{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5806111696522656,
  "eval_steps": 10,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01053740779768177,
      "grad_norm": 3.3616116046905518,
      "learning_rate": 4.982437653670531e-05,
      "loss": 1.7881,
      "step": 10
    },
    {
      "epoch": 0.01053740779768177,
      "eval_loss": 0.6833231449127197,
      "eval_runtime": 13.3064,
      "eval_samples_per_second": 36.073,
      "eval_steps_per_second": 2.255,
      "step": 10
    },
    {
      "epoch": 0.02107481559536354,
      "grad_norm": 1.3658663034439087,
      "learning_rate": 4.964875307341061e-05,
      "loss": 0.5571,
      "step": 20
    },
    {
      "epoch": 0.02107481559536354,
      "eval_loss": 0.5293903350830078,
      "eval_runtime": 13.4286,
      "eval_samples_per_second": 35.745,
      "eval_steps_per_second": 2.234,
      "step": 20
    },
    {
      "epoch": 0.03161222339304531,
      "grad_norm": 1.0748717784881592,
      "learning_rate": 4.947312961011591e-05,
      "loss": 0.4482,
      "step": 30
    },
    {
      "epoch": 0.03161222339304531,
      "eval_loss": 0.47912800312042236,
      "eval_runtime": 13.4918,
      "eval_samples_per_second": 35.577,
      "eval_steps_per_second": 2.224,
      "step": 30
    },
    {
      "epoch": 0.04214963119072708,
      "grad_norm": 0.890529215335846,
      "learning_rate": 4.929750614682122e-05,
      "loss": 0.3722,
      "step": 40
    },
    {
      "epoch": 0.04214963119072708,
      "eval_loss": 0.44374603033065796,
      "eval_runtime": 13.5807,
      "eval_samples_per_second": 35.344,
      "eval_steps_per_second": 2.209,
      "step": 40
    },
    {
      "epoch": 0.05268703898840885,
      "grad_norm": 0.8233364224433899,
      "learning_rate": 4.9121882683526524e-05,
      "loss": 0.3534,
      "step": 50
    },
    {
      "epoch": 0.05268703898840885,
      "eval_loss": 0.4318523108959198,
      "eval_runtime": 13.6096,
      "eval_samples_per_second": 35.269,
      "eval_steps_per_second": 2.204,
      "step": 50
    },
    {
      "epoch": 0.06322444678609063,
      "grad_norm": 0.7144497036933899,
      "learning_rate": 4.894625922023183e-05,
      "loss": 0.3512,
      "step": 60
    },
    {
      "epoch": 0.06322444678609063,
      "eval_loss": 0.41558387875556946,
      "eval_runtime": 13.6488,
      "eval_samples_per_second": 35.168,
      "eval_steps_per_second": 2.198,
      "step": 60
    },
    {
      "epoch": 0.0737618545837724,
      "grad_norm": 0.6796255707740784,
      "learning_rate": 4.877063575693713e-05,
      "loss": 0.3588,
      "step": 70
    },
    {
      "epoch": 0.0737618545837724,
      "eval_loss": 0.4085235297679901,
      "eval_runtime": 13.6512,
      "eval_samples_per_second": 35.162,
      "eval_steps_per_second": 2.198,
      "step": 70
    },
    {
      "epoch": 0.08429926238145416,
      "grad_norm": 0.6846384406089783,
      "learning_rate": 4.8595012293642434e-05,
      "loss": 0.3504,
      "step": 80
    },
    {
      "epoch": 0.08429926238145416,
      "eval_loss": 0.3993188142776489,
      "eval_runtime": 13.6377,
      "eval_samples_per_second": 35.197,
      "eval_steps_per_second": 2.2,
      "step": 80
    },
    {
      "epoch": 0.09483667017913593,
      "grad_norm": 0.7391332983970642,
      "learning_rate": 4.841938883034774e-05,
      "loss": 0.3299,
      "step": 90
    },
    {
      "epoch": 0.09483667017913593,
      "eval_loss": 0.3922707140445709,
      "eval_runtime": 13.6059,
      "eval_samples_per_second": 35.279,
      "eval_steps_per_second": 2.205,
      "step": 90
    },
    {
      "epoch": 0.1053740779768177,
      "grad_norm": 0.6717132925987244,
      "learning_rate": 4.824376536705304e-05,
      "loss": 0.3358,
      "step": 100
    },
    {
      "epoch": 0.1053740779768177,
      "eval_loss": 0.390600323677063,
      "eval_runtime": 13.6022,
      "eval_samples_per_second": 35.288,
      "eval_steps_per_second": 2.206,
      "step": 100
    },
    {
      "epoch": 0.11591148577449947,
      "grad_norm": 0.6522348523139954,
      "learning_rate": 4.8068141903758344e-05,
      "loss": 0.311,
      "step": 110
    },
    {
      "epoch": 0.11591148577449947,
      "eval_loss": 0.3844388425350189,
      "eval_runtime": 13.5921,
      "eval_samples_per_second": 35.315,
      "eval_steps_per_second": 2.207,
      "step": 110
    },
    {
      "epoch": 0.12644889357218125,
      "grad_norm": 0.5558998584747314,
      "learning_rate": 4.789251844046364e-05,
      "loss": 0.3161,
      "step": 120
    },
    {
      "epoch": 0.12644889357218125,
      "eval_loss": 0.3852270245552063,
      "eval_runtime": 13.5961,
      "eval_samples_per_second": 35.304,
      "eval_steps_per_second": 2.207,
      "step": 120
    },
    {
      "epoch": 0.136986301369863,
      "grad_norm": 0.639998197555542,
      "learning_rate": 4.7716894977168955e-05,
      "loss": 0.3037,
      "step": 130
    },
    {
      "epoch": 0.136986301369863,
      "eval_loss": 0.3841981887817383,
      "eval_runtime": 13.6057,
      "eval_samples_per_second": 35.279,
      "eval_steps_per_second": 2.205,
      "step": 130
    },
    {
      "epoch": 0.1475237091675448,
      "grad_norm": 0.583459198474884,
      "learning_rate": 4.754127151387426e-05,
      "loss": 0.3232,
      "step": 140
    },
    {
      "epoch": 0.1475237091675448,
      "eval_loss": 0.3742731213569641,
      "eval_runtime": 13.5955,
      "eval_samples_per_second": 35.306,
      "eval_steps_per_second": 2.207,
      "step": 140
    },
    {
      "epoch": 0.15806111696522657,
      "grad_norm": 0.5834677219390869,
      "learning_rate": 4.736564805057956e-05,
      "loss": 0.317,
      "step": 150
    },
    {
      "epoch": 0.15806111696522657,
      "eval_loss": 0.3765258193016052,
      "eval_runtime": 13.5995,
      "eval_samples_per_second": 35.295,
      "eval_steps_per_second": 2.206,
      "step": 150
    },
    {
      "epoch": 0.16859852476290832,
      "grad_norm": 0.5975239872932434,
      "learning_rate": 4.7190024587284866e-05,
      "loss": 0.2987,
      "step": 160
    },
    {
      "epoch": 0.16859852476290832,
      "eval_loss": 0.37060925364494324,
      "eval_runtime": 13.5971,
      "eval_samples_per_second": 35.302,
      "eval_steps_per_second": 2.206,
      "step": 160
    },
    {
      "epoch": 0.1791359325605901,
      "grad_norm": 0.6860706806182861,
      "learning_rate": 4.7014401123990165e-05,
      "loss": 0.3166,
      "step": 170
    },
    {
      "epoch": 0.1791359325605901,
      "eval_loss": 0.37034350633621216,
      "eval_runtime": 13.5923,
      "eval_samples_per_second": 35.314,
      "eval_steps_per_second": 2.207,
      "step": 170
    },
    {
      "epoch": 0.18967334035827185,
      "grad_norm": 0.7161134481430054,
      "learning_rate": 4.683877766069547e-05,
      "loss": 0.3036,
      "step": 180
    },
    {
      "epoch": 0.18967334035827185,
      "eval_loss": 0.3633898198604584,
      "eval_runtime": 13.6104,
      "eval_samples_per_second": 35.267,
      "eval_steps_per_second": 2.204,
      "step": 180
    },
    {
      "epoch": 0.20021074815595363,
      "grad_norm": 0.8212491273880005,
      "learning_rate": 4.6663154197400776e-05,
      "loss": 0.3023,
      "step": 190
    },
    {
      "epoch": 0.20021074815595363,
      "eval_loss": 0.36356136202812195,
      "eval_runtime": 13.5967,
      "eval_samples_per_second": 35.303,
      "eval_steps_per_second": 2.206,
      "step": 190
    },
    {
      "epoch": 0.2107481559536354,
      "grad_norm": 0.5965659022331238,
      "learning_rate": 4.6487530734106075e-05,
      "loss": 0.2999,
      "step": 200
    },
    {
      "epoch": 0.2107481559536354,
      "eval_loss": 0.3622604310512543,
      "eval_runtime": 13.5997,
      "eval_samples_per_second": 35.295,
      "eval_steps_per_second": 2.206,
      "step": 200
    },
    {
      "epoch": 0.22128556375131717,
      "grad_norm": 0.6638055443763733,
      "learning_rate": 4.631190727081138e-05,
      "loss": 0.2924,
      "step": 210
    },
    {
      "epoch": 0.22128556375131717,
      "eval_loss": 0.36028483510017395,
      "eval_runtime": 13.6089,
      "eval_samples_per_second": 35.271,
      "eval_steps_per_second": 2.204,
      "step": 210
    },
    {
      "epoch": 0.23182297154899895,
      "grad_norm": 0.4681139886379242,
      "learning_rate": 4.6136283807516686e-05,
      "loss": 0.3012,
      "step": 220
    },
    {
      "epoch": 0.23182297154899895,
      "eval_loss": 0.3585042357444763,
      "eval_runtime": 13.6099,
      "eval_samples_per_second": 35.269,
      "eval_steps_per_second": 2.204,
      "step": 220
    },
    {
      "epoch": 0.24236037934668073,
      "grad_norm": 0.6495727300643921,
      "learning_rate": 4.596066034422199e-05,
      "loss": 0.2884,
      "step": 230
    },
    {
      "epoch": 0.24236037934668073,
      "eval_loss": 0.36549657583236694,
      "eval_runtime": 13.6085,
      "eval_samples_per_second": 35.272,
      "eval_steps_per_second": 2.205,
      "step": 230
    },
    {
      "epoch": 0.2528977871443625,
      "grad_norm": 0.4928750991821289,
      "learning_rate": 4.57850368809273e-05,
      "loss": 0.2838,
      "step": 240
    },
    {
      "epoch": 0.2528977871443625,
      "eval_loss": 0.35502439737319946,
      "eval_runtime": 13.6002,
      "eval_samples_per_second": 35.293,
      "eval_steps_per_second": 2.206,
      "step": 240
    },
    {
      "epoch": 0.26343519494204426,
      "grad_norm": 0.6839736700057983,
      "learning_rate": 4.5609413417632596e-05,
      "loss": 0.2861,
      "step": 250
    },
    {
      "epoch": 0.26343519494204426,
      "eval_loss": 0.3560730516910553,
      "eval_runtime": 13.6056,
      "eval_samples_per_second": 35.28,
      "eval_steps_per_second": 2.205,
      "step": 250
    },
    {
      "epoch": 0.273972602739726,
      "grad_norm": 0.5583881735801697,
      "learning_rate": 4.54337899543379e-05,
      "loss": 0.3019,
      "step": 260
    },
    {
      "epoch": 0.273972602739726,
      "eval_loss": 0.3566754162311554,
      "eval_runtime": 13.5986,
      "eval_samples_per_second": 35.298,
      "eval_steps_per_second": 2.206,
      "step": 260
    },
    {
      "epoch": 0.2845100105374078,
      "grad_norm": 0.5884770750999451,
      "learning_rate": 4.525816649104321e-05,
      "loss": 0.3112,
      "step": 270
    },
    {
      "epoch": 0.2845100105374078,
      "eval_loss": 0.3552027940750122,
      "eval_runtime": 13.6456,
      "eval_samples_per_second": 35.176,
      "eval_steps_per_second": 2.199,
      "step": 270
    },
    {
      "epoch": 0.2950474183350896,
      "grad_norm": 0.540367603302002,
      "learning_rate": 4.5082543027748506e-05,
      "loss": 0.2805,
      "step": 280
    },
    {
      "epoch": 0.2950474183350896,
      "eval_loss": 0.3551066815853119,
      "eval_runtime": 13.59,
      "eval_samples_per_second": 35.32,
      "eval_steps_per_second": 2.208,
      "step": 280
    },
    {
      "epoch": 0.3055848261327713,
      "grad_norm": 0.4952965974807739,
      "learning_rate": 4.490691956445381e-05,
      "loss": 0.318,
      "step": 290
    },
    {
      "epoch": 0.3055848261327713,
      "eval_loss": 0.35322481393814087,
      "eval_runtime": 13.5781,
      "eval_samples_per_second": 35.351,
      "eval_steps_per_second": 2.209,
      "step": 290
    },
    {
      "epoch": 0.31612223393045313,
      "grad_norm": 0.48861801624298096,
      "learning_rate": 4.473129610115912e-05,
      "loss": 0.2878,
      "step": 300
    },
    {
      "epoch": 0.31612223393045313,
      "eval_loss": 0.3530054986476898,
      "eval_runtime": 13.5721,
      "eval_samples_per_second": 35.367,
      "eval_steps_per_second": 2.21,
      "step": 300
    },
    {
      "epoch": 0.3266596417281349,
      "grad_norm": 0.523873507976532,
      "learning_rate": 4.455567263786442e-05,
      "loss": 0.3084,
      "step": 310
    },
    {
      "epoch": 0.3266596417281349,
      "eval_loss": 0.35032400488853455,
      "eval_runtime": 13.5686,
      "eval_samples_per_second": 35.376,
      "eval_steps_per_second": 2.211,
      "step": 310
    },
    {
      "epoch": 0.33719704952581664,
      "grad_norm": 0.6253566741943359,
      "learning_rate": 4.438004917456973e-05,
      "loss": 0.2947,
      "step": 320
    },
    {
      "epoch": 0.33719704952581664,
      "eval_loss": 0.3499966263771057,
      "eval_runtime": 13.5767,
      "eval_samples_per_second": 35.355,
      "eval_steps_per_second": 2.21,
      "step": 320
    },
    {
      "epoch": 0.34773445732349845,
      "grad_norm": 0.5352979898452759,
      "learning_rate": 4.420442571127503e-05,
      "loss": 0.2884,
      "step": 330
    },
    {
      "epoch": 0.34773445732349845,
      "eval_loss": 0.34622183442115784,
      "eval_runtime": 13.5616,
      "eval_samples_per_second": 35.394,
      "eval_steps_per_second": 2.212,
      "step": 330
    },
    {
      "epoch": 0.3582718651211802,
      "grad_norm": 0.4931572675704956,
      "learning_rate": 4.4028802247980333e-05,
      "loss": 0.3085,
      "step": 340
    },
    {
      "epoch": 0.3582718651211802,
      "eval_loss": 0.3468419909477234,
      "eval_runtime": 13.5639,
      "eval_samples_per_second": 35.388,
      "eval_steps_per_second": 2.212,
      "step": 340
    },
    {
      "epoch": 0.36880927291886195,
      "grad_norm": 0.5033659934997559,
      "learning_rate": 4.385317878468563e-05,
      "loss": 0.3084,
      "step": 350
    },
    {
      "epoch": 0.36880927291886195,
      "eval_loss": 0.34584841132164,
      "eval_runtime": 13.5544,
      "eval_samples_per_second": 35.413,
      "eval_steps_per_second": 2.213,
      "step": 350
    },
    {
      "epoch": 0.3793466807165437,
      "grad_norm": 0.5806519389152527,
      "learning_rate": 4.367755532139094e-05,
      "loss": 0.2761,
      "step": 360
    },
    {
      "epoch": 0.3793466807165437,
      "eval_loss": 0.34370240569114685,
      "eval_runtime": 13.5617,
      "eval_samples_per_second": 35.394,
      "eval_steps_per_second": 2.212,
      "step": 360
    },
    {
      "epoch": 0.3898840885142255,
      "grad_norm": 0.6204981803894043,
      "learning_rate": 4.3501931858096244e-05,
      "loss": 0.3024,
      "step": 370
    },
    {
      "epoch": 0.3898840885142255,
      "eval_loss": 0.3421412408351898,
      "eval_runtime": 13.5312,
      "eval_samples_per_second": 35.474,
      "eval_steps_per_second": 2.217,
      "step": 370
    },
    {
      "epoch": 0.40042149631190727,
      "grad_norm": 0.48026618361473083,
      "learning_rate": 4.332630839480154e-05,
      "loss": 0.2788,
      "step": 380
    },
    {
      "epoch": 0.40042149631190727,
      "eval_loss": 0.3414471745491028,
      "eval_runtime": 13.5435,
      "eval_samples_per_second": 35.441,
      "eval_steps_per_second": 2.215,
      "step": 380
    },
    {
      "epoch": 0.410958904109589,
      "grad_norm": 0.48742201924324036,
      "learning_rate": 4.3150684931506855e-05,
      "loss": 0.2708,
      "step": 390
    },
    {
      "epoch": 0.410958904109589,
      "eval_loss": 0.3414710462093353,
      "eval_runtime": 13.5593,
      "eval_samples_per_second": 35.4,
      "eval_steps_per_second": 2.213,
      "step": 390
    },
    {
      "epoch": 0.4214963119072708,
      "grad_norm": 0.5737293362617493,
      "learning_rate": 4.297506146821216e-05,
      "loss": 0.2735,
      "step": 400
    },
    {
      "epoch": 0.4214963119072708,
      "eval_loss": 0.34133002161979675,
      "eval_runtime": 13.5644,
      "eval_samples_per_second": 35.387,
      "eval_steps_per_second": 2.212,
      "step": 400
    },
    {
      "epoch": 0.4320337197049526,
      "grad_norm": 0.5055080652236938,
      "learning_rate": 4.279943800491746e-05,
      "loss": 0.2846,
      "step": 410
    },
    {
      "epoch": 0.4320337197049526,
      "eval_loss": 0.3391459584236145,
      "eval_runtime": 13.5574,
      "eval_samples_per_second": 35.405,
      "eval_steps_per_second": 2.213,
      "step": 410
    },
    {
      "epoch": 0.44257112750263433,
      "grad_norm": 0.5293213129043579,
      "learning_rate": 4.2623814541622765e-05,
      "loss": 0.2757,
      "step": 420
    },
    {
      "epoch": 0.44257112750263433,
      "eval_loss": 0.3393440246582031,
      "eval_runtime": 13.5643,
      "eval_samples_per_second": 35.387,
      "eval_steps_per_second": 2.212,
      "step": 420
    },
    {
      "epoch": 0.45310853530031614,
      "grad_norm": 0.6627737283706665,
      "learning_rate": 4.2448191078328064e-05,
      "loss": 0.2829,
      "step": 430
    },
    {
      "epoch": 0.45310853530031614,
      "eval_loss": 0.34091198444366455,
      "eval_runtime": 13.5863,
      "eval_samples_per_second": 35.33,
      "eval_steps_per_second": 2.208,
      "step": 430
    },
    {
      "epoch": 0.4636459430979979,
      "grad_norm": 0.5125611424446106,
      "learning_rate": 4.227256761503337e-05,
      "loss": 0.2632,
      "step": 440
    },
    {
      "epoch": 0.4636459430979979,
      "eval_loss": 0.3409743010997772,
      "eval_runtime": 13.572,
      "eval_samples_per_second": 35.367,
      "eval_steps_per_second": 2.21,
      "step": 440
    },
    {
      "epoch": 0.47418335089567965,
      "grad_norm": 0.613191545009613,
      "learning_rate": 4.2096944151738675e-05,
      "loss": 0.2709,
      "step": 450
    },
    {
      "epoch": 0.47418335089567965,
      "eval_loss": 0.3384413421154022,
      "eval_runtime": 13.5818,
      "eval_samples_per_second": 35.341,
      "eval_steps_per_second": 2.209,
      "step": 450
    },
    {
      "epoch": 0.48472075869336145,
      "grad_norm": 0.4908188581466675,
      "learning_rate": 4.1921320688443974e-05,
      "loss": 0.2609,
      "step": 460
    },
    {
      "epoch": 0.48472075869336145,
      "eval_loss": 0.33855584263801575,
      "eval_runtime": 13.5848,
      "eval_samples_per_second": 35.334,
      "eval_steps_per_second": 2.208,
      "step": 460
    },
    {
      "epoch": 0.4952581664910432,
      "grad_norm": 0.5873326659202576,
      "learning_rate": 4.174569722514928e-05,
      "loss": 0.2709,
      "step": 470
    },
    {
      "epoch": 0.4952581664910432,
      "eval_loss": 0.3383730947971344,
      "eval_runtime": 13.591,
      "eval_samples_per_second": 35.317,
      "eval_steps_per_second": 2.207,
      "step": 470
    },
    {
      "epoch": 0.505795574288725,
      "grad_norm": 0.5365350246429443,
      "learning_rate": 4.1570073761854585e-05,
      "loss": 0.2848,
      "step": 480
    },
    {
      "epoch": 0.505795574288725,
      "eval_loss": 0.3378843367099762,
      "eval_runtime": 13.5963,
      "eval_samples_per_second": 35.304,
      "eval_steps_per_second": 2.206,
      "step": 480
    },
    {
      "epoch": 0.5163329820864068,
      "grad_norm": 0.5642485022544861,
      "learning_rate": 4.139445029855989e-05,
      "loss": 0.2911,
      "step": 490
    },
    {
      "epoch": 0.5163329820864068,
      "eval_loss": 0.3355303406715393,
      "eval_runtime": 13.6025,
      "eval_samples_per_second": 35.288,
      "eval_steps_per_second": 2.205,
      "step": 490
    },
    {
      "epoch": 0.5268703898840885,
      "grad_norm": 0.6058154702186584,
      "learning_rate": 4.12188268352652e-05,
      "loss": 0.262,
      "step": 500
    },
    {
      "epoch": 0.5268703898840885,
      "eval_loss": 0.3364686667919159,
      "eval_runtime": 13.6033,
      "eval_samples_per_second": 35.286,
      "eval_steps_per_second": 2.205,
      "step": 500
    },
    {
      "epoch": 0.5374077976817703,
      "grad_norm": 0.666671633720398,
      "learning_rate": 4.1043203371970496e-05,
      "loss": 0.2797,
      "step": 510
    },
    {
      "epoch": 0.5374077976817703,
      "eval_loss": 0.33817631006240845,
      "eval_runtime": 13.6129,
      "eval_samples_per_second": 35.261,
      "eval_steps_per_second": 2.204,
      "step": 510
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 0.5990138053894043,
      "learning_rate": 4.08675799086758e-05,
      "loss": 0.2606,
      "step": 520
    },
    {
      "epoch": 0.547945205479452,
      "eval_loss": 0.3355501890182495,
      "eval_runtime": 13.6034,
      "eval_samples_per_second": 35.285,
      "eval_steps_per_second": 2.205,
      "step": 520
    },
    {
      "epoch": 0.5584826132771338,
      "grad_norm": 0.5905662775039673,
      "learning_rate": 4.069195644538111e-05,
      "loss": 0.2687,
      "step": 530
    },
    {
      "epoch": 0.5584826132771338,
      "eval_loss": 0.3357114791870117,
      "eval_runtime": 13.5907,
      "eval_samples_per_second": 35.318,
      "eval_steps_per_second": 2.207,
      "step": 530
    },
    {
      "epoch": 0.5690200210748156,
      "grad_norm": 0.6308056116104126,
      "learning_rate": 4.0516332982086406e-05,
      "loss": 0.2964,
      "step": 540
    },
    {
      "epoch": 0.5690200210748156,
      "eval_loss": 0.3342372477054596,
      "eval_runtime": 13.5897,
      "eval_samples_per_second": 35.321,
      "eval_steps_per_second": 2.208,
      "step": 540
    },
    {
      "epoch": 0.5795574288724974,
      "grad_norm": 0.5483158230781555,
      "learning_rate": 4.034070951879171e-05,
      "loss": 0.262,
      "step": 550
    },
    {
      "epoch": 0.5795574288724974,
      "eval_loss": 0.33271318674087524,
      "eval_runtime": 13.5802,
      "eval_samples_per_second": 35.346,
      "eval_steps_per_second": 2.209,
      "step": 550
    },
    {
      "epoch": 0.5900948366701791,
      "grad_norm": 0.5433160662651062,
      "learning_rate": 4.016508605549702e-05,
      "loss": 0.2575,
      "step": 560
    },
    {
      "epoch": 0.5900948366701791,
      "eval_loss": 0.3352707326412201,
      "eval_runtime": 13.5772,
      "eval_samples_per_second": 35.353,
      "eval_steps_per_second": 2.21,
      "step": 560
    },
    {
      "epoch": 0.6006322444678609,
      "grad_norm": 0.7090951800346375,
      "learning_rate": 3.998946259220232e-05,
      "loss": 0.2624,
      "step": 570
    },
    {
      "epoch": 0.6006322444678609,
      "eval_loss": 0.33547133207321167,
      "eval_runtime": 13.5717,
      "eval_samples_per_second": 35.368,
      "eval_steps_per_second": 2.21,
      "step": 570
    },
    {
      "epoch": 0.6111696522655427,
      "grad_norm": 0.7720335721969604,
      "learning_rate": 3.981383912890763e-05,
      "loss": 0.2628,
      "step": 580
    },
    {
      "epoch": 0.6111696522655427,
      "eval_loss": 0.3337227702140808,
      "eval_runtime": 13.5578,
      "eval_samples_per_second": 35.404,
      "eval_steps_per_second": 2.213,
      "step": 580
    },
    {
      "epoch": 0.6217070600632244,
      "grad_norm": 0.5641891360282898,
      "learning_rate": 3.963821566561293e-05,
      "loss": 0.2617,
      "step": 590
    },
    {
      "epoch": 0.6217070600632244,
      "eval_loss": 0.3328714668750763,
      "eval_runtime": 13.5205,
      "eval_samples_per_second": 35.502,
      "eval_steps_per_second": 2.219,
      "step": 590
    },
    {
      "epoch": 0.6322444678609063,
      "grad_norm": 0.5985394716262817,
      "learning_rate": 3.946259220231823e-05,
      "loss": 0.2543,
      "step": 600
    },
    {
      "epoch": 0.6322444678609063,
      "eval_loss": 0.3321264386177063,
      "eval_runtime": 13.5572,
      "eval_samples_per_second": 35.406,
      "eval_steps_per_second": 2.213,
      "step": 600
    },
    {
      "epoch": 0.642781875658588,
      "grad_norm": 0.6638147234916687,
      "learning_rate": 3.928696873902353e-05,
      "loss": 0.2772,
      "step": 610
    },
    {
      "epoch": 0.642781875658588,
      "eval_loss": 0.33292463421821594,
      "eval_runtime": 13.5485,
      "eval_samples_per_second": 35.428,
      "eval_steps_per_second": 2.214,
      "step": 610
    },
    {
      "epoch": 0.6533192834562698,
      "grad_norm": 0.6401410102844238,
      "learning_rate": 3.911134527572884e-05,
      "loss": 0.2725,
      "step": 620
    },
    {
      "epoch": 0.6533192834562698,
      "eval_loss": 0.3313460648059845,
      "eval_runtime": 13.5546,
      "eval_samples_per_second": 35.412,
      "eval_steps_per_second": 2.213,
      "step": 620
    },
    {
      "epoch": 0.6638566912539515,
      "grad_norm": 0.5968373417854309,
      "learning_rate": 3.893572181243414e-05,
      "loss": 0.2626,
      "step": 630
    },
    {
      "epoch": 0.6638566912539515,
      "eval_loss": 0.3308354616165161,
      "eval_runtime": 13.5566,
      "eval_samples_per_second": 35.407,
      "eval_steps_per_second": 2.213,
      "step": 630
    },
    {
      "epoch": 0.6743940990516333,
      "grad_norm": 0.4671958088874817,
      "learning_rate": 3.876009834913945e-05,
      "loss": 0.2667,
      "step": 640
    },
    {
      "epoch": 0.6743940990516333,
      "eval_loss": 0.33001866936683655,
      "eval_runtime": 13.5739,
      "eval_samples_per_second": 35.362,
      "eval_steps_per_second": 2.21,
      "step": 640
    },
    {
      "epoch": 0.684931506849315,
      "grad_norm": 0.5390825867652893,
      "learning_rate": 3.8584474885844754e-05,
      "loss": 0.2599,
      "step": 650
    },
    {
      "epoch": 0.684931506849315,
      "eval_loss": 0.3301464915275574,
      "eval_runtime": 13.5678,
      "eval_samples_per_second": 35.378,
      "eval_steps_per_second": 2.211,
      "step": 650
    },
    {
      "epoch": 0.6954689146469969,
      "grad_norm": 0.5784376263618469,
      "learning_rate": 3.840885142255005e-05,
      "loss": 0.2452,
      "step": 660
    },
    {
      "epoch": 0.6954689146469969,
      "eval_loss": 0.33094659447669983,
      "eval_runtime": 13.5767,
      "eval_samples_per_second": 35.355,
      "eval_steps_per_second": 2.21,
      "step": 660
    },
    {
      "epoch": 0.7060063224446786,
      "grad_norm": 0.5976417660713196,
      "learning_rate": 3.823322795925536e-05,
      "loss": 0.2688,
      "step": 670
    },
    {
      "epoch": 0.7060063224446786,
      "eval_loss": 0.3312263488769531,
      "eval_runtime": 13.5986,
      "eval_samples_per_second": 35.298,
      "eval_steps_per_second": 2.206,
      "step": 670
    },
    {
      "epoch": 0.7165437302423604,
      "grad_norm": 0.5902723073959351,
      "learning_rate": 3.8057604495960664e-05,
      "loss": 0.2596,
      "step": 680
    },
    {
      "epoch": 0.7165437302423604,
      "eval_loss": 0.32888591289520264,
      "eval_runtime": 13.597,
      "eval_samples_per_second": 35.302,
      "eval_steps_per_second": 2.206,
      "step": 680
    },
    {
      "epoch": 0.7270811380400422,
      "grad_norm": 0.6102134585380554,
      "learning_rate": 3.788198103266596e-05,
      "loss": 0.2623,
      "step": 690
    },
    {
      "epoch": 0.7270811380400422,
      "eval_loss": 0.3287771940231323,
      "eval_runtime": 13.6069,
      "eval_samples_per_second": 35.276,
      "eval_steps_per_second": 2.205,
      "step": 690
    },
    {
      "epoch": 0.7376185458377239,
      "grad_norm": 0.5212683081626892,
      "learning_rate": 3.770635756937127e-05,
      "loss": 0.2497,
      "step": 700
    },
    {
      "epoch": 0.7376185458377239,
      "eval_loss": 0.330812007188797,
      "eval_runtime": 13.6008,
      "eval_samples_per_second": 35.292,
      "eval_steps_per_second": 2.206,
      "step": 700
    },
    {
      "epoch": 0.7481559536354057,
      "grad_norm": 0.614651620388031,
      "learning_rate": 3.7530734106076575e-05,
      "loss": 0.2568,
      "step": 710
    },
    {
      "epoch": 0.7481559536354057,
      "eval_loss": 0.3295523524284363,
      "eval_runtime": 13.6076,
      "eval_samples_per_second": 35.274,
      "eval_steps_per_second": 2.205,
      "step": 710
    },
    {
      "epoch": 0.7586933614330874,
      "grad_norm": 0.4703381061553955,
      "learning_rate": 3.7355110642781874e-05,
      "loss": 0.2325,
      "step": 720
    },
    {
      "epoch": 0.7586933614330874,
      "eval_loss": 0.32799476385116577,
      "eval_runtime": 13.6031,
      "eval_samples_per_second": 35.286,
      "eval_steps_per_second": 2.205,
      "step": 720
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.5316283702850342,
      "learning_rate": 3.717948717948718e-05,
      "loss": 0.2427,
      "step": 730
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 0.32939499616622925,
      "eval_runtime": 13.602,
      "eval_samples_per_second": 35.289,
      "eval_steps_per_second": 2.206,
      "step": 730
    },
    {
      "epoch": 0.779768177028451,
      "grad_norm": 0.5144335627555847,
      "learning_rate": 3.7003863716192485e-05,
      "loss": 0.2563,
      "step": 740
    },
    {
      "epoch": 0.779768177028451,
      "eval_loss": 0.3298059403896332,
      "eval_runtime": 13.5958,
      "eval_samples_per_second": 35.305,
      "eval_steps_per_second": 2.207,
      "step": 740
    },
    {
      "epoch": 0.7903055848261328,
      "grad_norm": 0.5247990489006042,
      "learning_rate": 3.682824025289779e-05,
      "loss": 0.2358,
      "step": 750
    },
    {
      "epoch": 0.7903055848261328,
      "eval_loss": 0.32794949412345886,
      "eval_runtime": 13.6027,
      "eval_samples_per_second": 35.287,
      "eval_steps_per_second": 2.205,
      "step": 750
    },
    {
      "epoch": 0.8008429926238145,
      "grad_norm": 0.5819652676582336,
      "learning_rate": 3.6652616789603096e-05,
      "loss": 0.2423,
      "step": 760
    },
    {
      "epoch": 0.8008429926238145,
      "eval_loss": 0.3263894319534302,
      "eval_runtime": 13.588,
      "eval_samples_per_second": 35.325,
      "eval_steps_per_second": 2.208,
      "step": 760
    },
    {
      "epoch": 0.8113804004214963,
      "grad_norm": 0.5112130045890808,
      "learning_rate": 3.6476993326308395e-05,
      "loss": 0.2638,
      "step": 770
    },
    {
      "epoch": 0.8113804004214963,
      "eval_loss": 0.3270197808742523,
      "eval_runtime": 13.5821,
      "eval_samples_per_second": 35.341,
      "eval_steps_per_second": 2.209,
      "step": 770
    },
    {
      "epoch": 0.821917808219178,
      "grad_norm": 0.5146061778068542,
      "learning_rate": 3.63013698630137e-05,
      "loss": 0.2375,
      "step": 780
    },
    {
      "epoch": 0.821917808219178,
      "eval_loss": 0.3265083432197571,
      "eval_runtime": 13.5816,
      "eval_samples_per_second": 35.342,
      "eval_steps_per_second": 2.209,
      "step": 780
    },
    {
      "epoch": 0.8324552160168599,
      "grad_norm": 0.5121347904205322,
      "learning_rate": 3.6125746399719e-05,
      "loss": 0.2468,
      "step": 790
    },
    {
      "epoch": 0.8324552160168599,
      "eval_loss": 0.32619577646255493,
      "eval_runtime": 13.5874,
      "eval_samples_per_second": 35.327,
      "eval_steps_per_second": 2.208,
      "step": 790
    },
    {
      "epoch": 0.8429926238145417,
      "grad_norm": 0.5837659239768982,
      "learning_rate": 3.5950122936424305e-05,
      "loss": 0.2392,
      "step": 800
    },
    {
      "epoch": 0.8429926238145417,
      "eval_loss": 0.3283963203430176,
      "eval_runtime": 13.5978,
      "eval_samples_per_second": 35.3,
      "eval_steps_per_second": 2.206,
      "step": 800
    },
    {
      "epoch": 0.8535300316122234,
      "grad_norm": 0.6293447613716125,
      "learning_rate": 3.577449947312961e-05,
      "loss": 0.2415,
      "step": 810
    },
    {
      "epoch": 0.8535300316122234,
      "eval_loss": 0.32890576124191284,
      "eval_runtime": 13.5797,
      "eval_samples_per_second": 35.347,
      "eval_steps_per_second": 2.209,
      "step": 810
    },
    {
      "epoch": 0.8640674394099052,
      "grad_norm": 0.5239388942718506,
      "learning_rate": 3.5598876009834916e-05,
      "loss": 0.2459,
      "step": 820
    },
    {
      "epoch": 0.8640674394099052,
      "eval_loss": 0.3269563317298889,
      "eval_runtime": 13.5785,
      "eval_samples_per_second": 35.35,
      "eval_steps_per_second": 2.209,
      "step": 820
    },
    {
      "epoch": 0.8746048472075869,
      "grad_norm": 0.5803388953208923,
      "learning_rate": 3.542325254654022e-05,
      "loss": 0.2561,
      "step": 830
    },
    {
      "epoch": 0.8746048472075869,
      "eval_loss": 0.32730814814567566,
      "eval_runtime": 13.5623,
      "eval_samples_per_second": 35.392,
      "eval_steps_per_second": 2.212,
      "step": 830
    },
    {
      "epoch": 0.8851422550052687,
      "grad_norm": 0.541553795337677,
      "learning_rate": 3.524762908324553e-05,
      "loss": 0.2407,
      "step": 840
    },
    {
      "epoch": 0.8851422550052687,
      "eval_loss": 0.32772594690322876,
      "eval_runtime": 13.5543,
      "eval_samples_per_second": 35.413,
      "eval_steps_per_second": 2.213,
      "step": 840
    },
    {
      "epoch": 0.8956796628029505,
      "grad_norm": 0.547488272190094,
      "learning_rate": 3.507200561995083e-05,
      "loss": 0.2512,
      "step": 850
    },
    {
      "epoch": 0.8956796628029505,
      "eval_loss": 0.32760006189346313,
      "eval_runtime": 13.5629,
      "eval_samples_per_second": 35.391,
      "eval_steps_per_second": 2.212,
      "step": 850
    },
    {
      "epoch": 0.9062170706006323,
      "grad_norm": 0.5485065579414368,
      "learning_rate": 3.489638215665613e-05,
      "loss": 0.2341,
      "step": 860
    },
    {
      "epoch": 0.9062170706006323,
      "eval_loss": 0.327595055103302,
      "eval_runtime": 13.5606,
      "eval_samples_per_second": 35.397,
      "eval_steps_per_second": 2.212,
      "step": 860
    },
    {
      "epoch": 0.916754478398314,
      "grad_norm": 0.5831624865531921,
      "learning_rate": 3.472075869336143e-05,
      "loss": 0.2465,
      "step": 870
    },
    {
      "epoch": 0.916754478398314,
      "eval_loss": 0.325825572013855,
      "eval_runtime": 13.5618,
      "eval_samples_per_second": 35.394,
      "eval_steps_per_second": 2.212,
      "step": 870
    },
    {
      "epoch": 0.9272918861959958,
      "grad_norm": 0.528672456741333,
      "learning_rate": 3.454513523006674e-05,
      "loss": 0.2355,
      "step": 880
    },
    {
      "epoch": 0.9272918861959958,
      "eval_loss": 0.32486769556999207,
      "eval_runtime": 13.553,
      "eval_samples_per_second": 35.417,
      "eval_steps_per_second": 2.214,
      "step": 880
    },
    {
      "epoch": 0.9378292939936775,
      "grad_norm": 0.5654621124267578,
      "learning_rate": 3.436951176677204e-05,
      "loss": 0.244,
      "step": 890
    },
    {
      "epoch": 0.9378292939936775,
      "eval_loss": 0.32602888345718384,
      "eval_runtime": 13.5581,
      "eval_samples_per_second": 35.403,
      "eval_steps_per_second": 2.213,
      "step": 890
    },
    {
      "epoch": 0.9483667017913593,
      "grad_norm": 0.5009713172912598,
      "learning_rate": 3.419388830347735e-05,
      "loss": 0.2299,
      "step": 900
    },
    {
      "epoch": 0.9483667017913593,
      "eval_loss": 0.32582351565361023,
      "eval_runtime": 13.5546,
      "eval_samples_per_second": 35.412,
      "eval_steps_per_second": 2.213,
      "step": 900
    },
    {
      "epoch": 0.958904109589041,
      "grad_norm": 0.5322644114494324,
      "learning_rate": 3.4018264840182654e-05,
      "loss": 0.2382,
      "step": 910
    },
    {
      "epoch": 0.958904109589041,
      "eval_loss": 0.32392945885658264,
      "eval_runtime": 13.5582,
      "eval_samples_per_second": 35.403,
      "eval_steps_per_second": 2.213,
      "step": 910
    },
    {
      "epoch": 0.9694415173867229,
      "grad_norm": 0.5522642731666565,
      "learning_rate": 3.384264137688795e-05,
      "loss": 0.2478,
      "step": 920
    },
    {
      "epoch": 0.9694415173867229,
      "eval_loss": 0.3241512179374695,
      "eval_runtime": 13.5666,
      "eval_samples_per_second": 35.381,
      "eval_steps_per_second": 2.211,
      "step": 920
    },
    {
      "epoch": 0.9799789251844047,
      "grad_norm": 0.5779546499252319,
      "learning_rate": 3.366701791359326e-05,
      "loss": 0.2534,
      "step": 930
    },
    {
      "epoch": 0.9799789251844047,
      "eval_loss": 0.3243790864944458,
      "eval_runtime": 13.524,
      "eval_samples_per_second": 35.492,
      "eval_steps_per_second": 2.218,
      "step": 930
    },
    {
      "epoch": 0.9905163329820864,
      "grad_norm": 0.6141674518585205,
      "learning_rate": 3.3491394450298564e-05,
      "loss": 0.2401,
      "step": 940
    },
    {
      "epoch": 0.9905163329820864,
      "eval_loss": 0.3262033462524414,
      "eval_runtime": 13.5269,
      "eval_samples_per_second": 35.485,
      "eval_steps_per_second": 2.218,
      "step": 940
    },
    {
      "epoch": 1.0010537407797682,
      "grad_norm": 0.42756763100624084,
      "learning_rate": 3.331577098700386e-05,
      "loss": 0.2338,
      "step": 950
    },
    {
      "epoch": 1.0010537407797682,
      "eval_loss": 0.3266028165817261,
      "eval_runtime": 13.5168,
      "eval_samples_per_second": 35.511,
      "eval_steps_per_second": 2.219,
      "step": 950
    },
    {
      "epoch": 1.01159114857745,
      "grad_norm": 0.5351940989494324,
      "learning_rate": 3.314014752370917e-05,
      "loss": 0.1934,
      "step": 960
    },
    {
      "epoch": 1.01159114857745,
      "eval_loss": 0.3327268064022064,
      "eval_runtime": 13.5225,
      "eval_samples_per_second": 35.497,
      "eval_steps_per_second": 2.219,
      "step": 960
    },
    {
      "epoch": 1.0221285563751317,
      "grad_norm": 0.5101184248924255,
      "learning_rate": 3.2964524060414474e-05,
      "loss": 0.208,
      "step": 970
    },
    {
      "epoch": 1.0221285563751317,
      "eval_loss": 0.3375646471977234,
      "eval_runtime": 13.5136,
      "eval_samples_per_second": 35.52,
      "eval_steps_per_second": 2.22,
      "step": 970
    },
    {
      "epoch": 1.0326659641728135,
      "grad_norm": 0.6187830567359924,
      "learning_rate": 3.278890059711977e-05,
      "loss": 0.2145,
      "step": 980
    },
    {
      "epoch": 1.0326659641728135,
      "eval_loss": 0.3343217074871063,
      "eval_runtime": 13.5266,
      "eval_samples_per_second": 35.486,
      "eval_steps_per_second": 2.218,
      "step": 980
    },
    {
      "epoch": 1.0432033719704952,
      "grad_norm": 0.669201672077179,
      "learning_rate": 3.261327713382508e-05,
      "loss": 0.2141,
      "step": 990
    },
    {
      "epoch": 1.0432033719704952,
      "eval_loss": 0.33267149329185486,
      "eval_runtime": 13.5144,
      "eval_samples_per_second": 35.518,
      "eval_steps_per_second": 2.22,
      "step": 990
    },
    {
      "epoch": 1.053740779768177,
      "grad_norm": 0.5544168949127197,
      "learning_rate": 3.2437653670530384e-05,
      "loss": 0.2074,
      "step": 1000
    },
    {
      "epoch": 1.053740779768177,
      "eval_loss": 0.33638790249824524,
      "eval_runtime": 13.5127,
      "eval_samples_per_second": 35.522,
      "eval_steps_per_second": 2.22,
      "step": 1000
    },
    {
      "epoch": 1.064278187565859,
      "grad_norm": 0.6323490738868713,
      "learning_rate": 3.226203020723569e-05,
      "loss": 0.2012,
      "step": 1010
    },
    {
      "epoch": 1.064278187565859,
      "eval_loss": 0.3400089144706726,
      "eval_runtime": 13.5093,
      "eval_samples_per_second": 35.531,
      "eval_steps_per_second": 2.221,
      "step": 1010
    },
    {
      "epoch": 1.0748155953635405,
      "grad_norm": 0.5988993048667908,
      "learning_rate": 3.2086406743940996e-05,
      "loss": 0.1954,
      "step": 1020
    },
    {
      "epoch": 1.0748155953635405,
      "eval_loss": 0.3393259048461914,
      "eval_runtime": 13.5107,
      "eval_samples_per_second": 35.527,
      "eval_steps_per_second": 2.22,
      "step": 1020
    },
    {
      "epoch": 1.0853530031612224,
      "grad_norm": 0.6308508515357971,
      "learning_rate": 3.1910783280646294e-05,
      "loss": 0.2066,
      "step": 1030
    },
    {
      "epoch": 1.0853530031612224,
      "eval_loss": 0.337782084941864,
      "eval_runtime": 13.5075,
      "eval_samples_per_second": 35.536,
      "eval_steps_per_second": 2.221,
      "step": 1030
    },
    {
      "epoch": 1.095890410958904,
      "grad_norm": 0.5316015481948853,
      "learning_rate": 3.17351598173516e-05,
      "loss": 0.2074,
      "step": 1040
    },
    {
      "epoch": 1.095890410958904,
      "eval_loss": 0.3380493223667145,
      "eval_runtime": 13.531,
      "eval_samples_per_second": 35.474,
      "eval_steps_per_second": 2.217,
      "step": 1040
    },
    {
      "epoch": 1.106427818756586,
      "grad_norm": 0.6207646131515503,
      "learning_rate": 3.15595363540569e-05,
      "loss": 0.2088,
      "step": 1050
    },
    {
      "epoch": 1.106427818756586,
      "eval_loss": 0.340431809425354,
      "eval_runtime": 13.5161,
      "eval_samples_per_second": 35.513,
      "eval_steps_per_second": 2.22,
      "step": 1050
    },
    {
      "epoch": 1.1169652265542676,
      "grad_norm": 0.5499827265739441,
      "learning_rate": 3.1383912890762205e-05,
      "loss": 0.2049,
      "step": 1060
    },
    {
      "epoch": 1.1169652265542676,
      "eval_loss": 0.34178677201271057,
      "eval_runtime": 13.5149,
      "eval_samples_per_second": 35.516,
      "eval_steps_per_second": 2.22,
      "step": 1060
    },
    {
      "epoch": 1.1275026343519494,
      "grad_norm": 0.5874269604682922,
      "learning_rate": 3.120828942746751e-05,
      "loss": 0.2027,
      "step": 1070
    },
    {
      "epoch": 1.1275026343519494,
      "eval_loss": 0.3401419222354889,
      "eval_runtime": 13.5078,
      "eval_samples_per_second": 35.535,
      "eval_steps_per_second": 2.221,
      "step": 1070
    },
    {
      "epoch": 1.1380400421496313,
      "grad_norm": 0.5852048397064209,
      "learning_rate": 3.1032665964172816e-05,
      "loss": 0.1999,
      "step": 1080
    },
    {
      "epoch": 1.1380400421496313,
      "eval_loss": 0.3392138183116913,
      "eval_runtime": 13.5107,
      "eval_samples_per_second": 35.527,
      "eval_steps_per_second": 2.22,
      "step": 1080
    },
    {
      "epoch": 1.148577449947313,
      "grad_norm": 0.6016051769256592,
      "learning_rate": 3.085704250087812e-05,
      "loss": 0.2044,
      "step": 1090
    },
    {
      "epoch": 1.148577449947313,
      "eval_loss": 0.34129372239112854,
      "eval_runtime": 13.5118,
      "eval_samples_per_second": 35.525,
      "eval_steps_per_second": 2.22,
      "step": 1090
    },
    {
      "epoch": 1.1591148577449948,
      "grad_norm": 0.5797489285469055,
      "learning_rate": 3.068141903758342e-05,
      "loss": 0.1974,
      "step": 1100
    },
    {
      "epoch": 1.1591148577449948,
      "eval_loss": 0.34007471799850464,
      "eval_runtime": 13.5066,
      "eval_samples_per_second": 35.538,
      "eval_steps_per_second": 2.221,
      "step": 1100
    },
    {
      "epoch": 1.1696522655426764,
      "grad_norm": 0.6308450102806091,
      "learning_rate": 3.0505795574288726e-05,
      "loss": 0.2037,
      "step": 1110
    },
    {
      "epoch": 1.1696522655426764,
      "eval_loss": 0.3387734293937683,
      "eval_runtime": 13.5225,
      "eval_samples_per_second": 35.496,
      "eval_steps_per_second": 2.219,
      "step": 1110
    },
    {
      "epoch": 1.1801896733403583,
      "grad_norm": 0.6952560544013977,
      "learning_rate": 3.0330172110994032e-05,
      "loss": 0.2009,
      "step": 1120
    },
    {
      "epoch": 1.1801896733403583,
      "eval_loss": 0.3363548815250397,
      "eval_runtime": 13.515,
      "eval_samples_per_second": 35.516,
      "eval_steps_per_second": 2.22,
      "step": 1120
    },
    {
      "epoch": 1.1907270811380402,
      "grad_norm": 0.6670855283737183,
      "learning_rate": 3.015454864769933e-05,
      "loss": 0.2046,
      "step": 1130
    },
    {
      "epoch": 1.1907270811380402,
      "eval_loss": 0.3387907147407532,
      "eval_runtime": 13.5274,
      "eval_samples_per_second": 35.484,
      "eval_steps_per_second": 2.218,
      "step": 1130
    },
    {
      "epoch": 1.2012644889357218,
      "grad_norm": 0.7167879343032837,
      "learning_rate": 2.997892518440464e-05,
      "loss": 0.2039,
      "step": 1140
    },
    {
      "epoch": 1.2012644889357218,
      "eval_loss": 0.33721429109573364,
      "eval_runtime": 13.5247,
      "eval_samples_per_second": 35.491,
      "eval_steps_per_second": 2.218,
      "step": 1140
    },
    {
      "epoch": 1.2118018967334037,
      "grad_norm": 0.5917449593544006,
      "learning_rate": 2.9803301721109945e-05,
      "loss": 0.1997,
      "step": 1150
    },
    {
      "epoch": 1.2118018967334037,
      "eval_loss": 0.33842217922210693,
      "eval_runtime": 13.5215,
      "eval_samples_per_second": 35.499,
      "eval_steps_per_second": 2.219,
      "step": 1150
    },
    {
      "epoch": 1.2223393045310853,
      "grad_norm": 0.6303601264953613,
      "learning_rate": 2.9627678257815244e-05,
      "loss": 0.2063,
      "step": 1160
    },
    {
      "epoch": 1.2223393045310853,
      "eval_loss": 0.3402339816093445,
      "eval_runtime": 13.5168,
      "eval_samples_per_second": 35.511,
      "eval_steps_per_second": 2.219,
      "step": 1160
    },
    {
      "epoch": 1.2328767123287672,
      "grad_norm": 0.7068540453910828,
      "learning_rate": 2.945205479452055e-05,
      "loss": 0.1959,
      "step": 1170
    },
    {
      "epoch": 1.2328767123287672,
      "eval_loss": 0.3410193622112274,
      "eval_runtime": 13.5262,
      "eval_samples_per_second": 35.487,
      "eval_steps_per_second": 2.218,
      "step": 1170
    },
    {
      "epoch": 1.2434141201264488,
      "grad_norm": 0.6145871877670288,
      "learning_rate": 2.9276431331225852e-05,
      "loss": 0.1979,
      "step": 1180
    },
    {
      "epoch": 1.2434141201264488,
      "eval_loss": 0.34399738907814026,
      "eval_runtime": 13.5178,
      "eval_samples_per_second": 35.509,
      "eval_steps_per_second": 2.219,
      "step": 1180
    },
    {
      "epoch": 1.2539515279241307,
      "grad_norm": 0.6426234841346741,
      "learning_rate": 2.9100807867931158e-05,
      "loss": 0.2077,
      "step": 1190
    },
    {
      "epoch": 1.2539515279241307,
      "eval_loss": 0.339756041765213,
      "eval_runtime": 13.5275,
      "eval_samples_per_second": 35.483,
      "eval_steps_per_second": 2.218,
      "step": 1190
    },
    {
      "epoch": 1.2644889357218125,
      "grad_norm": 0.6372554302215576,
      "learning_rate": 2.8925184404636463e-05,
      "loss": 0.1938,
      "step": 1200
    },
    {
      "epoch": 1.2644889357218125,
      "eval_loss": 0.3394372761249542,
      "eval_runtime": 13.5262,
      "eval_samples_per_second": 35.487,
      "eval_steps_per_second": 2.218,
      "step": 1200
    },
    {
      "epoch": 1.2750263435194942,
      "grad_norm": 0.49700266122817993,
      "learning_rate": 2.8749560941341762e-05,
      "loss": 0.1973,
      "step": 1210
    },
    {
      "epoch": 1.2750263435194942,
      "eval_loss": 0.34174108505249023,
      "eval_runtime": 13.5375,
      "eval_samples_per_second": 35.457,
      "eval_steps_per_second": 2.216,
      "step": 1210
    },
    {
      "epoch": 1.285563751317176,
      "grad_norm": 0.6628106832504272,
      "learning_rate": 2.8573937478047068e-05,
      "loss": 0.1892,
      "step": 1220
    },
    {
      "epoch": 1.285563751317176,
      "eval_loss": 0.3427627980709076,
      "eval_runtime": 13.5801,
      "eval_samples_per_second": 35.346,
      "eval_steps_per_second": 2.209,
      "step": 1220
    },
    {
      "epoch": 1.2961011591148577,
      "grad_norm": 0.6381083130836487,
      "learning_rate": 2.839831401475237e-05,
      "loss": 0.2023,
      "step": 1230
    },
    {
      "epoch": 1.2961011591148577,
      "eval_loss": 0.3431331515312195,
      "eval_runtime": 13.5304,
      "eval_samples_per_second": 35.476,
      "eval_steps_per_second": 2.217,
      "step": 1230
    },
    {
      "epoch": 1.3066385669125395,
      "grad_norm": 0.6037564873695374,
      "learning_rate": 2.8222690551457676e-05,
      "loss": 0.1925,
      "step": 1240
    },
    {
      "epoch": 1.3066385669125395,
      "eval_loss": 0.3404982388019562,
      "eval_runtime": 13.5282,
      "eval_samples_per_second": 35.481,
      "eval_steps_per_second": 2.218,
      "step": 1240
    },
    {
      "epoch": 1.3171759747102212,
      "grad_norm": 0.6601694822311401,
      "learning_rate": 2.804706708816298e-05,
      "loss": 0.1926,
      "step": 1250
    },
    {
      "epoch": 1.3171759747102212,
      "eval_loss": 0.34181320667266846,
      "eval_runtime": 13.5621,
      "eval_samples_per_second": 35.393,
      "eval_steps_per_second": 2.212,
      "step": 1250
    },
    {
      "epoch": 1.327713382507903,
      "grad_norm": 0.5349809527397156,
      "learning_rate": 2.7871443624868284e-05,
      "loss": 0.1936,
      "step": 1260
    },
    {
      "epoch": 1.327713382507903,
      "eval_loss": 0.34073132276535034,
      "eval_runtime": 13.6477,
      "eval_samples_per_second": 35.171,
      "eval_steps_per_second": 2.198,
      "step": 1260
    },
    {
      "epoch": 1.338250790305585,
      "grad_norm": 0.6491771340370178,
      "learning_rate": 2.769582016157359e-05,
      "loss": 0.1993,
      "step": 1270
    },
    {
      "epoch": 1.338250790305585,
      "eval_loss": 0.34483978152275085,
      "eval_runtime": 13.5329,
      "eval_samples_per_second": 35.469,
      "eval_steps_per_second": 2.217,
      "step": 1270
    },
    {
      "epoch": 1.3487881981032666,
      "grad_norm": 0.6270915865898132,
      "learning_rate": 2.7520196698278895e-05,
      "loss": 0.2007,
      "step": 1280
    },
    {
      "epoch": 1.3487881981032666,
      "eval_loss": 0.34615617990493774,
      "eval_runtime": 13.5185,
      "eval_samples_per_second": 35.507,
      "eval_steps_per_second": 2.219,
      "step": 1280
    },
    {
      "epoch": 1.3593256059009484,
      "grad_norm": 0.6853300333023071,
      "learning_rate": 2.7344573234984194e-05,
      "loss": 0.2131,
      "step": 1290
    },
    {
      "epoch": 1.3593256059009484,
      "eval_loss": 0.3406980335712433,
      "eval_runtime": 13.5247,
      "eval_samples_per_second": 35.491,
      "eval_steps_per_second": 2.218,
      "step": 1290
    },
    {
      "epoch": 1.36986301369863,
      "grad_norm": 0.5863353610038757,
      "learning_rate": 2.71689497716895e-05,
      "loss": 0.1846,
      "step": 1300
    },
    {
      "epoch": 1.36986301369863,
      "eval_loss": 0.34115493297576904,
      "eval_runtime": 13.5104,
      "eval_samples_per_second": 35.528,
      "eval_steps_per_second": 2.221,
      "step": 1300
    },
    {
      "epoch": 1.380400421496312,
      "grad_norm": 0.7511035203933716,
      "learning_rate": 2.6993326308394802e-05,
      "loss": 0.1911,
      "step": 1310
    },
    {
      "epoch": 1.380400421496312,
      "eval_loss": 0.34829092025756836,
      "eval_runtime": 13.5081,
      "eval_samples_per_second": 35.534,
      "eval_steps_per_second": 2.221,
      "step": 1310
    },
    {
      "epoch": 1.3909378292939936,
      "grad_norm": 0.6482399106025696,
      "learning_rate": 2.6817702845100107e-05,
      "loss": 0.2013,
      "step": 1320
    },
    {
      "epoch": 1.3909378292939936,
      "eval_loss": 0.3465197682380676,
      "eval_runtime": 13.5316,
      "eval_samples_per_second": 35.473,
      "eval_steps_per_second": 2.217,
      "step": 1320
    },
    {
      "epoch": 1.4014752370916754,
      "grad_norm": 0.570846438407898,
      "learning_rate": 2.6642079381805413e-05,
      "loss": 0.1889,
      "step": 1330
    },
    {
      "epoch": 1.4014752370916754,
      "eval_loss": 0.34688571095466614,
      "eval_runtime": 13.5157,
      "eval_samples_per_second": 35.514,
      "eval_steps_per_second": 2.22,
      "step": 1330
    },
    {
      "epoch": 1.4120126448893573,
      "grad_norm": 0.6336379051208496,
      "learning_rate": 2.6466455918510712e-05,
      "loss": 0.1994,
      "step": 1340
    },
    {
      "epoch": 1.4120126448893573,
      "eval_loss": 0.3466794788837433,
      "eval_runtime": 13.5071,
      "eval_samples_per_second": 35.537,
      "eval_steps_per_second": 2.221,
      "step": 1340
    },
    {
      "epoch": 1.422550052687039,
      "grad_norm": 0.653731644153595,
      "learning_rate": 2.6290832455216018e-05,
      "loss": 0.1894,
      "step": 1350
    },
    {
      "epoch": 1.422550052687039,
      "eval_loss": 0.34787434339523315,
      "eval_runtime": 13.5062,
      "eval_samples_per_second": 35.539,
      "eval_steps_per_second": 2.221,
      "step": 1350
    },
    {
      "epoch": 1.4330874604847208,
      "grad_norm": 0.6545577049255371,
      "learning_rate": 2.611520899192132e-05,
      "loss": 0.1981,
      "step": 1360
    },
    {
      "epoch": 1.4330874604847208,
      "eval_loss": 0.34745165705680847,
      "eval_runtime": 13.4977,
      "eval_samples_per_second": 35.562,
      "eval_steps_per_second": 2.223,
      "step": 1360
    },
    {
      "epoch": 1.4436248682824027,
      "grad_norm": 0.7751778364181519,
      "learning_rate": 2.5939585528626625e-05,
      "loss": 0.183,
      "step": 1370
    },
    {
      "epoch": 1.4436248682824027,
      "eval_loss": 0.3481937348842621,
      "eval_runtime": 13.5303,
      "eval_samples_per_second": 35.476,
      "eval_steps_per_second": 2.217,
      "step": 1370
    },
    {
      "epoch": 1.4541622760800843,
      "grad_norm": 0.7329134345054626,
      "learning_rate": 2.576396206533193e-05,
      "loss": 0.1906,
      "step": 1380
    },
    {
      "epoch": 1.4541622760800843,
      "eval_loss": 0.3493328094482422,
      "eval_runtime": 13.5155,
      "eval_samples_per_second": 35.515,
      "eval_steps_per_second": 2.22,
      "step": 1380
    },
    {
      "epoch": 1.464699683877766,
      "grad_norm": 0.7945630550384521,
      "learning_rate": 2.5588338602037233e-05,
      "loss": 0.1913,
      "step": 1390
    },
    {
      "epoch": 1.464699683877766,
      "eval_loss": 0.3451732397079468,
      "eval_runtime": 13.5276,
      "eval_samples_per_second": 35.483,
      "eval_steps_per_second": 2.218,
      "step": 1390
    },
    {
      "epoch": 1.4752370916754478,
      "grad_norm": 0.7154361009597778,
      "learning_rate": 2.541271513874254e-05,
      "loss": 0.1853,
      "step": 1400
    },
    {
      "epoch": 1.4752370916754478,
      "eval_loss": 0.3447299301624298,
      "eval_runtime": 13.5329,
      "eval_samples_per_second": 35.469,
      "eval_steps_per_second": 2.217,
      "step": 1400
    },
    {
      "epoch": 1.4857744994731297,
      "grad_norm": 0.6370890140533447,
      "learning_rate": 2.5237091675447845e-05,
      "loss": 0.1822,
      "step": 1410
    },
    {
      "epoch": 1.4857744994731297,
      "eval_loss": 0.34814396500587463,
      "eval_runtime": 13.5296,
      "eval_samples_per_second": 35.478,
      "eval_steps_per_second": 2.217,
      "step": 1410
    },
    {
      "epoch": 1.4963119072708113,
      "grad_norm": 0.7020454406738281,
      "learning_rate": 2.5061468212153144e-05,
      "loss": 0.1918,
      "step": 1420
    },
    {
      "epoch": 1.4963119072708113,
      "eval_loss": 0.35007527470588684,
      "eval_runtime": 13.5235,
      "eval_samples_per_second": 35.494,
      "eval_steps_per_second": 2.218,
      "step": 1420
    },
    {
      "epoch": 1.5068493150684932,
      "grad_norm": 0.7104793787002563,
      "learning_rate": 2.4885844748858446e-05,
      "loss": 0.2008,
      "step": 1430
    },
    {
      "epoch": 1.5068493150684932,
      "eval_loss": 0.3479911684989929,
      "eval_runtime": 13.5547,
      "eval_samples_per_second": 35.412,
      "eval_steps_per_second": 2.213,
      "step": 1430
    },
    {
      "epoch": 1.517386722866175,
      "grad_norm": 0.8473650217056274,
      "learning_rate": 2.4710221285563755e-05,
      "loss": 0.2042,
      "step": 1440
    },
    {
      "epoch": 1.517386722866175,
      "eval_loss": 0.34348005056381226,
      "eval_runtime": 13.5204,
      "eval_samples_per_second": 35.502,
      "eval_steps_per_second": 2.219,
      "step": 1440
    },
    {
      "epoch": 1.5279241306638567,
      "grad_norm": 0.6656286716461182,
      "learning_rate": 2.4534597822269057e-05,
      "loss": 0.1968,
      "step": 1450
    },
    {
      "epoch": 1.5279241306638567,
      "eval_loss": 0.3460903763771057,
      "eval_runtime": 13.5325,
      "eval_samples_per_second": 35.47,
      "eval_steps_per_second": 2.217,
      "step": 1450
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.6165732145309448,
      "learning_rate": 2.435897435897436e-05,
      "loss": 0.1948,
      "step": 1460
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.3462531864643097,
      "eval_runtime": 13.5278,
      "eval_samples_per_second": 35.483,
      "eval_steps_per_second": 2.218,
      "step": 1460
    },
    {
      "epoch": 1.5489989462592202,
      "grad_norm": 0.8803408741950989,
      "learning_rate": 2.418335089567966e-05,
      "loss": 0.1814,
      "step": 1470
    },
    {
      "epoch": 1.5489989462592202,
      "eval_loss": 0.3515132665634155,
      "eval_runtime": 13.5309,
      "eval_samples_per_second": 35.474,
      "eval_steps_per_second": 2.217,
      "step": 1470
    },
    {
      "epoch": 1.559536354056902,
      "grad_norm": 0.628629744052887,
      "learning_rate": 2.4007727432384967e-05,
      "loss": 0.1894,
      "step": 1480
    },
    {
      "epoch": 1.559536354056902,
      "eval_loss": 0.3480067551136017,
      "eval_runtime": 13.5314,
      "eval_samples_per_second": 35.473,
      "eval_steps_per_second": 2.217,
      "step": 1480
    },
    {
      "epoch": 1.5700737618545837,
      "grad_norm": 0.6435084939002991,
      "learning_rate": 2.3832103969090273e-05,
      "loss": 0.1898,
      "step": 1490
    },
    {
      "epoch": 1.5700737618545837,
      "eval_loss": 0.3466918170452118,
      "eval_runtime": 13.5167,
      "eval_samples_per_second": 35.512,
      "eval_steps_per_second": 2.219,
      "step": 1490
    },
    {
      "epoch": 1.5806111696522656,
      "grad_norm": 0.5330148935317993,
      "learning_rate": 2.3656480505795575e-05,
      "loss": 0.188,
      "step": 1500
    },
    {
      "epoch": 1.5806111696522656,
      "eval_loss": 0.3500882387161255,
      "eval_runtime": 13.516,
      "eval_samples_per_second": 35.514,
      "eval_steps_per_second": 2.22,
      "step": 1500
    }
  ],
  "logging_steps": 10,
  "max_steps": 2847,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.797694680218665e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}