| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9878213802435725, | |
| "eval_steps": 50, | |
| "global_step": 552, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02706359945872801, | |
| "grad_norm": 18.82761565702094, | |
| "learning_rate": 5e-07, | |
| "loss": 1.734, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05412719891745602, | |
| "grad_norm": 12.867825529466979, | |
| "learning_rate": 1e-06, | |
| "loss": 1.6027, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08119079837618404, | |
| "grad_norm": 7.146838469131762, | |
| "learning_rate": 9.997900331216397e-07, | |
| "loss": 1.2943, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10825439783491204, | |
| "grad_norm": 4.096985428996853, | |
| "learning_rate": 9.991603088309193e-07, | |
| "loss": 1.1214, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13531799729364005, | |
| "grad_norm": 4.030931956641265, | |
| "learning_rate": 9.981113560128126e-07, | |
| "loss": 1.0701, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16238159675236807, | |
| "grad_norm": 3.9631259966731793, | |
| "learning_rate": 9.966440556487147e-07, | |
| "loss": 1.0514, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18944519621109607, | |
| "grad_norm": 3.7182444654950806, | |
| "learning_rate": 9.947596400765342e-07, | |
| "loss": 0.9816, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2165087956698241, | |
| "grad_norm": 3.8553362221283574, | |
| "learning_rate": 9.924596919556916e-07, | |
| "loss": 0.9671, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2435723951285521, | |
| "grad_norm": 3.48008462855318, | |
| "learning_rate": 9.897461429378964e-07, | |
| "loss": 0.9488, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "grad_norm": 3.5713771414265274, | |
| "learning_rate": 9.866212720448147e-07, | |
| "loss": 0.9428, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "eval_loss": 0.9270144701004028, | |
| "eval_runtime": 93.2201, | |
| "eval_samples_per_second": 56.318, | |
| "eval_steps_per_second": 0.89, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2976995940460081, | |
| "grad_norm": 3.5829158119832303, | |
| "learning_rate": 9.830877037539935e-07, | |
| "loss": 0.9318, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.32476319350473615, | |
| "grad_norm": 3.4021832899344626, | |
| "learning_rate": 9.791484057946465e-07, | |
| "loss": 0.9079, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35182679296346414, | |
| "grad_norm": 3.391325595052156, | |
| "learning_rate": 9.748066866551555e-07, | |
| "loss": 0.8999, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.37889039242219213, | |
| "grad_norm": 3.515379759871202, | |
| "learning_rate": 9.700661928043785e-07, | |
| "loss": 0.9047, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4059539918809202, | |
| "grad_norm": 3.44913794411675, | |
| "learning_rate": 9.649309056290984e-07, | |
| "loss": 0.8954, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4330175913396482, | |
| "grad_norm": 3.5207352414980435, | |
| "learning_rate": 9.594051380901858e-07, | |
| "loss": 0.8984, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.46008119079837617, | |
| "grad_norm": 3.7404288010932287, | |
| "learning_rate": 9.534935311002834e-07, | |
| "loss": 0.9018, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.4871447902571042, | |
| "grad_norm": 3.435250970708313, | |
| "learning_rate": 9.472010496260544e-07, | |
| "loss": 0.888, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5142083897158322, | |
| "grad_norm": 3.5926141044891455, | |
| "learning_rate": 9.405329785182678e-07, | |
| "loss": 0.8828, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "grad_norm": 3.8742294578209013, | |
| "learning_rate": 9.334949180732244e-07, | |
| "loss": 0.8747, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "eval_loss": 0.8708707690238953, | |
| "eval_runtime": 92.2649, | |
| "eval_samples_per_second": 56.901, | |
| "eval_steps_per_second": 0.9, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5683355886332883, | |
| "grad_norm": 3.592933241886934, | |
| "learning_rate": 9.260927793292497e-07, | |
| "loss": 0.8665, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.5953991880920162, | |
| "grad_norm": 3.5073835258263855, | |
| "learning_rate": 9.183327791022047e-07, | |
| "loss": 0.8528, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6224627875507442, | |
| "grad_norm": 3.460111360598304, | |
| "learning_rate": 9.102214347641843e-07, | |
| "loss": 0.8865, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.6495263870094723, | |
| "grad_norm": 3.3741983776754854, | |
| "learning_rate": 9.017655587697883e-07, | |
| "loss": 0.8648, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6765899864682002, | |
| "grad_norm": 3.5787989483258675, | |
| "learning_rate": 8.929722529345623e-07, | |
| "loss": 0.8484, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7036535859269283, | |
| "grad_norm": 3.532632305089824, | |
| "learning_rate": 8.83848902470413e-07, | |
| "loss": 0.8496, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7307171853856563, | |
| "grad_norm": 3.800249509395461, | |
| "learning_rate": 8.744031697830088e-07, | |
| "loss": 0.8592, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.7577807848443843, | |
| "grad_norm": 3.471168235847642, | |
| "learning_rate": 8.646429880363746e-07, | |
| "loss": 0.8624, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7848443843031123, | |
| "grad_norm": 3.220664028161556, | |
| "learning_rate": 8.545765544900846e-07, | |
| "loss": 0.8516, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "grad_norm": 3.5792636736018637, | |
| "learning_rate": 8.442123236146508e-07, | |
| "loss": 0.8698, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "eval_loss": 0.8459715247154236, | |
| "eval_runtime": 92.3333, | |
| "eval_samples_per_second": 56.859, | |
| "eval_steps_per_second": 0.899, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8389715832205683, | |
| "grad_norm": 3.6444829594963184, | |
| "learning_rate": 8.33558999990887e-07, | |
| "loss": 0.844, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.8660351826792964, | |
| "grad_norm": 3.619543017726238, | |
| "learning_rate": 8.22625530999215e-07, | |
| "loss": 0.829, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8930987821380244, | |
| "grad_norm": 3.4305250980444866, | |
| "learning_rate": 8.114210993050502e-07, | |
| "loss": 0.859, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.9201623815967523, | |
| "grad_norm": 3.4515883823343816, | |
| "learning_rate": 7.999551151465791e-07, | |
| "loss": 0.8373, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9472259810554804, | |
| "grad_norm": 3.4061397273113494, | |
| "learning_rate": 7.88237208431406e-07, | |
| "loss": 0.8301, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9742895805142084, | |
| "grad_norm": 3.450097547042426, | |
| "learning_rate": 7.762772206487065e-07, | |
| "loss": 0.824, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0013531799729365, | |
| "grad_norm": 3.7734602033825495, | |
| "learning_rate": 7.640851966036805e-07, | |
| "loss": 0.8525, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.0284167794316643, | |
| "grad_norm": 3.540725853109944, | |
| "learning_rate": 7.516713759812464e-07, | |
| "loss": 0.778, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0554803788903924, | |
| "grad_norm": 3.5399554970338456, | |
| "learning_rate": 7.390461847460628e-07, | |
| "loss": 0.7658, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "grad_norm": 3.6301828104234843, | |
| "learning_rate": 7.262202263860988e-07, | |
| "loss": 0.7652, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "eval_loss": 0.833743155002594, | |
| "eval_runtime": 92.2579, | |
| "eval_samples_per_second": 56.906, | |
| "eval_steps_per_second": 0.9, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1096075778078485, | |
| "grad_norm": 3.458697572657953, | |
| "learning_rate": 7.1320427300711e-07, | |
| "loss": 0.7688, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.1366711772665765, | |
| "grad_norm": 3.48579357869599, | |
| "learning_rate": 7.000092562854959e-07, | |
| "loss": 0.762, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1637347767253043, | |
| "grad_norm": 3.364903450939617, | |
| "learning_rate": 6.866462582871401e-07, | |
| "loss": 0.7764, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.1907983761840324, | |
| "grad_norm": 3.6608165573448352, | |
| "learning_rate": 6.731265021599436e-07, | |
| "loss": 0.772, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2178619756427604, | |
| "grad_norm": 3.3685745523262587, | |
| "learning_rate": 6.594613427078674e-07, | |
| "loss": 0.7738, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.2449255751014885, | |
| "grad_norm": 3.8004909414967263, | |
| "learning_rate": 6.456622568544011e-07, | |
| "loss": 0.78, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2719891745602165, | |
| "grad_norm": 3.507999447214229, | |
| "learning_rate": 6.317408340034684e-07, | |
| "loss": 0.7832, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.2990527740189446, | |
| "grad_norm": 3.622195105708189, | |
| "learning_rate": 6.177087663058625e-07, | |
| "loss": 0.776, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.3261163734776726, | |
| "grad_norm": 3.5978192394759616, | |
| "learning_rate": 6.035778388393893e-07, | |
| "loss": 0.7556, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "grad_norm": 3.3885549115018554, | |
| "learning_rate": 5.893599197109624e-07, | |
| "loss": 0.7822, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "eval_loss": 0.8234151005744934, | |
| "eval_runtime": 92.2768, | |
| "eval_samples_per_second": 56.894, | |
| "eval_steps_per_second": 0.899, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3802435723951285, | |
| "grad_norm": 3.6453760206706898, | |
| "learning_rate": 5.750669500889666e-07, | |
| "loss": 0.7841, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.4073071718538566, | |
| "grad_norm": 3.6248461961460254, | |
| "learning_rate": 5.607109341742578e-07, | |
| "loss": 0.7684, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.4343707713125846, | |
| "grad_norm": 3.482156523315791, | |
| "learning_rate": 5.463039291182256e-07, | |
| "loss": 0.7704, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.4614343707713127, | |
| "grad_norm": 3.402395974373717, | |
| "learning_rate": 5.318580348963825e-07, | |
| "loss": 0.749, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4884979702300405, | |
| "grad_norm": 3.6300127193307237, | |
| "learning_rate": 5.173853841459877e-07, | |
| "loss": 0.7643, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.5155615696887685, | |
| "grad_norm": 3.699178765966641, | |
| "learning_rate": 5.028981319762399e-07, | |
| "loss": 0.7564, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.5426251691474966, | |
| "grad_norm": 3.7257683305221967, | |
| "learning_rate": 4.884084457595956e-07, | |
| "loss": 0.7512, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.5696887686062246, | |
| "grad_norm": 3.443950941720584, | |
| "learning_rate": 4.7392849491278817e-07, | |
| "loss": 0.7729, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5967523680649527, | |
| "grad_norm": 3.485674476973515, | |
| "learning_rate": 4.5947044067613e-07, | |
| "loss": 0.7479, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "grad_norm": 3.6693587030820574, | |
| "learning_rate": 4.4504642589968217e-07, | |
| "loss": 0.7628, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "eval_loss": 0.816413402557373, | |
| "eval_runtime": 92.3489, | |
| "eval_samples_per_second": 56.85, | |
| "eval_steps_per_second": 0.899, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6508795669824088, | |
| "grad_norm": 3.6447287635482084, | |
| "learning_rate": 4.3066856484486847e-07, | |
| "loss": 0.7534, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.6779431664411368, | |
| "grad_norm": 3.643734092885536, | |
| "learning_rate": 4.1634893301010165e-07, | |
| "loss": 0.7607, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.7050067658998647, | |
| "grad_norm": 3.6916619314799797, | |
| "learning_rate": 4.0209955698896445e-07, | |
| "loss": 0.7514, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.7320703653585927, | |
| "grad_norm": 3.6681332975803977, | |
| "learning_rate": 3.8793240436946385e-07, | |
| "loss": 0.7597, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.7591339648173205, | |
| "grad_norm": 3.6184885278786663, | |
| "learning_rate": 3.738593736828426e-07, | |
| "loss": 0.7649, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.7861975642760486, | |
| "grad_norm": 3.737975444433867, | |
| "learning_rate": 3.598922844103902e-07, | |
| "loss": 0.7599, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.8132611637347766, | |
| "grad_norm": 3.456980655495973, | |
| "learning_rate": 3.4604286705664397e-07, | |
| "loss": 0.7498, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.8403247631935047, | |
| "grad_norm": 3.479321682385595, | |
| "learning_rate": 3.323227532973193e-07, | |
| "loss": 0.7689, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.8673883626522327, | |
| "grad_norm": 3.670587230199258, | |
| "learning_rate": 3.187434662102434e-07, | |
| "loss": 0.7727, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "grad_norm": 3.659280365639458, | |
| "learning_rate": 3.0531641059749634e-07, | |
| "loss": 0.768, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "eval_loss": 0.810582160949707, | |
| "eval_runtime": 92.3521, | |
| "eval_samples_per_second": 56.848, | |
| "eval_steps_per_second": 0.899, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.9215155615696888, | |
| "grad_norm": 3.515883526758233, | |
| "learning_rate": 2.920528634068885e-07, | |
| "loss": 0.7411, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.9485791610284169, | |
| "grad_norm": 3.5334817509226593, | |
| "learning_rate": 2.789639642608184e-07, | |
| "loss": 0.7385, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.975642760487145, | |
| "grad_norm": 3.698753767478549, | |
| "learning_rate": 2.6606070610046526e-07, | |
| "loss": 0.7573, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.002706359945873, | |
| "grad_norm": 3.3551617194703063, | |
| "learning_rate": 2.533539259531757e-07, | |
| "loss": 0.7353, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.029769959404601, | |
| "grad_norm": 3.6997168009784684, | |
| "learning_rate": 2.408542958307957e-07, | |
| "loss": 0.7006, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.0568335588633286, | |
| "grad_norm": 3.636840131305413, | |
| "learning_rate": 2.2857231376659514e-07, | |
| "loss": 0.7166, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0838971583220567, | |
| "grad_norm": 3.972386455839474, | |
| "learning_rate": 2.1651829499831043e-07, | |
| "loss": 0.7113, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.1109607577807847, | |
| "grad_norm": 3.696729837743398, | |
| "learning_rate": 2.0470236330471126e-07, | |
| "loss": 0.709, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.138024357239513, | |
| "grad_norm": 3.7691871086231017, | |
| "learning_rate": 1.9313444250296846e-07, | |
| "loss": 0.7132, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "grad_norm": 3.6764401143556014, | |
| "learning_rate": 1.818242481139613e-07, | |
| "loss": 0.7075, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "eval_loss": 0.8137385249137878, | |
| "eval_runtime": 92.2291, | |
| "eval_samples_per_second": 56.923, | |
| "eval_steps_per_second": 0.9, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.192151556156969, | |
| "grad_norm": 3.722594968646584, | |
| "learning_rate": 1.7078127920252783e-07, | |
| "loss": 0.7064, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.219215155615697, | |
| "grad_norm": 3.9827901528666434, | |
| "learning_rate": 1.600148103995087e-07, | |
| "loss": 0.7055, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.246278755074425, | |
| "grad_norm": 3.848365017568979, | |
| "learning_rate": 1.4953388411228602e-07, | |
| "loss": 0.7147, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.273342354533153, | |
| "grad_norm": 3.813956302279095, | |
| "learning_rate": 1.3934730293035936e-07, | |
| "loss": 0.6951, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.300405953991881, | |
| "grad_norm": 3.816112318327775, | |
| "learning_rate": 1.2946362223233614e-07, | |
| "loss": 0.709, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.3274695534506087, | |
| "grad_norm": 3.9531391193741667, | |
| "learning_rate": 1.198911430005478e-07, | |
| "loss": 0.7197, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.3545331529093367, | |
| "grad_norm": 3.9313649848564913, | |
| "learning_rate": 1.1063790484932462e-07, | |
| "loss": 0.7163, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.381596752368065, | |
| "grad_norm": 3.7946532677122455, | |
| "learning_rate": 1.0171167927278368e-07, | |
| "loss": 0.6888, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.408660351826793, | |
| "grad_norm": 3.6866229175987737, | |
| "learning_rate": 9.311996311780446e-08, | |
| "loss": 0.7105, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "grad_norm": 3.6331365766503136, | |
| "learning_rate": 8.486997228767012e-08, | |
| "loss": 0.71, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "eval_loss": 0.8126546144485474, | |
| "eval_runtime": 92.1857, | |
| "eval_samples_per_second": 56.95, | |
| "eval_steps_per_second": 0.9, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.462787550744249, | |
| "grad_norm": 3.81903858751765, | |
| "learning_rate": 7.696863568166518e-08, | |
| "loss": 0.6992, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.489851150202977, | |
| "grad_norm": 3.8485837389475273, | |
| "learning_rate": 6.942258937571771e-08, | |
| "loss": 0.6948, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.516914749661705, | |
| "grad_norm": 3.7712141765804583, | |
| "learning_rate": 6.2238171048975e-08, | |
| "loss": 0.6949, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.543978349120433, | |
| "grad_norm": 3.584093640205327, | |
| "learning_rate": 5.5421414660992705e-08, | |
| "loss": 0.733, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.571041948579161, | |
| "grad_norm": 3.864716126183329, | |
| "learning_rate": 4.8978045384008125e-08, | |
| "loss": 0.7058, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.598105548037889, | |
| "grad_norm": 4.221053360392504, | |
| "learning_rate": 4.2913474794554036e-08, | |
| "loss": 0.7229, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.6251691474966172, | |
| "grad_norm": 3.7216448526253267, | |
| "learning_rate": 3.723279632845155e-08, | |
| "loss": 0.6851, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.6522327469553453, | |
| "grad_norm": 3.714335585514238, | |
| "learning_rate": 3.194078100299863e-08, | |
| "loss": 0.7011, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.6792963464140733, | |
| "grad_norm": 3.822561595275015, | |
| "learning_rate": 2.7041873409947734e-08, | |
| "loss": 0.6853, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "grad_norm": 3.626297678435798, | |
| "learning_rate": 2.2540187982637627e-08, | |
| "loss": 0.7069, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "eval_loss": 0.8118973970413208, | |
| "eval_runtime": 92.3179, | |
| "eval_samples_per_second": 56.869, | |
| "eval_steps_per_second": 0.899, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.733423545331529, | |
| "grad_norm": 3.6886544451935412, | |
| "learning_rate": 1.8439505540414458e-08, | |
| "loss": 0.7075, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.760487144790257, | |
| "grad_norm": 3.9389896132103237, | |
| "learning_rate": 1.4743270113244277e-08, | |
| "loss": 0.7051, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.787550744248985, | |
| "grad_norm": 3.803689145741291, | |
| "learning_rate": 1.1454586049184589e-08, | |
| "loss": 0.6957, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.814614343707713, | |
| "grad_norm": 3.7647580840059702, | |
| "learning_rate": 8.576215407142651e-09, | |
| "loss": 0.7189, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.841677943166441, | |
| "grad_norm": 3.731265168222702, | |
| "learning_rate": 6.110575637112425e-09, | |
| "loss": 0.7245, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.8687415426251692, | |
| "grad_norm": 3.6148858260407883, | |
| "learning_rate": 4.059737549836517e-09, | |
| "loss": 0.7008, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.8958051420838973, | |
| "grad_norm": 4.036765550791644, | |
| "learning_rate": 2.425423577599783e-09, | |
| "loss": 0.7199, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.9228687415426253, | |
| "grad_norm": 3.8323701179806386, | |
| "learning_rate": 1.209006327614226e-09, | |
| "loss": 0.6992, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.949932341001353, | |
| "grad_norm": 3.7653155106882643, | |
| "learning_rate": 4.115074292109777e-10, | |
| "loss": 0.7005, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "grad_norm": 3.9157330059593396, | |
| "learning_rate": 3.3596675806824013e-11, | |
| "loss": 0.7088, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "eval_loss": 0.8117214441299438, | |
| "eval_runtime": 92.3355, | |
| "eval_samples_per_second": 56.858, | |
| "eval_steps_per_second": 0.899, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.9878213802435725, | |
| "step": 552, | |
| "total_flos": 3254608239525888.0, | |
| "train_loss": 0.8072819815597673, | |
| "train_runtime": 8267.6947, | |
| "train_samples_per_second": 17.145, | |
| "train_steps_per_second": 0.067 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 552, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3254608239525888.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |