{ "best_metric": null, "best_model_checkpoint": null, "epoch": 8.807339449541285, "eval_steps": 500, "global_step": 1080, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.08154943934760449, "grad_norm": 13.197553634643555, "learning_rate": 9.259259259259259e-07, "loss": 1.1931, "step": 10 }, { "epoch": 0.16309887869520898, "grad_norm": 15.166808128356934, "learning_rate": 1.8518518518518519e-06, "loss": 1.123, "step": 20 }, { "epoch": 0.24464831804281345, "grad_norm": 16.447948455810547, "learning_rate": 2.7777777777777783e-06, "loss": 1.1451, "step": 30 }, { "epoch": 0.32619775739041795, "grad_norm": 9.410321235656738, "learning_rate": 3.7037037037037037e-06, "loss": 1.0174, "step": 40 }, { "epoch": 0.4077471967380224, "grad_norm": 13.654789924621582, "learning_rate": 4.62962962962963e-06, "loss": 1.0555, "step": 50 }, { "epoch": 0.4892966360856269, "grad_norm": 8.732908248901367, "learning_rate": 5.555555555555557e-06, "loss": 1.1169, "step": 60 }, { "epoch": 0.5708460754332314, "grad_norm": 16.178895950317383, "learning_rate": 6.481481481481482e-06, "loss": 1.1143, "step": 70 }, { "epoch": 0.6523955147808359, "grad_norm": 20.069887161254883, "learning_rate": 7.4074074074074075e-06, "loss": 0.9055, "step": 80 }, { "epoch": 0.7339449541284404, "grad_norm": 7.063018798828125, "learning_rate": 8.333333333333334e-06, "loss": 1.0901, "step": 90 }, { "epoch": 0.8154943934760448, "grad_norm": 10.402606964111328, "learning_rate": 9.25925925925926e-06, "loss": 0.98, "step": 100 }, { "epoch": 0.8970438328236493, "grad_norm": 11.173501014709473, "learning_rate": 9.999895536228031e-06, "loss": 1.1819, "step": 110 }, { "epoch": 0.9785932721712538, "grad_norm": 9.713117599487305, "learning_rate": 9.996239762521152e-06, "loss": 1.019, "step": 120 }, { "epoch": 1.0601427115188584, "grad_norm": 6.5643534660339355, "learning_rate": 9.987365164467767e-06, "loss": 0.8775, "step": 130 }, { "epoch": 1.1416921508664628, "grad_norm": 7.599371910095215, "learning_rate": 9.973281012033009e-06, "loss": 0.7782, "step": 140 }, { "epoch": 1.2232415902140672, "grad_norm": 7.510239601135254, "learning_rate": 9.954002016824226e-06, "loss": 0.7959, "step": 150 }, { "epoch": 1.3047910295616718, "grad_norm": 5.231692790985107, "learning_rate": 9.929548316723983e-06, "loss": 0.7244, "step": 160 }, { "epoch": 1.3863404689092762, "grad_norm": 15.204421043395996, "learning_rate": 9.899945454855007e-06, "loss": 0.6964, "step": 170 }, { "epoch": 1.4678899082568808, "grad_norm": 6.420439720153809, "learning_rate": 9.86522435289912e-06, "loss": 0.6273, "step": 180 }, { "epoch": 1.5494393476044852, "grad_norm": 5.670159816741943, "learning_rate": 9.825421278797984e-06, "loss": 0.6795, "step": 190 }, { "epoch": 1.6309887869520896, "grad_norm": 7.807967185974121, "learning_rate": 9.7805778088694e-06, "loss": 0.7074, "step": 200 }, { "epoch": 1.7125382262996942, "grad_norm": 5.227024078369141, "learning_rate": 9.730740784378755e-06, "loss": 0.7545, "step": 210 }, { "epoch": 1.7940876656472988, "grad_norm": 10.772171020507812, "learning_rate": 9.67596226261095e-06, "loss": 0.7335, "step": 220 }, { "epoch": 1.8756371049949032, "grad_norm": 5.900475025177002, "learning_rate": 9.616299462493952e-06, "loss": 0.7046, "step": 230 }, { "epoch": 1.9571865443425076, "grad_norm": 8.141891479492188, "learning_rate": 9.551814704830734e-06, "loss": 0.6498, "step": 240 }, { "epoch": 2.038735983690112, "grad_norm": 7.0482659339904785, "learning_rate": 9.482575347202047e-06, "loss": 0.6264, "step": 250 }, { "epoch": 2.120285423037717, "grad_norm": 9.526334762573242, "learning_rate": 9.40865371360804e-06, "loss": 0.4823, "step": 260 }, { "epoch": 2.2018348623853212, "grad_norm": 4.857308864593506, "learning_rate": 9.330127018922195e-06, "loss": 0.4391, "step": 270 }, { "epoch": 2.2833843017329256, "grad_norm": 15.281082153320312, "learning_rate": 9.247077288236488e-06, "loss": 0.3913, "step": 280 }, { "epoch": 2.36493374108053, "grad_norm": 9.266241073608398, "learning_rate": 9.159591271182058e-06, "loss": 0.4327, "step": 290 }, { "epoch": 2.4464831804281344, "grad_norm": 4.62817907333374, "learning_rate": 9.067760351314838e-06, "loss": 0.4115, "step": 300 }, { "epoch": 2.528032619775739, "grad_norm": 6.626443862915039, "learning_rate": 8.97168045066082e-06, "loss": 0.3849, "step": 310 }, { "epoch": 2.6095820591233436, "grad_norm": 7.996133804321289, "learning_rate": 8.871451929520662e-06, "loss": 0.4089, "step": 320 }, { "epoch": 2.691131498470948, "grad_norm": 7.03031587600708, "learning_rate": 8.767179481638303e-06, "loss": 0.451, "step": 330 }, { "epoch": 2.7726809378185524, "grad_norm": 5.80501127243042, "learning_rate": 8.658972024843063e-06, "loss": 0.4384, "step": 340 }, { "epoch": 2.8542303771661572, "grad_norm": 7.129272937774658, "learning_rate": 8.546942587279465e-06, "loss": 0.4124, "step": 350 }, { "epoch": 2.9357798165137616, "grad_norm": 6.861177921295166, "learning_rate": 8.43120818934367e-06, "loss": 0.4244, "step": 360 }, { "epoch": 3.017329255861366, "grad_norm": 4.206913948059082, "learning_rate": 8.31188972144974e-06, "loss": 0.3352, "step": 370 }, { "epoch": 3.0988786952089704, "grad_norm": 6.867356300354004, "learning_rate": 8.18911181775353e-06, "loss": 0.2296, "step": 380 }, { "epoch": 3.180428134556575, "grad_norm": 6.237820625305176, "learning_rate": 8.063002725966014e-06, "loss": 0.2352, "step": 390 }, { "epoch": 3.261977573904179, "grad_norm": 5.08056640625, "learning_rate": 7.93369417339209e-06, "loss": 0.2583, "step": 400 }, { "epoch": 3.343527013251784, "grad_norm": 15.075705528259277, "learning_rate": 7.801321229334764e-06, "loss": 0.2715, "step": 410 }, { "epoch": 3.4250764525993884, "grad_norm": 3.862241744995117, "learning_rate": 7.666022164008458e-06, "loss": 0.2239, "step": 420 }, { "epoch": 3.506625891946993, "grad_norm": 3.4025380611419678, "learning_rate": 7.527938304108795e-06, "loss": 0.2292, "step": 430 }, { "epoch": 3.588175331294597, "grad_norm": 3.6665425300598145, "learning_rate": 7.387213885189746e-06, "loss": 0.2294, "step": 440 }, { "epoch": 3.669724770642202, "grad_norm": 6.751482009887695, "learning_rate": 7.243995901002312e-06, "loss": 0.2348, "step": 450 }, { "epoch": 3.7512742099898064, "grad_norm": 2.690695285797119, "learning_rate": 7.098433949952146e-06, "loss": 0.2471, "step": 460 }, { "epoch": 3.832823649337411, "grad_norm": 3.8786795139312744, "learning_rate": 6.950680078836475e-06, "loss": 0.2518, "step": 470 }, { "epoch": 3.914373088685015, "grad_norm": 4.775713920593262, "learning_rate": 6.800888624023552e-06, "loss": 0.2274, "step": 480 }, { "epoch": 3.9959225280326196, "grad_norm": 2.959533929824829, "learning_rate": 6.649216050240539e-06, "loss": 0.245, "step": 490 }, { "epoch": 4.077471967380224, "grad_norm": 1.844783067703247, "learning_rate": 6.495820787138209e-06, "loss": 0.1411, "step": 500 }, { "epoch": 4.077471967380224, "eval_loss": 1.580026388168335, "eval_runtime": 3.9804, "eval_samples_per_second": 27.636, "eval_steps_per_second": 27.636, "step": 500 }, { "epoch": 4.159021406727828, "grad_norm": 6.747076988220215, "learning_rate": 6.340863063803187e-06, "loss": 0.1365, "step": 510 }, { "epoch": 4.240570846075434, "grad_norm": 2.562960624694824, "learning_rate": 6.184504741390596e-06, "loss": 0.1192, "step": 520 }, { "epoch": 4.322120285423038, "grad_norm": 5.0884552001953125, "learning_rate": 6.02690914405191e-06, "loss": 0.1517, "step": 530 }, { "epoch": 4.4036697247706424, "grad_norm": 3.955160617828369, "learning_rate": 5.8682408883346535e-06, "loss": 0.0783, "step": 540 }, { "epoch": 4.485219164118247, "grad_norm": 6.160187721252441, "learning_rate": 5.708665711232103e-06, "loss": 0.1147, "step": 550 }, { "epoch": 4.566768603465851, "grad_norm": 1.819947361946106, "learning_rate": 5.548350297062659e-06, "loss": 0.1061, "step": 560 }, { "epoch": 4.648318042813456, "grad_norm": 5.095085620880127, "learning_rate": 5.387462103359655e-06, "loss": 0.1298, "step": 570 }, { "epoch": 4.72986748216106, "grad_norm": 5.54387092590332, "learning_rate": 5.2261691859535325e-06, "loss": 0.1029, "step": 580 }, { "epoch": 4.811416921508664, "grad_norm": 1.481745958328247, "learning_rate": 5.064640023429042e-06, "loss": 0.0996, "step": 590 }, { "epoch": 4.892966360856269, "grad_norm": 3.6181232929229736, "learning_rate": 4.903043341140879e-06, "loss": 0.1103, "step": 600 }, { "epoch": 4.974515800203873, "grad_norm": 5.307178497314453, "learning_rate": 4.741547934971528e-06, "loss": 0.126, "step": 610 }, { "epoch": 5.0560652395514785, "grad_norm": 2.1945080757141113, "learning_rate": 4.580322495015466e-06, "loss": 0.0616, "step": 620 }, { "epoch": 5.137614678899083, "grad_norm": 3.431023597717285, "learning_rate": 4.4195354293738484e-06, "loss": 0.049, "step": 630 }, { "epoch": 5.219164118246687, "grad_norm": 4.72199010848999, "learning_rate": 4.259354688243758e-06, "loss": 0.0563, "step": 640 }, { "epoch": 5.300713557594292, "grad_norm": 2.1474854946136475, "learning_rate": 4.099947588485744e-06, "loss": 0.0464, "step": 650 }, { "epoch": 5.382262996941896, "grad_norm": 1.2775018215179443, "learning_rate": 3.941480638852948e-06, "loss": 0.0454, "step": 660 }, { "epoch": 5.4638124362895, "grad_norm": 2.0239715576171875, "learning_rate": 3.784119366064293e-06, "loss": 0.0776, "step": 670 }, { "epoch": 5.545361875637105, "grad_norm": 9.16295051574707, "learning_rate": 3.6280281419034934e-06, "loss": 0.0565, "step": 680 }, { "epoch": 5.626911314984709, "grad_norm": 3.2734322547912598, "learning_rate": 3.473370011524435e-06, "loss": 0.0624, "step": 690 }, { "epoch": 5.708460754332314, "grad_norm": 2.0991153717041016, "learning_rate": 3.3203065231422904e-06, "loss": 0.0506, "step": 700 }, { "epoch": 5.790010193679919, "grad_norm": 2.030103921890259, "learning_rate": 3.1689975592882603e-06, "loss": 0.0482, "step": 710 }, { "epoch": 5.871559633027523, "grad_norm": 3.2920758724212646, "learning_rate": 3.019601169804216e-06, "loss": 0.062, "step": 720 }, { "epoch": 5.953109072375128, "grad_norm": 8.777363777160645, "learning_rate": 2.8722734067516637e-06, "loss": 0.0573, "step": 730 }, { "epoch": 6.034658511722732, "grad_norm": 1.1958752870559692, "learning_rate": 2.7271681614074973e-06, "loss": 0.0389, "step": 740 }, { "epoch": 6.116207951070336, "grad_norm": 5.468116283416748, "learning_rate": 2.5844370035168077e-06, "loss": 0.0331, "step": 750 }, { "epoch": 6.197757390417941, "grad_norm": 0.45254144072532654, "learning_rate": 2.4442290229706344e-06, "loss": 0.0248, "step": 760 }, { "epoch": 6.279306829765545, "grad_norm": 1.1617120504379272, "learning_rate": 2.3066906740740626e-06, "loss": 0.023, "step": 770 }, { "epoch": 6.36085626911315, "grad_norm": 1.583760380744934, "learning_rate": 2.171965622567308e-06, "loss": 0.0167, "step": 780 }, { "epoch": 6.442405708460754, "grad_norm": 0.8195653557777405, "learning_rate": 2.0401945955596206e-06, "loss": 0.0369, "step": 790 }, { "epoch": 6.523955147808358, "grad_norm": 1.8565921783447266, "learning_rate": 1.9115152345327154e-06, "loss": 0.0183, "step": 800 }, { "epoch": 6.605504587155964, "grad_norm": 1.3874293565750122, "learning_rate": 1.7860619515673034e-06, "loss": 0.0191, "step": 810 }, { "epoch": 6.687054026503568, "grad_norm": 1.5542582273483276, "learning_rate": 1.6639657889429017e-06, "loss": 0.0183, "step": 820 }, { "epoch": 6.7686034658511725, "grad_norm": 1.0783458948135376, "learning_rate": 1.5453542822575624e-06, "loss": 0.02, "step": 830 }, { "epoch": 6.850152905198777, "grad_norm": 1.0901105403900146, "learning_rate": 1.4303513272105057e-06, "loss": 0.0197, "step": 840 }, { "epoch": 6.931702344546381, "grad_norm": 1.554459810256958, "learning_rate": 1.3190770501868243e-06, "loss": 0.0211, "step": 850 }, { "epoch": 7.013251783893986, "grad_norm": 3.050743818283081, "learning_rate": 1.2116476827794104e-06, "loss": 0.0148, "step": 860 }, { "epoch": 7.09480122324159, "grad_norm": 0.37844064831733704, "learning_rate": 1.1081754403792e-06, "loss": 0.0063, "step": 870 }, { "epoch": 7.176350662589194, "grad_norm": 0.4527202844619751, "learning_rate": 1.008768404960535e-06, "loss": 0.0087, "step": 880 }, { "epoch": 7.257900101936799, "grad_norm": 0.7698076963424683, "learning_rate": 9.135304121840976e-07, "loss": 0.0086, "step": 890 }, { "epoch": 7.339449541284404, "grad_norm": 0.1463124006986618, "learning_rate": 8.225609429353187e-07, "loss": 0.0064, "step": 900 }, { "epoch": 7.4209989806320085, "grad_norm": 0.4381123483181, "learning_rate": 7.35955019411585e-07, "loss": 0.0105, "step": 910 }, { "epoch": 7.502548419979613, "grad_norm": 1.877506971359253, "learning_rate": 6.53803105866761e-07, "loss": 0.008, "step": 920 }, { "epoch": 7.584097859327217, "grad_norm": 1.034546971321106, "learning_rate": 5.76191014116711e-07, "loss": 0.0106, "step": 930 }, { "epoch": 7.665647298674822, "grad_norm": 0.48700782656669617, "learning_rate": 5.031998139045352e-07, "loss": 0.0094, "step": 940 }, { "epoch": 7.747196738022426, "grad_norm": 1.1844407320022583, "learning_rate": 4.349057482191299e-07, "loss": 0.0108, "step": 950 }, { "epoch": 7.82874617737003, "grad_norm": 1.7279776334762573, "learning_rate": 3.7138015365554834e-07, "loss": 0.0143, "step": 960 }, { "epoch": 7.910295616717635, "grad_norm": 0.49893057346343994, "learning_rate": 3.1268938590032495e-07, "loss": 0.0091, "step": 970 }, { "epoch": 7.991845056065239, "grad_norm": 0.29836803674697876, "learning_rate": 2.5889475041961767e-07, "loss": 0.0063, "step": 980 }, { "epoch": 8.073394495412844, "grad_norm": 0.30645838379859924, "learning_rate": 2.1005243842255552e-07, "loss": 0.015, "step": 990 }, { "epoch": 8.154943934760448, "grad_norm": 0.2830670475959778, "learning_rate": 1.6621346816668993e-07, "loss": 0.0056, "step": 1000 }, { "epoch": 8.154943934760448, "eval_loss": 2.430070638656616, "eval_runtime": 3.8698, "eval_samples_per_second": 28.425, "eval_steps_per_second": 28.425, "step": 1000 }, { "epoch": 8.236493374108052, "grad_norm": 0.15629835426807404, "learning_rate": 1.2742363166685035e-07, "loss": 0.006, "step": 1010 }, { "epoch": 8.318042813455657, "grad_norm": 0.0809037983417511, "learning_rate": 9.372344686307655e-08, "loss": 0.0047, "step": 1020 }, { "epoch": 8.399592252803261, "grad_norm": 0.775462806224823, "learning_rate": 6.514811529758747e-08, "loss": 0.009, "step": 1030 }, { "epoch": 8.481141692150867, "grad_norm": 0.14626573026180267, "learning_rate": 4.172748534499449e-08, "loss": 0.008, "step": 1040 }, { "epoch": 8.562691131498472, "grad_norm": 0.44388526678085327, "learning_rate": 2.3486021034170857e-08, "loss": 0.0059, "step": 1050 }, { "epoch": 8.644240570846076, "grad_norm": 0.3372404873371124, "learning_rate": 1.044277649433989e-08, "loss": 0.006, "step": 1060 }, { "epoch": 8.72579001019368, "grad_norm": 0.5315991640090942, "learning_rate": 2.611376052073511e-09, "loss": 0.007, "step": 1070 }, { "epoch": 8.807339449541285, "grad_norm": 0.2758943736553192, "learning_rate": 0.0, "loss": 0.0048, "step": 1080 }, { "epoch": 8.807339449541285, "step": 1080, "total_flos": 1.2056852760477696e+16, "train_loss": 0.3034052159499239, "train_runtime": 1803.2161, "train_samples_per_second": 4.791, "train_steps_per_second": 0.599 } ], "logging_steps": 10, "max_steps": 1080, "num_input_tokens_seen": 0, "num_train_epochs": 9, "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": 1.2056852760477696e+16, "train_batch_size": 1, "trial_name": null, "trial_params": null }