{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 1182, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.025380710659898477, "grad_norm": 2.13912296295166, "learning_rate": 4.957698815566836e-05, "loss": 0.4596, "step": 10 }, { "epoch": 0.050761421319796954, "grad_norm": 1.5265041589736938, "learning_rate": 4.9153976311336716e-05, "loss": 0.5119, "step": 20 }, { "epoch": 0.07614213197969544, "grad_norm": 3.402026891708374, "learning_rate": 4.873096446700508e-05, "loss": 0.5408, "step": 30 }, { "epoch": 0.10152284263959391, "grad_norm": 1.8265776634216309, "learning_rate": 4.8307952622673436e-05, "loss": 0.5605, "step": 40 }, { "epoch": 0.12690355329949238, "grad_norm": 1.4988059997558594, "learning_rate": 4.7884940778341796e-05, "loss": 0.5919, "step": 50 }, { "epoch": 0.15228426395939088, "grad_norm": 1.6790786981582642, "learning_rate": 4.746192893401015e-05, "loss": 0.5605, "step": 60 }, { "epoch": 0.17766497461928935, "grad_norm": 2.108412742614746, "learning_rate": 4.7038917089678517e-05, "loss": 0.5024, "step": 70 }, { "epoch": 0.20304568527918782, "grad_norm": 1.2667797803878784, "learning_rate": 4.661590524534687e-05, "loss": 0.6062, "step": 80 }, { "epoch": 0.22842639593908629, "grad_norm": 1.2963333129882812, "learning_rate": 4.619289340101523e-05, "loss": 0.5623, "step": 90 }, { "epoch": 0.25380710659898476, "grad_norm": 4.575770378112793, "learning_rate": 4.576988155668359e-05, "loss": 0.5304, "step": 100 }, { "epoch": 0.27918781725888325, "grad_norm": 1.6193996667861938, "learning_rate": 4.534686971235195e-05, "loss": 0.5276, "step": 110 }, { "epoch": 0.30456852791878175, "grad_norm": 1.5291751623153687, "learning_rate": 4.492385786802031e-05, "loss": 0.4513, "step": 120 }, { "epoch": 0.3299492385786802, "grad_norm": 2.916069984436035, "learning_rate": 4.4500846023688664e-05, "loss": 0.4674, "step": 130 }, { "epoch": 0.3553299492385787, "grad_norm": 2.216944456100464, "learning_rate": 4.4077834179357024e-05, "loss": 0.455, "step": 140 }, { "epoch": 0.38071065989847713, "grad_norm": 2.572436571121216, "learning_rate": 4.365482233502538e-05, "loss": 0.5409, "step": 150 }, { "epoch": 0.40609137055837563, "grad_norm": 2.5171351432800293, "learning_rate": 4.3231810490693744e-05, "loss": 0.5328, "step": 160 }, { "epoch": 0.43147208121827413, "grad_norm": 2.056858777999878, "learning_rate": 4.2808798646362104e-05, "loss": 0.515, "step": 170 }, { "epoch": 0.45685279187817257, "grad_norm": 1.575071930885315, "learning_rate": 4.238578680203046e-05, "loss": 0.5201, "step": 180 }, { "epoch": 0.48223350253807107, "grad_norm": 3.0955426692962646, "learning_rate": 4.196277495769882e-05, "loss": 0.4181, "step": 190 }, { "epoch": 0.5076142131979695, "grad_norm": 2.798530340194702, "learning_rate": 4.153976311336718e-05, "loss": 0.3292, "step": 200 }, { "epoch": 0.5329949238578681, "grad_norm": 2.3137614727020264, "learning_rate": 4.111675126903554e-05, "loss": 0.6242, "step": 210 }, { "epoch": 0.5583756345177665, "grad_norm": 1.7211440801620483, "learning_rate": 4.069373942470389e-05, "loss": 0.5459, "step": 220 }, { "epoch": 0.583756345177665, "grad_norm": 1.9022053480148315, "learning_rate": 4.027072758037225e-05, "loss": 0.4379, "step": 230 }, { "epoch": 0.6091370558375635, "grad_norm": 1.4369803667068481, "learning_rate": 3.9847715736040605e-05, "loss": 0.5092, "step": 240 }, { "epoch": 0.6345177664974619, "grad_norm": 1.8029989004135132, "learning_rate": 3.942470389170897e-05, "loss": 0.5678, "step": 250 }, { "epoch": 0.6598984771573604, "grad_norm": 2.690417528152466, "learning_rate": 3.900169204737733e-05, "loss": 0.4386, "step": 260 }, { "epoch": 0.6852791878172588, "grad_norm": 3.036839485168457, "learning_rate": 3.8578680203045685e-05, "loss": 0.5936, "step": 270 }, { "epoch": 0.7106598984771574, "grad_norm": 1.6595145463943481, "learning_rate": 3.8155668358714046e-05, "loss": 0.5406, "step": 280 }, { "epoch": 0.7360406091370558, "grad_norm": 4.433804035186768, "learning_rate": 3.7732656514382406e-05, "loss": 0.4874, "step": 290 }, { "epoch": 0.7614213197969543, "grad_norm": 2.043555498123169, "learning_rate": 3.7309644670050766e-05, "loss": 0.4603, "step": 300 }, { "epoch": 0.7868020304568528, "grad_norm": 2.553321123123169, "learning_rate": 3.688663282571912e-05, "loss": 0.5858, "step": 310 }, { "epoch": 0.8121827411167513, "grad_norm": 2.677241563796997, "learning_rate": 3.646362098138748e-05, "loss": 0.4095, "step": 320 }, { "epoch": 0.8375634517766497, "grad_norm": 3.408170461654663, "learning_rate": 3.604060913705584e-05, "loss": 0.6228, "step": 330 }, { "epoch": 0.8629441624365483, "grad_norm": 3.3228750228881836, "learning_rate": 3.56175972927242e-05, "loss": 0.4735, "step": 340 }, { "epoch": 0.8883248730964467, "grad_norm": 4.007043838500977, "learning_rate": 3.519458544839256e-05, "loss": 0.51, "step": 350 }, { "epoch": 0.9137055837563451, "grad_norm": 1.5816287994384766, "learning_rate": 3.477157360406091e-05, "loss": 0.5176, "step": 360 }, { "epoch": 0.9390862944162437, "grad_norm": 1.4616796970367432, "learning_rate": 3.434856175972927e-05, "loss": 0.286, "step": 370 }, { "epoch": 0.9644670050761421, "grad_norm": 4.507604122161865, "learning_rate": 3.3925549915397633e-05, "loss": 0.6015, "step": 380 }, { "epoch": 0.9898477157360406, "grad_norm": 1.7000436782836914, "learning_rate": 3.3502538071065994e-05, "loss": 0.3593, "step": 390 }, { "epoch": 1.0, "eval_loss": 0.512408971786499, "eval_runtime": 8.5129, "eval_samples_per_second": 92.565, "eval_steps_per_second": 11.629, "step": 394 }, { "epoch": 1.015228426395939, "grad_norm": 3.039973497390747, "learning_rate": 3.307952622673435e-05, "loss": 0.5087, "step": 400 }, { "epoch": 1.0406091370558375, "grad_norm": 2.5285613536834717, "learning_rate": 3.265651438240271e-05, "loss": 0.4335, "step": 410 }, { "epoch": 1.0659898477157361, "grad_norm": 1.5140420198440552, "learning_rate": 3.223350253807107e-05, "loss": 0.3759, "step": 420 }, { "epoch": 1.0913705583756346, "grad_norm": 2.0796151161193848, "learning_rate": 3.181049069373943e-05, "loss": 0.6431, "step": 430 }, { "epoch": 1.116751269035533, "grad_norm": 3.370028018951416, "learning_rate": 3.138747884940779e-05, "loss": 0.4801, "step": 440 }, { "epoch": 1.1421319796954315, "grad_norm": 7.150363445281982, "learning_rate": 3.096446700507614e-05, "loss": 0.4856, "step": 450 }, { "epoch": 1.16751269035533, "grad_norm": 4.6761860847473145, "learning_rate": 3.05414551607445e-05, "loss": 0.5973, "step": 460 }, { "epoch": 1.1928934010152283, "grad_norm": 1.6876248121261597, "learning_rate": 3.0118443316412858e-05, "loss": 0.4265, "step": 470 }, { "epoch": 1.218274111675127, "grad_norm": 2.2751359939575195, "learning_rate": 2.969543147208122e-05, "loss": 0.4307, "step": 480 }, { "epoch": 1.2436548223350254, "grad_norm": 4.351472854614258, "learning_rate": 2.927241962774958e-05, "loss": 0.4978, "step": 490 }, { "epoch": 1.2690355329949239, "grad_norm": 3.3015713691711426, "learning_rate": 2.8849407783417938e-05, "loss": 0.4159, "step": 500 }, { "epoch": 1.2944162436548223, "grad_norm": 1.4904918670654297, "learning_rate": 2.84263959390863e-05, "loss": 0.512, "step": 510 }, { "epoch": 1.3197969543147208, "grad_norm": 2.306755542755127, "learning_rate": 2.800338409475465e-05, "loss": 0.5745, "step": 520 }, { "epoch": 1.3451776649746192, "grad_norm": 3.0485565662384033, "learning_rate": 2.7580372250423015e-05, "loss": 0.4243, "step": 530 }, { "epoch": 1.3705583756345177, "grad_norm": 5.009148597717285, "learning_rate": 2.715736040609137e-05, "loss": 0.5187, "step": 540 }, { "epoch": 1.3959390862944163, "grad_norm": 3.1525375843048096, "learning_rate": 2.6734348561759732e-05, "loss": 0.4693, "step": 550 }, { "epoch": 1.4213197969543148, "grad_norm": 2.2495410442352295, "learning_rate": 2.6311336717428085e-05, "loss": 0.5011, "step": 560 }, { "epoch": 1.4467005076142132, "grad_norm": 2.390629291534424, "learning_rate": 2.588832487309645e-05, "loss": 0.6348, "step": 570 }, { "epoch": 1.4720812182741116, "grad_norm": 1.7083393335342407, "learning_rate": 2.546531302876481e-05, "loss": 0.4737, "step": 580 }, { "epoch": 1.49746192893401, "grad_norm": 2.4492344856262207, "learning_rate": 2.5042301184433166e-05, "loss": 0.5164, "step": 590 }, { "epoch": 1.5228426395939088, "grad_norm": 8.137948989868164, "learning_rate": 2.4619289340101523e-05, "loss": 0.4653, "step": 600 }, { "epoch": 1.548223350253807, "grad_norm": 2.3943252563476562, "learning_rate": 2.4196277495769883e-05, "loss": 0.4019, "step": 610 }, { "epoch": 1.5736040609137056, "grad_norm": 1.8179367780685425, "learning_rate": 2.3773265651438243e-05, "loss": 0.3428, "step": 620 }, { "epoch": 1.598984771573604, "grad_norm": 2.3452229499816895, "learning_rate": 2.33502538071066e-05, "loss": 0.5092, "step": 630 }, { "epoch": 1.6243654822335025, "grad_norm": 1.6915607452392578, "learning_rate": 2.292724196277496e-05, "loss": 0.5964, "step": 640 }, { "epoch": 1.649746192893401, "grad_norm": 3.0646092891693115, "learning_rate": 2.2504230118443317e-05, "loss": 0.4237, "step": 650 }, { "epoch": 1.6751269035532994, "grad_norm": 2.2048497200012207, "learning_rate": 2.2081218274111677e-05, "loss": 0.3713, "step": 660 }, { "epoch": 1.700507614213198, "grad_norm": 3.50895094871521, "learning_rate": 2.1658206429780033e-05, "loss": 0.4798, "step": 670 }, { "epoch": 1.7258883248730963, "grad_norm": 2.0043256282806396, "learning_rate": 2.1235194585448394e-05, "loss": 0.583, "step": 680 }, { "epoch": 1.751269035532995, "grad_norm": 1.845831274986267, "learning_rate": 2.0812182741116754e-05, "loss": 0.5429, "step": 690 }, { "epoch": 1.7766497461928934, "grad_norm": 1.7165807485580444, "learning_rate": 2.038917089678511e-05, "loss": 0.4464, "step": 700 }, { "epoch": 1.8020304568527918, "grad_norm": 1.361329436302185, "learning_rate": 1.996615905245347e-05, "loss": 0.4246, "step": 710 }, { "epoch": 1.8274111675126905, "grad_norm": 2.407137870788574, "learning_rate": 1.9543147208121827e-05, "loss": 0.4645, "step": 720 }, { "epoch": 1.8527918781725887, "grad_norm": 2.324209213256836, "learning_rate": 1.9120135363790187e-05, "loss": 0.5234, "step": 730 }, { "epoch": 1.8781725888324874, "grad_norm": 3.296576738357544, "learning_rate": 1.8697123519458544e-05, "loss": 0.557, "step": 740 }, { "epoch": 1.9035532994923858, "grad_norm": 2.2372303009033203, "learning_rate": 1.8274111675126904e-05, "loss": 0.4279, "step": 750 }, { "epoch": 1.9289340101522843, "grad_norm": 4.05513334274292, "learning_rate": 1.785109983079526e-05, "loss": 0.4958, "step": 760 }, { "epoch": 1.9543147208121827, "grad_norm": 1.2267847061157227, "learning_rate": 1.7428087986463625e-05, "loss": 0.4711, "step": 770 }, { "epoch": 1.9796954314720812, "grad_norm": 2.452728033065796, "learning_rate": 1.700507614213198e-05, "loss": 0.4696, "step": 780 }, { "epoch": 2.0, "eval_loss": 0.47782018780708313, "eval_runtime": 8.5194, "eval_samples_per_second": 92.495, "eval_steps_per_second": 11.621, "step": 788 }, { "epoch": 2.00507614213198, "grad_norm": 1.9264096021652222, "learning_rate": 1.658206429780034e-05, "loss": 0.4061, "step": 790 }, { "epoch": 2.030456852791878, "grad_norm": 1.578764796257019, "learning_rate": 1.6159052453468698e-05, "loss": 0.4168, "step": 800 }, { "epoch": 2.0558375634517767, "grad_norm": 5.699263095855713, "learning_rate": 1.5736040609137055e-05, "loss": 0.4339, "step": 810 }, { "epoch": 2.081218274111675, "grad_norm": 3.1477041244506836, "learning_rate": 1.5313028764805415e-05, "loss": 0.3623, "step": 820 }, { "epoch": 2.1065989847715736, "grad_norm": 2.8351192474365234, "learning_rate": 1.4890016920473774e-05, "loss": 0.446, "step": 830 }, { "epoch": 2.1319796954314723, "grad_norm": 6.005784511566162, "learning_rate": 1.4467005076142132e-05, "loss": 0.3802, "step": 840 }, { "epoch": 2.1573604060913705, "grad_norm": 3.5085108280181885, "learning_rate": 1.404399323181049e-05, "loss": 0.4335, "step": 850 }, { "epoch": 2.182741116751269, "grad_norm": 2.457703113555908, "learning_rate": 1.362098138747885e-05, "loss": 0.3866, "step": 860 }, { "epoch": 2.2081218274111674, "grad_norm": 6.848071098327637, "learning_rate": 1.3197969543147209e-05, "loss": 0.3146, "step": 870 }, { "epoch": 2.233502538071066, "grad_norm": 2.1162843704223633, "learning_rate": 1.2774957698815568e-05, "loss": 0.3747, "step": 880 }, { "epoch": 2.2588832487309647, "grad_norm": 2.40061354637146, "learning_rate": 1.2351945854483926e-05, "loss": 0.5355, "step": 890 }, { "epoch": 2.284263959390863, "grad_norm": 12.88298225402832, "learning_rate": 1.1928934010152284e-05, "loss": 0.3908, "step": 900 }, { "epoch": 2.3096446700507616, "grad_norm": 6.1054463386535645, "learning_rate": 1.1505922165820643e-05, "loss": 0.3276, "step": 910 }, { "epoch": 2.33502538071066, "grad_norm": 16.432769775390625, "learning_rate": 1.1082910321489003e-05, "loss": 0.2436, "step": 920 }, { "epoch": 2.3604060913705585, "grad_norm": 9.989404678344727, "learning_rate": 1.0659898477157361e-05, "loss": 0.5455, "step": 930 }, { "epoch": 2.3857868020304567, "grad_norm": 3.592336893081665, "learning_rate": 1.023688663282572e-05, "loss": 0.467, "step": 940 }, { "epoch": 2.4111675126903553, "grad_norm": 4.606255531311035, "learning_rate": 9.813874788494078e-06, "loss": 0.4826, "step": 950 }, { "epoch": 2.436548223350254, "grad_norm": 2.08891224861145, "learning_rate": 9.390862944162437e-06, "loss": 0.2995, "step": 960 }, { "epoch": 2.4619289340101522, "grad_norm": 4.196601867675781, "learning_rate": 8.967851099830795e-06, "loss": 0.5598, "step": 970 }, { "epoch": 2.487309644670051, "grad_norm": 2.653444290161133, "learning_rate": 8.544839255499154e-06, "loss": 0.4471, "step": 980 }, { "epoch": 2.512690355329949, "grad_norm": 1.8868201971054077, "learning_rate": 8.121827411167512e-06, "loss": 0.4616, "step": 990 }, { "epoch": 2.5380710659898478, "grad_norm": 5.2379045486450195, "learning_rate": 7.698815566835872e-06, "loss": 0.2389, "step": 1000 }, { "epoch": 2.563451776649746, "grad_norm": 3.2069485187530518, "learning_rate": 7.275803722504231e-06, "loss": 0.3617, "step": 1010 }, { "epoch": 2.5888324873096447, "grad_norm": 7.2437286376953125, "learning_rate": 6.852791878172589e-06, "loss": 0.6099, "step": 1020 }, { "epoch": 2.6142131979695433, "grad_norm": 1.056522250175476, "learning_rate": 6.429780033840948e-06, "loss": 0.1708, "step": 1030 }, { "epoch": 2.6395939086294415, "grad_norm": 7.495356559753418, "learning_rate": 6.006768189509306e-06, "loss": 0.3881, "step": 1040 }, { "epoch": 2.66497461928934, "grad_norm": 4.80796480178833, "learning_rate": 5.583756345177665e-06, "loss": 0.3701, "step": 1050 }, { "epoch": 2.6903553299492384, "grad_norm": 4.566274642944336, "learning_rate": 5.160744500846024e-06, "loss": 0.2963, "step": 1060 }, { "epoch": 2.715736040609137, "grad_norm": 5.139068603515625, "learning_rate": 4.737732656514383e-06, "loss": 0.4711, "step": 1070 }, { "epoch": 2.7411167512690353, "grad_norm": 5.687432289123535, "learning_rate": 4.3147208121827415e-06, "loss": 0.3127, "step": 1080 }, { "epoch": 2.766497461928934, "grad_norm": 2.3243467807769775, "learning_rate": 3.8917089678511e-06, "loss": 0.3814, "step": 1090 }, { "epoch": 2.7918781725888326, "grad_norm": 3.6257083415985107, "learning_rate": 3.4686971235194584e-06, "loss": 0.3983, "step": 1100 }, { "epoch": 2.817258883248731, "grad_norm": 7.288547992706299, "learning_rate": 3.0456852791878177e-06, "loss": 0.3005, "step": 1110 }, { "epoch": 2.8426395939086295, "grad_norm": 2.3332388401031494, "learning_rate": 2.622673434856176e-06, "loss": 0.4473, "step": 1120 }, { "epoch": 2.868020304568528, "grad_norm": 14.024263381958008, "learning_rate": 2.199661590524535e-06, "loss": 0.4144, "step": 1130 }, { "epoch": 2.8934010152284264, "grad_norm": 4.453007698059082, "learning_rate": 1.7766497461928936e-06, "loss": 0.2824, "step": 1140 }, { "epoch": 2.9187817258883246, "grad_norm": 19.20343017578125, "learning_rate": 1.353637901861252e-06, "loss": 0.4371, "step": 1150 }, { "epoch": 2.9441624365482233, "grad_norm": 8.754419326782227, "learning_rate": 9.306260575296109e-07, "loss": 0.3963, "step": 1160 }, { "epoch": 2.969543147208122, "grad_norm": 4.8408589363098145, "learning_rate": 5.076142131979695e-07, "loss": 0.3653, "step": 1170 }, { "epoch": 2.99492385786802, "grad_norm": 3.477670431137085, "learning_rate": 8.460236886632826e-08, "loss": 0.4175, "step": 1180 } ], "logging_steps": 10, "max_steps": 1182, "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": 1252611721691136.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }