{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2130013831258646,
  "eval_steps": 50,
  "global_step": 2700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04098150709492342,
      "grad_norm": 0.5333279371261597,
      "learning_rate": 3.663003663003663e-06,
      "loss": 2.2117,
      "step": 50
    },
    {
      "epoch": 0.04098150709492342,
      "eval_loss": 2.7143771648406982,
      "eval_runtime": 282.6413,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 50
    },
    {
      "epoch": 0.08196301418984683,
      "grad_norm": 0.42295441031455994,
      "learning_rate": 7.326007326007326e-06,
      "loss": 2.1609,
      "step": 100
    },
    {
      "epoch": 0.08196301418984683,
      "eval_loss": 2.6825826168060303,
      "eval_runtime": 283.0578,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 100
    },
    {
      "epoch": 0.12294452128477025,
      "grad_norm": 0.277256041765213,
      "learning_rate": 1.098901098901099e-05,
      "loss": 2.0212,
      "step": 150
    },
    {
      "epoch": 0.12294452128477025,
      "eval_loss": 2.602968215942383,
      "eval_runtime": 282.5654,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 150
    },
    {
      "epoch": 0.16392602837969367,
      "grad_norm": 0.2273157685995102,
      "learning_rate": 1.4652014652014653e-05,
      "loss": 1.8938,
      "step": 200
    },
    {
      "epoch": 0.16392602837969367,
      "eval_loss": 2.506340980529785,
      "eval_runtime": 283.8897,
      "eval_samples_per_second": 0.764,
      "eval_steps_per_second": 0.099,
      "step": 200
    },
    {
      "epoch": 0.20490753547461707,
      "grad_norm": 0.21983110904693604,
      "learning_rate": 1.8315018315018315e-05,
      "loss": 1.7918,
      "step": 250
    },
    {
      "epoch": 0.20490753547461707,
      "eval_loss": 2.4188530445098877,
      "eval_runtime": 283.2047,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 250
    },
    {
      "epoch": 0.2458890425695405,
      "grad_norm": 0.20148395001888275,
      "learning_rate": 1.999406558079547e-05,
      "loss": 1.7115,
      "step": 300
    },
    {
      "epoch": 0.2458890425695405,
      "eval_loss": 2.3436834812164307,
      "eval_runtime": 282.2728,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 300
    },
    {
      "epoch": 0.2868705496644639,
      "grad_norm": 0.20601870119571686,
      "learning_rate": 1.9951769064396967e-05,
      "loss": 1.6595,
      "step": 350
    },
    {
      "epoch": 0.2868705496644639,
      "eval_loss": 2.3192338943481445,
      "eval_runtime": 282.1213,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 350
    },
    {
      "epoch": 0.32785205675938733,
      "grad_norm": 0.19198212027549744,
      "learning_rate": 1.986897612915546e-05,
      "loss": 1.6081,
      "step": 400
    },
    {
      "epoch": 0.32785205675938733,
      "eval_loss": 2.3058478832244873,
      "eval_runtime": 282.7227,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 400
    },
    {
      "epoch": 0.36883356385431076,
      "grad_norm": 0.19531460106372833,
      "learning_rate": 1.9746023681741606e-05,
      "loss": 1.6127,
      "step": 450
    },
    {
      "epoch": 0.36883356385431076,
      "eval_loss": 2.2954719066619873,
      "eval_runtime": 283.3445,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 450
    },
    {
      "epoch": 0.40981507094923414,
      "grad_norm": 0.2188873440027237,
      "learning_rate": 1.9583412048657773e-05,
      "loss": 1.5999,
      "step": 500
    },
    {
      "epoch": 0.40981507094923414,
      "eval_loss": 2.2856028079986572,
      "eval_runtime": 282.6955,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 500
    },
    {
      "epoch": 0.45079657804415757,
      "grad_norm": 0.2119743824005127,
      "learning_rate": 1.9381802940275198e-05,
      "loss": 1.6074,
      "step": 550
    },
    {
      "epoch": 0.45079657804415757,
      "eval_loss": 2.2773895263671875,
      "eval_runtime": 283.317,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 550
    },
    {
      "epoch": 0.491778085139081,
      "grad_norm": 0.21725280582904816,
      "learning_rate": 1.914201675815694e-05,
      "loss": 1.5989,
      "step": 600
    },
    {
      "epoch": 0.491778085139081,
      "eval_loss": 2.2710676193237305,
      "eval_runtime": 283.5744,
      "eval_samples_per_second": 0.765,
      "eval_steps_per_second": 0.099,
      "step": 600
    },
    {
      "epoch": 0.5327595922340044,
      "grad_norm": 0.22087362408638,
      "learning_rate": 1.8865029256623765e-05,
      "loss": 1.5708,
      "step": 650
    },
    {
      "epoch": 0.5327595922340044,
      "eval_loss": 2.2653110027313232,
      "eval_runtime": 282.6403,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 650
    },
    {
      "epoch": 0.5737410993289278,
      "grad_norm": 0.24597273766994476,
      "learning_rate": 1.855196757214796e-05,
      "loss": 1.5981,
      "step": 700
    },
    {
      "epoch": 0.5737410993289278,
      "eval_loss": 2.2589895725250244,
      "eval_runtime": 282.6053,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 700
    },
    {
      "epoch": 0.6147226064238512,
      "grad_norm": 0.20500172674655914,
      "learning_rate": 1.8204105636732604e-05,
      "loss": 1.5859,
      "step": 750
    },
    {
      "epoch": 0.6147226064238512,
      "eval_loss": 2.2537271976470947,
      "eval_runtime": 282.8109,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 750
    },
    {
      "epoch": 0.6557041135187747,
      "grad_norm": 0.24009792506694794,
      "learning_rate": 1.782285899394034e-05,
      "loss": 1.5765,
      "step": 800
    },
    {
      "epoch": 0.6557041135187747,
      "eval_loss": 2.250108242034912,
      "eval_runtime": 282.295,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 800
    },
    {
      "epoch": 0.6966856206136981,
      "grad_norm": 0.23266170918941498,
      "learning_rate": 1.74097790386668e-05,
      "loss": 1.5676,
      "step": 850
    },
    {
      "epoch": 0.6966856206136981,
      "eval_loss": 2.2453207969665527,
      "eval_runtime": 281.9055,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 850
    },
    {
      "epoch": 0.7376671277086215,
      "grad_norm": 0.2351984679698944,
      "learning_rate": 1.6966546704098455e-05,
      "loss": 1.5688,
      "step": 900
    },
    {
      "epoch": 0.7376671277086215,
      "eval_loss": 2.2418150901794434,
      "eval_runtime": 281.9821,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 900
    },
    {
      "epoch": 0.7786486348035448,
      "grad_norm": 0.237622931599617,
      "learning_rate": 1.6494965621544403e-05,
      "loss": 1.5643,
      "step": 950
    },
    {
      "epoch": 0.7786486348035448,
      "eval_loss": 2.2379164695739746,
      "eval_runtime": 282.2084,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 950
    },
    {
      "epoch": 0.8196301418984683,
      "grad_norm": 0.24924997985363007,
      "learning_rate": 1.5996954780976568e-05,
      "loss": 1.5346,
      "step": 1000
    },
    {
      "epoch": 0.8196301418984683,
      "eval_loss": 2.2346854209899902,
      "eval_runtime": 283.4225,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 1000
    },
    {
      "epoch": 0.8606116489933917,
      "grad_norm": 0.24776747822761536,
      "learning_rate": 1.547454072214457e-05,
      "loss": 1.5507,
      "step": 1050
    },
    {
      "epoch": 0.8606116489933917,
      "eval_loss": 2.2320806980133057,
      "eval_runtime": 282.9227,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 1050
    },
    {
      "epoch": 0.9015931560883151,
      "grad_norm": 0.2700960040092468,
      "learning_rate": 1.4929849288041656e-05,
      "loss": 1.5582,
      "step": 1100
    },
    {
      "epoch": 0.9015931560883151,
      "eval_loss": 2.2285408973693848,
      "eval_runtime": 283.0683,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 1100
    },
    {
      "epoch": 0.9425746631832386,
      "grad_norm": 0.26280835270881653,
      "learning_rate": 1.4365096974279093e-05,
      "loss": 1.5275,
      "step": 1150
    },
    {
      "epoch": 0.9425746631832386,
      "eval_loss": 2.2259719371795654,
      "eval_runtime": 282.6889,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 1150
    },
    {
      "epoch": 0.983556170278162,
      "grad_norm": 0.2555929124355316,
      "learning_rate": 1.3782581909570757e-05,
      "loss": 1.523,
      "step": 1200
    },
    {
      "epoch": 0.983556170278162,
      "eval_loss": 2.2228543758392334,
      "eval_runtime": 284.1011,
      "eval_samples_per_second": 0.764,
      "eval_steps_per_second": 0.099,
      "step": 1200
    },
    {
      "epoch": 1.0245376773730854,
      "grad_norm": 0.27351436018943787,
      "learning_rate": 1.3184674504030679e-05,
      "loss": 1.5354,
      "step": 1250
    },
    {
      "epoch": 1.0245376773730854,
      "eval_loss": 2.2217090129852295,
      "eval_runtime": 284.3948,
      "eval_samples_per_second": 0.763,
      "eval_steps_per_second": 0.098,
      "step": 1250
    },
    {
      "epoch": 1.0655191844680088,
      "grad_norm": 0.2628322243690491,
      "learning_rate": 1.2573807803338216e-05,
      "loss": 1.5386,
      "step": 1300
    },
    {
      "epoch": 1.0655191844680088,
      "eval_loss": 2.2189579010009766,
      "eval_runtime": 282.9448,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 1300
    },
    {
      "epoch": 1.1065006915629323,
      "grad_norm": 0.27614346146583557,
      "learning_rate": 1.1952467588022282e-05,
      "loss": 1.5338,
      "step": 1350
    },
    {
      "epoch": 1.1065006915629323,
      "eval_loss": 2.21720814704895,
      "eval_runtime": 282.859,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 1350
    },
    {
      "epoch": 1.1474821986578556,
      "grad_norm": 0.27579498291015625,
      "learning_rate": 1.1323182258153314e-05,
      "loss": 1.5292,
      "step": 1400
    },
    {
      "epoch": 1.1474821986578556,
      "eval_loss": 2.2148287296295166,
      "eval_runtime": 283.7073,
      "eval_samples_per_second": 0.765,
      "eval_steps_per_second": 0.099,
      "step": 1400
    },
    {
      "epoch": 1.1884637057527792,
      "grad_norm": 0.2812260687351227,
      "learning_rate": 1.0688512544604915e-05,
      "loss": 1.5376,
      "step": 1450
    },
    {
      "epoch": 1.1884637057527792,
      "eval_loss": 2.2129642963409424,
      "eval_runtime": 281.4833,
      "eval_samples_per_second": 0.771,
      "eval_steps_per_second": 0.099,
      "step": 1450
    },
    {
      "epoch": 1.2294452128477025,
      "grad_norm": 0.28713124990463257,
      "learning_rate": 1.005104108875275e-05,
      "loss": 1.5273,
      "step": 1500
    },
    {
      "epoch": 1.2294452128477025,
      "eval_loss": 2.2116787433624268,
      "eval_runtime": 283.0969,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 1500
    },
    {
      "epoch": 1.270426719942626,
      "grad_norm": 0.27820634841918945,
      "learning_rate": 9.41336193301377e-06,
      "loss": 1.526,
      "step": 1550
    },
    {
      "epoch": 1.270426719942626,
      "eval_loss": 2.2110674381256104,
      "eval_runtime": 281.8783,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 1550
    },
    {
      "epoch": 1.3114082270375493,
      "grad_norm": 0.26735347509384155,
      "learning_rate": 8.778069964991484e-06,
      "loss": 1.537,
      "step": 1600
    },
    {
      "epoch": 1.3114082270375493,
      "eval_loss": 2.2088816165924072,
      "eval_runtime": 281.9837,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 1600
    },
    {
      "epoch": 1.3523897341324727,
      "grad_norm": 0.2734984755516052,
      "learning_rate": 8.147750358182e-06,
      "loss": 1.5431,
      "step": 1650
    },
    {
      "epoch": 1.3523897341324727,
      "eval_loss": 2.207979917526245,
      "eval_runtime": 282.6878,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 1650
    },
    {
      "epoch": 1.3933712412273962,
      "grad_norm": 0.3110925257205963,
      "learning_rate": 7.524968052209331e-06,
      "loss": 1.5401,
      "step": 1700
    },
    {
      "epoch": 1.3933712412273962,
      "eval_loss": 2.2074739933013916,
      "eval_runtime": 282.1594,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 1700
    },
    {
      "epoch": 1.4343527483223195,
      "grad_norm": 0.3202177882194519,
      "learning_rate": 6.912257315397784e-06,
      "loss": 1.5331,
      "step": 1750
    },
    {
      "epoch": 1.4343527483223195,
      "eval_loss": 2.205904245376587,
      "eval_runtime": 280.7829,
      "eval_samples_per_second": 0.773,
      "eval_steps_per_second": 0.1,
      "step": 1750
    },
    {
      "epoch": 1.475334255417243,
      "grad_norm": 0.28285861015319824,
      "learning_rate": 6.312111432154074e-06,
      "loss": 1.5395,
      "step": 1800
    },
    {
      "epoch": 1.475334255417243,
      "eval_loss": 2.2046620845794678,
      "eval_runtime": 282.5413,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 1800
    },
    {
      "epoch": 1.5163157625121664,
      "grad_norm": 0.2836764454841614,
      "learning_rate": 5.726972557124022e-06,
      "loss": 1.542,
      "step": 1850
    },
    {
      "epoch": 1.5163157625121664,
      "eval_loss": 2.203850507736206,
      "eval_runtime": 283.3938,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 1850
    },
    {
      "epoch": 1.5572972696070897,
      "grad_norm": 0.2900823652744293,
      "learning_rate": 5.159221777409953e-06,
      "loss": 1.502,
      "step": 1900
    },
    {
      "epoch": 1.5572972696070897,
      "eval_loss": 2.2032361030578613,
      "eval_runtime": 281.099,
      "eval_samples_per_second": 0.772,
      "eval_steps_per_second": 0.1,
      "step": 1900
    },
    {
      "epoch": 1.5982787767020132,
      "grad_norm": 0.282176673412323,
      "learning_rate": 4.611169423288323e-06,
      "loss": 1.5267,
      "step": 1950
    },
    {
      "epoch": 1.5982787767020132,
      "eval_loss": 2.202617883682251,
      "eval_runtime": 283.2145,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 1950
    },
    {
      "epoch": 1.6392602837969368,
      "grad_norm": 0.29563507437705994,
      "learning_rate": 4.085045666855846e-06,
      "loss": 1.51,
      "step": 2000
    },
    {
      "epoch": 1.6392602837969368,
      "eval_loss": 2.2023086547851562,
      "eval_runtime": 282.4408,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 2000
    },
    {
      "epoch": 1.68024179089186,
      "grad_norm": 0.2923285663127899,
      "learning_rate": 3.5829914468607874e-06,
      "loss": 1.5319,
      "step": 2050
    },
    {
      "epoch": 1.68024179089186,
      "eval_loss": 2.2018585205078125,
      "eval_runtime": 281.9707,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 2050
    },
    {
      "epoch": 1.7212232979867834,
      "grad_norm": 0.2978394031524658,
      "learning_rate": 3.1070497566486825e-06,
      "loss": 1.5267,
      "step": 2100
    },
    {
      "epoch": 1.7212232979867834,
      "eval_loss": 2.2010927200317383,
      "eval_runtime": 282.5046,
      "eval_samples_per_second": 0.768,
      "eval_steps_per_second": 0.099,
      "step": 2100
    },
    {
      "epoch": 1.7622048050817067,
      "grad_norm": 0.2987017035484314,
      "learning_rate": 2.6591573306741704e-06,
      "loss": 1.5201,
      "step": 2150
    },
    {
      "epoch": 1.7622048050817067,
      "eval_loss": 2.200831174850464,
      "eval_runtime": 283.0038,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 2150
    },
    {
      "epoch": 1.8031863121766303,
      "grad_norm": 0.28967171907424927,
      "learning_rate": 2.241136763408801e-06,
      "loss": 1.5204,
      "step": 2200
    },
    {
      "epoch": 1.8031863121766303,
      "eval_loss": 2.2002053260803223,
      "eval_runtime": 281.6629,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 2200
    },
    {
      "epoch": 1.8441678192715538,
      "grad_norm": 0.321614146232605,
      "learning_rate": 1.8546890927150273e-06,
      "loss": 1.5094,
      "step": 2250
    },
    {
      "epoch": 1.8441678192715538,
      "eval_loss": 2.2001166343688965,
      "eval_runtime": 282.843,
      "eval_samples_per_second": 0.767,
      "eval_steps_per_second": 0.099,
      "step": 2250
    },
    {
      "epoch": 1.8851493263664771,
      "grad_norm": 0.2887614965438843,
      "learning_rate": 1.501386877866694e-06,
      "loss": 1.5268,
      "step": 2300
    },
    {
      "epoch": 1.8851493263664771,
      "eval_loss": 2.199751615524292,
      "eval_runtime": 282.0209,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 2300
    },
    {
      "epoch": 1.9261308334614005,
      "grad_norm": 0.30499428510665894,
      "learning_rate": 1.1826678003833402e-06,
      "loss": 1.513,
      "step": 2350
    },
    {
      "epoch": 1.9261308334614005,
      "eval_loss": 2.1995420455932617,
      "eval_runtime": 283.2079,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.099,
      "step": 2350
    },
    {
      "epoch": 1.967112340556324,
      "grad_norm": 0.29722264409065247,
      "learning_rate": 8.998288137183209e-07,
      "loss": 1.5263,
      "step": 2400
    },
    {
      "epoch": 1.967112340556324,
      "eval_loss": 2.199302911758423,
      "eval_runtime": 281.4307,
      "eval_samples_per_second": 0.771,
      "eval_steps_per_second": 0.099,
      "step": 2400
    },
    {
      "epoch": 2.0080938476512475,
      "grad_norm": 0.30117130279541016,
      "learning_rate": 6.540208656071601e-07,
      "loss": 1.5291,
      "step": 2450
    },
    {
      "epoch": 2.0080938476512475,
      "eval_loss": 2.199272394180298,
      "eval_runtime": 281.4783,
      "eval_samples_per_second": 0.771,
      "eval_steps_per_second": 0.099,
      "step": 2450
    },
    {
      "epoch": 2.049075354746171,
      "grad_norm": 0.29823198914527893,
      "learning_rate": 4.4624421455236156e-07,
      "loss": 1.5187,
      "step": 2500
    },
    {
      "epoch": 2.049075354746171,
      "eval_loss": 2.1992440223693848,
      "eval_runtime": 280.9197,
      "eval_samples_per_second": 0.772,
      "eval_steps_per_second": 0.1,
      "step": 2500
    },
    {
      "epoch": 2.090056861841094,
      "grad_norm": 0.3053443729877472,
      "learning_rate": 2.7734435950315663e-07,
      "loss": 1.5324,
      "step": 2550
    },
    {
      "epoch": 2.090056861841094,
      "eval_loss": 2.1991806030273438,
      "eval_runtime": 281.8919,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 2550
    },
    {
      "epoch": 2.1310383689360175,
      "grad_norm": 0.31491023302078247,
      "learning_rate": 1.4800859929338218e-07,
      "loss": 1.5314,
      "step": 2600
    },
    {
      "epoch": 2.1310383689360175,
      "eval_loss": 2.199115037918091,
      "eval_runtime": 282.019,
      "eval_samples_per_second": 0.769,
      "eval_steps_per_second": 0.099,
      "step": 2600
    },
    {
      "epoch": 2.172019876030941,
      "grad_norm": 0.300465852022171,
      "learning_rate": 5.876323583810184e-08,
      "loss": 1.5357,
      "step": 2650
    },
    {
      "epoch": 2.172019876030941,
      "eval_loss": 2.19909930229187,
      "eval_runtime": 281.0522,
      "eval_samples_per_second": 0.772,
      "eval_steps_per_second": 0.1,
      "step": 2650
    },
    {
      "epoch": 2.2130013831258646,
      "grad_norm": 0.3176944851875305,
      "learning_rate": 9.971432469871866e-09,
      "loss": 1.507,
      "step": 2700
    },
    {
      "epoch": 2.2130013831258646,
      "eval_loss": 2.1991000175476074,
      "eval_runtime": 281.6411,
      "eval_samples_per_second": 0.77,
      "eval_steps_per_second": 0.099,
      "step": 2700
    }
  ],
  "logging_steps": 50,
  "max_steps": 2735,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "total_flos": 1.100581152473088e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}