{
  "best_metric": 0.4225879484500174,
  "best_model_checkpoint": "/groups/claytonm/enoriega/kw_pubmed/kw_pubmed_5000_0.00006/checkpoint-585",
  "epoch": 4.770047108764301,
  "global_step": 735,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 5.961038961038961e-05,
      "loss": 4.0597,
      "step": 5
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.922077922077922e-05,
      "loss": 3.63,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.883116883116883e-05,
      "loss": 3.467,
      "step": 15
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.3901079763148729,
      "eval_loss": 3.7511544227600098,
      "eval_runtime": 16.5637,
      "eval_samples_per_second": 603.729,
      "eval_steps_per_second": 18.897,
      "step": 15
    },
    {
      "epoch": 0.13,
      "learning_rate": 5.844155844155844e-05,
      "loss": 3.3347,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 5.805194805194805e-05,
      "loss": 3.2169,
      "step": 25
    },
    {
      "epoch": 0.19,
      "learning_rate": 5.7662337662337664e-05,
      "loss": 3.1359,
      "step": 30
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.39041274817136884,
      "eval_loss": 3.7756261825561523,
      "eval_runtime": 16.5475,
      "eval_samples_per_second": 604.322,
      "eval_steps_per_second": 18.915,
      "step": 30
    },
    {
      "epoch": 0.23,
      "learning_rate": 5.7272727272727274e-05,
      "loss": 3.076,
      "step": 35
    },
    {
      "epoch": 0.26,
      "learning_rate": 5.6883116883116884e-05,
      "loss": 2.9729,
      "step": 40
    },
    {
      "epoch": 0.29,
      "learning_rate": 5.6493506493506495e-05,
      "loss": 2.931,
      "step": 45
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.39467955416231276,
      "eval_loss": 3.798929452896118,
      "eval_runtime": 16.648,
      "eval_samples_per_second": 600.672,
      "eval_steps_per_second": 18.801,
      "step": 45
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.6103896103896105e-05,
      "loss": 2.8682,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 5.5714285714285715e-05,
      "loss": 2.8255,
      "step": 55
    },
    {
      "epoch": 0.39,
      "learning_rate": 5.5324675324675326e-05,
      "loss": 2.7794,
      "step": 60
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.39803204458376873,
      "eval_loss": 3.7885875701904297,
      "eval_runtime": 16.6293,
      "eval_samples_per_second": 601.348,
      "eval_steps_per_second": 18.822,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 5.4935064935064936e-05,
      "loss": 2.7529,
      "step": 65
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 2.699,
      "step": 70
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.4155844155844157e-05,
      "loss": 2.6464,
      "step": 75
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.4050853361198189,
      "eval_loss": 3.7896740436553955,
      "eval_runtime": 17.133,
      "eval_samples_per_second": 583.668,
      "eval_steps_per_second": 18.269,
      "step": 75
    },
    {
      "epoch": 0.52,
      "learning_rate": 5.376623376623377e-05,
      "loss": 2.6402,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 5.337662337662338e-05,
      "loss": 2.6216,
      "step": 85
    },
    {
      "epoch": 0.58,
      "learning_rate": 5.298701298701299e-05,
      "loss": 2.5798,
      "step": 90
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.40316962730755834,
      "eval_loss": 3.832270383834839,
      "eval_runtime": 17.2068,
      "eval_samples_per_second": 581.165,
      "eval_steps_per_second": 18.19,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 5.267532467532468e-05,
      "loss": 2.5716,
      "step": 95
    },
    {
      "epoch": 0.65,
      "learning_rate": 5.228571428571429e-05,
      "loss": 2.5615,
      "step": 100
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.18961038961039e-05,
      "loss": 2.5126,
      "step": 105
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.40687042842215254,
      "eval_loss": 3.8271889686584473,
      "eval_runtime": 17.1827,
      "eval_samples_per_second": 581.981,
      "eval_steps_per_second": 18.216,
      "step": 105
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.150649350649351e-05,
      "loss": 2.4727,
      "step": 110
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.111688311688312e-05,
      "loss": 2.5236,
      "step": 115
    },
    {
      "epoch": 0.78,
      "learning_rate": 5.072727272727273e-05,
      "loss": 2.4631,
      "step": 120
    },
    {
      "epoch": 0.78,
      "eval_accuracy": 0.39885928247997215,
      "eval_loss": 3.866598129272461,
      "eval_runtime": 17.205,
      "eval_samples_per_second": 581.228,
      "eval_steps_per_second": 18.192,
      "step": 120
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.033766233766233e-05,
      "loss": 2.4563,
      "step": 125
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.994805194805195e-05,
      "loss": 2.4166,
      "step": 130
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.955844155844156e-05,
      "loss": 2.4321,
      "step": 135
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.4076541274817137,
      "eval_loss": 3.8538074493408203,
      "eval_runtime": 17.5232,
      "eval_samples_per_second": 570.673,
      "eval_steps_per_second": 17.862,
      "step": 135
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.916883116883117e-05,
      "loss": 2.3705,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.877922077922078e-05,
      "loss": 2.3907,
      "step": 145
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.838961038961039e-05,
      "loss": 2.357,
      "step": 150
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.4062173458725183,
      "eval_loss": 3.8510184288024902,
      "eval_runtime": 17.1923,
      "eval_samples_per_second": 581.656,
      "eval_steps_per_second": 18.206,
      "step": 150
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.8e-05,
      "loss": 2.5669,
      "step": 155
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.761038961038961e-05,
      "loss": 2.2904,
      "step": 160
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.722077922077922e-05,
      "loss": 2.2463,
      "step": 165
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.4090909090909091,
      "eval_loss": 3.870206117630005,
      "eval_runtime": 17.2036,
      "eval_samples_per_second": 581.273,
      "eval_steps_per_second": 18.194,
      "step": 165
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.683116883116883e-05,
      "loss": 2.2434,
      "step": 170
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.644155844155844e-05,
      "loss": 2.2517,
      "step": 175
    },
    {
      "epoch": 1.17,
      "learning_rate": 4.605194805194805e-05,
      "loss": 2.2312,
      "step": 180
    },
    {
      "epoch": 1.17,
      "eval_accuracy": 0.4103099965168931,
      "eval_loss": 3.878140687942505,
      "eval_runtime": 17.2005,
      "eval_samples_per_second": 581.377,
      "eval_steps_per_second": 18.197,
      "step": 180
    },
    {
      "epoch": 1.2,
      "learning_rate": 4.566233766233766e-05,
      "loss": 2.2025,
      "step": 185
    },
    {
      "epoch": 1.23,
      "learning_rate": 4.5272727272727274e-05,
      "loss": 2.199,
      "step": 190
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.4883116883116884e-05,
      "loss": 2.1948,
      "step": 195
    },
    {
      "epoch": 1.27,
      "eval_accuracy": 0.4127917102055033,
      "eval_loss": 3.8671493530273438,
      "eval_runtime": 17.2222,
      "eval_samples_per_second": 580.645,
      "eval_steps_per_second": 18.174,
      "step": 195
    },
    {
      "epoch": 1.3,
      "learning_rate": 4.4493506493506494e-05,
      "loss": 2.1707,
      "step": 200
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.4103896103896105e-05,
      "loss": 2.1603,
      "step": 205
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.3714285714285715e-05,
      "loss": 2.1735,
      "step": 210
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.4157088122605364,
      "eval_loss": 3.863778829574585,
      "eval_runtime": 17.1906,
      "eval_samples_per_second": 581.713,
      "eval_steps_per_second": 18.208,
      "step": 210
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.3324675324675325e-05,
      "loss": 2.1833,
      "step": 215
    },
    {
      "epoch": 1.43,
      "learning_rate": 4.2935064935064936e-05,
      "loss": 2.1442,
      "step": 220
    },
    {
      "epoch": 1.46,
      "learning_rate": 4.2545454545454546e-05,
      "loss": 2.1541,
      "step": 225
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.4085249042145594,
      "eval_loss": 3.8899924755096436,
      "eval_runtime": 17.2073,
      "eval_samples_per_second": 581.148,
      "eval_steps_per_second": 18.19,
      "step": 225
    },
    {
      "epoch": 1.49,
      "learning_rate": 4.2155844155844156e-05,
      "loss": 2.1361,
      "step": 230
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.1766233766233767e-05,
      "loss": 2.1254,
      "step": 235
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.137662337662338e-05,
      "loss": 2.1117,
      "step": 240
    },
    {
      "epoch": 1.56,
      "eval_accuracy": 0.414272030651341,
      "eval_loss": 3.9167368412017822,
      "eval_runtime": 17.1975,
      "eval_samples_per_second": 581.48,
      "eval_steps_per_second": 18.2,
      "step": 240
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.098701298701299e-05,
      "loss": 2.1208,
      "step": 245
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.05974025974026e-05,
      "loss": 2.0991,
      "step": 250
    },
    {
      "epoch": 1.65,
      "learning_rate": 4.020779220779221e-05,
      "loss": 2.1094,
      "step": 255
    },
    {
      "epoch": 1.65,
      "eval_accuracy": 0.4169278996865204,
      "eval_loss": 3.8918027877807617,
      "eval_runtime": 17.2198,
      "eval_samples_per_second": 580.727,
      "eval_steps_per_second": 18.177,
      "step": 255
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.9818181818181825e-05,
      "loss": 2.1038,
      "step": 260
    },
    {
      "epoch": 1.72,
      "learning_rate": 3.942857142857143e-05,
      "loss": 2.0806,
      "step": 265
    },
    {
      "epoch": 1.75,
      "learning_rate": 3.9038961038961046e-05,
      "loss": 2.0917,
      "step": 270
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.41300940438871475,
      "eval_loss": 3.920091152191162,
      "eval_runtime": 17.181,
      "eval_samples_per_second": 582.037,
      "eval_steps_per_second": 18.218,
      "step": 270
    },
    {
      "epoch": 1.78,
      "learning_rate": 3.864935064935065e-05,
      "loss": 2.0582,
      "step": 275
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.8259740259740266e-05,
      "loss": 2.0541,
      "step": 280
    },
    {
      "epoch": 1.85,
      "learning_rate": 3.787012987012987e-05,
      "loss": 2.0453,
      "step": 285
    },
    {
      "epoch": 1.85,
      "eval_accuracy": 0.41814698711250436,
      "eval_loss": 3.8712852001190186,
      "eval_runtime": 17.1826,
      "eval_samples_per_second": 581.984,
      "eval_steps_per_second": 18.216,
      "step": 285
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.748051948051949e-05,
      "loss": 2.0351,
      "step": 290
    },
    {
      "epoch": 1.91,
      "learning_rate": 3.709090909090909e-05,
      "loss": 2.0527,
      "step": 295
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.670129870129871e-05,
      "loss": 2.0247,
      "step": 300
    },
    {
      "epoch": 1.94,
      "eval_accuracy": 0.41762452107279696,
      "eval_loss": 3.8774173259735107,
      "eval_runtime": 17.2528,
      "eval_samples_per_second": 579.615,
      "eval_steps_per_second": 18.142,
      "step": 300
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.631168831168831e-05,
      "loss": 2.0146,
      "step": 305
    },
    {
      "epoch": 2.01,
      "learning_rate": 3.592207792207792e-05,
      "loss": 2.2296,
      "step": 310
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.553246753246753e-05,
      "loss": 1.9703,
      "step": 315
    },
    {
      "epoch": 2.05,
      "eval_accuracy": 0.4164925113200975,
      "eval_loss": 3.891786813735962,
      "eval_runtime": 17.1846,
      "eval_samples_per_second": 581.915,
      "eval_steps_per_second": 18.214,
      "step": 315
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.514285714285714e-05,
      "loss": 1.9429,
      "step": 320
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.475324675324675e-05,
      "loss": 1.9408,
      "step": 325
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.436363636363636e-05,
      "loss": 1.9492,
      "step": 330
    },
    {
      "epoch": 2.14,
      "eval_accuracy": 0.41610066179031696,
      "eval_loss": 3.928813934326172,
      "eval_runtime": 17.2139,
      "eval_samples_per_second": 580.927,
      "eval_steps_per_second": 18.183,
      "step": 330
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.397402597402597e-05,
      "loss": 1.9459,
      "step": 335
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.3584415584415584e-05,
      "loss": 1.9532,
      "step": 340
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.3194805194805194e-05,
      "loss": 1.9351,
      "step": 345
    },
    {
      "epoch": 2.24,
      "eval_accuracy": 0.4128352490421456,
      "eval_loss": 3.947727680206299,
      "eval_runtime": 17.2191,
      "eval_samples_per_second": 580.75,
      "eval_steps_per_second": 18.177,
      "step": 345
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.2805194805194804e-05,
      "loss": 1.9483,
      "step": 350
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.2415584415584415e-05,
      "loss": 1.9433,
      "step": 355
    },
    {
      "epoch": 2.34,
      "learning_rate": 3.2025974025974025e-05,
      "loss": 1.9001,
      "step": 360
    },
    {
      "epoch": 2.34,
      "eval_accuracy": 0.4186259143155695,
      "eval_loss": 3.9299826622009277,
      "eval_runtime": 17.1998,
      "eval_samples_per_second": 581.401,
      "eval_steps_per_second": 18.198,
      "step": 360
    },
    {
      "epoch": 2.37,
      "learning_rate": 3.1636363636363635e-05,
      "loss": 1.9127,
      "step": 365
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.1246753246753246e-05,
      "loss": 1.9051,
      "step": 370
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.0857142857142856e-05,
      "loss": 1.9082,
      "step": 375
    },
    {
      "epoch": 2.43,
      "eval_accuracy": 0.4161442006269592,
      "eval_loss": 3.9622690677642822,
      "eval_runtime": 17.2078,
      "eval_samples_per_second": 581.133,
      "eval_steps_per_second": 18.189,
      "step": 375
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.046753246753247e-05,
      "loss": 1.8929,
      "step": 380
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.0077922077922077e-05,
      "loss": 1.912,
      "step": 385
    },
    {
      "epoch": 2.53,
      "learning_rate": 2.968831168831169e-05,
      "loss": 1.8804,
      "step": 390
    },
    {
      "epoch": 2.53,
      "eval_accuracy": 0.4174503657262278,
      "eval_loss": 3.9447922706604004,
      "eval_runtime": 17.219,
      "eval_samples_per_second": 580.753,
      "eval_steps_per_second": 18.178,
      "step": 390
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.92987012987013e-05,
      "loss": 1.8821,
      "step": 395
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.890909090909091e-05,
      "loss": 1.8962,
      "step": 400
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.851948051948052e-05,
      "loss": 1.8962,
      "step": 405
    },
    {
      "epoch": 2.63,
      "eval_accuracy": 0.4183646812957158,
      "eval_loss": 3.939225673675537,
      "eval_runtime": 17.2072,
      "eval_samples_per_second": 581.153,
      "eval_steps_per_second": 18.19,
      "step": 405
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.812987012987013e-05,
      "loss": 1.8903,
      "step": 410
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.774025974025974e-05,
      "loss": 1.8881,
      "step": 415
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.735064935064935e-05,
      "loss": 1.884,
      "step": 420
    },
    {
      "epoch": 2.72,
      "eval_accuracy": 0.4125740160222919,
      "eval_loss": 3.95487380027771,
      "eval_runtime": 17.2137,
      "eval_samples_per_second": 580.934,
      "eval_steps_per_second": 18.183,
      "step": 420
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.696103896103896e-05,
      "loss": 1.8694,
      "step": 425
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.657142857142857e-05,
      "loss": 1.872,
      "step": 430
    },
    {
      "epoch": 2.82,
      "learning_rate": 2.618181818181818e-05,
      "loss": 1.8689,
      "step": 435
    },
    {
      "epoch": 2.82,
      "eval_accuracy": 0.4190613026819923,
      "eval_loss": 3.909341812133789,
      "eval_runtime": 17.2189,
      "eval_samples_per_second": 580.758,
      "eval_steps_per_second": 18.178,
      "step": 435
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.579220779220779e-05,
      "loss": 1.877,
      "step": 440
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.54025974025974e-05,
      "loss": 1.8685,
      "step": 445
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.501298701298701e-05,
      "loss": 1.8522,
      "step": 450
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.4203239289446186,
      "eval_loss": 3.9448978900909424,
      "eval_runtime": 17.2358,
      "eval_samples_per_second": 580.188,
      "eval_steps_per_second": 18.16,
      "step": 450
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.4623376623376625e-05,
      "loss": 1.8654,
      "step": 455
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.4233766233766235e-05,
      "loss": 1.8605,
      "step": 460
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.3844155844155845e-05,
      "loss": 2.037,
      "step": 465
    },
    {
      "epoch": 3.02,
      "eval_accuracy": 0.41705851619644724,
      "eval_loss": 3.981081008911133,
      "eval_runtime": 17.206,
      "eval_samples_per_second": 581.194,
      "eval_steps_per_second": 18.191,
      "step": 465
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.3454545454545456e-05,
      "loss": 1.8046,
      "step": 470
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.3064935064935066e-05,
      "loss": 1.8011,
      "step": 475
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.2675324675324676e-05,
      "loss": 1.8038,
      "step": 480
    },
    {
      "epoch": 3.12,
      "eval_accuracy": 0.4150992685475444,
      "eval_loss": 3.9613595008850098,
      "eval_runtime": 17.1973,
      "eval_samples_per_second": 581.488,
      "eval_steps_per_second": 18.201,
      "step": 480
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.2285714285714287e-05,
      "loss": 1.8073,
      "step": 485
    },
    {
      "epoch": 3.18,
      "learning_rate": 2.1896103896103897e-05,
      "loss": 1.783,
      "step": 490
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.1506493506493507e-05,
      "loss": 1.805,
      "step": 495
    },
    {
      "epoch": 3.21,
      "eval_accuracy": 0.41819052594914663,
      "eval_loss": 3.9464197158813477,
      "eval_runtime": 17.2204,
      "eval_samples_per_second": 580.705,
      "eval_steps_per_second": 18.176,
      "step": 495
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.1116883116883118e-05,
      "loss": 1.7972,
      "step": 500
    },
    {
      "epoch": 3.28,
      "learning_rate": 2.0727272727272728e-05,
      "loss": 1.7782,
      "step": 505
    },
    {
      "epoch": 3.31,
      "learning_rate": 2.033766233766234e-05,
      "loss": 1.8135,
      "step": 510
    },
    {
      "epoch": 3.31,
      "eval_accuracy": 0.41592650644374785,
      "eval_loss": 3.9678966999053955,
      "eval_runtime": 17.1877,
      "eval_samples_per_second": 581.811,
      "eval_steps_per_second": 18.211,
      "step": 510
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.994805194805195e-05,
      "loss": 1.7835,
      "step": 515
    },
    {
      "epoch": 3.38,
      "learning_rate": 1.955844155844156e-05,
      "loss": 1.782,
      "step": 520
    },
    {
      "epoch": 3.41,
      "learning_rate": 1.916883116883117e-05,
      "loss": 1.7927,
      "step": 525
    },
    {
      "epoch": 3.41,
      "eval_accuracy": 0.419932079414838,
      "eval_loss": 3.9554903507232666,
      "eval_runtime": 17.1692,
      "eval_samples_per_second": 582.437,
      "eval_steps_per_second": 18.23,
      "step": 525
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.877922077922078e-05,
      "loss": 1.7979,
      "step": 530
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.838961038961039e-05,
      "loss": 1.7894,
      "step": 535
    },
    {
      "epoch": 3.5,
      "learning_rate": 1.8e-05,
      "loss": 1.7723,
      "step": 540
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.41936607453848834,
      "eval_loss": 3.970630645751953,
      "eval_runtime": 17.1943,
      "eval_samples_per_second": 581.588,
      "eval_steps_per_second": 18.204,
      "step": 540
    },
    {
      "epoch": 3.54,
      "learning_rate": 1.761038961038961e-05,
      "loss": 1.797,
      "step": 545
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.722077922077922e-05,
      "loss": 1.7836,
      "step": 550
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.683116883116883e-05,
      "loss": 1.7893,
      "step": 555
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.4174068268895855,
      "eval_loss": 3.984471321105957,
      "eval_runtime": 17.1471,
      "eval_samples_per_second": 583.19,
      "eval_steps_per_second": 18.254,
      "step": 555
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.6441558441558442e-05,
      "loss": 1.7891,
      "step": 560
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.6051948051948052e-05,
      "loss": 1.7564,
      "step": 565
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.5662337662337662e-05,
      "loss": 1.7725,
      "step": 570
    },
    {
      "epoch": 3.7,
      "eval_accuracy": 0.41762452107279696,
      "eval_loss": 3.9673519134521484,
      "eval_runtime": 16.8716,
      "eval_samples_per_second": 592.713,
      "eval_steps_per_second": 18.552,
      "step": 570
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.5272727272727273e-05,
      "loss": 1.7631,
      "step": 575
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.4883116883116883e-05,
      "loss": 1.7441,
      "step": 580
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.4493506493506493e-05,
      "loss": 1.7721,
      "step": 585
    },
    {
      "epoch": 3.8,
      "eval_accuracy": 0.4225879484500174,
      "eval_loss": 3.9382729530334473,
      "eval_runtime": 17.13,
      "eval_samples_per_second": 583.773,
      "eval_steps_per_second": 18.272,
      "step": 585
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.4103896103896104e-05,
      "loss": 1.7317,
      "step": 590
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.3714285714285714e-05,
      "loss": 1.7738,
      "step": 595
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.3324675324675324e-05,
      "loss": 1.7533,
      "step": 600
    },
    {
      "epoch": 3.89,
      "eval_accuracy": 0.4196708463949843,
      "eval_loss": 3.96612811088562,
      "eval_runtime": 17.1432,
      "eval_samples_per_second": 583.32,
      "eval_steps_per_second": 18.258,
      "step": 600
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.2935064935064935e-05,
      "loss": 1.7658,
      "step": 605
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.2545454545454545e-05,
      "loss": 1.7635,
      "step": 610
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.2155844155844157e-05,
      "loss": 1.7437,
      "step": 615
    },
    {
      "epoch": 3.99,
      "eval_accuracy": 0.41910484151863464,
      "eval_loss": 3.972932815551758,
      "eval_runtime": 17.2356,
      "eval_samples_per_second": 580.195,
      "eval_steps_per_second": 18.16,
      "step": 615
    },
    {
      "epoch": 4.03,
      "learning_rate": 1.1766233766233768e-05,
      "loss": 1.9054,
      "step": 620
    },
    {
      "epoch": 4.06,
      "learning_rate": 1.1376623376623378e-05,
      "loss": 1.7217,
      "step": 625
    },
    {
      "epoch": 4.09,
      "learning_rate": 1.0987012987012988e-05,
      "loss": 1.7418,
      "step": 630
    },
    {
      "epoch": 4.09,
      "eval_accuracy": 0.4188000696621386,
      "eval_loss": 4.000043869018555,
      "eval_runtime": 17.2412,
      "eval_samples_per_second": 580.005,
      "eval_steps_per_second": 18.154,
      "step": 630
    },
    {
      "epoch": 4.12,
      "learning_rate": 1.0597402597402599e-05,
      "loss": 1.6958,
      "step": 635
    },
    {
      "epoch": 4.16,
      "learning_rate": 1.0207792207792209e-05,
      "loss": 1.7156,
      "step": 640
    },
    {
      "epoch": 4.19,
      "learning_rate": 9.81818181818182e-06,
      "loss": 1.7237,
      "step": 645
    },
    {
      "epoch": 4.19,
      "eval_accuracy": 0.42106408916753746,
      "eval_loss": 3.971540927886963,
      "eval_runtime": 17.1943,
      "eval_samples_per_second": 581.589,
      "eval_steps_per_second": 18.204,
      "step": 645
    },
    {
      "epoch": 4.22,
      "learning_rate": 9.428571428571428e-06,
      "loss": 1.7111,
      "step": 650
    },
    {
      "epoch": 4.25,
      "learning_rate": 9.038961038961038e-06,
      "loss": 1.7228,
      "step": 655
    },
    {
      "epoch": 4.28,
      "learning_rate": 8.649350649350649e-06,
      "loss": 1.7029,
      "step": 660
    },
    {
      "epoch": 4.28,
      "eval_accuracy": 0.41919191919191917,
      "eval_loss": 4.000442981719971,
      "eval_runtime": 17.2363,
      "eval_samples_per_second": 580.172,
      "eval_steps_per_second": 18.159,
      "step": 660
    },
    {
      "epoch": 4.32,
      "learning_rate": 8.259740259740259e-06,
      "loss": 1.7336,
      "step": 665
    },
    {
      "epoch": 4.35,
      "learning_rate": 7.87012987012987e-06,
      "loss": 1.7146,
      "step": 670
    },
    {
      "epoch": 4.38,
      "learning_rate": 7.48051948051948e-06,
      "loss": 1.7096,
      "step": 675
    },
    {
      "epoch": 4.38,
      "eval_accuracy": 0.4215430163706026,
      "eval_loss": 3.9922502040863037,
      "eval_runtime": 17.1842,
      "eval_samples_per_second": 581.93,
      "eval_steps_per_second": 18.214,
      "step": 675
    },
    {
      "epoch": 4.41,
      "learning_rate": 7.090909090909091e-06,
      "loss": 1.7183,
      "step": 680
    },
    {
      "epoch": 4.45,
      "learning_rate": 6.701298701298701e-06,
      "loss": 1.7323,
      "step": 685
    },
    {
      "epoch": 4.48,
      "learning_rate": 6.311688311688311e-06,
      "loss": 1.7087,
      "step": 690
    },
    {
      "epoch": 4.48,
      "eval_accuracy": 0.42254440961337514,
      "eval_loss": 3.9751763343811035,
      "eval_runtime": 17.2048,
      "eval_samples_per_second": 581.233,
      "eval_steps_per_second": 18.193,
      "step": 690
    },
    {
      "epoch": 4.51,
      "learning_rate": 5.9220779220779226e-06,
      "loss": 1.7267,
      "step": 695
    },
    {
      "epoch": 4.54,
      "learning_rate": 5.532467532467533e-06,
      "loss": 1.6924,
      "step": 700
    },
    {
      "epoch": 4.58,
      "learning_rate": 5.142857142857143e-06,
      "loss": 1.7101,
      "step": 705
    },
    {
      "epoch": 4.58,
      "eval_accuracy": 0.42080285614768376,
      "eval_loss": 3.9924566745758057,
      "eval_runtime": 17.2206,
      "eval_samples_per_second": 580.699,
      "eval_steps_per_second": 18.176,
      "step": 705
    },
    {
      "epoch": 4.61,
      "learning_rate": 4.7532467532467536e-06,
      "loss": 1.7133,
      "step": 710
    },
    {
      "epoch": 4.64,
      "learning_rate": 4.363636363636364e-06,
      "loss": 1.7042,
      "step": 715
    },
    {
      "epoch": 4.67,
      "learning_rate": 3.974025974025973e-06,
      "loss": 1.7061,
      "step": 720
    },
    {
      "epoch": 4.67,
      "eval_accuracy": 0.4183211424590735,
      "eval_loss": 3.9925851821899414,
      "eval_runtime": 17.1622,
      "eval_samples_per_second": 582.675,
      "eval_steps_per_second": 18.238,
      "step": 720
    },
    {
      "epoch": 4.71,
      "learning_rate": 3.5844155844155846e-06,
      "loss": 1.7194,
      "step": 725
    },
    {
      "epoch": 4.74,
      "learning_rate": 3.194805194805195e-06,
      "loss": 1.7283,
      "step": 730
    },
    {
      "epoch": 4.77,
      "learning_rate": 2.8051948051948052e-06,
      "loss": 1.7227,
      "step": 735
    },
    {
      "epoch": 4.77,
      "eval_accuracy": 0.41671020550330895,
      "eval_loss": 4.021413326263428,
      "eval_runtime": 17.2026,
      "eval_samples_per_second": 581.308,
      "eval_steps_per_second": 18.195,
      "step": 735
    },
    {
      "epoch": 4.77,
      "step": 735,
      "total_flos": 4.8027460072854234e+17,
      "train_loss": 2.085782461101506,
      "train_runtime": 24097.4857,
      "train_samples_per_second": 256.517,
      "train_steps_per_second": 0.032
    },
    {
      "epoch": 4.77,
      "eval_accuracy": 0.4171891327063741,
      "eval_loss": 3.979358434677124,
      "eval_runtime": 17.2044,
      "eval_samples_per_second": 581.246,
      "eval_steps_per_second": 18.193,
      "step": 735
    }
  ],
  "max_steps": 770,
  "num_train_epochs": 5,
  "total_flos": 4.8027460072854234e+17,
  "trial_name": null,
  "trial_params": null
}