| { | |
| "best_metric": 2.031247138977051, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_reverse_r32/checkpoint-8", | |
| "epoch": 0.9903687285915777, | |
| "eval_steps": 8, | |
| "global_step": 384, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025790852307072333, | |
| "grad_norm": 23.795961380004883, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.2237, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010316340922828933, | |
| "grad_norm": 10.762740135192871, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0557, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "grad_norm": 12.864238739013672, | |
| "learning_rate": 0.0001, | |
| "loss": 1.991, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "eval_loss": 2.031247138977051, | |
| "eval_runtime": 50.8376, | |
| "eval_samples_per_second": 4.839, | |
| "eval_steps_per_second": 4.839, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0309490227684868, | |
| "grad_norm": 20.236906051635742, | |
| "learning_rate": 9.997251843068762e-05, | |
| "loss": 2.0347, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "grad_norm": 11.911772727966309, | |
| "learning_rate": 9.989010393221656e-05, | |
| "loss": 2.0461, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "eval_loss": 2.0334763526916504, | |
| "eval_runtime": 69.6741, | |
| "eval_samples_per_second": 3.531, | |
| "eval_steps_per_second": 3.531, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05158170461414467, | |
| "grad_norm": 13.50040340423584, | |
| "learning_rate": 9.97528470997769e-05, | |
| "loss": 1.995, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "grad_norm": 12.473864555358887, | |
| "learning_rate": 9.956089881469482e-05, | |
| "loss": 2.0456, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "eval_loss": 2.060091018676758, | |
| "eval_runtime": 49.4489, | |
| "eval_samples_per_second": 4.975, | |
| "eval_steps_per_second": 4.975, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07221438645980253, | |
| "grad_norm": 16.583091735839844, | |
| "learning_rate": 9.931447007857432e-05, | |
| "loss": 2.0845, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "grad_norm": 38.525352478027344, | |
| "learning_rate": 9.901383178135113e-05, | |
| "loss": 2.0584, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "eval_loss": 2.087923765182495, | |
| "eval_runtime": 68.1674, | |
| "eval_samples_per_second": 3.609, | |
| "eval_steps_per_second": 3.609, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09284706830546041, | |
| "grad_norm": 18.212785720825195, | |
| "learning_rate": 9.865931440351337e-05, | |
| "loss": 2.0671, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "grad_norm": 18.464550018310547, | |
| "learning_rate": 9.825130765281668e-05, | |
| "loss": 2.1123, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "eval_loss": 2.080946922302246, | |
| "eval_runtime": 51.3512, | |
| "eval_samples_per_second": 4.791, | |
| "eval_steps_per_second": 4.791, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11347975015111827, | |
| "grad_norm": 18.683780670166016, | |
| "learning_rate": 9.779026003589304e-05, | |
| "loss": 2.0638, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "grad_norm": 20.039987564086914, | |
| "learning_rate": 9.727667836522407e-05, | |
| "loss": 2.0666, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "eval_loss": 2.0889713764190674, | |
| "eval_runtime": 49.6263, | |
| "eval_samples_per_second": 4.957, | |
| "eval_steps_per_second": 4.957, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13411243199677614, | |
| "grad_norm": 23.5998477935791, | |
| "learning_rate": 9.6711127202021e-05, | |
| "loss": 2.1188, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "grad_norm": 16.58580780029297, | |
| "learning_rate": 9.609422823562345e-05, | |
| "loss": 2.0733, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "eval_loss": 2.0954222679138184, | |
| "eval_runtime": 68.3777, | |
| "eval_samples_per_second": 3.598, | |
| "eval_steps_per_second": 3.598, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.154745113842434, | |
| "grad_norm": 17.798566818237305, | |
| "learning_rate": 9.542665960009959e-05, | |
| "loss": 2.0855, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "grad_norm": 16.25016212463379, | |
| "learning_rate": 9.470915512879852e-05, | |
| "loss": 2.1236, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "eval_loss": 2.0970985889434814, | |
| "eval_runtime": 50.5579, | |
| "eval_samples_per_second": 4.866, | |
| "eval_steps_per_second": 4.866, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1753777956880919, | |
| "grad_norm": 14.97128677368164, | |
| "learning_rate": 9.394250354767467e-05, | |
| "loss": 2.104, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "grad_norm": 18.22063636779785, | |
| "learning_rate": 9.312754760827061e-05, | |
| "loss": 2.1103, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "eval_loss": 2.1007986068725586, | |
| "eval_runtime": 68.15, | |
| "eval_samples_per_second": 3.61, | |
| "eval_steps_per_second": 3.61, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19601047753374976, | |
| "grad_norm": 15.88463306427002, | |
| "learning_rate": 9.226518316131176e-05, | |
| "loss": 2.1123, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "grad_norm": 16.19556999206543, | |
| "learning_rate": 9.1356358171931e-05, | |
| "loss": 2.0876, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "eval_loss": 2.104221820831299, | |
| "eval_runtime": 51.0209, | |
| "eval_samples_per_second": 4.822, | |
| "eval_steps_per_second": 4.822, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21664315937940762, | |
| "grad_norm": 21.194753646850586, | |
| "learning_rate": 9.040207167760586e-05, | |
| "loss": 2.0851, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "grad_norm": 16.5502986907959, | |
| "learning_rate": 8.940337268995385e-05, | |
| "loss": 2.1107, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "eval_loss": 2.115513324737549, | |
| "eval_runtime": 71.0739, | |
| "eval_samples_per_second": 3.461, | |
| "eval_steps_per_second": 3.461, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23727584122506548, | |
| "grad_norm": 16.366037368774414, | |
| "learning_rate": 8.836135904159302e-05, | |
| "loss": 2.1165, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "grad_norm": 19.07808494567871, | |
| "learning_rate": 8.727717617933544e-05, | |
| "loss": 2.0889, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "eval_loss": 2.1082587242126465, | |
| "eval_runtime": 51.4649, | |
| "eval_samples_per_second": 4.78, | |
| "eval_steps_per_second": 4.78, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.25790852307072337, | |
| "grad_norm": 16.868947982788086, | |
| "learning_rate": 8.615201590504017e-05, | |
| "loss": 2.1333, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "grad_norm": 18.22587013244629, | |
| "learning_rate": 8.498711506550983e-05, | |
| "loss": 2.097, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "eval_loss": 2.1185896396636963, | |
| "eval_runtime": 50.4577, | |
| "eval_samples_per_second": 4.875, | |
| "eval_steps_per_second": 4.875, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27854120491638124, | |
| "grad_norm": 17.453964233398438, | |
| "learning_rate": 8.378375419287099e-05, | |
| "loss": 2.1481, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "grad_norm": 15.08243465423584, | |
| "learning_rate": 8.25432560969328e-05, | |
| "loss": 2.0962, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "eval_loss": 2.120164394378662, | |
| "eval_runtime": 68.8529, | |
| "eval_samples_per_second": 3.573, | |
| "eval_steps_per_second": 3.573, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2991738867620391, | |
| "grad_norm": 15.695137977600098, | |
| "learning_rate": 8.126698441107146e-05, | |
| "loss": 2.1125, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "grad_norm": 28.060436248779297, | |
| "learning_rate": 7.995634209323886e-05, | |
| "loss": 2.1415, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "eval_loss": 2.1305339336395264, | |
| "eval_runtime": 50.6817, | |
| "eval_samples_per_second": 4.854, | |
| "eval_steps_per_second": 4.854, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31980656860769696, | |
| "grad_norm": 17.550668716430664, | |
| "learning_rate": 7.861276988374302e-05, | |
| "loss": 2.141, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "grad_norm": 20.923986434936523, | |
| "learning_rate": 7.723774472149601e-05, | |
| "loss": 2.1294, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "eval_loss": 2.1169350147247314, | |
| "eval_runtime": 68.8987, | |
| "eval_samples_per_second": 3.57, | |
| "eval_steps_per_second": 3.57, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3404392504533548, | |
| "grad_norm": 18.64177131652832, | |
| "learning_rate": 7.583277812046993e-05, | |
| "loss": 2.1473, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "grad_norm": 20.397233963012695, | |
| "learning_rate": 7.439941450814591e-05, | |
| "loss": 2.1476, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "eval_loss": 2.13000226020813, | |
| "eval_runtime": 52.251, | |
| "eval_samples_per_second": 4.708, | |
| "eval_steps_per_second": 4.708, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3610719322990127, | |
| "grad_norm": 14.290401458740234, | |
| "learning_rate": 7.293922952778239e-05, | |
| "loss": 2.1519, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "grad_norm": 19.956518173217773, | |
| "learning_rate": 7.145382830636924e-05, | |
| "loss": 2.1725, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "eval_loss": 2.124537467956543, | |
| "eval_runtime": 66.2352, | |
| "eval_samples_per_second": 3.714, | |
| "eval_steps_per_second": 3.714, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.38170461414467055, | |
| "grad_norm": 18.11756134033203, | |
| "learning_rate": 6.994484369017143e-05, | |
| "loss": 2.131, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "grad_norm": 15.948026657104492, | |
| "learning_rate": 6.841393444980177e-05, | |
| "loss": 2.1159, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "eval_loss": 2.117161273956299, | |
| "eval_runtime": 65.9204, | |
| "eval_samples_per_second": 3.732, | |
| "eval_steps_per_second": 3.732, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4023372959903284, | |
| "grad_norm": 17.677406311035156, | |
| "learning_rate": 6.686278345679625e-05, | |
| "loss": 2.0999, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "grad_norm": 18.509098052978516, | |
| "learning_rate": 6.529309583369605e-05, | |
| "loss": 2.0921, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "eval_loss": 2.122069835662842, | |
| "eval_runtime": 52.1961, | |
| "eval_samples_per_second": 4.713, | |
| "eval_steps_per_second": 4.713, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4229699778359863, | |
| "grad_norm": 19.927303314208984, | |
| "learning_rate": 6.370659707966967e-05, | |
| "loss": 2.1235, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "grad_norm": 17.947538375854492, | |
| "learning_rate": 6.2105031173736e-05, | |
| "loss": 2.141, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "eval_loss": 2.1334285736083984, | |
| "eval_runtime": 66.1797, | |
| "eval_samples_per_second": 3.717, | |
| "eval_steps_per_second": 3.717, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.44360265968164414, | |
| "grad_norm": 17.444217681884766, | |
| "learning_rate": 6.049015865767318e-05, | |
| "loss": 2.0906, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "grad_norm": 16.887006759643555, | |
| "learning_rate": 5.88637547007204e-05, | |
| "loss": 2.1312, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "eval_loss": 2.125943183898926, | |
| "eval_runtime": 50.8568, | |
| "eval_samples_per_second": 4.837, | |
| "eval_steps_per_second": 4.837, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46423534152730206, | |
| "grad_norm": 16.4300537109375, | |
| "learning_rate": 5.722760714820057e-05, | |
| "loss": 2.1426, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "grad_norm": 17.936267852783203, | |
| "learning_rate": 5.5583514556208514e-05, | |
| "loss": 2.106, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "eval_loss": 2.126936912536621, | |
| "eval_runtime": 72.4228, | |
| "eval_samples_per_second": 3.397, | |
| "eval_steps_per_second": 3.397, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4848680233729599, | |
| "grad_norm": 15.06416130065918, | |
| "learning_rate": 5.393328421452514e-05, | |
| "loss": 2.1054, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "grad_norm": 16.073657989501953, | |
| "learning_rate": 5.2278730159931076e-05, | |
| "loss": 2.1015, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "eval_loss": 2.1197259426116943, | |
| "eval_runtime": 50.2504, | |
| "eval_samples_per_second": 4.895, | |
| "eval_steps_per_second": 4.895, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5055007052186178, | |
| "grad_norm": 18.07906723022461, | |
| "learning_rate": 5.062167118210367e-05, | |
| "loss": 2.1588, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "grad_norm": 16.21997833251953, | |
| "learning_rate": 4.896392882428901e-05, | |
| "loss": 2.1368, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "eval_loss": 2.1163711547851562, | |
| "eval_runtime": 51.0039, | |
| "eval_samples_per_second": 4.823, | |
| "eval_steps_per_second": 4.823, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5261333870642756, | |
| "grad_norm": 17.681896209716797, | |
| "learning_rate": 4.730732538094749e-05, | |
| "loss": 2.1108, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "grad_norm": 14.08202838897705, | |
| "learning_rate": 4.565368189457313e-05, | |
| "loss": 2.0751, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "eval_loss": 2.1104061603546143, | |
| "eval_runtime": 70.8566, | |
| "eval_samples_per_second": 3.472, | |
| "eval_steps_per_second": 3.472, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5467660689099335, | |
| "grad_norm": 17.15287971496582, | |
| "learning_rate": 4.400481615388948e-05, | |
| "loss": 2.0737, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "grad_norm": 14.733731269836426, | |
| "learning_rate": 4.236254069562213e-05, | |
| "loss": 2.135, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "eval_loss": 2.1105358600616455, | |
| "eval_runtime": 50.3866, | |
| "eval_samples_per_second": 4.882, | |
| "eval_steps_per_second": 4.882, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5673987507555914, | |
| "grad_norm": 12.141791343688965, | |
| "learning_rate": 4.0728660812044536e-05, | |
| "loss": 2.1395, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "grad_norm": 13.20874309539795, | |
| "learning_rate": 3.910497256648742e-05, | |
| "loss": 2.0718, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "eval_loss": 2.100306749343872, | |
| "eval_runtime": 68.0529, | |
| "eval_samples_per_second": 3.615, | |
| "eval_steps_per_second": 3.615, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5880314326012492, | |
| "grad_norm": 14.548846244812012, | |
| "learning_rate": 3.749326081899329e-05, | |
| "loss": 2.0944, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "grad_norm": 15.402515411376953, | |
| "learning_rate": 3.589529726428615e-05, | |
| "loss": 2.0393, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "eval_loss": 2.1025285720825195, | |
| "eval_runtime": 51.9996, | |
| "eval_samples_per_second": 4.731, | |
| "eval_steps_per_second": 4.731, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6086641144469072, | |
| "grad_norm": 13.062746047973633, | |
| "learning_rate": 3.431283848421347e-05, | |
| "loss": 2.1215, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "grad_norm": 16.821453094482422, | |
| "learning_rate": 3.274762401680124e-05, | |
| "loss": 2.1034, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "eval_loss": 2.0945794582366943, | |
| "eval_runtime": 69.0514, | |
| "eval_samples_per_second": 3.563, | |
| "eval_steps_per_second": 3.563, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.629296796292565, | |
| "grad_norm": 13.247776985168457, | |
| "learning_rate": 3.120137444404442e-05, | |
| "loss": 2.0619, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "grad_norm": 12.850509643554688, | |
| "learning_rate": 2.9675789500535328e-05, | |
| "loss": 2.045, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "eval_loss": 2.093926191329956, | |
| "eval_runtime": 49.7964, | |
| "eval_samples_per_second": 4.94, | |
| "eval_steps_per_second": 4.94, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6499294781382229, | |
| "grad_norm": 15.139781951904297, | |
| "learning_rate": 2.8172546205008683e-05, | |
| "loss": 2.1207, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "grad_norm": 13.269444465637207, | |
| "learning_rate": 2.6693297016857188e-05, | |
| "loss": 2.077, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "eval_loss": 2.081383466720581, | |
| "eval_runtime": 52.4381, | |
| "eval_samples_per_second": 4.691, | |
| "eval_steps_per_second": 4.691, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6705621599838807, | |
| "grad_norm": 11.657261848449707, | |
| "learning_rate": 2.523966801964468e-05, | |
| "loss": 2.0693, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "grad_norm": 11.285372734069824, | |
| "learning_rate": 2.3813257133612827e-05, | |
| "loss": 2.0514, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "eval_loss": 2.0800182819366455, | |
| "eval_runtime": 66.6605, | |
| "eval_samples_per_second": 3.69, | |
| "eval_steps_per_second": 3.69, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6911948418295386, | |
| "grad_norm": 13.049285888671875, | |
| "learning_rate": 2.2415632359146856e-05, | |
| "loss": 2.0855, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "grad_norm": 11.901713371276855, | |
| "learning_rate": 2.104833005313131e-05, | |
| "loss": 2.0222, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "eval_loss": 2.0774030685424805, | |
| "eval_runtime": 51.2508, | |
| "eval_samples_per_second": 4.8, | |
| "eval_steps_per_second": 4.8, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7118275236751964, | |
| "grad_norm": 12.342489242553711, | |
| "learning_rate": 1.971285324008994e-05, | |
| "loss": 2.1345, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "grad_norm": 13.865422248840332, | |
| "learning_rate": 1.84106699599668e-05, | |
| "loss": 2.075, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "eval_loss": 2.074871301651001, | |
| "eval_runtime": 63.1191, | |
| "eval_samples_per_second": 3.897, | |
| "eval_steps_per_second": 3.897, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7324602055208543, | |
| "grad_norm": 12.347137451171875, | |
| "learning_rate": 1.7143211654364762e-05, | |
| "loss": 2.0861, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "grad_norm": 15.943358421325684, | |
| "learning_rate": 1.5911871593014837e-05, | |
| "loss": 2.1013, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "eval_loss": 2.070469379425049, | |
| "eval_runtime": 50.0079, | |
| "eval_samples_per_second": 4.919, | |
| "eval_steps_per_second": 4.919, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7530928873665121, | |
| "grad_norm": 11.594297409057617, | |
| "learning_rate": 1.4718003342206722e-05, | |
| "loss": 2.053, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "grad_norm": 13.693614959716797, | |
| "learning_rate": 1.3562919276863844e-05, | |
| "loss": 2.0929, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "eval_loss": 2.064333438873291, | |
| "eval_runtime": 51.108, | |
| "eval_samples_per_second": 4.813, | |
| "eval_steps_per_second": 4.813, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7737255692121701, | |
| "grad_norm": 12.406514167785645, | |
| "learning_rate": 1.2447889137898293e-05, | |
| "loss": 2.1147, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "grad_norm": 14.044355392456055, | |
| "learning_rate": 1.1374138636432053e-05, | |
| "loss": 2.0996, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "eval_loss": 2.0692458152770996, | |
| "eval_runtime": 67.0744, | |
| "eval_samples_per_second": 3.668, | |
| "eval_steps_per_second": 3.668, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.794358251057828, | |
| "grad_norm": 10.247350692749023, | |
| "learning_rate": 1.0342848106418368e-05, | |
| "loss": 2.0892, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "grad_norm": 10.73900032043457, | |
| "learning_rate": 9.35515120714447e-06, | |
| "loss": 2.0507, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "eval_loss": 2.0587587356567383, | |
| "eval_runtime": 49.942, | |
| "eval_samples_per_second": 4.926, | |
| "eval_steps_per_second": 4.926, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8149909329034858, | |
| "grad_norm": 10.756080627441406, | |
| "learning_rate": 8.41213367704224e-06, | |
| "loss": 2.0867, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "grad_norm": 14.330180168151855, | |
| "learning_rate": 7.51483214017637e-06, | |
| "loss": 2.0353, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "eval_loss": 2.0574405193328857, | |
| "eval_runtime": 68.5693, | |
| "eval_samples_per_second": 3.588, | |
| "eval_steps_per_second": 3.588, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8356236147491437, | |
| "grad_norm": 10.09432601928711, | |
| "learning_rate": 6.664232966721995e-06, | |
| "loss": 2.0535, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "grad_norm": 12.156997680664062, | |
| "learning_rate": 5.8612711886848196e-06, | |
| "loss": 2.0128, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "eval_loss": 2.056994915008545, | |
| "eval_runtime": 49.8867, | |
| "eval_samples_per_second": 4.931, | |
| "eval_steps_per_second": 4.931, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8562562965948015, | |
| "grad_norm": 13.915254592895508, | |
| "learning_rate": 5.106829472055202e-06, | |
| "loss": 2.0233, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "grad_norm": 14.312435150146484, | |
| "learning_rate": 4.401737146526219e-06, | |
| "loss": 2.0508, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "eval_loss": 2.050326108932495, | |
| "eval_runtime": 67.6182, | |
| "eval_samples_per_second": 3.638, | |
| "eval_steps_per_second": 3.638, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8768889784404594, | |
| "grad_norm": 12.76665210723877, | |
| "learning_rate": 3.7467692938425057e-06, | |
| "loss": 2.0105, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "grad_norm": 17.83547019958496, | |
| "learning_rate": 3.142645895781715e-06, | |
| "loss": 2.067, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "eval_loss": 2.0471653938293457, | |
| "eval_runtime": 157.8315, | |
| "eval_samples_per_second": 1.559, | |
| "eval_steps_per_second": 1.559, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8975216602861172, | |
| "grad_norm": 10.910913467407227, | |
| "learning_rate": 2.5900310427053044e-06, | |
| "loss": 2.0453, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "grad_norm": 9.634235382080078, | |
| "learning_rate": 2.089532203548794e-06, | |
| "loss": 2.0821, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "eval_loss": 2.047574281692505, | |
| "eval_runtime": 229.272, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 1.073, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9181543421317752, | |
| "grad_norm": 11.296228408813477, | |
| "learning_rate": 1.6416995580537664e-06, | |
| "loss": 2.0581, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "grad_norm": 12.595250129699707, | |
| "learning_rate": 1.247025391975698e-06, | |
| "loss": 2.0461, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "eval_loss": 2.04707932472229, | |
| "eval_runtime": 278.4472, | |
| "eval_samples_per_second": 0.883, | |
| "eval_steps_per_second": 0.883, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.938787023977433, | |
| "grad_norm": 11.871248245239258, | |
| "learning_rate": 9.059435559326257e-07, | |
| "loss": 2.0658, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "grad_norm": 11.920108795166016, | |
| "learning_rate": 6.188289884893062e-07, | |
| "loss": 2.0666, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "eval_loss": 2.046149492263794, | |
| "eval_runtime": 241.2948, | |
| "eval_samples_per_second": 1.019, | |
| "eval_steps_per_second": 1.019, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9594197058230909, | |
| "grad_norm": 12.083566665649414, | |
| "learning_rate": 3.8599730400115107e-07, | |
| "loss": 2.0008, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "grad_norm": 11.261472702026367, | |
| "learning_rate": 2.0770444567118075e-07, | |
| "loss": 2.0639, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "eval_loss": 2.0458359718322754, | |
| "eval_runtime": 377.6181, | |
| "eval_samples_per_second": 0.651, | |
| "eval_steps_per_second": 0.651, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9800523876687487, | |
| "grad_norm": 10.003210067749023, | |
| "learning_rate": 8.414640420116305e-08, | |
| "loss": 2.0595, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "grad_norm": 9.608034133911133, | |
| "learning_rate": 1.5459002346324135e-08, | |
| "loss": 1.9859, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "eval_loss": 2.04579496383667, | |
| "eval_runtime": 243.787, | |
| "eval_samples_per_second": 1.009, | |
| "eval_steps_per_second": 1.009, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.501033446014976e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |