| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 833, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012004801920768308, | |
| "grad_norm": 0.49814867973327637, | |
| "learning_rate": 9.879951980792317e-05, | |
| "loss": 2.3071, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.024009603841536616, | |
| "grad_norm": 0.4931434690952301, | |
| "learning_rate": 9.759903961584634e-05, | |
| "loss": 2.1442, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03601440576230492, | |
| "grad_norm": 0.9165758490562439, | |
| "learning_rate": 9.639855942376951e-05, | |
| "loss": 2.0314, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04801920768307323, | |
| "grad_norm": 0.4459473192691803, | |
| "learning_rate": 9.519807923169268e-05, | |
| "loss": 2.0196, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.060024009603841535, | |
| "grad_norm": 0.7976562976837158, | |
| "learning_rate": 9.399759903961585e-05, | |
| "loss": 1.9785, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07202881152460984, | |
| "grad_norm": 0.6203379034996033, | |
| "learning_rate": 9.279711884753903e-05, | |
| "loss": 1.9363, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08403361344537816, | |
| "grad_norm": 0.5313595533370972, | |
| "learning_rate": 9.159663865546218e-05, | |
| "loss": 1.9209, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09603841536614646, | |
| "grad_norm": 0.6249675750732422, | |
| "learning_rate": 9.039615846338536e-05, | |
| "loss": 1.986, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10804321728691477, | |
| "grad_norm": 0.5185940265655518, | |
| "learning_rate": 8.919567827130852e-05, | |
| "loss": 1.9521, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12004801920768307, | |
| "grad_norm": 0.4526331126689911, | |
| "learning_rate": 8.79951980792317e-05, | |
| "loss": 1.8528, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12004801920768307, | |
| "eval_loss": 1.8883458375930786, | |
| "eval_runtime": 38.5366, | |
| "eval_samples_per_second": 25.949, | |
| "eval_steps_per_second": 4.334, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13205282112845138, | |
| "grad_norm": 0.3635704815387726, | |
| "learning_rate": 8.679471788715487e-05, | |
| "loss": 1.8359, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14405762304921968, | |
| "grad_norm": 0.4598451554775238, | |
| "learning_rate": 8.559423769507804e-05, | |
| "loss": 1.9101, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15606242496998798, | |
| "grad_norm": 0.43790119886398315, | |
| "learning_rate": 8.43937575030012e-05, | |
| "loss": 1.8116, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 0.43906617164611816, | |
| "learning_rate": 8.319327731092437e-05, | |
| "loss": 1.8962, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.18007202881152462, | |
| "grad_norm": 0.43742913007736206, | |
| "learning_rate": 8.199279711884754e-05, | |
| "loss": 1.8382, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.19207683073229292, | |
| "grad_norm": 0.4784979224205017, | |
| "learning_rate": 8.079231692677071e-05, | |
| "loss": 1.8842, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.20408163265306123, | |
| "grad_norm": 0.4014301598072052, | |
| "learning_rate": 7.959183673469388e-05, | |
| "loss": 1.863, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.21608643457382953, | |
| "grad_norm": 0.5035635828971863, | |
| "learning_rate": 7.839135654261706e-05, | |
| "loss": 1.8979, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.22809123649459784, | |
| "grad_norm": 0.534414529800415, | |
| "learning_rate": 7.719087635054022e-05, | |
| "loss": 1.9105, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.24009603841536614, | |
| "grad_norm": 0.4861285388469696, | |
| "learning_rate": 7.599039615846338e-05, | |
| "loss": 1.7796, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24009603841536614, | |
| "eval_loss": 1.8688795566558838, | |
| "eval_runtime": 38.5301, | |
| "eval_samples_per_second": 25.954, | |
| "eval_steps_per_second": 4.334, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.25210084033613445, | |
| "grad_norm": 0.34365198016166687, | |
| "learning_rate": 7.478991596638657e-05, | |
| "loss": 1.84, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.26410564225690275, | |
| "grad_norm": 0.42961230874061584, | |
| "learning_rate": 7.358943577430972e-05, | |
| "loss": 1.9065, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.27611044417767105, | |
| "grad_norm": 0.389285683631897, | |
| "learning_rate": 7.23889555822329e-05, | |
| "loss": 1.8709, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.28811524609843936, | |
| "grad_norm": 0.4626094698905945, | |
| "learning_rate": 7.118847539015606e-05, | |
| "loss": 1.8709, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.30012004801920766, | |
| "grad_norm": 0.46512117981910706, | |
| "learning_rate": 6.998799519807924e-05, | |
| "loss": 1.8201, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.31212484993997597, | |
| "grad_norm": 0.5081538558006287, | |
| "learning_rate": 6.878751500600241e-05, | |
| "loss": 1.8761, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3241296518607443, | |
| "grad_norm": 0.7665489315986633, | |
| "learning_rate": 6.758703481392558e-05, | |
| "loss": 1.8218, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.4920073449611664, | |
| "learning_rate": 6.638655462184874e-05, | |
| "loss": 1.9478, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.34813925570228094, | |
| "grad_norm": 0.5283176302909851, | |
| "learning_rate": 6.518607442977191e-05, | |
| "loss": 1.8343, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.36014405762304924, | |
| "grad_norm": 0.5958182215690613, | |
| "learning_rate": 6.398559423769508e-05, | |
| "loss": 1.8715, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.36014405762304924, | |
| "eval_loss": 1.8636847734451294, | |
| "eval_runtime": 38.4834, | |
| "eval_samples_per_second": 25.985, | |
| "eval_steps_per_second": 4.34, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37214885954381755, | |
| "grad_norm": 0.4687075912952423, | |
| "learning_rate": 6.278511404561825e-05, | |
| "loss": 1.8762, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.38415366146458585, | |
| "grad_norm": 0.5822851657867432, | |
| "learning_rate": 6.158463385354142e-05, | |
| "loss": 1.8943, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.39615846338535415, | |
| "grad_norm": 0.3719615042209625, | |
| "learning_rate": 6.038415366146459e-05, | |
| "loss": 1.8973, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.40816326530612246, | |
| "grad_norm": 0.36546796560287476, | |
| "learning_rate": 5.918367346938776e-05, | |
| "loss": 1.894, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.42016806722689076, | |
| "grad_norm": 0.38791441917419434, | |
| "learning_rate": 5.7983193277310935e-05, | |
| "loss": 1.8221, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.43217286914765907, | |
| "grad_norm": 0.40212976932525635, | |
| "learning_rate": 5.6782713085234096e-05, | |
| "loss": 1.8472, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.44417767106842737, | |
| "grad_norm": 0.43383002281188965, | |
| "learning_rate": 5.558223289315727e-05, | |
| "loss": 1.9356, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.4561824729891957, | |
| "grad_norm": 0.7097159028053284, | |
| "learning_rate": 5.438175270108043e-05, | |
| "loss": 1.9236, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.468187274909964, | |
| "grad_norm": 0.9919241666793823, | |
| "learning_rate": 5.31812725090036e-05, | |
| "loss": 1.8988, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4801920768307323, | |
| "grad_norm": 0.365247517824173, | |
| "learning_rate": 5.1980792316926776e-05, | |
| "loss": 1.9492, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4801920768307323, | |
| "eval_loss": 1.8609040975570679, | |
| "eval_runtime": 38.8163, | |
| "eval_samples_per_second": 25.762, | |
| "eval_steps_per_second": 4.302, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4921968787515006, | |
| "grad_norm": 0.44502708315849304, | |
| "learning_rate": 5.078031212484994e-05, | |
| "loss": 1.8592, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.5130265951156616, | |
| "learning_rate": 4.957983193277311e-05, | |
| "loss": 1.9291, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.5162064825930373, | |
| "grad_norm": 0.4663325250148773, | |
| "learning_rate": 4.837935174069628e-05, | |
| "loss": 1.858, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.5282112845138055, | |
| "grad_norm": 0.7875126600265503, | |
| "learning_rate": 4.717887154861945e-05, | |
| "loss": 1.882, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5402160864345739, | |
| "grad_norm": 0.5712985396385193, | |
| "learning_rate": 4.5978391356542624e-05, | |
| "loss": 1.8705, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5522208883553421, | |
| "grad_norm": 0.48575717210769653, | |
| "learning_rate": 4.477791116446579e-05, | |
| "loss": 1.9747, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5642256902761105, | |
| "grad_norm": 0.521267831325531, | |
| "learning_rate": 4.3577430972388954e-05, | |
| "loss": 1.8128, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5762304921968787, | |
| "grad_norm": 0.43215957283973694, | |
| "learning_rate": 4.237695078031212e-05, | |
| "loss": 1.8486, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 0.4327848255634308, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 1.8379, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.6002400960384153, | |
| "grad_norm": 0.5328503847122192, | |
| "learning_rate": 3.9975990396158466e-05, | |
| "loss": 1.8718, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6002400960384153, | |
| "eval_loss": 1.8586153984069824, | |
| "eval_runtime": 38.4963, | |
| "eval_samples_per_second": 25.977, | |
| "eval_steps_per_second": 4.338, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6122448979591837, | |
| "grad_norm": 0.5828935503959656, | |
| "learning_rate": 3.8775510204081634e-05, | |
| "loss": 1.8329, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.6242496998799519, | |
| "grad_norm": 0.7918308973312378, | |
| "learning_rate": 3.75750300120048e-05, | |
| "loss": 1.8729, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.6362545018007203, | |
| "grad_norm": 0.40867704153060913, | |
| "learning_rate": 3.637454981992797e-05, | |
| "loss": 1.8365, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.6482593037214885, | |
| "grad_norm": 0.426268994808197, | |
| "learning_rate": 3.517406962785114e-05, | |
| "loss": 1.8463, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.6602641056422569, | |
| "grad_norm": 0.8128913044929504, | |
| "learning_rate": 3.3973589435774314e-05, | |
| "loss": 1.9473, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.35315391421318054, | |
| "learning_rate": 3.277310924369748e-05, | |
| "loss": 1.8123, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.6842737094837935, | |
| "grad_norm": 0.5528764724731445, | |
| "learning_rate": 3.157262905162065e-05, | |
| "loss": 1.8663, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6962785114045619, | |
| "grad_norm": 0.4432496726512909, | |
| "learning_rate": 3.037214885954382e-05, | |
| "loss": 1.8923, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.7082833133253301, | |
| "grad_norm": 0.415711373090744, | |
| "learning_rate": 2.917166866746699e-05, | |
| "loss": 1.8289, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.7202881152460985, | |
| "grad_norm": 0.42969274520874023, | |
| "learning_rate": 2.797118847539016e-05, | |
| "loss": 1.8472, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7202881152460985, | |
| "eval_loss": 1.8567453622817993, | |
| "eval_runtime": 38.6325, | |
| "eval_samples_per_second": 25.885, | |
| "eval_steps_per_second": 4.323, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7322929171668667, | |
| "grad_norm": 0.5071336030960083, | |
| "learning_rate": 2.6770708283313327e-05, | |
| "loss": 1.8568, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.7442977190876351, | |
| "grad_norm": 0.4259433150291443, | |
| "learning_rate": 2.5570228091236498e-05, | |
| "loss": 1.8289, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.7563025210084033, | |
| "grad_norm": 0.522114634513855, | |
| "learning_rate": 2.4369747899159663e-05, | |
| "loss": 1.9236, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.7683073229291717, | |
| "grad_norm": 0.4996345341205597, | |
| "learning_rate": 2.3169267707082835e-05, | |
| "loss": 1.8676, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.78031212484994, | |
| "grad_norm": 0.5358065962791443, | |
| "learning_rate": 2.1968787515006003e-05, | |
| "loss": 1.9048, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7923169267707083, | |
| "grad_norm": 0.4080793261528015, | |
| "learning_rate": 2.076830732292917e-05, | |
| "loss": 1.9375, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.8043217286914766, | |
| "grad_norm": 0.4291868209838867, | |
| "learning_rate": 1.9567827130852343e-05, | |
| "loss": 1.8131, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "grad_norm": 0.4137159287929535, | |
| "learning_rate": 1.836734693877551e-05, | |
| "loss": 1.8438, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.8283313325330132, | |
| "grad_norm": 0.46769535541534424, | |
| "learning_rate": 1.7166866746698683e-05, | |
| "loss": 1.8608, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.3832717537879944, | |
| "learning_rate": 1.5966386554621848e-05, | |
| "loss": 1.7676, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "eval_loss": 1.8551292419433594, | |
| "eval_runtime": 38.7937, | |
| "eval_samples_per_second": 25.777, | |
| "eval_steps_per_second": 4.305, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8523409363745498, | |
| "grad_norm": 0.6028046011924744, | |
| "learning_rate": 1.4765906362545018e-05, | |
| "loss": 1.8747, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.8643457382953181, | |
| "grad_norm": 0.4539136290550232, | |
| "learning_rate": 1.3565426170468188e-05, | |
| "loss": 1.8388, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.8763505402160864, | |
| "grad_norm": 0.5354043245315552, | |
| "learning_rate": 1.2364945978391356e-05, | |
| "loss": 1.8492, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.8883553421368547, | |
| "grad_norm": 0.5466801524162292, | |
| "learning_rate": 1.1164465786314526e-05, | |
| "loss": 1.8355, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.9003601440576231, | |
| "grad_norm": 0.7095188498497009, | |
| "learning_rate": 9.963985594237696e-06, | |
| "loss": 1.8558, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.9123649459783914, | |
| "grad_norm": 0.5179396867752075, | |
| "learning_rate": 8.763505402160866e-06, | |
| "loss": 1.9008, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.9243697478991597, | |
| "grad_norm": 0.41170573234558105, | |
| "learning_rate": 7.563025210084033e-06, | |
| "loss": 1.772, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.936374549819928, | |
| "grad_norm": 0.4213010370731354, | |
| "learning_rate": 6.362545018007203e-06, | |
| "loss": 1.8478, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.9483793517406963, | |
| "grad_norm": 0.42276838421821594, | |
| "learning_rate": 5.162064825930372e-06, | |
| "loss": 1.8858, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.9603841536614646, | |
| "grad_norm": 0.4341106712818146, | |
| "learning_rate": 3.9615846338535415e-06, | |
| "loss": 1.8898, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9603841536614646, | |
| "eval_loss": 1.8548470735549927, | |
| "eval_runtime": 38.7458, | |
| "eval_samples_per_second": 25.809, | |
| "eval_steps_per_second": 4.31, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9723889555822329, | |
| "grad_norm": 0.42987948656082153, | |
| "learning_rate": 2.7611044417767106e-06, | |
| "loss": 1.8521, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.9843937575030012, | |
| "grad_norm": 0.5700000524520874, | |
| "learning_rate": 1.56062424969988e-06, | |
| "loss": 1.8504, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.9963985594237695, | |
| "grad_norm": 0.4523591101169586, | |
| "learning_rate": 3.601440576230492e-07, | |
| "loss": 1.9274, | |
| "step": 830 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 833, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3742818208228966e+17, | |
| "train_batch_size": 12, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |