| { | |
| "best_metric": 2.037567615509033, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_default_r32/checkpoint-16", | |
| "epoch": 0.9981059842836993, | |
| "eval_steps": 8, | |
| "global_step": 387, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025790852307072333, | |
| "grad_norm": 23.78647804260254, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.2237, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010316340922828933, | |
| "grad_norm": 10.68380355834961, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0556, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "grad_norm": 13.343311309814453, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9915, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "eval_loss": 2.0385255813598633, | |
| "eval_runtime": 251.2319, | |
| "eval_samples_per_second": 0.979, | |
| "eval_steps_per_second": 0.979, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0309490227684868, | |
| "grad_norm": 19.941059112548828, | |
| "learning_rate": 9.997251843068762e-05, | |
| "loss": 2.0389, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "grad_norm": 13.265397071838379, | |
| "learning_rate": 9.989010393221656e-05, | |
| "loss": 2.054, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "eval_loss": 2.037567615509033, | |
| "eval_runtime": 356.5096, | |
| "eval_samples_per_second": 0.69, | |
| "eval_steps_per_second": 0.69, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05158170461414467, | |
| "grad_norm": 12.110239028930664, | |
| "learning_rate": 9.97528470997769e-05, | |
| "loss": 1.9974, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "grad_norm": 15.224806785583496, | |
| "learning_rate": 9.956089881469482e-05, | |
| "loss": 2.0356, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "eval_loss": 2.060382127761841, | |
| "eval_runtime": 260.5723, | |
| "eval_samples_per_second": 0.944, | |
| "eval_steps_per_second": 0.944, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07221438645980253, | |
| "grad_norm": 15.310386657714844, | |
| "learning_rate": 9.931447007857432e-05, | |
| "loss": 2.0835, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "grad_norm": 14.485530853271484, | |
| "learning_rate": 9.901383178135113e-05, | |
| "loss": 2.0385, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "eval_loss": 2.0639231204986572, | |
| "eval_runtime": 347.075, | |
| "eval_samples_per_second": 0.709, | |
| "eval_steps_per_second": 0.709, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09284706830546041, | |
| "grad_norm": 23.205764770507812, | |
| "learning_rate": 9.865931440351337e-05, | |
| "loss": 2.0658, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "grad_norm": 20.585407257080078, | |
| "learning_rate": 9.825130765281668e-05, | |
| "loss": 2.1223, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "eval_loss": 2.083315849304199, | |
| "eval_runtime": 244.456, | |
| "eval_samples_per_second": 1.006, | |
| "eval_steps_per_second": 1.006, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11347975015111827, | |
| "grad_norm": 23.029006958007812, | |
| "learning_rate": 9.779026003589304e-05, | |
| "loss": 2.0669, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "grad_norm": 27.581724166870117, | |
| "learning_rate": 9.727667836522407e-05, | |
| "loss": 2.0677, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "eval_loss": 2.0909621715545654, | |
| "eval_runtime": 258.3491, | |
| "eval_samples_per_second": 0.952, | |
| "eval_steps_per_second": 0.952, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13411243199677614, | |
| "grad_norm": 20.641569137573242, | |
| "learning_rate": 9.6711127202021e-05, | |
| "loss": 2.118, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "grad_norm": 15.354959487915039, | |
| "learning_rate": 9.609422823562345e-05, | |
| "loss": 2.0729, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "eval_loss": 2.0872063636779785, | |
| "eval_runtime": 366.7103, | |
| "eval_samples_per_second": 0.671, | |
| "eval_steps_per_second": 0.671, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.154745113842434, | |
| "grad_norm": 13.397254943847656, | |
| "learning_rate": 9.542665960009959e-05, | |
| "loss": 2.0806, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "grad_norm": 14.432162284851074, | |
| "learning_rate": 9.470915512879852e-05, | |
| "loss": 2.1197, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "eval_loss": 2.0972707271575928, | |
| "eval_runtime": 241.6226, | |
| "eval_samples_per_second": 1.018, | |
| "eval_steps_per_second": 1.018, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1753777956880919, | |
| "grad_norm": 16.219799041748047, | |
| "learning_rate": 9.394250354767467e-05, | |
| "loss": 2.1002, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "grad_norm": 19.845632553100586, | |
| "learning_rate": 9.312754760827061e-05, | |
| "loss": 2.1053, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "eval_loss": 2.091938018798828, | |
| "eval_runtime": 374.7664, | |
| "eval_samples_per_second": 0.656, | |
| "eval_steps_per_second": 0.656, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19601047753374976, | |
| "grad_norm": 19.960819244384766, | |
| "learning_rate": 9.226518316131176e-05, | |
| "loss": 2.1111, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "grad_norm": 22.90218734741211, | |
| "learning_rate": 9.1356358171931e-05, | |
| "loss": 2.0848, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "eval_loss": 2.1034836769104004, | |
| "eval_runtime": 250.3178, | |
| "eval_samples_per_second": 0.983, | |
| "eval_steps_per_second": 0.983, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21664315937940762, | |
| "grad_norm": 20.233325958251953, | |
| "learning_rate": 9.040207167760586e-05, | |
| "loss": 2.0845, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "grad_norm": 18.64703941345215, | |
| "learning_rate": 8.940337268995385e-05, | |
| "loss": 2.1015, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "eval_loss": 2.1114280223846436, | |
| "eval_runtime": 378.994, | |
| "eval_samples_per_second": 0.649, | |
| "eval_steps_per_second": 0.649, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23727584122506548, | |
| "grad_norm": 12.925492286682129, | |
| "learning_rate": 8.836135904159302e-05, | |
| "loss": 2.1163, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "grad_norm": 17.775686264038086, | |
| "learning_rate": 8.727717617933544e-05, | |
| "loss": 2.0872, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "eval_loss": 2.113255023956299, | |
| "eval_runtime": 253.955, | |
| "eval_samples_per_second": 0.969, | |
| "eval_steps_per_second": 0.969, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.25790852307072337, | |
| "grad_norm": 13.843049049377441, | |
| "learning_rate": 8.615201590504017e-05, | |
| "loss": 2.1374, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "grad_norm": 17.85464859008789, | |
| "learning_rate": 8.498711506550983e-05, | |
| "loss": 2.0948, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "eval_loss": 2.1220598220825195, | |
| "eval_runtime": 250.0959, | |
| "eval_samples_per_second": 0.984, | |
| "eval_steps_per_second": 0.984, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27854120491638124, | |
| "grad_norm": 14.508893966674805, | |
| "learning_rate": 8.378375419287099e-05, | |
| "loss": 2.1539, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "grad_norm": 14.467598915100098, | |
| "learning_rate": 8.25432560969328e-05, | |
| "loss": 2.097, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "eval_loss": 2.1218836307525635, | |
| "eval_runtime": 366.7235, | |
| "eval_samples_per_second": 0.671, | |
| "eval_steps_per_second": 0.671, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2991738867620391, | |
| "grad_norm": 25.095067977905273, | |
| "learning_rate": 8.126698441107146e-05, | |
| "loss": 2.1127, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "grad_norm": 16.594051361083984, | |
| "learning_rate": 7.995634209323886e-05, | |
| "loss": 2.147, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "eval_loss": 2.1240272521972656, | |
| "eval_runtime": 255.4843, | |
| "eval_samples_per_second": 0.963, | |
| "eval_steps_per_second": 0.963, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31980656860769696, | |
| "grad_norm": 14.561055183410645, | |
| "learning_rate": 7.861276988374302e-05, | |
| "loss": 2.1397, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "grad_norm": 16.39865493774414, | |
| "learning_rate": 7.723774472149601e-05, | |
| "loss": 2.1315, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "eval_loss": 2.1188597679138184, | |
| "eval_runtime": 389.6053, | |
| "eval_samples_per_second": 0.631, | |
| "eval_steps_per_second": 0.631, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3404392504533548, | |
| "grad_norm": 15.670085906982422, | |
| "learning_rate": 7.583277812046993e-05, | |
| "loss": 2.1466, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "grad_norm": 22.689796447753906, | |
| "learning_rate": 7.439941450814591e-05, | |
| "loss": 2.1563, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "eval_loss": 2.136768341064453, | |
| "eval_runtime": 252.2156, | |
| "eval_samples_per_second": 0.975, | |
| "eval_steps_per_second": 0.975, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3610719322990127, | |
| "grad_norm": 18.863473892211914, | |
| "learning_rate": 7.293922952778239e-05, | |
| "loss": 2.1586, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "grad_norm": 17.896759033203125, | |
| "learning_rate": 7.145382830636924e-05, | |
| "loss": 2.1836, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "eval_loss": 2.127108335494995, | |
| "eval_runtime": 364.7213, | |
| "eval_samples_per_second": 0.674, | |
| "eval_steps_per_second": 0.674, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.38170461414467055, | |
| "grad_norm": 21.21641731262207, | |
| "learning_rate": 6.994484369017143e-05, | |
| "loss": 2.1371, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "grad_norm": 15.671196937561035, | |
| "learning_rate": 6.841393444980177e-05, | |
| "loss": 2.1245, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "eval_loss": 2.1197926998138428, | |
| "eval_runtime": 395.7933, | |
| "eval_samples_per_second": 0.622, | |
| "eval_steps_per_second": 0.622, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4023372959903284, | |
| "grad_norm": 16.22295379638672, | |
| "learning_rate": 6.686278345679625e-05, | |
| "loss": 2.1008, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "grad_norm": 14.019622802734375, | |
| "learning_rate": 6.529309583369605e-05, | |
| "loss": 2.0947, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "eval_loss": 2.123974561691284, | |
| "eval_runtime": 257.3823, | |
| "eval_samples_per_second": 0.956, | |
| "eval_steps_per_second": 0.956, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4229699778359863, | |
| "grad_norm": 18.07195281982422, | |
| "learning_rate": 6.370659707966967e-05, | |
| "loss": 2.1187, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "grad_norm": 19.1149959564209, | |
| "learning_rate": 6.2105031173736e-05, | |
| "loss": 2.1472, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "eval_loss": 2.13541579246521, | |
| "eval_runtime": 358.8473, | |
| "eval_samples_per_second": 0.686, | |
| "eval_steps_per_second": 0.686, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.44360265968164414, | |
| "grad_norm": 17.229467391967773, | |
| "learning_rate": 6.049015865767318e-05, | |
| "loss": 2.0992, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "grad_norm": 14.356927871704102, | |
| "learning_rate": 5.88637547007204e-05, | |
| "loss": 2.1348, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "eval_loss": 2.1260793209075928, | |
| "eval_runtime": 248.3974, | |
| "eval_samples_per_second": 0.99, | |
| "eval_steps_per_second": 0.99, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46423534152730206, | |
| "grad_norm": 16.18288803100586, | |
| "learning_rate": 5.722760714820057e-05, | |
| "loss": 2.1451, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "grad_norm": 16.129526138305664, | |
| "learning_rate": 5.5583514556208514e-05, | |
| "loss": 2.1099, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "eval_loss": 2.127509355545044, | |
| "eval_runtime": 357.9954, | |
| "eval_samples_per_second": 0.687, | |
| "eval_steps_per_second": 0.687, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4848680233729599, | |
| "grad_norm": 17.3072509765625, | |
| "learning_rate": 5.393328421452514e-05, | |
| "loss": 2.108, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "grad_norm": 15.436331748962402, | |
| "learning_rate": 5.2278730159931076e-05, | |
| "loss": 2.1006, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "eval_loss": 2.1196277141571045, | |
| "eval_runtime": 266.5248, | |
| "eval_samples_per_second": 0.923, | |
| "eval_steps_per_second": 0.923, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5055007052186178, | |
| "grad_norm": 13.889219284057617, | |
| "learning_rate": 5.062167118210367e-05, | |
| "loss": 2.1588, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "grad_norm": 15.053701400756836, | |
| "learning_rate": 4.896392882428901e-05, | |
| "loss": 2.1339, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "eval_loss": 2.1169850826263428, | |
| "eval_runtime": 243.5045, | |
| "eval_samples_per_second": 1.01, | |
| "eval_steps_per_second": 1.01, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5261333870642756, | |
| "grad_norm": 15.929015159606934, | |
| "learning_rate": 4.730732538094749e-05, | |
| "loss": 2.1116, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "grad_norm": 13.288613319396973, | |
| "learning_rate": 4.565368189457313e-05, | |
| "loss": 2.0841, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "eval_loss": 2.1104753017425537, | |
| "eval_runtime": 365.3009, | |
| "eval_samples_per_second": 0.673, | |
| "eval_steps_per_second": 0.673, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5467660689099335, | |
| "grad_norm": 15.282868385314941, | |
| "learning_rate": 4.400481615388948e-05, | |
| "loss": 2.0709, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "grad_norm": 16.86932945251465, | |
| "learning_rate": 4.236254069562213e-05, | |
| "loss": 2.1344, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "eval_loss": 2.1079087257385254, | |
| "eval_runtime": 251.0859, | |
| "eval_samples_per_second": 0.98, | |
| "eval_steps_per_second": 0.98, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5673987507555914, | |
| "grad_norm": 15.028175354003906, | |
| "learning_rate": 4.0728660812044536e-05, | |
| "loss": 2.1397, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "grad_norm": 14.224735260009766, | |
| "learning_rate": 3.910497256648742e-05, | |
| "loss": 2.0732, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "eval_loss": 2.104301929473877, | |
| "eval_runtime": 366.8962, | |
| "eval_samples_per_second": 0.67, | |
| "eval_steps_per_second": 0.67, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5880314326012492, | |
| "grad_norm": 15.107568740844727, | |
| "learning_rate": 3.749326081899329e-05, | |
| "loss": 2.1003, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "grad_norm": 12.49576473236084, | |
| "learning_rate": 3.589529726428615e-05, | |
| "loss": 2.0417, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "eval_loss": 2.103517532348633, | |
| "eval_runtime": 243.1722, | |
| "eval_samples_per_second": 1.012, | |
| "eval_steps_per_second": 1.012, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6086641144469072, | |
| "grad_norm": 10.86776065826416, | |
| "learning_rate": 3.431283848421347e-05, | |
| "loss": 2.1209, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "grad_norm": 15.035432815551758, | |
| "learning_rate": 3.274762401680124e-05, | |
| "loss": 2.1003, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "eval_loss": 2.0966618061065674, | |
| "eval_runtime": 354.4771, | |
| "eval_samples_per_second": 0.694, | |
| "eval_steps_per_second": 0.694, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.629296796292565, | |
| "grad_norm": 19.571407318115234, | |
| "learning_rate": 3.120137444404442e-05, | |
| "loss": 2.0631, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "grad_norm": 12.437437057495117, | |
| "learning_rate": 2.9675789500535328e-05, | |
| "loss": 2.0501, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "eval_loss": 2.1006603240966797, | |
| "eval_runtime": 242.766, | |
| "eval_samples_per_second": 1.013, | |
| "eval_steps_per_second": 1.013, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6499294781382229, | |
| "grad_norm": 14.765650749206543, | |
| "learning_rate": 2.8172546205008683e-05, | |
| "loss": 2.1215, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "grad_norm": 12.97276496887207, | |
| "learning_rate": 2.6693297016857188e-05, | |
| "loss": 2.078, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "eval_loss": 2.0861904621124268, | |
| "eval_runtime": 251.7783, | |
| "eval_samples_per_second": 0.977, | |
| "eval_steps_per_second": 0.977, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6705621599838807, | |
| "grad_norm": 13.271310806274414, | |
| "learning_rate": 2.523966801964468e-05, | |
| "loss": 2.0733, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "grad_norm": 13.07889461517334, | |
| "learning_rate": 2.3813257133612827e-05, | |
| "loss": 2.0507, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "eval_loss": 2.084022045135498, | |
| "eval_runtime": 351.4005, | |
| "eval_samples_per_second": 0.7, | |
| "eval_steps_per_second": 0.7, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6911948418295386, | |
| "grad_norm": 12.636960983276367, | |
| "learning_rate": 2.2415632359146856e-05, | |
| "loss": 2.0855, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "grad_norm": 16.52528190612793, | |
| "learning_rate": 2.104833005313131e-05, | |
| "loss": 2.0235, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "eval_loss": 2.076244354248047, | |
| "eval_runtime": 240.3826, | |
| "eval_samples_per_second": 1.023, | |
| "eval_steps_per_second": 1.023, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7118275236751964, | |
| "grad_norm": 11.037178039550781, | |
| "learning_rate": 1.971285324008994e-05, | |
| "loss": 2.1384, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "grad_norm": 13.414417266845703, | |
| "learning_rate": 1.84106699599668e-05, | |
| "loss": 2.0743, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "eval_loss": 2.072314500808716, | |
| "eval_runtime": 366.894, | |
| "eval_samples_per_second": 0.67, | |
| "eval_steps_per_second": 0.67, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7324602055208543, | |
| "grad_norm": 12.51091480255127, | |
| "learning_rate": 1.7143211654364762e-05, | |
| "loss": 2.0863, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "grad_norm": 15.719615936279297, | |
| "learning_rate": 1.5911871593014837e-05, | |
| "loss": 2.1028, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "eval_loss": 2.0720996856689453, | |
| "eval_runtime": 240.5658, | |
| "eval_samples_per_second": 1.023, | |
| "eval_steps_per_second": 1.023, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7530928873665121, | |
| "grad_norm": 11.165517807006836, | |
| "learning_rate": 1.4718003342206722e-05, | |
| "loss": 2.0551, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "grad_norm": 13.10633659362793, | |
| "learning_rate": 1.3562919276863844e-05, | |
| "loss": 2.0987, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "eval_loss": 2.066223382949829, | |
| "eval_runtime": 248.9943, | |
| "eval_samples_per_second": 0.988, | |
| "eval_steps_per_second": 0.988, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7737255692121701, | |
| "grad_norm": 15.164176940917969, | |
| "learning_rate": 1.2447889137898293e-05, | |
| "loss": 2.113, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "grad_norm": 13.534417152404785, | |
| "learning_rate": 1.1374138636432053e-05, | |
| "loss": 2.0985, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "eval_loss": 2.066317558288574, | |
| "eval_runtime": 373.4925, | |
| "eval_samples_per_second": 0.659, | |
| "eval_steps_per_second": 0.659, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.794358251057828, | |
| "grad_norm": 11.72275161743164, | |
| "learning_rate": 1.0342848106418368e-05, | |
| "loss": 2.0898, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "grad_norm": 12.691418647766113, | |
| "learning_rate": 9.35515120714447e-06, | |
| "loss": 2.0548, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "eval_loss": 2.0601634979248047, | |
| "eval_runtime": 252.2425, | |
| "eval_samples_per_second": 0.975, | |
| "eval_steps_per_second": 0.975, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8149909329034858, | |
| "grad_norm": 9.275248527526855, | |
| "learning_rate": 8.41213367704224e-06, | |
| "loss": 2.0907, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "grad_norm": 13.456718444824219, | |
| "learning_rate": 7.51483214017637e-06, | |
| "loss": 2.0365, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "eval_loss": 2.0563478469848633, | |
| "eval_runtime": 378.3713, | |
| "eval_samples_per_second": 0.65, | |
| "eval_steps_per_second": 0.65, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8356236147491437, | |
| "grad_norm": 10.312747955322266, | |
| "learning_rate": 6.664232966721995e-06, | |
| "loss": 2.0537, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "grad_norm": 11.122403144836426, | |
| "learning_rate": 5.8612711886848196e-06, | |
| "loss": 2.0102, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "eval_loss": 2.056385040283203, | |
| "eval_runtime": 247.0489, | |
| "eval_samples_per_second": 0.996, | |
| "eval_steps_per_second": 0.996, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8562562965948015, | |
| "grad_norm": 12.165043830871582, | |
| "learning_rate": 5.106829472055202e-06, | |
| "loss": 2.0219, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "grad_norm": 14.259943008422852, | |
| "learning_rate": 4.401737146526219e-06, | |
| "loss": 2.0497, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "eval_loss": 2.0522236824035645, | |
| "eval_runtime": 366.3478, | |
| "eval_samples_per_second": 0.671, | |
| "eval_steps_per_second": 0.671, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8768889784404594, | |
| "grad_norm": 12.8046293258667, | |
| "learning_rate": 3.7467692938425057e-06, | |
| "loss": 2.0136, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "grad_norm": 15.552796363830566, | |
| "learning_rate": 3.142645895781715e-06, | |
| "loss": 2.0721, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "eval_loss": 2.0470504760742188, | |
| "eval_runtime": 245.8972, | |
| "eval_samples_per_second": 1.0, | |
| "eval_steps_per_second": 1.0, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8975216602861172, | |
| "grad_norm": 10.629472732543945, | |
| "learning_rate": 2.5900310427053044e-06, | |
| "loss": 2.0459, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "grad_norm": 12.79671859741211, | |
| "learning_rate": 2.089532203548794e-06, | |
| "loss": 2.0812, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "eval_loss": 2.0468461513519287, | |
| "eval_runtime": 246.3251, | |
| "eval_samples_per_second": 0.999, | |
| "eval_steps_per_second": 0.999, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9181543421317752, | |
| "grad_norm": 10.172734260559082, | |
| "learning_rate": 1.6416995580537664e-06, | |
| "loss": 2.0588, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "grad_norm": 13.063249588012695, | |
| "learning_rate": 1.247025391975698e-06, | |
| "loss": 2.0475, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "eval_loss": 2.0462241172790527, | |
| "eval_runtime": 383.7451, | |
| "eval_samples_per_second": 0.641, | |
| "eval_steps_per_second": 0.641, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.938787023977433, | |
| "grad_norm": 9.2144193649292, | |
| "learning_rate": 9.059435559326257e-07, | |
| "loss": 2.0669, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "grad_norm": 11.035449028015137, | |
| "learning_rate": 6.188289884893062e-07, | |
| "loss": 2.0687, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "eval_loss": 2.045191526412964, | |
| "eval_runtime": 250.014, | |
| "eval_samples_per_second": 0.984, | |
| "eval_steps_per_second": 0.984, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9594197058230909, | |
| "grad_norm": 11.38818073272705, | |
| "learning_rate": 3.8599730400115107e-07, | |
| "loss": 2.0007, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "grad_norm": 11.021515846252441, | |
| "learning_rate": 2.0770444567118075e-07, | |
| "loss": 2.065, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "eval_loss": 2.044952392578125, | |
| "eval_runtime": 364.9331, | |
| "eval_samples_per_second": 0.674, | |
| "eval_steps_per_second": 0.674, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9800523876687487, | |
| "grad_norm": 14.076632499694824, | |
| "learning_rate": 8.414640420116305e-08, | |
| "loss": 2.0618, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "grad_norm": 15.729289054870605, | |
| "learning_rate": 1.5459002346324135e-08, | |
| "loss": 1.991, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "eval_loss": 2.04483962059021, | |
| "eval_runtime": 253.6132, | |
| "eval_samples_per_second": 0.97, | |
| "eval_steps_per_second": 0.97, | |
| "step": 384 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 387, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.5374998322774016e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |