step": 10100 }, { "epoch": 2.2513234884368907, "eval_loss": 1.6091859340667725, "eval_runtime": 63.9924, "eval_samples_per_second": 208.9, "eval_steps_per_second": 13.064, "step": 10100 }, { "epoch": 2.2736138200055724, "grad_norm": 405977.28125, "learning_rate": 7.542830683747492e-06, "loss": 1.7439, "step": 10200 }, { "epoch": 2.2736138200055724, "eval_loss": 1.6095374822616577, "eval_runtime": 63.1561, "eval_samples_per_second": 211.666, "eval_steps_per_second": 13.237, "step": 10200 }, { "epoch": 2.2959041515742546, "grad_norm": 441095.4375, "learning_rate": 7.31131347430159e-06, "loss": 1.7417, "step": 10300 }, { "epoch": 2.2959041515742546, "eval_loss": 1.6078873872756958, "eval_runtime": 62.9452, "eval_samples_per_second": 212.375, "eval_steps_per_second": 13.281, "step": 10300 }, { "epoch": 2.318194483142937, "grad_norm": 450342.375, "learning_rate": 7.079796264855688e-06, "loss": 1.7482, "step": 10400 }, { "epoch": 2.318194483142937, "eval_loss": 1.6116305589675903, "eval_runtime": 63.4969, "eval_samples_per_second": 210.53, "eval_steps_per_second": 13.166, "step": 10400 }, { "epoch": 2.340484814711619, "grad_norm": 462851.03125, "learning_rate": 6.848279055409786e-06, "loss": 1.7391, "step": 10500 }, { "epoch": 2.340484814711619, "eval_loss": 1.6078118085861206, "eval_runtime": 63.3093, "eval_samples_per_second": 211.154, "eval_steps_per_second": 13.205, "step": 10500 }, { "epoch": 2.362775146280301, "grad_norm": 405519.75, "learning_rate": 6.616761845963883e-06, "loss": 1.7345, "step": 10600 }, { "epoch": 2.362775146280301, "eval_loss": 1.6122682094573975, "eval_runtime": 63.3463, "eval_samples_per_second": 211.031, "eval_steps_per_second": 13.197, "step": 10600 }, { "epoch": 2.385065477848983, "grad_norm": 471771.9375, "learning_rate": 6.385244636517981e-06, "loss": 1.7365, "step": 10700 }, { "epoch": 2.385065477848983, "eval_loss": 1.6120107173919678, "eval_runtime": 63.2512, "eval_samples_per_second": 211.348, "eval_steps_per_second": 13.217, "step": 10700 }, { "epoch": 2.4073558094176652, "grad_norm": 416676.4375, "learning_rate": 6.153727427072079e-06, "loss": 1.7446, "step": 10800 }, { "epoch": 2.4073558094176652, "eval_loss": 1.6103585958480835, "eval_runtime": 63.4362, "eval_samples_per_second": 210.731, "eval_steps_per_second": 13.179, "step": 10800 }, { "epoch": 2.429646140986347, "grad_norm": 409846.5, "learning_rate": 5.9222102176261775e-06, "loss": 1.7273, "step": 10900 }, { "epoch": 2.429646140986347, "eval_loss": 1.6118780374526978, "eval_runtime": 63.3423, "eval_samples_per_second": 211.044, "eval_steps_per_second": 13.198, "step": 10900 }, { "epoch": 2.4519364725550292, "grad_norm": 400530.59375, "learning_rate": 5.690693008180275e-06, "loss": 1.7494, "step": 11000 }, { "epoch": 2.4519364725550292, "eval_loss": 1.6059410572052002, "eval_runtime": 63.1379, "eval_samples_per_second": 211.727, "eval_steps_per_second": 13.241, "step": 11000 }, { "epoch": 2.4742268041237114, "grad_norm": 427720.5625, "learning_rate": 5.459175798734372e-06, "loss": 1.7332, "step": 11100 }, { "epoch": 2.4742268041237114, "eval_loss": 1.6064603328704834, "eval_runtime": 63.7921, "eval_samples_per_second": 209.556, "eval_steps_per_second": 13.105, "step": 11100 }, { "epoch": 2.496517135692393, "grad_norm": 450690.875, "learning_rate": 5.22765858928847e-06, "loss": 1.7285, "step": 11200 }, { "epoch": 2.496517135692393, "eval_loss": 1.6082170009613037, "eval_runtime": 62.5222, "eval_samples_per_second": 213.812, "eval_steps_per_second": 13.371, "step": 11200 }, { "epoch": 2.5188074672610754, "grad_norm": 412636.5, "learning_rate": 4.996141379842568e-06, "loss": 1.7319, "step": 11300 }, { "epoch": 2.5188074672610754, "eval_loss": 1.6035947799682617, "eval_runtime": 63.3474, "eval_samples_per_second": 211.027, "eval_steps_per_second": 13.197, "step": 11300 }, { "epoch": 2.5410977988297576, "grad_norm": 403399.875, "learning_rate": 4.764624170396667e-06, "loss": 1.7245, "step": 11400 }, { "epoch": 2.5410977988297576, "eval_loss": 1.6081373691558838, "eval_runtime": 63.6595, "eval_samples_per_second": 209.992, "eval_steps_per_second": 13.132, "step": 11400 }, { "epoch": 2.56338813039844, "grad_norm": 432179.4375, "learning_rate": 4.5331069609507645e-06, "loss": 1.7379, "step": 11500 }, { "epoch": 2.56338813039844, "eval_loss": 1.6043556928634644, "eval_runtime": 63.1427, "eval_samples_per_second": 211.711, "eval_steps_per_second": 13.24, "step": 11500 }, { "epoch": 2.585678461967122, "grad_norm": 398148.96875, "learning_rate": 4.3015897515048615e-06, "loss": 1.7349, "step": 11600 }, { "epoch": 2.585678461967122, "eval_loss": 1.6080468893051147, "eval_runtime": 63.1768, "eval_samples_per_second": 211.597, "eval_steps_per_second": 13.233, "step": 11600 }, { "epoch": 2.607968793535804, "grad_norm": 406493.65625, "learning_rate": 4.070072542058959e-06, "loss": 1.7408, "step": 11700 }, { "epoch": 2.607968793535804, "eval_loss": 1.6043606996536255, "eval_runtime": 63.7306, "eval_samples_per_second": 209.758, "eval_steps_per_second": 13.118, "step": 11700 }, { "epoch": 2.630259125104486, "grad_norm": 415316.5625, "learning_rate": 3.838555332613058e-06, "loss": 1.7317, "step": 11800 }, { "epoch": 2.630259125104486, "eval_loss": 1.6045427322387695, "eval_runtime": 63.3899, "eval_samples_per_second": 210.885, "eval_steps_per_second": 13.188, "step": 11800 }, { "epoch": 2.652549456673168, "grad_norm": 433577.84375, "learning_rate": 3.6070381231671554e-06, "loss": 1.7288, "step": 11900 }, { "epoch": 2.652549456673168, "eval_loss": 1.603334903717041, "eval_runtime": 63.1008, "eval_samples_per_second": 211.851, "eval_steps_per_second": 13.249, "step": 11900 }, { "epoch": 2.67483978824185, "grad_norm": 408270.8125, "learning_rate": 3.3755209137212533e-06, "loss": 1.7327, "step": 12000 }, { "epoch": 2.67483978824185, "eval_loss": 1.6032938957214355, "eval_runtime": 62.9232, "eval_samples_per_second": 212.45, "eval_steps_per_second": 13.286, "step": 12000 }, { "epoch": 2.697130119810532, "grad_norm": 412520.28125, "learning_rate": 3.144003704275351e-06, "loss": 1.7349, "step": 12100 }, { "epoch": 2.697130119810532, "eval_loss": 1.6040232181549072, "eval_runtime": 63.2685, "eval_samples_per_second": 211.29, "eval_steps_per_second": 13.214, "step": 12100 }, { "epoch": 2.7194204513792144, "grad_norm": 430910.1875, "learning_rate": 2.9124864948294494e-06, "loss": 1.7378, "step": 12200 }, { "epoch": 2.7194204513792144, "eval_loss": 1.6026825904846191, "eval_runtime": 63.0275, "eval_samples_per_second": 212.098, "eval_steps_per_second": 13.264, "step": 12200 }, { "epoch": 2.741710782947896, "grad_norm": 426524.9375, "learning_rate": 2.680969285383547e-06, "loss": 1.7438, "step": 12300 }, { "epoch": 2.741710782947896, "eval_loss": 1.6039636135101318, "eval_runtime": 63.8654, "eval_samples_per_second": 209.315, "eval_steps_per_second": 13.09, "step": 12300 }, { "epoch": 2.7640011145165784, "grad_norm": 439803.8125, "learning_rate": 2.449452075937645e-06, "loss": 1.7357, "step": 12400 }, { "epoch": 2.7640011145165784, "eval_loss": 1.6037323474884033, "eval_runtime": 63.1113, "eval_samples_per_second": 211.816, "eval_steps_per_second": 13.246, "step": 12400 }, { "epoch": 2.7862914460852606, "grad_norm": 442227.34375, "learning_rate": 2.2179348664917425e-06, "loss": 1.7333, "step": 12500 }, { "epoch": 2.7862914460852606, "eval_loss": 1.6014666557312012, "eval_runtime": 63.3883, "eval_samples_per_second": 210.891, "eval_steps_per_second": 13.189, "step": 12500 }, { "epoch": 2.8085817776539423, "grad_norm": 409348.65625, "learning_rate": 1.9864176570458403e-06, "loss": 1.7344, "step": 12600 }, { "epoch": 2.8085817776539423, "eval_loss": 1.6025654077529907, "eval_runtime": 63.562, "eval_samples_per_second": 210.315, "eval_steps_per_second": 13.153, "step": 12600 }, { "epoch": 2.8308721092226246, "grad_norm": 438774.5, "learning_rate": 1.7549004475999384e-06, "loss": 1.7356, "step": 12700 }, { "epoch": 2.8308721092226246, "eval_loss": 1.6015514135360718, "eval_runtime": 63.4784, "eval_samples_per_second": 210.591, "eval_steps_per_second": 13.17, "step": 12700 }, { "epoch": 2.8531624407913068, "grad_norm": 425528.09375, "learning_rate": 1.523383238154036e-06, "loss": 1.7494, "step": 12800 }, { "epoch": 2.8531624407913068, "eval_loss": 1.602927565574646, "eval_runtime": 63.9753, "eval_samples_per_second": 208.956, "eval_steps_per_second": 13.068, "step": 12800 }, { "epoch": 2.875452772359989, "grad_norm": 402082.0625, "learning_rate": 1.291866028708134e-06, "loss": 1.7414, "step": 12900 }, { "epoch": 2.875452772359989, "eval_loss": 1.6020632982254028, "eval_runtime": 63.2754, "eval_samples_per_second": 211.267, "eval_steps_per_second": 13.212, "step": 12900 }, { "epoch": 2.897743103928671, "grad_norm": 409632.125, "learning_rate": 1.0603488192622319e-06, "loss": 1.7337, "step": 13000 }, { "epoch": 2.897743103928671, "eval_loss": 1.6021629571914673, "eval_runtime": 63.0484, "eval_samples_per_second": 212.028, "eval_steps_per_second": 13.26, "step": 13000 }, { "epoch": 2.920033435497353, "grad_norm": 431921.65625, "learning_rate": 8.288316098163297e-07, "loss": 1.7239, "step": 13100 }, { "epoch": 2.920033435497353, "eval_loss": 1.6020002365112305, "eval_runtime": 63.0873, "eval_samples_per_second": 211.897, "eval_steps_per_second": 13.251, "step": 13100 }, { "epoch": 2.942323767066035, "grad_norm": 436490.65625, "learning_rate": 5.973144003704276e-07, "loss": 1.7461, "step": 13200 }, { "epoch": 2.942323767066035, "eval_loss": 1.6018171310424805, "eval_runtime": 63.2016, "eval_samples_per_second": 211.514, "eval_steps_per_second": 13.228, "step": 13200 }, { "epoch": 2.9646140986347174, "grad_norm": 396898.375, "learning_rate": 3.657971909245254e-07, "loss": 1.7358, "step": 13300 }, { "epoch": 2.9646140986347174, "eval_loss": 1.601110577583313, "eval_runtime": 63.1463, "eval_samples_per_second": 211.699, "eval_steps_per_second": 13.239, "step": 13300 }, { "epoch": 2.986904430203399, "grad_norm": 419936.46875, "learning_rate": 1.3427998147862324e-07, "loss": 1.7293, "step": 13400 }, { "epoch": 2.986904430203399, "eval_loss": 1.6014801263809204, "eval_runtime": 63.6722, "eval_samples_per_second": 209.95, "eval_steps_per_second": 13.13, "step": 13400 } ], "logging_steps": 100, "max_steps": 13458, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 5.251647920091955e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }