{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7376185458377239,
  "eval_steps": 50,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01053740779768177,
      "grad_norm": 3.070249080657959,
      "learning_rate": 4.982437653670531e-05,
      "loss": 1.7879,
      "step": 10
    },
    {
      "epoch": 0.02107481559536354,
      "grad_norm": 1.702326774597168,
      "learning_rate": 4.964875307341061e-05,
      "loss": 0.5567,
      "step": 20
    },
    {
      "epoch": 0.03161222339304531,
      "grad_norm": 1.1947294473648071,
      "learning_rate": 4.947312961011591e-05,
      "loss": 0.4493,
      "step": 30
    },
    {
      "epoch": 0.04214963119072708,
      "grad_norm": 0.9556658267974854,
      "learning_rate": 4.929750614682122e-05,
      "loss": 0.3728,
      "step": 40
    },
    {
      "epoch": 0.05268703898840885,
      "grad_norm": 0.7952510714530945,
      "learning_rate": 4.9121882683526524e-05,
      "loss": 0.3535,
      "step": 50
    },
    {
      "epoch": 0.05268703898840885,
      "eval_loss": 0.4311191439628601,
      "eval_runtime": 13.6539,
      "eval_samples_per_second": 35.155,
      "eval_steps_per_second": 2.197,
      "step": 50
    },
    {
      "epoch": 0.06322444678609063,
      "grad_norm": 0.6962826251983643,
      "learning_rate": 4.894625922023183e-05,
      "loss": 0.3507,
      "step": 60
    },
    {
      "epoch": 0.0737618545837724,
      "grad_norm": 0.6941961646080017,
      "learning_rate": 4.877063575693713e-05,
      "loss": 0.3585,
      "step": 70
    },
    {
      "epoch": 0.08429926238145416,
      "grad_norm": 0.6864392757415771,
      "learning_rate": 4.8595012293642434e-05,
      "loss": 0.3496,
      "step": 80
    },
    {
      "epoch": 0.09483667017913593,
      "grad_norm": 0.7322937846183777,
      "learning_rate": 4.841938883034774e-05,
      "loss": 0.3295,
      "step": 90
    },
    {
      "epoch": 0.1053740779768177,
      "grad_norm": 0.6921488046646118,
      "learning_rate": 4.824376536705304e-05,
      "loss": 0.3357,
      "step": 100
    },
    {
      "epoch": 0.1053740779768177,
      "eval_loss": 0.39120009541511536,
      "eval_runtime": 13.7031,
      "eval_samples_per_second": 35.029,
      "eval_steps_per_second": 2.189,
      "step": 100
    },
    {
      "epoch": 0.11591148577449947,
      "grad_norm": 0.6553240418434143,
      "learning_rate": 4.8068141903758344e-05,
      "loss": 0.3105,
      "step": 110
    },
    {
      "epoch": 0.12644889357218125,
      "grad_norm": 0.5637819170951843,
      "learning_rate": 4.789251844046364e-05,
      "loss": 0.3164,
      "step": 120
    },
    {
      "epoch": 0.136986301369863,
      "grad_norm": 0.6341928839683533,
      "learning_rate": 4.7716894977168955e-05,
      "loss": 0.304,
      "step": 130
    },
    {
      "epoch": 0.1475237091675448,
      "grad_norm": 0.5917785167694092,
      "learning_rate": 4.754127151387426e-05,
      "loss": 0.3234,
      "step": 140
    },
    {
      "epoch": 0.15806111696522657,
      "grad_norm": 0.5884453654289246,
      "learning_rate": 4.736564805057956e-05,
      "loss": 0.317,
      "step": 150
    },
    {
      "epoch": 0.15806111696522657,
      "eval_loss": 0.37688738107681274,
      "eval_runtime": 13.6535,
      "eval_samples_per_second": 35.156,
      "eval_steps_per_second": 2.197,
      "step": 150
    },
    {
      "epoch": 0.16859852476290832,
      "grad_norm": 0.5819964408874512,
      "learning_rate": 4.7190024587284866e-05,
      "loss": 0.2992,
      "step": 160
    },
    {
      "epoch": 0.1791359325605901,
      "grad_norm": 0.689468264579773,
      "learning_rate": 4.7014401123990165e-05,
      "loss": 0.3168,
      "step": 170
    },
    {
      "epoch": 0.18967334035827185,
      "grad_norm": 0.6950872540473938,
      "learning_rate": 4.683877766069547e-05,
      "loss": 0.3041,
      "step": 180
    },
    {
      "epoch": 0.20021074815595363,
      "grad_norm": 0.8322122097015381,
      "learning_rate": 4.6663154197400776e-05,
      "loss": 0.3028,
      "step": 190
    },
    {
      "epoch": 0.2107481559536354,
      "grad_norm": 0.5850774645805359,
      "learning_rate": 4.6487530734106075e-05,
      "loss": 0.2992,
      "step": 200
    },
    {
      "epoch": 0.2107481559536354,
      "eval_loss": 0.36230018734931946,
      "eval_runtime": 13.6165,
      "eval_samples_per_second": 35.251,
      "eval_steps_per_second": 2.203,
      "step": 200
    },
    {
      "epoch": 0.22128556375131717,
      "grad_norm": 0.6668715476989746,
      "learning_rate": 4.631190727081138e-05,
      "loss": 0.2924,
      "step": 210
    },
    {
      "epoch": 0.23182297154899895,
      "grad_norm": 0.4749641418457031,
      "learning_rate": 4.6136283807516686e-05,
      "loss": 0.3017,
      "step": 220
    },
    {
      "epoch": 0.24236037934668073,
      "grad_norm": 0.6381515860557556,
      "learning_rate": 4.596066034422199e-05,
      "loss": 0.2887,
      "step": 230
    },
    {
      "epoch": 0.2528977871443625,
      "grad_norm": 0.49952977895736694,
      "learning_rate": 4.57850368809273e-05,
      "loss": 0.2833,
      "step": 240
    },
    {
      "epoch": 0.26343519494204426,
      "grad_norm": 0.699518084526062,
      "learning_rate": 4.5609413417632596e-05,
      "loss": 0.2859,
      "step": 250
    },
    {
      "epoch": 0.26343519494204426,
      "eval_loss": 0.3563433885574341,
      "eval_runtime": 13.6275,
      "eval_samples_per_second": 35.223,
      "eval_steps_per_second": 2.201,
      "step": 250
    },
    {
      "epoch": 0.273972602739726,
      "grad_norm": 0.5912085175514221,
      "learning_rate": 4.54337899543379e-05,
      "loss": 0.302,
      "step": 260
    },
    {
      "epoch": 0.2845100105374078,
      "grad_norm": 0.6353363990783691,
      "learning_rate": 4.525816649104321e-05,
      "loss": 0.3112,
      "step": 270
    },
    {
      "epoch": 0.2950474183350896,
      "grad_norm": 0.5483567118644714,
      "learning_rate": 4.5082543027748506e-05,
      "loss": 0.2808,
      "step": 280
    },
    {
      "epoch": 0.3055848261327713,
      "grad_norm": 0.5003193616867065,
      "learning_rate": 4.490691956445381e-05,
      "loss": 0.3185,
      "step": 290
    },
    {
      "epoch": 0.31612223393045313,
      "grad_norm": 0.4919240176677704,
      "learning_rate": 4.473129610115912e-05,
      "loss": 0.2883,
      "step": 300
    },
    {
      "epoch": 0.31612223393045313,
      "eval_loss": 0.35315924882888794,
      "eval_runtime": 13.6133,
      "eval_samples_per_second": 35.26,
      "eval_steps_per_second": 2.204,
      "step": 300
    },
    {
      "epoch": 0.3266596417281349,
      "grad_norm": 0.5145038366317749,
      "learning_rate": 4.455567263786442e-05,
      "loss": 0.3084,
      "step": 310
    },
    {
      "epoch": 0.33719704952581664,
      "grad_norm": 0.6343855261802673,
      "learning_rate": 4.438004917456973e-05,
      "loss": 0.295,
      "step": 320
    },
    {
      "epoch": 0.34773445732349845,
      "grad_norm": 0.5336400270462036,
      "learning_rate": 4.420442571127503e-05,
      "loss": 0.2882,
      "step": 330
    },
    {
      "epoch": 0.3582718651211802,
      "grad_norm": 0.49547308683395386,
      "learning_rate": 4.4028802247980333e-05,
      "loss": 0.3084,
      "step": 340
    },
    {
      "epoch": 0.36880927291886195,
      "grad_norm": 0.5037292242050171,
      "learning_rate": 4.385317878468563e-05,
      "loss": 0.3084,
      "step": 350
    },
    {
      "epoch": 0.36880927291886195,
      "eval_loss": 0.34608179330825806,
      "eval_runtime": 13.6537,
      "eval_samples_per_second": 35.155,
      "eval_steps_per_second": 2.197,
      "step": 350
    },
    {
      "epoch": 0.3793466807165437,
      "grad_norm": 0.5956543684005737,
      "learning_rate": 4.367755532139094e-05,
      "loss": 0.2763,
      "step": 360
    },
    {
      "epoch": 0.3898840885142255,
      "grad_norm": 0.6263634562492371,
      "learning_rate": 4.3501931858096244e-05,
      "loss": 0.3025,
      "step": 370
    },
    {
      "epoch": 0.40042149631190727,
      "grad_norm": 0.4832920730113983,
      "learning_rate": 4.332630839480154e-05,
      "loss": 0.2792,
      "step": 380
    },
    {
      "epoch": 0.410958904109589,
      "grad_norm": 0.4969714879989624,
      "learning_rate": 4.3150684931506855e-05,
      "loss": 0.2704,
      "step": 390
    },
    {
      "epoch": 0.4214963119072708,
      "grad_norm": 0.568900465965271,
      "learning_rate": 4.297506146821216e-05,
      "loss": 0.2734,
      "step": 400
    },
    {
      "epoch": 0.4214963119072708,
      "eval_loss": 0.34151414036750793,
      "eval_runtime": 13.664,
      "eval_samples_per_second": 35.129,
      "eval_steps_per_second": 2.196,
      "step": 400
    },
    {
      "epoch": 0.4320337197049526,
      "grad_norm": 0.5023282170295715,
      "learning_rate": 4.279943800491746e-05,
      "loss": 0.2845,
      "step": 410
    },
    {
      "epoch": 0.44257112750263433,
      "grad_norm": 0.531538188457489,
      "learning_rate": 4.2623814541622765e-05,
      "loss": 0.276,
      "step": 420
    },
    {
      "epoch": 0.45310853530031614,
      "grad_norm": 0.6686979532241821,
      "learning_rate": 4.2448191078328064e-05,
      "loss": 0.283,
      "step": 430
    },
    {
      "epoch": 0.4636459430979979,
      "grad_norm": 0.5194190144538879,
      "learning_rate": 4.227256761503337e-05,
      "loss": 0.2636,
      "step": 440
    },
    {
      "epoch": 0.47418335089567965,
      "grad_norm": 0.6105541586875916,
      "learning_rate": 4.2096944151738675e-05,
      "loss": 0.2713,
      "step": 450
    },
    {
      "epoch": 0.47418335089567965,
      "eval_loss": 0.33879777789115906,
      "eval_runtime": 13.6666,
      "eval_samples_per_second": 35.122,
      "eval_steps_per_second": 2.195,
      "step": 450
    },
    {
      "epoch": 0.48472075869336145,
      "grad_norm": 0.4929138123989105,
      "learning_rate": 4.1921320688443974e-05,
      "loss": 0.261,
      "step": 460
    },
    {
      "epoch": 0.4952581664910432,
      "grad_norm": 0.584095299243927,
      "learning_rate": 4.174569722514928e-05,
      "loss": 0.2714,
      "step": 470
    },
    {
      "epoch": 0.505795574288725,
      "grad_norm": 0.5386167764663696,
      "learning_rate": 4.1570073761854585e-05,
      "loss": 0.2855,
      "step": 480
    },
    {
      "epoch": 0.5163329820864068,
      "grad_norm": 0.5819774270057678,
      "learning_rate": 4.139445029855989e-05,
      "loss": 0.2911,
      "step": 490
    },
    {
      "epoch": 0.5268703898840885,
      "grad_norm": 0.5875944495201111,
      "learning_rate": 4.12188268352652e-05,
      "loss": 0.2619,
      "step": 500
    },
    {
      "epoch": 0.5268703898840885,
      "eval_loss": 0.3364439010620117,
      "eval_runtime": 13.6731,
      "eval_samples_per_second": 35.105,
      "eval_steps_per_second": 2.194,
      "step": 500
    },
    {
      "epoch": 0.5374077976817703,
      "grad_norm": 0.6705812811851501,
      "learning_rate": 4.1043203371970496e-05,
      "loss": 0.2792,
      "step": 510
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 0.5975854396820068,
      "learning_rate": 4.08675799086758e-05,
      "loss": 0.2611,
      "step": 520
    },
    {
      "epoch": 0.5584826132771338,
      "grad_norm": 0.5905255675315857,
      "learning_rate": 4.069195644538111e-05,
      "loss": 0.2688,
      "step": 530
    },
    {
      "epoch": 0.5690200210748156,
      "grad_norm": 0.6222755312919617,
      "learning_rate": 4.0516332982086406e-05,
      "loss": 0.2968,
      "step": 540
    },
    {
      "epoch": 0.5795574288724974,
      "grad_norm": 0.5604814291000366,
      "learning_rate": 4.034070951879171e-05,
      "loss": 0.2623,
      "step": 550
    },
    {
      "epoch": 0.5795574288724974,
      "eval_loss": 0.3329848647117615,
      "eval_runtime": 13.6573,
      "eval_samples_per_second": 35.146,
      "eval_steps_per_second": 2.197,
      "step": 550
    },
    {
      "epoch": 0.5900948366701791,
      "grad_norm": 0.552557647228241,
      "learning_rate": 4.016508605549702e-05,
      "loss": 0.2579,
      "step": 560
    },
    {
      "epoch": 0.6006322444678609,
      "grad_norm": 0.7151882648468018,
      "learning_rate": 3.998946259220232e-05,
      "loss": 0.263,
      "step": 570
    },
    {
      "epoch": 0.6111696522655427,
      "grad_norm": 0.7065399289131165,
      "learning_rate": 3.981383912890763e-05,
      "loss": 0.2626,
      "step": 580
    },
    {
      "epoch": 0.6217070600632244,
      "grad_norm": 0.5590985417366028,
      "learning_rate": 3.963821566561293e-05,
      "loss": 0.2614,
      "step": 590
    },
    {
      "epoch": 0.6322444678609063,
      "grad_norm": 0.6018164157867432,
      "learning_rate": 3.946259220231823e-05,
      "loss": 0.2547,
      "step": 600
    },
    {
      "epoch": 0.6322444678609063,
      "eval_loss": 0.3322373926639557,
      "eval_runtime": 13.6386,
      "eval_samples_per_second": 35.194,
      "eval_steps_per_second": 2.2,
      "step": 600
    },
    {
      "epoch": 0.642781875658588,
      "grad_norm": 0.6659051179885864,
      "learning_rate": 3.928696873902353e-05,
      "loss": 0.2768,
      "step": 610
    },
    {
      "epoch": 0.6533192834562698,
      "grad_norm": 0.6509780883789062,
      "learning_rate": 3.911134527572884e-05,
      "loss": 0.272,
      "step": 620
    },
    {
      "epoch": 0.6638566912539515,
      "grad_norm": 0.5752527713775635,
      "learning_rate": 3.893572181243414e-05,
      "loss": 0.2625,
      "step": 630
    },
    {
      "epoch": 0.6743940990516333,
      "grad_norm": 0.4737485349178314,
      "learning_rate": 3.876009834913945e-05,
      "loss": 0.2668,
      "step": 640
    },
    {
      "epoch": 0.684931506849315,
      "grad_norm": 0.548322319984436,
      "learning_rate": 3.8584474885844754e-05,
      "loss": 0.2597,
      "step": 650
    },
    {
      "epoch": 0.684931506849315,
      "eval_loss": 0.3307412266731262,
      "eval_runtime": 13.6407,
      "eval_samples_per_second": 35.189,
      "eval_steps_per_second": 2.199,
      "step": 650
    },
    {
      "epoch": 0.6954689146469969,
      "grad_norm": 0.5635099411010742,
      "learning_rate": 3.840885142255005e-05,
      "loss": 0.2453,
      "step": 660
    },
    {
      "epoch": 0.7060063224446786,
      "grad_norm": 0.6012313365936279,
      "learning_rate": 3.823322795925536e-05,
      "loss": 0.2695,
      "step": 670
    },
    {
      "epoch": 0.7165437302423604,
      "grad_norm": 0.6025511026382446,
      "learning_rate": 3.8057604495960664e-05,
      "loss": 0.2599,
      "step": 680
    },
    {
      "epoch": 0.7270811380400422,
      "grad_norm": 0.5996225476264954,
      "learning_rate": 3.788198103266596e-05,
      "loss": 0.2624,
      "step": 690
    },
    {
      "epoch": 0.7376185458377239,
      "grad_norm": 0.5269689559936523,
      "learning_rate": 3.770635756937127e-05,
      "loss": 0.2498,
      "step": 700
    },
    {
      "epoch": 0.7376185458377239,
      "eval_loss": 0.3312172591686249,
      "eval_runtime": 13.6169,
      "eval_samples_per_second": 35.25,
      "eval_steps_per_second": 2.203,
      "step": 700
    }
  ],
  "logging_steps": 10,
  "max_steps": 2847,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.394121343808307e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}