| { | |
| "best_metric": 0.05613677854233956, | |
| "best_model_checkpoint": "./whisper-large-v3-turbo-finetuned-lora/checkpoint-500", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 11.033143997192383, | |
| "learning_rate": 8.333333333333334e-07, | |
| "loss": 3.7365, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 10.168129920959473, | |
| "learning_rate": 1.875e-06, | |
| "loss": 3.6756, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 10.131425857543945, | |
| "learning_rate": 2.916666666666667e-06, | |
| "loss": 3.6681, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 9.962166786193848, | |
| "learning_rate": 3.958333333333334e-06, | |
| "loss": 3.6567, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 9.637451171875, | |
| "learning_rate": 4.791666666666667e-06, | |
| "loss": 3.5903, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 9.27942943572998, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 3.4592, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 9.690427780151367, | |
| "learning_rate": 6.875e-06, | |
| "loss": 3.299, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 8.123926162719727, | |
| "learning_rate": 7.916666666666668e-06, | |
| "loss": 3.2058, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 6.938026428222656, | |
| "learning_rate": 8.958333333333334e-06, | |
| "loss": 3.0613, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 6.615925312042236, | |
| "learning_rate": 1e-05, | |
| "loss": 2.8859, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 5.712332725524902, | |
| "learning_rate": 1.1041666666666666e-05, | |
| "loss": 2.6746, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 4.229877471923828, | |
| "learning_rate": 1.2083333333333333e-05, | |
| "loss": 2.4948, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 3.9951322078704834, | |
| "learning_rate": 1.3125e-05, | |
| "loss": 2.3496, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 4.010512351989746, | |
| "learning_rate": 1.4166666666666666e-05, | |
| "loss": 2.2345, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 3.2869503498077393, | |
| "learning_rate": 1.5208333333333335e-05, | |
| "loss": 2.0418, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.47694993019104, | |
| "learning_rate": 1.6250000000000002e-05, | |
| "loss": 1.8212, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.761810779571533, | |
| "learning_rate": 1.7291666666666666e-05, | |
| "loss": 1.7471, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.83661150932312, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 1.6647, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 2.7371621131896973, | |
| "learning_rate": 1.9375e-05, | |
| "loss": 1.5239, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.5980722904205322, | |
| "learning_rate": 2.0416666666666667e-05, | |
| "loss": 1.3501, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 2.8566689491271973, | |
| "learning_rate": 2.1458333333333334e-05, | |
| "loss": 1.3153, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 2.052793264389038, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 1.175, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 2.08168363571167, | |
| "learning_rate": 2.3541666666666665e-05, | |
| "loss": 1.0395, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.830390453338623, | |
| "learning_rate": 2.4583333333333332e-05, | |
| "loss": 0.9517, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.9608633518218994, | |
| "learning_rate": 2.5625e-05, | |
| "loss": 0.8602, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 2.2221925258636475, | |
| "learning_rate": 2.666666666666667e-05, | |
| "loss": 0.7687, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.185246706008911, | |
| "learning_rate": 2.7708333333333334e-05, | |
| "loss": 0.6957, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 2.3892769813537598, | |
| "learning_rate": 2.875e-05, | |
| "loss": 0.6452, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 2.2923471927642822, | |
| "learning_rate": 2.9791666666666668e-05, | |
| "loss": 0.6001, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.059990882873535, | |
| "learning_rate": 3.0833333333333335e-05, | |
| "loss": 0.5505, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 2.1722524166107178, | |
| "learning_rate": 3.1875e-05, | |
| "loss": 0.513, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.4212610721588135, | |
| "learning_rate": 3.291666666666666e-05, | |
| "loss": 0.4868, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 2.297727584838867, | |
| "learning_rate": 3.3958333333333337e-05, | |
| "loss": 0.4696, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.7377690076828003, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 0.4174, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.821341872215271, | |
| "learning_rate": 3.6041666666666664e-05, | |
| "loss": 0.4204, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.0993902683258057, | |
| "learning_rate": 3.708333333333334e-05, | |
| "loss": 0.3846, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 2.5224227905273438, | |
| "learning_rate": 3.8125e-05, | |
| "loss": 0.3499, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.8540211915969849, | |
| "learning_rate": 3.916666666666667e-05, | |
| "loss": 0.3414, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.9813562631607056, | |
| "learning_rate": 4.020833333333333e-05, | |
| "loss": 0.3274, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.385871171951294, | |
| "learning_rate": 4.125e-05, | |
| "loss": 0.2907, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 2.0511081218719482, | |
| "learning_rate": 4.229166666666667e-05, | |
| "loss": 0.2735, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 2.0850329399108887, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.2384, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 2.191450595855713, | |
| "learning_rate": 4.4375e-05, | |
| "loss": 0.2244, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 3.4809000492095947, | |
| "learning_rate": 4.541666666666667e-05, | |
| "loss": 0.223, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.4290976524353027, | |
| "learning_rate": 4.645833333333333e-05, | |
| "loss": 0.194, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.8528721332550049, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.1817, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 1.4630467891693115, | |
| "learning_rate": 4.854166666666666e-05, | |
| "loss": 0.1728, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6305458545684814, | |
| "learning_rate": 4.958333333333334e-05, | |
| "loss": 0.1859, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 1.455244779586792, | |
| "learning_rate": 5.0625000000000004e-05, | |
| "loss": 0.1531, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.8049136400222778, | |
| "learning_rate": 5.1666666666666664e-05, | |
| "loss": 0.1412, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 1.577805519104004, | |
| "learning_rate": 5.270833333333334e-05, | |
| "loss": 0.1384, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.020803213119507, | |
| "learning_rate": 5.375e-05, | |
| "loss": 0.1398, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 1.4404263496398926, | |
| "learning_rate": 5.479166666666667e-05, | |
| "loss": 0.1484, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 2.497192621231079, | |
| "learning_rate": 5.583333333333333e-05, | |
| "loss": 0.1353, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.2407219409942627, | |
| "learning_rate": 5.6875e-05, | |
| "loss": 0.1433, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.4051156044006348, | |
| "learning_rate": 5.791666666666667e-05, | |
| "loss": 0.1181, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1400000000000001, | |
| "grad_norm": 1.4117530584335327, | |
| "learning_rate": 5.8958333333333334e-05, | |
| "loss": 0.1252, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 1.6360172033309937, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1233, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.225799560546875, | |
| "learning_rate": 6.104166666666667e-05, | |
| "loss": 0.1084, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.4874345064163208, | |
| "learning_rate": 6.208333333333333e-05, | |
| "loss": 0.125, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 1.3238331079483032, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1132, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 2.354384183883667, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0993, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 2.2216718196868896, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1325, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.026408076286316, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1035, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 2.0583767890930176, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1208, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.9303004741668701, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1119, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 1.7043157815933228, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1023, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 1.3245861530303955, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1008, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 1.541318655014038, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0977, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.8400285243988037, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0887, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 1.0839234590530396, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0912, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.269062876701355, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0959, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.0546581745147705, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0875, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 1.683465838432312, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0933, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.470189094543457, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0986, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.183585524559021, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0979, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 1.3886022567749023, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1031, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 1.105749487876892, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0995, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 1.0494953393936157, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0755, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.7028089761734009, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0981, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.5404858589172363, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0917, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 1.6659576892852783, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0891, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6600000000000001, | |
| "grad_norm": 1.353579044342041, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0889, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 2.1539247035980225, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.1092, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 1.4106309413909912, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0951, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 1.1167716979980469, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0899, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 1.2029541730880737, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0902, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 1.0979869365692139, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.089, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 1.1568419933319092, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0859, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.2472410202026367, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0877, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8199999999999998, | |
| "grad_norm": 1.2323497533798218, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0865, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 1.2814995050430298, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0831, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.8599999999999999, | |
| "grad_norm": 0.912714421749115, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0813, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.2273714542388916, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0884, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.5928541421890259, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0852, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9074931740760803, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0792, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.9795681834220886, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0781, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 1.4303114414215088, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0757, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.0313260555267334, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0881, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1294418573379517, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0785, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.05613677854233956, | |
| "eval_loss": 0.08199143409729004, | |
| "eval_runtime": 495.5562, | |
| "eval_samples_per_second": 2.018, | |
| "eval_steps_per_second": 0.504, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.376040517632e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |