| { | |
| "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": 18.65970657349142, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7373, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05412719891745602, | |
| "grad_norm": 12.123851504493118, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5834, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08119079837618404, | |
| "grad_norm": 7.621307641306264, | |
| "learning_rate": 9.997900331216397e-07, | |
| "loss": 1.2794, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10825439783491204, | |
| "grad_norm": 4.267249289625919, | |
| "learning_rate": 9.991603088309193e-07, | |
| "loss": 1.1358, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13531799729364005, | |
| "grad_norm": 3.917907714316784, | |
| "learning_rate": 9.981113560128126e-07, | |
| "loss": 1.0736, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16238159675236807, | |
| "grad_norm": 3.7859846750996846, | |
| "learning_rate": 9.966440556487147e-07, | |
| "loss": 1.035, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18944519621109607, | |
| "grad_norm": 3.7025125719803813, | |
| "learning_rate": 9.947596400765342e-07, | |
| "loss": 1.0091, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2165087956698241, | |
| "grad_norm": 3.370534729114571, | |
| "learning_rate": 9.924596919556916e-07, | |
| "loss": 0.9798, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2435723951285521, | |
| "grad_norm": 3.494521506906822, | |
| "learning_rate": 9.897461429378964e-07, | |
| "loss": 0.9557, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "grad_norm": 3.2135593780998435, | |
| "learning_rate": 9.866212720448147e-07, | |
| "loss": 0.9274, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2706359945872801, | |
| "eval_loss": 0.9371856451034546, | |
| "eval_runtime": 92.1867, | |
| "eval_samples_per_second": 56.95, | |
| "eval_steps_per_second": 0.9, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2976995940460081, | |
| "grad_norm": 3.6113816994688874, | |
| "learning_rate": 9.830877037539935e-07, | |
| "loss": 0.9243, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.32476319350473615, | |
| "grad_norm": 3.3612765440218966, | |
| "learning_rate": 9.791484057946465e-07, | |
| "loss": 0.9356, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35182679296346414, | |
| "grad_norm": 3.4973229455252706, | |
| "learning_rate": 9.748066866551555e-07, | |
| "loss": 0.9136, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.37889039242219213, | |
| "grad_norm": 3.515361088136603, | |
| "learning_rate": 9.700661928043785e-07, | |
| "loss": 0.9077, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4059539918809202, | |
| "grad_norm": 3.6572315608142527, | |
| "learning_rate": 9.649309056290984e-07, | |
| "loss": 0.9102, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4330175913396482, | |
| "grad_norm": 3.84703859120007, | |
| "learning_rate": 9.594051380901858e-07, | |
| "loss": 0.8847, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.46008119079837617, | |
| "grad_norm": 3.429617493706216, | |
| "learning_rate": 9.534935311002834e-07, | |
| "loss": 0.8902, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.4871447902571042, | |
| "grad_norm": 3.5683074263183254, | |
| "learning_rate": 9.472010496260544e-07, | |
| "loss": 0.8873, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5142083897158322, | |
| "grad_norm": 3.575422711992299, | |
| "learning_rate": 9.405329785182678e-07, | |
| "loss": 0.881, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "grad_norm": 3.669206503451801, | |
| "learning_rate": 9.334949180732244e-07, | |
| "loss": 0.8893, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5412719891745602, | |
| "eval_loss": 0.8818467855453491, | |
| "eval_runtime": 91.9423, | |
| "eval_samples_per_second": 57.101, | |
| "eval_steps_per_second": 0.903, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5683355886332883, | |
| "grad_norm": 3.914942398465802, | |
| "learning_rate": 9.260927793292497e-07, | |
| "loss": 0.8724, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.5953991880920162, | |
| "grad_norm": 3.5380899440058564, | |
| "learning_rate": 9.183327791022047e-07, | |
| "loss": 0.8944, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6224627875507442, | |
| "grad_norm": 3.4498048009589968, | |
| "learning_rate": 9.102214347641843e-07, | |
| "loss": 0.876, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.6495263870094723, | |
| "grad_norm": 3.7226989372850356, | |
| "learning_rate": 9.017655587697883e-07, | |
| "loss": 0.8822, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6765899864682002, | |
| "grad_norm": 3.698396988748353, | |
| "learning_rate": 8.929722529345623e-07, | |
| "loss": 0.8689, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7036535859269283, | |
| "grad_norm": 3.7148159292863028, | |
| "learning_rate": 8.83848902470413e-07, | |
| "loss": 0.8583, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7307171853856563, | |
| "grad_norm": 3.5813361420522827, | |
| "learning_rate": 8.744031697830088e-07, | |
| "loss": 0.8662, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.7577807848443843, | |
| "grad_norm": 3.135451465920447, | |
| "learning_rate": 8.646429880363746e-07, | |
| "loss": 0.8554, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7848443843031123, | |
| "grad_norm": 3.5589111641165405, | |
| "learning_rate": 8.545765544900846e-07, | |
| "loss": 0.859, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "grad_norm": 3.4347954694969465, | |
| "learning_rate": 8.442123236146508e-07, | |
| "loss": 0.8684, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8119079837618404, | |
| "eval_loss": 0.8576312065124512, | |
| "eval_runtime": 91.9678, | |
| "eval_samples_per_second": 57.085, | |
| "eval_steps_per_second": 0.902, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8389715832205683, | |
| "grad_norm": 3.42544344763707, | |
| "learning_rate": 8.33558999990887e-07, | |
| "loss": 0.8259, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.8660351826792964, | |
| "grad_norm": 3.748411527272942, | |
| "learning_rate": 8.22625530999215e-07, | |
| "loss": 0.8591, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8930987821380244, | |
| "grad_norm": 3.3899159655842626, | |
| "learning_rate": 8.114210993050502e-07, | |
| "loss": 0.8411, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.9201623815967523, | |
| "grad_norm": 3.4095342832234445, | |
| "learning_rate": 7.999551151465791e-07, | |
| "loss": 0.8509, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9472259810554804, | |
| "grad_norm": 3.486474439832698, | |
| "learning_rate": 7.88237208431406e-07, | |
| "loss": 0.8407, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9742895805142084, | |
| "grad_norm": 3.236038581839862, | |
| "learning_rate": 7.762772206487065e-07, | |
| "loss": 0.8106, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0013531799729365, | |
| "grad_norm": 3.4903154970716432, | |
| "learning_rate": 7.640851966036805e-07, | |
| "loss": 0.84, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.0284167794316643, | |
| "grad_norm": 3.5863433980686557, | |
| "learning_rate": 7.516713759812464e-07, | |
| "loss": 0.7865, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0554803788903924, | |
| "grad_norm": 3.362482872884403, | |
| "learning_rate": 7.390461847460628e-07, | |
| "loss": 0.7834, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "grad_norm": 3.510512158864217, | |
| "learning_rate": 7.262202263860988e-07, | |
| "loss": 0.7885, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0825439783491204, | |
| "eval_loss": 0.8443654775619507, | |
| "eval_runtime": 91.8778, | |
| "eval_samples_per_second": 57.141, | |
| "eval_steps_per_second": 0.903, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1096075778078485, | |
| "grad_norm": 3.4220527353768824, | |
| "learning_rate": 7.1320427300711e-07, | |
| "loss": 0.7751, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.1366711772665765, | |
| "grad_norm": 3.52648434908435, | |
| "learning_rate": 7.000092562854959e-07, | |
| "loss": 0.7738, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1637347767253043, | |
| "grad_norm": 3.4548324057171147, | |
| "learning_rate": 6.866462582871401e-07, | |
| "loss": 0.7716, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.1907983761840324, | |
| "grad_norm": 3.459942778474347, | |
| "learning_rate": 6.731265021599436e-07, | |
| "loss": 0.7802, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2178619756427604, | |
| "grad_norm": 3.597273563743811, | |
| "learning_rate": 6.594613427078674e-07, | |
| "loss": 0.7696, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.2449255751014885, | |
| "grad_norm": 3.8266943969656317, | |
| "learning_rate": 6.456622568544011e-07, | |
| "loss": 0.7735, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2719891745602165, | |
| "grad_norm": 3.780461192372918, | |
| "learning_rate": 6.317408340034684e-07, | |
| "loss": 0.7793, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.2990527740189446, | |
| "grad_norm": 3.4361198088375673, | |
| "learning_rate": 6.177087663058625e-07, | |
| "loss": 0.7539, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.3261163734776726, | |
| "grad_norm": 3.7110817877487383, | |
| "learning_rate": 6.035778388393893e-07, | |
| "loss": 0.7605, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "grad_norm": 3.7117782049022483, | |
| "learning_rate": 5.893599197109624e-07, | |
| "loss": 0.7588, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3531799729364005, | |
| "eval_loss": 0.8342949748039246, | |
| "eval_runtime": 91.9385, | |
| "eval_samples_per_second": 57.103, | |
| "eval_steps_per_second": 0.903, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3802435723951285, | |
| "grad_norm": 3.6963553360558032, | |
| "learning_rate": 5.750669500889666e-07, | |
| "loss": 0.7692, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.4073071718538566, | |
| "grad_norm": 3.50195752763911, | |
| "learning_rate": 5.607109341742578e-07, | |
| "loss": 0.7721, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.4343707713125846, | |
| "grad_norm": 3.7055534906664165, | |
| "learning_rate": 5.463039291182256e-07, | |
| "loss": 0.7651, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.4614343707713127, | |
| "grad_norm": 3.5346162965379695, | |
| "learning_rate": 5.318580348963825e-07, | |
| "loss": 0.7785, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4884979702300405, | |
| "grad_norm": 3.5657020524687986, | |
| "learning_rate": 5.173853841459877e-07, | |
| "loss": 0.763, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.5155615696887685, | |
| "grad_norm": 3.5651769161124074, | |
| "learning_rate": 5.028981319762399e-07, | |
| "loss": 0.755, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.5426251691474966, | |
| "grad_norm": 3.4198506759540233, | |
| "learning_rate": 4.884084457595956e-07, | |
| "loss": 0.755, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.5696887686062246, | |
| "grad_norm": 3.5240822496367556, | |
| "learning_rate": 4.7392849491278817e-07, | |
| "loss": 0.772, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5967523680649527, | |
| "grad_norm": 3.8692497705578335, | |
| "learning_rate": 4.5947044067613e-07, | |
| "loss": 0.7595, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "grad_norm": 3.471893348202751, | |
| "learning_rate": 4.4504642589968217e-07, | |
| "loss": 0.7736, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6238159675236807, | |
| "eval_loss": 0.8263227343559265, | |
| "eval_runtime": 91.9869, | |
| "eval_samples_per_second": 57.073, | |
| "eval_steps_per_second": 0.902, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6508795669824088, | |
| "grad_norm": 3.5822544768320417, | |
| "learning_rate": 4.3066856484486847e-07, | |
| "loss": 0.751, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.6779431664411368, | |
| "grad_norm": 3.5379322852796924, | |
| "learning_rate": 4.1634893301010165e-07, | |
| "loss": 0.7659, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.7050067658998647, | |
| "grad_norm": 3.6759115180695465, | |
| "learning_rate": 4.0209955698896445e-07, | |
| "loss": 0.7859, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.7320703653585927, | |
| "grad_norm": 3.325644714682765, | |
| "learning_rate": 3.8793240436946385e-07, | |
| "loss": 0.7501, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.7591339648173205, | |
| "grad_norm": 3.7071013179144257, | |
| "learning_rate": 3.738593736828426e-07, | |
| "loss": 0.7652, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.7861975642760486, | |
| "grad_norm": 3.7366222242171303, | |
| "learning_rate": 3.598922844103902e-07, | |
| "loss": 0.7653, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.8132611637347766, | |
| "grad_norm": 3.699329762945061, | |
| "learning_rate": 3.4604286705664397e-07, | |
| "loss": 0.7667, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.8403247631935047, | |
| "grad_norm": 3.452541395212277, | |
| "learning_rate": 3.323227532973193e-07, | |
| "loss": 0.7478, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.8673883626522327, | |
| "grad_norm": 3.665296534909908, | |
| "learning_rate": 3.187434662102434e-07, | |
| "loss": 0.7678, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "grad_norm": 3.842830417177158, | |
| "learning_rate": 3.0531641059749634e-07, | |
| "loss": 0.7499, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8944519621109608, | |
| "eval_loss": 0.8207802176475525, | |
| "eval_runtime": 91.9974, | |
| "eval_samples_per_second": 57.067, | |
| "eval_steps_per_second": 0.902, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.9215155615696888, | |
| "grad_norm": 3.561775197313456, | |
| "learning_rate": 2.920528634068885e-07, | |
| "loss": 0.7482, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.9485791610284169, | |
| "grad_norm": 3.83958742018717, | |
| "learning_rate": 2.789639642608184e-07, | |
| "loss": 0.7501, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.975642760487145, | |
| "grad_norm": 3.798708193385439, | |
| "learning_rate": 2.6606070610046526e-07, | |
| "loss": 0.7639, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.002706359945873, | |
| "grad_norm": 3.686196257239308, | |
| "learning_rate": 2.533539259531757e-07, | |
| "loss": 0.7545, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.029769959404601, | |
| "grad_norm": 3.66828576372959, | |
| "learning_rate": 2.408542958307957e-07, | |
| "loss": 0.7153, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.0568335588633286, | |
| "grad_norm": 3.7385059480327767, | |
| "learning_rate": 2.2857231376659514e-07, | |
| "loss": 0.7111, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0838971583220567, | |
| "grad_norm": 3.738542782092733, | |
| "learning_rate": 2.1651829499831043e-07, | |
| "loss": 0.7091, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.1109607577807847, | |
| "grad_norm": 3.6847092660306067, | |
| "learning_rate": 2.0470236330471126e-07, | |
| "loss": 0.6938, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.138024357239513, | |
| "grad_norm": 3.8317927714979008, | |
| "learning_rate": 1.9313444250296846e-07, | |
| "loss": 0.725, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "grad_norm": 3.77803814777597, | |
| "learning_rate": 1.818242481139613e-07, | |
| "loss": 0.6949, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.165087956698241, | |
| "eval_loss": 0.823599100112915, | |
| "eval_runtime": 91.9806, | |
| "eval_samples_per_second": 57.077, | |
| "eval_steps_per_second": 0.902, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.192151556156969, | |
| "grad_norm": 3.6134354332095238, | |
| "learning_rate": 1.7078127920252783e-07, | |
| "loss": 0.7078, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.219215155615697, | |
| "grad_norm": 3.7411384166319346, | |
| "learning_rate": 1.600148103995087e-07, | |
| "loss": 0.7089, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.246278755074425, | |
| "grad_norm": 3.8076732272754406, | |
| "learning_rate": 1.4953388411228602e-07, | |
| "loss": 0.7351, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.273342354533153, | |
| "grad_norm": 3.561250617360428, | |
| "learning_rate": 1.3934730293035936e-07, | |
| "loss": 0.7127, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.300405953991881, | |
| "grad_norm": 3.857897941251079, | |
| "learning_rate": 1.2946362223233614e-07, | |
| "loss": 0.7155, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.3274695534506087, | |
| "grad_norm": 3.87623198142243, | |
| "learning_rate": 1.198911430005478e-07, | |
| "loss": 0.7248, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.3545331529093367, | |
| "grad_norm": 3.94878669421802, | |
| "learning_rate": 1.1063790484932462e-07, | |
| "loss": 0.7125, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.381596752368065, | |
| "grad_norm": 3.9196427575075288, | |
| "learning_rate": 1.0171167927278368e-07, | |
| "loss": 0.7087, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.408660351826793, | |
| "grad_norm": 3.8012419710191123, | |
| "learning_rate": 9.311996311780446e-08, | |
| "loss": 0.7013, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "grad_norm": 3.872302135767634, | |
| "learning_rate": 8.486997228767012e-08, | |
| "loss": 0.7056, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.435723951285521, | |
| "eval_loss": 0.8227203488349915, | |
| "eval_runtime": 91.9356, | |
| "eval_samples_per_second": 57.105, | |
| "eval_steps_per_second": 0.903, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.462787550744249, | |
| "grad_norm": 3.700203865295363, | |
| "learning_rate": 7.696863568166518e-08, | |
| "loss": 0.7216, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.489851150202977, | |
| "grad_norm": 3.6078275253514183, | |
| "learning_rate": 6.942258937571771e-08, | |
| "loss": 0.6922, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.516914749661705, | |
| "grad_norm": 3.8271435786761088, | |
| "learning_rate": 6.2238171048975e-08, | |
| "loss": 0.7127, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.543978349120433, | |
| "grad_norm": 4.00275667268857, | |
| "learning_rate": 5.5421414660992705e-08, | |
| "loss": 0.7072, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.571041948579161, | |
| "grad_norm": 4.262739343601447, | |
| "learning_rate": 4.8978045384008125e-08, | |
| "loss": 0.6983, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.598105548037889, | |
| "grad_norm": 3.627717683448711, | |
| "learning_rate": 4.2913474794554036e-08, | |
| "loss": 0.718, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.6251691474966172, | |
| "grad_norm": 3.627655916654691, | |
| "learning_rate": 3.723279632845155e-08, | |
| "loss": 0.6992, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.6522327469553453, | |
| "grad_norm": 4.157793979172056, | |
| "learning_rate": 3.194078100299863e-08, | |
| "loss": 0.7154, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.6792963464140733, | |
| "grad_norm": 3.6512611965864314, | |
| "learning_rate": 2.7041873409947734e-08, | |
| "loss": 0.715, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "grad_norm": 3.726128068146228, | |
| "learning_rate": 2.2540187982637627e-08, | |
| "loss": 0.7001, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.706359945872801, | |
| "eval_loss": 0.8219273686408997, | |
| "eval_runtime": 91.8886, | |
| "eval_samples_per_second": 57.134, | |
| "eval_steps_per_second": 0.903, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.733423545331529, | |
| "grad_norm": 3.728833279635095, | |
| "learning_rate": 1.8439505540414458e-08, | |
| "loss": 0.7019, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.760487144790257, | |
| "grad_norm": 3.6477799057801645, | |
| "learning_rate": 1.4743270113244277e-08, | |
| "loss": 0.6896, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.787550744248985, | |
| "grad_norm": 3.7765174476962233, | |
| "learning_rate": 1.1454586049184589e-08, | |
| "loss": 0.7095, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.814614343707713, | |
| "grad_norm": 3.844174292857732, | |
| "learning_rate": 8.576215407142651e-09, | |
| "loss": 0.6985, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.841677943166441, | |
| "grad_norm": 3.756024935666749, | |
| "learning_rate": 6.110575637112425e-09, | |
| "loss": 0.7077, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.8687415426251692, | |
| "grad_norm": 3.9062256497537864, | |
| "learning_rate": 4.059737549836517e-09, | |
| "loss": 0.707, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.8958051420838973, | |
| "grad_norm": 3.8224427448415494, | |
| "learning_rate": 2.425423577599783e-09, | |
| "loss": 0.6974, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.9228687415426253, | |
| "grad_norm": 3.7396928746940383, | |
| "learning_rate": 1.209006327614226e-09, | |
| "loss": 0.7088, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.949932341001353, | |
| "grad_norm": 3.6670690549497804, | |
| "learning_rate": 4.115074292109777e-10, | |
| "loss": 0.7058, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "grad_norm": 3.7461282178146655, | |
| "learning_rate": 3.3596675806824013e-11, | |
| "loss": 0.7045, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.976995940460081, | |
| "eval_loss": 0.8217999339103699, | |
| "eval_runtime": 91.9491, | |
| "eval_samples_per_second": 57.097, | |
| "eval_steps_per_second": 0.903, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.9878213802435725, | |
| "step": 552, | |
| "total_flos": 3254608239525888.0, | |
| "train_loss": 0.81000138948793, | |
| "train_runtime": 8272.8839, | |
| "train_samples_per_second": 17.134, | |
| "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 | |
| } | |