{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9995963944571504,
  "eval_steps": 500,
  "global_step": 11148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.9990073706856164e-05,
      "loss": 2.3809,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.99603027099283e-05,
      "loss": 0.1075,
      "step": 200
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.991071065046783e-05,
      "loss": 0.0487,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.984133690970033e-05,
      "loss": 0.0357,
      "step": 400
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.975223657755279e-05,
      "loss": 0.0271,
      "step": 500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.9643480408906496e-05,
      "loss": 0.0257,
      "step": 600
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.951515476741036e-05,
      "loss": 0.0231,
      "step": 700
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.936736155689918e-05,
      "loss": 0.0176,
      "step": 800
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.920021814047156e-05,
      "loss": 0.018,
      "step": 900
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.901385724729137e-05,
      "loss": 0.0184,
      "step": 1000
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.880842686718711e-05,
      "loss": 0.0181,
      "step": 1100
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.858409013313266e-05,
      "loss": 0.016,
      "step": 1200
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.834354792422293e-05,
      "loss": 0.0171,
      "step": 1300
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.8082132142108465e-05,
      "loss": 0.0141,
      "step": 1400
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.780238675919182e-05,
      "loss": 0.0137,
      "step": 1500
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.75045339222471e-05,
      "loss": 0.0156,
      "step": 1600
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.718881015724017e-05,
      "loss": 0.0136,
      "step": 1700
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.6855466181502544e-05,
      "loss": 0.0134,
      "step": 1800
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.65047667046359e-05,
      "loss": 0.0101,
      "step": 1900
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.613699021830524e-05,
      "loss": 0.0126,
      "step": 2000
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.575635648326851e-05,
      "loss": 0.0138,
      "step": 2100
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.5355478705672265e-05,
      "loss": 0.0125,
      "step": 2200
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.4938436572178774e-05,
      "loss": 0.0097,
      "step": 2300
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.450556125738566e-05,
      "loss": 0.0157,
      "step": 2400
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.4057196509074484e-05,
      "loss": 0.0091,
      "step": 2500
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.359369837523941e-05,
      "loss": 0.0131,
      "step": 2600
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.311543492134827e-05,
      "loss": 0.0094,
      "step": 2700
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.2622785938060536e-05,
      "loss": 0.0139,
      "step": 2800
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.211614263963423e-05,
      "loss": 0.0089,
      "step": 2900
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.1595907353261296e-05,
      "loss": 0.009,
      "step": 3000
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.106249319957826e-05,
      "loss": 0.0103,
      "step": 3100
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.051632376460562e-05,
      "loss": 0.0115,
      "step": 3200
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.9957832763376815e-05,
      "loss": 0.0106,
      "step": 3300
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.9387463695523555e-05,
      "loss": 0.0098,
      "step": 3400
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.880566949309125e-05,
      "loss": 0.0087,
      "step": 3500
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.82129121608641e-05,
      "loss": 0.0081,
      "step": 3600
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.7609662409485565e-05,
      "loss": 0.0082,
      "step": 3700
    },
    {
      "epoch": 1.02,
      "learning_rate": 3.699639928166532e-05,
      "loss": 0.0067,
      "step": 3800
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.6373609771769846e-05,
      "loss": 0.0071,
      "step": 3900
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.574178843909851e-05,
      "loss": 0.0081,
      "step": 4000
    },
    {
      "epoch": 1.1,
      "learning_rate": 3.5101437015152346e-05,
      "loss": 0.0075,
      "step": 4100
    },
    {
      "epoch": 1.13,
      "learning_rate": 3.445306400520726e-05,
      "loss": 0.0088,
      "step": 4200
    },
    {
      "epoch": 1.16,
      "learning_rate": 3.379718428450832e-05,
      "loss": 0.0066,
      "step": 4300
    },
    {
      "epoch": 1.18,
      "learning_rate": 3.313431868940551e-05,
      "loss": 0.0057,
      "step": 4400
    },
    {
      "epoch": 1.21,
      "learning_rate": 3.24649936037558e-05,
      "loss": 0.0072,
      "step": 4500
    },
    {
      "epoch": 1.24,
      "learning_rate": 3.1789740540919856e-05,
      "loss": 0.0068,
      "step": 4600
    },
    {
      "epoch": 1.26,
      "learning_rate": 3.1109095721685525e-05,
      "loss": 0.0068,
      "step": 4700
    },
    {
      "epoch": 1.29,
      "learning_rate": 3.0423599648453037e-05,
      "loss": 0.008,
      "step": 4800
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.9733796676020104e-05,
      "loss": 0.0076,
      "step": 4900
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.9040234579308025e-05,
      "loss": 0.0071,
      "step": 5000
    },
    {
      "epoch": 1.37,
      "learning_rate": 2.8343464118371603e-05,
      "loss": 0.0071,
      "step": 5100
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.7644038601038813e-05,
      "loss": 0.0082,
      "step": 5200
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.6942513443527008e-05,
      "loss": 0.0047,
      "step": 5300
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.6239445729385104e-05,
      "loss": 0.0065,
      "step": 5400
    },
    {
      "epoch": 1.48,
      "learning_rate": 2.5535393767111533e-05,
      "loss": 0.0066,
      "step": 5500
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.483091664679959e-05,
      "loss": 0.0035,
      "step": 5600
    },
    {
      "epoch": 1.53,
      "learning_rate": 2.4126573796162015e-05,
      "loss": 0.0048,
      "step": 5700
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.3422924536287548e-05,
      "loss": 0.0072,
      "step": 5800
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.272052763748209e-05,
      "loss": 0.0043,
      "step": 5900
    },
    {
      "epoch": 1.61,
      "learning_rate": 2.201994087554733e-05,
      "loss": 0.008,
      "step": 6000
    },
    {
      "epoch": 1.64,
      "learning_rate": 2.1321720588848995e-05,
      "loss": 0.0057,
      "step": 6100
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.0626421236526656e-05,
      "loss": 0.0072,
      "step": 6200
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.9934594958195834e-05,
      "loss": 0.0041,
      "step": 6300
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.9246791135491982e-05,
      "loss": 0.0065,
      "step": 6400
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.8563555955804683e-05,
      "loss": 0.0057,
      "step": 6500
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.7885431978548305e-05,
      "loss": 0.0032,
      "step": 6600
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.7212957704313752e-05,
      "loss": 0.0076,
      "step": 6700
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.6546667147243222e-05,
      "loss": 0.0056,
      "step": 6800
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.5887089410967808e-05,
      "loss": 0.005,
      "step": 6900
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.5234748268444419e-05,
      "loss": 0.004,
      "step": 7000
    },
    {
      "epoch": 1.91,
      "learning_rate": 1.4590161746025896e-05,
      "loss": 0.0062,
      "step": 7100
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.3953841712094482e-05,
      "loss": 0.0056,
      "step": 7200
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.3326293470585377e-05,
      "loss": 0.0063,
      "step": 7300
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.2708015359723158e-05,
      "loss": 0.0057,
      "step": 7400
    },
    {
      "epoch": 2.02,
      "learning_rate": 1.2099498356289662e-05,
      "loss": 0.0047,
      "step": 7500
    },
    {
      "epoch": 2.04,
      "learning_rate": 1.1501225685737607e-05,
      "loss": 0.0032,
      "step": 7600
    },
    {
      "epoch": 2.07,
      "learning_rate": 1.0913672438459612e-05,
      "loss": 0.0018,
      "step": 7700
    },
    {
      "epoch": 2.1,
      "learning_rate": 1.0337305192517294e-05,
      "loss": 0.0029,
      "step": 7800
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.772581643129997e-06,
      "loss": 0.0052,
      "step": 7900
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.219950239217432e-06,
      "loss": 0.0032,
      "step": 8000
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.679849827284852e-06,
      "loss": 0.0026,
      "step": 8100
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.152709302933523e-06,
      "loss": 0.0051,
      "step": 8200
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.638947270273291e-06,
      "loss": 0.0024,
      "step": 8300
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.138971709507527e-06,
      "loss": 0.0035,
      "step": 8400
    },
    {
      "epoch": 2.29,
      "learning_rate": 6.653179652954706e-06,
      "loss": 0.003,
      "step": 8500
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.181956869763645e-06,
      "loss": 0.002,
      "step": 8600
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.725677559572909e-06,
      "loss": 0.0036,
      "step": 8700
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.284704055357578e-06,
      "loss": 0.0013,
      "step": 8800
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.859386535699373e-06,
      "loss": 0.002,
      "step": 8900
    },
    {
      "epoch": 2.42,
      "learning_rate": 4.4500627467086245e-06,
      "loss": 0.0048,
      "step": 9000
    },
    {
      "epoch": 2.45,
      "learning_rate": 4.057057733818967e-06,
      "loss": 0.0026,
      "step": 9100
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.680683583667599e-06,
      "loss": 0.002,
      "step": 9200
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.321239176266211e-06,
      "loss": 0.0039,
      "step": 9300
    },
    {
      "epoch": 2.53,
      "learning_rate": 2.9790099476593393e-06,
      "loss": 0.0015,
      "step": 9400
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.654267663258628e-06,
      "loss": 0.0039,
      "step": 9500
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.3472702020329696e-06,
      "loss": 0.0036,
      "step": 9600
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.0582613517259377e-06,
      "loss": 0.0039,
      "step": 9700
    },
    {
      "epoch": 2.64,
      "learning_rate": 1.7874706152630705e-06,
      "loss": 0.0019,
      "step": 9800
    },
    {
      "epoch": 2.66,
      "learning_rate": 1.5351130285028214e-06,
      "loss": 0.0031,
      "step": 9900
    },
    {
      "epoch": 2.69,
      "learning_rate": 1.301388989475863e-06,
      "loss": 0.0034,
      "step": 10000
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.0864840992482778e-06,
      "loss": 0.0024,
      "step": 10100
    },
    {
      "epoch": 2.74,
      "learning_rate": 8.905690145351475e-07,
      "loss": 0.0029,
      "step": 10200
    },
    {
      "epoch": 2.77,
      "learning_rate": 7.137993121814402e-07,
      "loss": 0.0033,
      "step": 10300
    },
    {
      "epoch": 2.8,
      "learning_rate": 5.563153656179209e-07,
      "loss": 0.0025,
      "step": 10400
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.1952650833933515e-07,
      "loss": 0.0022,
      "step": 10500
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.007781383039976e-07,
      "loss": 0.0022,
      "step": 10600
    },
    {
      "epoch": 2.88,
      "learning_rate": 2.0164350559856714e-07,
      "loss": 0.0026,
      "step": 10700
    },
    {
      "epoch": 2.91,
      "learning_rate": 1.222013333770422e-07,
      "loss": 0.003,
      "step": 10800
    },
    {
      "epoch": 2.93,
      "learning_rate": 6.251470694257356e-08,
      "loss": 0.0023,
      "step": 10900
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.2631023651226157e-08,
      "loss": 0.0024,
      "step": 11000
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.581955273558978e-09,
      "loss": 0.004,
      "step": 11100
    },
    {
      "epoch": 3.0,
      "step": 11148,
      "total_flos": 5.393398537924706e+18,
      "train_loss": 0.03027582302646779,
      "train_runtime": 186831.816,
      "train_samples_per_second": 1.91,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 100,
  "max_steps": 11148,
  "num_train_epochs": 3,
  "save_steps": 5000,
  "total_flos": 5.393398537924706e+18,
  "trial_name": null,
  "trial_params": null
}