| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 444, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.033783783783783786, | |
| "grad_norm": 7.693039168386045, | |
| "learning_rate": 1.0869565217391305e-05, | |
| "loss": 0.5133, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 4.296719547848672, | |
| "learning_rate": 2.173913043478261e-05, | |
| "loss": 0.3581, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10135135135135136, | |
| "grad_norm": 3.2417658261755915, | |
| "learning_rate": 3.260869565217392e-05, | |
| "loss": 0.3547, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 2.8412337980778553, | |
| "learning_rate": 4.347826086956522e-05, | |
| "loss": 0.2666, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16891891891891891, | |
| "grad_norm": 4.90451510502916, | |
| "learning_rate": 4.9997494236918504e-05, | |
| "loss": 0.3582, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 22.780898735414855, | |
| "learning_rate": 4.996931081151707e-05, | |
| "loss": 0.3768, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23648648648648649, | |
| "grad_norm": 3.320519938493738, | |
| "learning_rate": 4.990985111773183e-05, | |
| "loss": 0.4268, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 2.5294359467709095, | |
| "learning_rate": 4.981919792077782e-05, | |
| "loss": 0.2873, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.30405405405405406, | |
| "grad_norm": 2.3280683716637873, | |
| "learning_rate": 4.969747740582118e-05, | |
| "loss": 0.2544, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.33783783783783783, | |
| "grad_norm": 2.8289139949817335, | |
| "learning_rate": 4.95448590023351e-05, | |
| "loss": 0.2697, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3716216216216216, | |
| "grad_norm": 2.1629521504260993, | |
| "learning_rate": 4.936155514826161e-05, | |
| "loss": 0.3052, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 1.9564591527643984, | |
| "learning_rate": 4.914782099430755e-05, | |
| "loss": 0.2319, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4391891891891892, | |
| "grad_norm": 2.6620448846059714, | |
| "learning_rate": 4.890395404878627e-05, | |
| "loss": 0.2613, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.47297297297297297, | |
| "grad_norm": 2.333611941701714, | |
| "learning_rate": 4.863029376349949e-05, | |
| "loss": 0.2595, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5067567567567568, | |
| "grad_norm": 2.5677617585089147, | |
| "learning_rate": 4.8327221061235635e-05, | |
| "loss": 0.2733, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 1.8945353614021259, | |
| "learning_rate": 4.799515780554253e-05, | |
| "loss": 0.2298, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5743243243243243, | |
| "grad_norm": 1.9687212656745774, | |
| "learning_rate": 4.763456621351229e-05, | |
| "loss": 0.2657, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 2.2237600068784653, | |
| "learning_rate": 4.724594821239601e-05, | |
| "loss": 0.244, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6418918918918919, | |
| "grad_norm": 2.1737571541215295, | |
| "learning_rate": 4.6829844740943586e-05, | |
| "loss": 0.2409, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 1.7570800060495113, | |
| "learning_rate": 4.6386834996441395e-05, | |
| "loss": 0.2054, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7094594594594594, | |
| "grad_norm": 1.5936369087529354, | |
| "learning_rate": 4.5917535628495714e-05, | |
| "loss": 0.2658, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7432432432432432, | |
| "grad_norm": 1.687954605448381, | |
| "learning_rate": 4.542259988068434e-05, | |
| "loss": 0.2306, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.777027027027027, | |
| "grad_norm": 1.6798383407257704, | |
| "learning_rate": 4.4902716681270805e-05, | |
| "loss": 0.2051, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 1.9894543465428474, | |
| "learning_rate": 4.435860968424745e-05, | |
| "loss": 0.2885, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8445945945945946, | |
| "grad_norm": 1.5663367660943097, | |
| "learning_rate": 4.379103626204153e-05, | |
| "loss": 0.2412, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8783783783783784, | |
| "grad_norm": 1.4832323634362194, | |
| "learning_rate": 4.320078645128699e-05, | |
| "loss": 0.2374, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9121621621621622, | |
| "grad_norm": 1.999609063337418, | |
| "learning_rate": 4.258868185312901e-05, | |
| "loss": 0.21, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 1.449925989442346, | |
| "learning_rate": 4.195557448959231e-05, | |
| "loss": 0.2533, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9797297297297297, | |
| "grad_norm": 1.278422978606916, | |
| "learning_rate": 4.130234561760477e-05, | |
| "loss": 0.235, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 1.1102776403418126, | |
| "learning_rate": 4.0629904502327556e-05, | |
| "loss": 0.2101, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0472972972972974, | |
| "grad_norm": 1.452689044129832, | |
| "learning_rate": 3.993918715149896e-05, | |
| "loss": 0.1268, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 2.564080697825406, | |
| "learning_rate": 3.923115501255381e-05, | |
| "loss": 0.1504, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.114864864864865, | |
| "grad_norm": 1.4284081142136367, | |
| "learning_rate": 3.8506793634331925e-05, | |
| "loss": 0.1501, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 1.3921371445923525, | |
| "learning_rate": 3.7767111295238555e-05, | |
| "loss": 0.1517, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1824324324324325, | |
| "grad_norm": 1.0189911635230715, | |
| "learning_rate": 3.701313759976626e-05, | |
| "loss": 0.1482, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 1.4942069614270297, | |
| "learning_rate": 3.624592204533184e-05, | |
| "loss": 0.1297, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.3920628246297202, | |
| "learning_rate": 3.546653256142321e-05, | |
| "loss": 0.1325, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.2837837837837838, | |
| "grad_norm": 1.0187887226878, | |
| "learning_rate": 3.467605402308966e-05, | |
| "loss": 0.142, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3175675675675675, | |
| "grad_norm": 1.1984277342370044, | |
| "learning_rate": 3.3875586740844675e-05, | |
| "loss": 0.1347, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 1.1170028929426064, | |
| "learning_rate": 3.3066244929083246e-05, | |
| "loss": 0.1304, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.385135135135135, | |
| "grad_norm": 1.0186976243308743, | |
| "learning_rate": 3.2249155155145665e-05, | |
| "loss": 0.141, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 1.2898941169402762, | |
| "learning_rate": 3.142545477118649e-05, | |
| "loss": 0.1354, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4527027027027026, | |
| "grad_norm": 0.9961209346671913, | |
| "learning_rate": 3.059629033103166e-05, | |
| "loss": 0.1302, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 0.9875250619369351, | |
| "learning_rate": 2.9762815994227135e-05, | |
| "loss": 0.1403, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5202702702702702, | |
| "grad_norm": 1.0982139109346354, | |
| "learning_rate": 2.8926191919500854e-05, | |
| "loss": 0.1235, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.554054054054054, | |
| "grad_norm": 0.9668045928793142, | |
| "learning_rate": 2.808758264987406e-05, | |
| "loss": 0.1257, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5878378378378377, | |
| "grad_norm": 0.8581177979692445, | |
| "learning_rate": 2.7248155491669854e-05, | |
| "loss": 0.1174, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 1.1046256528106388, | |
| "learning_rate": 2.6409078889675382e-05, | |
| "loss": 0.1214, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6554054054054053, | |
| "grad_norm": 0.880781318496101, | |
| "learning_rate": 2.5571520800719363e-05, | |
| "loss": 0.1376, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 0.8458236531048872, | |
| "learning_rate": 2.473664706792873e-05, | |
| "loss": 0.1026, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.722972972972973, | |
| "grad_norm": 3.38684923966137, | |
| "learning_rate": 2.390561979792763e-05, | |
| "loss": 0.1335, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 0.8747529589638384, | |
| "learning_rate": 2.3079595743237243e-05, | |
| "loss": 0.1089, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7905405405405406, | |
| "grad_norm": 1.061409594201986, | |
| "learning_rate": 2.2259724692128448e-05, | |
| "loss": 0.1211, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 0.7187465056471518, | |
| "learning_rate": 2.144714786816836e-05, | |
| "loss": 0.093, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8581081081081081, | |
| "grad_norm": 0.7427271054292853, | |
| "learning_rate": 2.0642996341688498e-05, | |
| "loss": 0.1061, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 0.9102492961843579, | |
| "learning_rate": 1.9848389455385845e-05, | |
| "loss": 0.1117, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9256756756756757, | |
| "grad_norm": 0.6300885493957281, | |
| "learning_rate": 1.9064433266248287e-05, | |
| "loss": 0.1089, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.9594594594594594, | |
| "grad_norm": 0.8142160296904185, | |
| "learning_rate": 1.829221900597305e-05, | |
| "loss": 0.1039, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.9932432432432432, | |
| "grad_norm": 2.6048425820022088, | |
| "learning_rate": 1.7532821562021373e-05, | |
| "loss": 0.1028, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 0.7413049752502284, | |
| "learning_rate": 1.6787297981423618e-05, | |
| "loss": 0.0681, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.060810810810811, | |
| "grad_norm": 1.1954219728615871, | |
| "learning_rate": 1.6056685999417336e-05, | |
| "loss": 0.0589, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 0.5252454560354707, | |
| "learning_rate": 1.5342002594966657e-05, | |
| "loss": 0.0606, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1283783783783785, | |
| "grad_norm": 0.6859669085189071, | |
| "learning_rate": 1.4644242575173363e-05, | |
| "loss": 0.056, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 0.7090167332441359, | |
| "learning_rate": 1.3964377190550165e-05, | |
| "loss": 0.0608, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.195945945945946, | |
| "grad_norm": 0.47669323187220364, | |
| "learning_rate": 1.330335278308384e-05, | |
| "loss": 0.0462, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 0.3967741627628627, | |
| "learning_rate": 1.2662089468969717e-05, | |
| "loss": 0.0506, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.2635135135135136, | |
| "grad_norm": 0.7073240580865503, | |
| "learning_rate": 1.2041479857851485e-05, | |
| "loss": 0.0584, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 0.534095245997379, | |
| "learning_rate": 1.14423878103487e-05, | |
| "loss": 0.0448, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.331081081081081, | |
| "grad_norm": 0.5592832360622514, | |
| "learning_rate": 1.086564723560177e-05, | |
| "loss": 0.0515, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.364864864864865, | |
| "grad_norm": 0.5012383371857895, | |
| "learning_rate": 1.031206093050798e-05, | |
| "loss": 0.0468, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3986486486486487, | |
| "grad_norm": 0.5873919693251372, | |
| "learning_rate": 9.78239946226439e-06, | |
| "loss": 0.0481, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 0.7663749270398796, | |
| "learning_rate": 9.277400095772979e-06, | |
| "loss": 0.051, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.4662162162162162, | |
| "grad_norm": 0.593558971064878, | |
| "learning_rate": 8.797765767401159e-06, | |
| "loss": 0.0554, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.5159284273389936, | |
| "learning_rate": 8.34416410652601e-06, | |
| "loss": 0.0448, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5337837837837838, | |
| "grad_norm": 0.6366287598455895, | |
| "learning_rate": 7.917226506224227e-06, | |
| "loss": 0.0472, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 0.9410891314952453, | |
| "learning_rate": 7.51754724440146e-06, | |
| "loss": 0.0502, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6013513513513513, | |
| "grad_norm": 0.47104911363203844, | |
| "learning_rate": 7.145682656584196e-06, | |
| "loss": 0.0494, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 0.48833508181607244, | |
| "learning_rate": 6.802150361525786e-06, | |
| "loss": 0.0454, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.668918918918919, | |
| "grad_norm": 0.5274580172661194, | |
| "learning_rate": 6.487428540704467e-06, | |
| "loss": 0.0405, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 0.4401735004442807, | |
| "learning_rate": 6.201955272716275e-06, | |
| "loss": 0.0387, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7364864864864864, | |
| "grad_norm": 0.5222808193914125, | |
| "learning_rate": 5.946127923489382e-06, | |
| "loss": 0.0501, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.77027027027027, | |
| "grad_norm": 0.885902051673082, | |
| "learning_rate": 5.720302593168628e-06, | |
| "loss": 0.05, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.804054054054054, | |
| "grad_norm": 0.4765687214481478, | |
| "learning_rate": 5.524793620440148e-06, | |
| "loss": 0.038, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 0.5896068685310154, | |
| "learning_rate": 5.3598731449861e-06, | |
| "loss": 0.0466, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8716216216216215, | |
| "grad_norm": 0.4787445681998641, | |
| "learning_rate": 5.225770728678475e-06, | |
| "loss": 0.0486, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.9054054054054053, | |
| "grad_norm": 0.7298521032646316, | |
| "learning_rate": 5.122673036039321e-06, | |
| "loss": 0.0534, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.939189189189189, | |
| "grad_norm": 0.5563651328125397, | |
| "learning_rate": 5.050723574412132e-06, | |
| "loss": 0.0399, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 0.42205363383969957, | |
| "learning_rate": 5.010022494206098e-06, | |
| "loss": 0.0405, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 444, | |
| "total_flos": 311859968540672.0, | |
| "train_loss": 0.15194854576576938, | |
| "train_runtime": 8344.2681, | |
| "train_samples_per_second": 3.401, | |
| "train_steps_per_second": 0.053 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 444, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 311859968540672.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |