{
  "best_metric": 0.718989971086903,
  "best_model_checkpoint": "20E-affecthq/checkpoint-2716",
  "epoch": 20.0,
  "global_step": 3880,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 2.577319587628866e-07,
      "loss": 2.0744,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.154639175257732e-07,
      "loss": 2.0746,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 7.731958762886599e-07,
      "loss": 2.0716,
      "step": 30
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.0309278350515464e-06,
      "loss": 2.0661,
      "step": 40
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.288659793814433e-06,
      "loss": 2.0648,
      "step": 50
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.5463917525773197e-06,
      "loss": 2.0612,
      "step": 60
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.8041237113402063e-06,
      "loss": 2.0582,
      "step": 70
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.061855670103093e-06,
      "loss": 2.0522,
      "step": 80
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.3195876288659796e-06,
      "loss": 2.0516,
      "step": 90
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.577319587628866e-06,
      "loss": 2.0432,
      "step": 100
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.8350515463917527e-06,
      "loss": 2.0425,
      "step": 110
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.0927835051546395e-06,
      "loss": 2.0368,
      "step": 120
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.350515463917526e-06,
      "loss": 2.0229,
      "step": 130
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.6082474226804126e-06,
      "loss": 2.0163,
      "step": 140
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.865979381443299e-06,
      "loss": 1.9999,
      "step": 150
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.123711340206186e-06,
      "loss": 1.9825,
      "step": 160
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.381443298969073e-06,
      "loss": 1.9702,
      "step": 170
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.639175257731959e-06,
      "loss": 1.9417,
      "step": 180
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.8969072164948455e-06,
      "loss": 1.9149,
      "step": 190
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.37504030957755563,
      "eval_f1": 0.30450075259542253,
      "eval_loss": 1.8887449502944946,
      "eval_precision": 0.34133971468533547,
      "eval_recall": 0.37504030957755563,
      "eval_runtime": 204.064,
      "eval_samples_per_second": 15.196,
      "eval_steps_per_second": 0.475,
      "step": 194
    },
    {
      "epoch": 1.03,
      "learning_rate": 5.154639175257732e-06,
      "loss": 1.8825,
      "step": 200
    },
    {
      "epoch": 1.08,
      "learning_rate": 5.412371134020619e-06,
      "loss": 1.8587,
      "step": 210
    },
    {
      "epoch": 1.13,
      "learning_rate": 5.670103092783505e-06,
      "loss": 1.7938,
      "step": 220
    },
    {
      "epoch": 1.19,
      "learning_rate": 5.927835051546393e-06,
      "loss": 1.7462,
      "step": 230
    },
    {
      "epoch": 1.24,
      "learning_rate": 6.185567010309279e-06,
      "loss": 1.6973,
      "step": 240
    },
    {
      "epoch": 1.29,
      "learning_rate": 6.443298969072166e-06,
      "loss": 1.6757,
      "step": 250
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.701030927835052e-06,
      "loss": 1.6265,
      "step": 260
    },
    {
      "epoch": 1.39,
      "learning_rate": 6.958762886597939e-06,
      "loss": 1.5975,
      "step": 270
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.216494845360825e-06,
      "loss": 1.5301,
      "step": 280
    },
    {
      "epoch": 1.49,
      "learning_rate": 7.474226804123712e-06,
      "loss": 1.5235,
      "step": 290
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.731958762886599e-06,
      "loss": 1.4745,
      "step": 300
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.989690721649486e-06,
      "loss": 1.4207,
      "step": 310
    },
    {
      "epoch": 1.65,
      "learning_rate": 8.247422680412371e-06,
      "loss": 1.4171,
      "step": 320
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.505154639175259e-06,
      "loss": 1.419,
      "step": 330
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.762886597938146e-06,
      "loss": 1.3424,
      "step": 340
    },
    {
      "epoch": 1.8,
      "learning_rate": 9.020618556701031e-06,
      "loss": 1.3403,
      "step": 350
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.278350515463918e-06,
      "loss": 1.3536,
      "step": 360
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.536082474226806e-06,
      "loss": 1.3036,
      "step": 370
    },
    {
      "epoch": 1.96,
      "learning_rate": 9.793814432989691e-06,
      "loss": 1.2903,
      "step": 380
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5791680103192518,
      "eval_f1": 0.5526365213062105,
      "eval_loss": 1.2485474348068237,
      "eval_precision": 0.5725810384525178,
      "eval_recall": 0.5791680103192518,
      "eval_runtime": 141.2499,
      "eval_samples_per_second": 21.954,
      "eval_steps_per_second": 0.687,
      "step": 388
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.994272623138604e-06,
      "loss": 1.2612,
      "step": 390
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.965635738831616e-06,
      "loss": 1.2295,
      "step": 400
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.93699885452463e-06,
      "loss": 1.2367,
      "step": 410
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.908361970217641e-06,
      "loss": 1.2015,
      "step": 420
    },
    {
      "epoch": 2.22,
      "learning_rate": 9.879725085910653e-06,
      "loss": 1.1918,
      "step": 430
    },
    {
      "epoch": 2.27,
      "learning_rate": 9.851088201603667e-06,
      "loss": 1.2132,
      "step": 440
    },
    {
      "epoch": 2.32,
      "learning_rate": 9.822451317296679e-06,
      "loss": 1.1727,
      "step": 450
    },
    {
      "epoch": 2.37,
      "learning_rate": 9.793814432989691e-06,
      "loss": 1.1598,
      "step": 460
    },
    {
      "epoch": 2.42,
      "learning_rate": 9.765177548682703e-06,
      "loss": 1.137,
      "step": 470
    },
    {
      "epoch": 2.47,
      "learning_rate": 9.736540664375717e-06,
      "loss": 1.167,
      "step": 480
    },
    {
      "epoch": 2.53,
      "learning_rate": 9.707903780068729e-06,
      "loss": 1.1318,
      "step": 490
    },
    {
      "epoch": 2.58,
      "learning_rate": 9.679266895761742e-06,
      "loss": 1.1026,
      "step": 500
    },
    {
      "epoch": 2.63,
      "learning_rate": 9.650630011454754e-06,
      "loss": 1.1069,
      "step": 510
    },
    {
      "epoch": 2.68,
      "learning_rate": 9.621993127147768e-06,
      "loss": 1.0721,
      "step": 520
    },
    {
      "epoch": 2.73,
      "learning_rate": 9.59335624284078e-06,
      "loss": 1.0903,
      "step": 530
    },
    {
      "epoch": 2.78,
      "learning_rate": 9.564719358533792e-06,
      "loss": 1.0951,
      "step": 540
    },
    {
      "epoch": 2.84,
      "learning_rate": 9.536082474226806e-06,
      "loss": 1.1402,
      "step": 550
    },
    {
      "epoch": 2.89,
      "learning_rate": 9.507445589919818e-06,
      "loss": 1.1012,
      "step": 560
    },
    {
      "epoch": 2.94,
      "learning_rate": 9.47880870561283e-06,
      "loss": 1.0767,
      "step": 570
    },
    {
      "epoch": 2.99,
      "learning_rate": 9.450171821305843e-06,
      "loss": 1.071,
      "step": 580
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6320541760722348,
      "eval_f1": 0.6228282492696803,
      "eval_loss": 1.0587016344070435,
      "eval_precision": 0.6258393540507952,
      "eval_recall": 0.6320541760722348,
      "eval_runtime": 156.4705,
      "eval_samples_per_second": 19.818,
      "eval_steps_per_second": 0.62,
      "step": 582
    },
    {
      "epoch": 3.04,
      "learning_rate": 9.421534936998855e-06,
      "loss": 1.0438,
      "step": 590
    },
    {
      "epoch": 3.09,
      "learning_rate": 9.392898052691867e-06,
      "loss": 1.0081,
      "step": 600
    },
    {
      "epoch": 3.14,
      "learning_rate": 9.36426116838488e-06,
      "loss": 1.0547,
      "step": 610
    },
    {
      "epoch": 3.2,
      "learning_rate": 9.335624284077893e-06,
      "loss": 1.0588,
      "step": 620
    },
    {
      "epoch": 3.25,
      "learning_rate": 9.306987399770906e-06,
      "loss": 1.0339,
      "step": 630
    },
    {
      "epoch": 3.3,
      "learning_rate": 9.278350515463918e-06,
      "loss": 1.0542,
      "step": 640
    },
    {
      "epoch": 3.35,
      "learning_rate": 9.249713631156932e-06,
      "loss": 1.0186,
      "step": 650
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.221076746849944e-06,
      "loss": 0.9947,
      "step": 660
    },
    {
      "epoch": 3.45,
      "learning_rate": 9.192439862542956e-06,
      "loss": 1.0304,
      "step": 670
    },
    {
      "epoch": 3.51,
      "learning_rate": 9.163802978235968e-06,
      "loss": 0.9644,
      "step": 680
    },
    {
      "epoch": 3.56,
      "learning_rate": 9.135166093928982e-06,
      "loss": 1.0023,
      "step": 690
    },
    {
      "epoch": 3.61,
      "learning_rate": 9.106529209621994e-06,
      "loss": 0.9797,
      "step": 700
    },
    {
      "epoch": 3.66,
      "learning_rate": 9.077892325315006e-06,
      "loss": 0.9779,
      "step": 710
    },
    {
      "epoch": 3.71,
      "learning_rate": 9.04925544100802e-06,
      "loss": 1.0429,
      "step": 720
    },
    {
      "epoch": 3.76,
      "learning_rate": 9.020618556701031e-06,
      "loss": 0.9848,
      "step": 730
    },
    {
      "epoch": 3.81,
      "learning_rate": 8.991981672394045e-06,
      "loss": 0.9981,
      "step": 740
    },
    {
      "epoch": 3.87,
      "learning_rate": 8.963344788087057e-06,
      "loss": 0.9571,
      "step": 750
    },
    {
      "epoch": 3.92,
      "learning_rate": 8.93470790378007e-06,
      "loss": 0.9786,
      "step": 760
    },
    {
      "epoch": 3.97,
      "learning_rate": 8.90607101947308e-06,
      "loss": 1.0185,
      "step": 770
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6617220251531764,
      "eval_f1": 0.6552882607911024,
      "eval_loss": 0.9817498326301575,
      "eval_precision": 0.6583817535052932,
      "eval_recall": 0.6617220251531764,
      "eval_runtime": 140.4425,
      "eval_samples_per_second": 22.08,
      "eval_steps_per_second": 0.691,
      "step": 776
    },
    {
      "epoch": 4.02,
      "learning_rate": 8.877434135166094e-06,
      "loss": 0.955,
      "step": 780
    },
    {
      "epoch": 4.07,
      "learning_rate": 8.848797250859106e-06,
      "loss": 0.9995,
      "step": 790
    },
    {
      "epoch": 4.12,
      "learning_rate": 8.82016036655212e-06,
      "loss": 0.9615,
      "step": 800
    },
    {
      "epoch": 4.18,
      "learning_rate": 8.791523482245132e-06,
      "loss": 0.8775,
      "step": 810
    },
    {
      "epoch": 4.23,
      "learning_rate": 8.762886597938146e-06,
      "loss": 0.948,
      "step": 820
    },
    {
      "epoch": 4.28,
      "learning_rate": 8.734249713631158e-06,
      "loss": 0.9255,
      "step": 830
    },
    {
      "epoch": 4.33,
      "learning_rate": 8.70561282932417e-06,
      "loss": 0.9125,
      "step": 840
    },
    {
      "epoch": 4.38,
      "learning_rate": 8.676975945017183e-06,
      "loss": 0.9153,
      "step": 850
    },
    {
      "epoch": 4.43,
      "learning_rate": 8.648339060710195e-06,
      "loss": 0.9227,
      "step": 860
    },
    {
      "epoch": 4.48,
      "learning_rate": 8.619702176403209e-06,
      "loss": 0.945,
      "step": 870
    },
    {
      "epoch": 4.54,
      "learning_rate": 8.591065292096221e-06,
      "loss": 0.9121,
      "step": 880
    },
    {
      "epoch": 4.59,
      "learning_rate": 8.562428407789233e-06,
      "loss": 0.9311,
      "step": 890
    },
    {
      "epoch": 4.64,
      "learning_rate": 8.533791523482245e-06,
      "loss": 0.8944,
      "step": 900
    },
    {
      "epoch": 4.69,
      "learning_rate": 8.505154639175259e-06,
      "loss": 0.9093,
      "step": 910
    },
    {
      "epoch": 4.74,
      "learning_rate": 8.47651775486827e-06,
      "loss": 0.8956,
      "step": 920
    },
    {
      "epoch": 4.79,
      "learning_rate": 8.447880870561284e-06,
      "loss": 0.963,
      "step": 930
    },
    {
      "epoch": 4.85,
      "learning_rate": 8.419243986254296e-06,
      "loss": 0.9149,
      "step": 940
    },
    {
      "epoch": 4.9,
      "learning_rate": 8.39060710194731e-06,
      "loss": 0.909,
      "step": 950
    },
    {
      "epoch": 4.95,
      "learning_rate": 8.361970217640322e-06,
      "loss": 0.8628,
      "step": 960
    },
    {
      "epoch": 5.0,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.894,
      "step": 970
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6868752015478877,
      "eval_f1": 0.6820473818962892,
      "eval_loss": 0.9292588829994202,
      "eval_precision": 0.6872338048544774,
      "eval_recall": 0.6868752015478877,
      "eval_runtime": 140.2334,
      "eval_samples_per_second": 22.113,
      "eval_steps_per_second": 0.692,
      "step": 970
    },
    {
      "epoch": 5.05,
      "learning_rate": 8.304696449026347e-06,
      "loss": 0.8574,
      "step": 980
    },
    {
      "epoch": 5.1,
      "learning_rate": 8.27605956471936e-06,
      "loss": 0.8418,
      "step": 990
    },
    {
      "epoch": 5.15,
      "learning_rate": 8.247422680412371e-06,
      "loss": 0.8805,
      "step": 1000
    },
    {
      "epoch": 5.21,
      "learning_rate": 8.218785796105383e-06,
      "loss": 0.8624,
      "step": 1010
    },
    {
      "epoch": 5.26,
      "learning_rate": 8.190148911798397e-06,
      "loss": 0.8462,
      "step": 1020
    },
    {
      "epoch": 5.31,
      "learning_rate": 8.161512027491409e-06,
      "loss": 0.8716,
      "step": 1030
    },
    {
      "epoch": 5.36,
      "learning_rate": 8.132875143184423e-06,
      "loss": 0.8289,
      "step": 1040
    },
    {
      "epoch": 5.41,
      "learning_rate": 8.104238258877435e-06,
      "loss": 0.8918,
      "step": 1050
    },
    {
      "epoch": 5.46,
      "learning_rate": 8.075601374570448e-06,
      "loss": 0.8383,
      "step": 1060
    },
    {
      "epoch": 5.52,
      "learning_rate": 8.04696449026346e-06,
      "loss": 0.8729,
      "step": 1070
    },
    {
      "epoch": 5.57,
      "learning_rate": 8.018327605956472e-06,
      "loss": 0.8548,
      "step": 1080
    },
    {
      "epoch": 5.62,
      "learning_rate": 7.989690721649486e-06,
      "loss": 0.8519,
      "step": 1090
    },
    {
      "epoch": 5.67,
      "learning_rate": 7.961053837342498e-06,
      "loss": 0.869,
      "step": 1100
    },
    {
      "epoch": 5.72,
      "learning_rate": 7.93241695303551e-06,
      "loss": 0.8671,
      "step": 1110
    },
    {
      "epoch": 5.77,
      "learning_rate": 7.903780068728523e-06,
      "loss": 0.8645,
      "step": 1120
    },
    {
      "epoch": 5.82,
      "learning_rate": 7.875143184421535e-06,
      "loss": 0.8585,
      "step": 1130
    },
    {
      "epoch": 5.88,
      "learning_rate": 7.846506300114547e-06,
      "loss": 0.8622,
      "step": 1140
    },
    {
      "epoch": 5.93,
      "learning_rate": 7.817869415807561e-06,
      "loss": 0.8516,
      "step": 1150
    },
    {
      "epoch": 5.98,
      "learning_rate": 7.789232531500573e-06,
      "loss": 0.8283,
      "step": 1160
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6936472105772331,
      "eval_f1": 0.6905055219394208,
      "eval_loss": 0.8881023526191711,
      "eval_precision": 0.6928623941074218,
      "eval_recall": 0.6936472105772331,
      "eval_runtime": 139.4491,
      "eval_samples_per_second": 22.238,
      "eval_steps_per_second": 0.696,
      "step": 1164
    },
    {
      "epoch": 6.03,
      "learning_rate": 7.760595647193587e-06,
      "loss": 0.8177,
      "step": 1170
    },
    {
      "epoch": 6.08,
      "learning_rate": 7.731958762886599e-06,
      "loss": 0.8277,
      "step": 1180
    },
    {
      "epoch": 6.13,
      "learning_rate": 7.703321878579612e-06,
      "loss": 0.7713,
      "step": 1190
    },
    {
      "epoch": 6.19,
      "learning_rate": 7.674684994272624e-06,
      "loss": 0.8008,
      "step": 1200
    },
    {
      "epoch": 6.24,
      "learning_rate": 7.646048109965636e-06,
      "loss": 0.823,
      "step": 1210
    },
    {
      "epoch": 6.29,
      "learning_rate": 7.617411225658648e-06,
      "loss": 0.8256,
      "step": 1220
    },
    {
      "epoch": 6.34,
      "learning_rate": 7.588774341351661e-06,
      "loss": 0.8161,
      "step": 1230
    },
    {
      "epoch": 6.39,
      "learning_rate": 7.560137457044674e-06,
      "loss": 0.7932,
      "step": 1240
    },
    {
      "epoch": 6.44,
      "learning_rate": 7.531500572737687e-06,
      "loss": 0.829,
      "step": 1250
    },
    {
      "epoch": 6.49,
      "learning_rate": 7.5028636884306995e-06,
      "loss": 0.853,
      "step": 1260
    },
    {
      "epoch": 6.55,
      "learning_rate": 7.474226804123712e-06,
      "loss": 0.8209,
      "step": 1270
    },
    {
      "epoch": 6.6,
      "learning_rate": 7.445589919816725e-06,
      "loss": 0.8186,
      "step": 1280
    },
    {
      "epoch": 6.65,
      "learning_rate": 7.416953035509737e-06,
      "loss": 0.7944,
      "step": 1290
    },
    {
      "epoch": 6.7,
      "learning_rate": 7.38831615120275e-06,
      "loss": 0.7968,
      "step": 1300
    },
    {
      "epoch": 6.75,
      "learning_rate": 7.359679266895763e-06,
      "loss": 0.7907,
      "step": 1310
    },
    {
      "epoch": 6.8,
      "learning_rate": 7.331042382588775e-06,
      "loss": 0.7876,
      "step": 1320
    },
    {
      "epoch": 6.86,
      "learning_rate": 7.3024054982817875e-06,
      "loss": 0.8202,
      "step": 1330
    },
    {
      "epoch": 6.91,
      "learning_rate": 7.2737686139747995e-06,
      "loss": 0.7491,
      "step": 1340
    },
    {
      "epoch": 6.96,
      "learning_rate": 7.245131729667812e-06,
      "loss": 0.8185,
      "step": 1350
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6981618832634634,
      "eval_f1": 0.6987690565427545,
      "eval_loss": 0.8659195303916931,
      "eval_precision": 0.7011005120637903,
      "eval_recall": 0.6981618832634634,
      "eval_runtime": 137.6906,
      "eval_samples_per_second": 22.522,
      "eval_steps_per_second": 0.704,
      "step": 1358
    },
    {
      "epoch": 7.01,
      "learning_rate": 7.216494845360825e-06,
      "loss": 0.783,
      "step": 1360
    },
    {
      "epoch": 7.06,
      "learning_rate": 7.187857961053838e-06,
      "loss": 0.7655,
      "step": 1370
    },
    {
      "epoch": 7.11,
      "learning_rate": 7.159221076746851e-06,
      "loss": 0.7166,
      "step": 1380
    },
    {
      "epoch": 7.16,
      "learning_rate": 7.130584192439864e-06,
      "loss": 0.7497,
      "step": 1390
    },
    {
      "epoch": 7.22,
      "learning_rate": 7.101947308132876e-06,
      "loss": 0.7692,
      "step": 1400
    },
    {
      "epoch": 7.27,
      "learning_rate": 7.073310423825888e-06,
      "loss": 0.7567,
      "step": 1410
    },
    {
      "epoch": 7.32,
      "learning_rate": 7.044673539518901e-06,
      "loss": 0.8078,
      "step": 1420
    },
    {
      "epoch": 7.37,
      "learning_rate": 7.016036655211913e-06,
      "loss": 0.7567,
      "step": 1430
    },
    {
      "epoch": 7.42,
      "learning_rate": 6.987399770904926e-06,
      "loss": 0.7553,
      "step": 1440
    },
    {
      "epoch": 7.47,
      "learning_rate": 6.958762886597939e-06,
      "loss": 0.7315,
      "step": 1450
    },
    {
      "epoch": 7.53,
      "learning_rate": 6.930126002290951e-06,
      "loss": 0.7585,
      "step": 1460
    },
    {
      "epoch": 7.58,
      "learning_rate": 6.901489117983964e-06,
      "loss": 0.7772,
      "step": 1470
    },
    {
      "epoch": 7.63,
      "learning_rate": 6.872852233676976e-06,
      "loss": 0.7388,
      "step": 1480
    },
    {
      "epoch": 7.68,
      "learning_rate": 6.844215349369989e-06,
      "loss": 0.7477,
      "step": 1490
    },
    {
      "epoch": 7.73,
      "learning_rate": 6.815578465063002e-06,
      "loss": 0.7693,
      "step": 1500
    },
    {
      "epoch": 7.78,
      "learning_rate": 6.786941580756015e-06,
      "loss": 0.7444,
      "step": 1510
    },
    {
      "epoch": 7.84,
      "learning_rate": 6.758304696449028e-06,
      "loss": 0.8008,
      "step": 1520
    },
    {
      "epoch": 7.89,
      "learning_rate": 6.72966781214204e-06,
      "loss": 0.7655,
      "step": 1530
    },
    {
      "epoch": 7.94,
      "learning_rate": 6.701030927835052e-06,
      "loss": 0.8314,
      "step": 1540
    },
    {
      "epoch": 7.99,
      "learning_rate": 6.672394043528064e-06,
      "loss": 0.7499,
      "step": 1550
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7046114156723637,
      "eval_f1": 0.7021460447498534,
      "eval_loss": 0.8557855486869812,
      "eval_precision": 0.7050032399049077,
      "eval_recall": 0.7046114156723637,
      "eval_runtime": 138.2055,
      "eval_samples_per_second": 22.438,
      "eval_steps_per_second": 0.702,
      "step": 1552
    },
    {
      "epoch": 8.04,
      "learning_rate": 6.643757159221077e-06,
      "loss": 0.7519,
      "step": 1560
    },
    {
      "epoch": 8.09,
      "learning_rate": 6.61512027491409e-06,
      "loss": 0.719,
      "step": 1570
    },
    {
      "epoch": 8.14,
      "learning_rate": 6.586483390607102e-06,
      "loss": 0.7143,
      "step": 1580
    },
    {
      "epoch": 8.2,
      "learning_rate": 6.557846506300115e-06,
      "loss": 0.7502,
      "step": 1590
    },
    {
      "epoch": 8.25,
      "learning_rate": 6.529209621993128e-06,
      "loss": 0.753,
      "step": 1600
    },
    {
      "epoch": 8.3,
      "learning_rate": 6.5005727376861405e-06,
      "loss": 0.7136,
      "step": 1610
    },
    {
      "epoch": 8.35,
      "learning_rate": 6.471935853379153e-06,
      "loss": 0.7153,
      "step": 1620
    },
    {
      "epoch": 8.4,
      "learning_rate": 6.443298969072166e-06,
      "loss": 0.7353,
      "step": 1630
    },
    {
      "epoch": 8.45,
      "learning_rate": 6.414662084765179e-06,
      "loss": 0.7612,
      "step": 1640
    },
    {
      "epoch": 8.51,
      "learning_rate": 6.38602520045819e-06,
      "loss": 0.7023,
      "step": 1650
    },
    {
      "epoch": 8.56,
      "learning_rate": 6.357388316151203e-06,
      "loss": 0.7415,
      "step": 1660
    },
    {
      "epoch": 8.61,
      "learning_rate": 6.328751431844216e-06,
      "loss": 0.7073,
      "step": 1670
    },
    {
      "epoch": 8.66,
      "learning_rate": 6.3001145475372285e-06,
      "loss": 0.7085,
      "step": 1680
    },
    {
      "epoch": 8.71,
      "learning_rate": 6.271477663230241e-06,
      "loss": 0.7469,
      "step": 1690
    },
    {
      "epoch": 8.76,
      "learning_rate": 6.242840778923253e-06,
      "loss": 0.7371,
      "step": 1700
    },
    {
      "epoch": 8.81,
      "learning_rate": 6.214203894616266e-06,
      "loss": 0.7228,
      "step": 1710
    },
    {
      "epoch": 8.87,
      "learning_rate": 6.185567010309279e-06,
      "loss": 0.7249,
      "step": 1720
    },
    {
      "epoch": 8.92,
      "learning_rate": 6.156930126002292e-06,
      "loss": 0.7462,
      "step": 1730
    },
    {
      "epoch": 8.97,
      "learning_rate": 6.1282932416953046e-06,
      "loss": 0.7219,
      "step": 1740
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7123508545630441,
      "eval_f1": 0.7127491151695062,
      "eval_loss": 0.8398727774620056,
      "eval_precision": 0.7164935273335432,
      "eval_recall": 0.7123508545630441,
      "eval_runtime": 138.5399,
      "eval_samples_per_second": 22.383,
      "eval_steps_per_second": 0.7,
      "step": 1746
    },
    {
      "epoch": 9.02,
      "learning_rate": 6.099656357388317e-06,
      "loss": 0.7051,
      "step": 1750
    },
    {
      "epoch": 9.07,
      "learning_rate": 6.0710194730813285e-06,
      "loss": 0.7038,
      "step": 1760
    },
    {
      "epoch": 9.12,
      "learning_rate": 6.042382588774341e-06,
      "loss": 0.6517,
      "step": 1770
    },
    {
      "epoch": 9.18,
      "learning_rate": 6.013745704467354e-06,
      "loss": 0.7214,
      "step": 1780
    },
    {
      "epoch": 9.23,
      "learning_rate": 5.985108820160367e-06,
      "loss": 0.698,
      "step": 1790
    },
    {
      "epoch": 9.28,
      "learning_rate": 5.95647193585338e-06,
      "loss": 0.6854,
      "step": 1800
    },
    {
      "epoch": 9.33,
      "learning_rate": 5.927835051546393e-06,
      "loss": 0.7,
      "step": 1810
    },
    {
      "epoch": 9.38,
      "learning_rate": 5.8991981672394046e-06,
      "loss": 0.684,
      "step": 1820
    },
    {
      "epoch": 9.43,
      "learning_rate": 5.870561282932417e-06,
      "loss": 0.693,
      "step": 1830
    },
    {
      "epoch": 9.48,
      "learning_rate": 5.84192439862543e-06,
      "loss": 0.6796,
      "step": 1840
    },
    {
      "epoch": 9.54,
      "learning_rate": 5.813287514318443e-06,
      "loss": 0.7016,
      "step": 1850
    },
    {
      "epoch": 9.59,
      "learning_rate": 5.784650630011456e-06,
      "loss": 0.7032,
      "step": 1860
    },
    {
      "epoch": 9.64,
      "learning_rate": 5.756013745704467e-06,
      "loss": 0.7215,
      "step": 1870
    },
    {
      "epoch": 9.69,
      "learning_rate": 5.72737686139748e-06,
      "loss": 0.7012,
      "step": 1880
    },
    {
      "epoch": 9.74,
      "learning_rate": 5.698739977090493e-06,
      "loss": 0.68,
      "step": 1890
    },
    {
      "epoch": 9.79,
      "learning_rate": 5.670103092783505e-06,
      "loss": 0.6938,
      "step": 1900
    },
    {
      "epoch": 9.85,
      "learning_rate": 5.641466208476518e-06,
      "loss": 0.6876,
      "step": 1910
    },
    {
      "epoch": 9.9,
      "learning_rate": 5.612829324169531e-06,
      "loss": 0.7367,
      "step": 1920
    },
    {
      "epoch": 9.95,
      "learning_rate": 5.584192439862544e-06,
      "loss": 0.6787,
      "step": 1930
    },
    {
      "epoch": 10.0,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.7382,
      "step": 1940
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7158980973879394,
      "eval_f1": 0.7145145595526569,
      "eval_loss": 0.8300238251686096,
      "eval_precision": 0.7183831182263194,
      "eval_recall": 0.7158980973879394,
      "eval_runtime": 142.0327,
      "eval_samples_per_second": 21.833,
      "eval_steps_per_second": 0.683,
      "step": 1940
    },
    {
      "epoch": 10.05,
      "learning_rate": 5.526918671248569e-06,
      "loss": 0.6854,
      "step": 1950
    },
    {
      "epoch": 10.1,
      "learning_rate": 5.4982817869415815e-06,
      "loss": 0.7064,
      "step": 1960
    },
    {
      "epoch": 10.15,
      "learning_rate": 5.469644902634594e-06,
      "loss": 0.6548,
      "step": 1970
    },
    {
      "epoch": 10.21,
      "learning_rate": 5.441008018327606e-06,
      "loss": 0.6742,
      "step": 1980
    },
    {
      "epoch": 10.26,
      "learning_rate": 5.412371134020619e-06,
      "loss": 0.6549,
      "step": 1990
    },
    {
      "epoch": 10.31,
      "learning_rate": 5.383734249713631e-06,
      "loss": 0.6867,
      "step": 2000
    },
    {
      "epoch": 10.36,
      "learning_rate": 5.355097365406644e-06,
      "loss": 0.6595,
      "step": 2010
    },
    {
      "epoch": 10.41,
      "learning_rate": 5.326460481099657e-06,
      "loss": 0.6721,
      "step": 2020
    },
    {
      "epoch": 10.46,
      "learning_rate": 5.2978235967926695e-06,
      "loss": 0.6652,
      "step": 2030
    },
    {
      "epoch": 10.52,
      "learning_rate": 5.269186712485682e-06,
      "loss": 0.6689,
      "step": 2040
    },
    {
      "epoch": 10.57,
      "learning_rate": 5.240549828178695e-06,
      "loss": 0.6372,
      "step": 2050
    },
    {
      "epoch": 10.62,
      "learning_rate": 5.211912943871708e-06,
      "loss": 0.7036,
      "step": 2060
    },
    {
      "epoch": 10.67,
      "learning_rate": 5.18327605956472e-06,
      "loss": 0.6809,
      "step": 2070
    },
    {
      "epoch": 10.72,
      "learning_rate": 5.154639175257732e-06,
      "loss": 0.7015,
      "step": 2080
    },
    {
      "epoch": 10.77,
      "learning_rate": 5.126002290950745e-06,
      "loss": 0.6779,
      "step": 2090
    },
    {
      "epoch": 10.82,
      "learning_rate": 5.0973654066437575e-06,
      "loss": 0.6516,
      "step": 2100
    },
    {
      "epoch": 10.88,
      "learning_rate": 5.06872852233677e-06,
      "loss": 0.6445,
      "step": 2110
    },
    {
      "epoch": 10.93,
      "learning_rate": 5.040091638029782e-06,
      "loss": 0.6939,
      "step": 2120
    },
    {
      "epoch": 10.98,
      "learning_rate": 5.011454753722795e-06,
      "loss": 0.6392,
      "step": 2130
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.708803611738149,
      "eval_f1": 0.7094972389297926,
      "eval_loss": 0.8328733444213867,
      "eval_precision": 0.7134851917714098,
      "eval_recall": 0.708803611738149,
      "eval_runtime": 136.2862,
      "eval_samples_per_second": 22.754,
      "eval_steps_per_second": 0.712,
      "step": 2134
    },
    {
      "epoch": 11.03,
      "learning_rate": 4.982817869415808e-06,
      "loss": 0.6682,
      "step": 2140
    },
    {
      "epoch": 11.08,
      "learning_rate": 4.954180985108821e-06,
      "loss": 0.6482,
      "step": 2150
    },
    {
      "epoch": 11.13,
      "learning_rate": 4.9255441008018336e-06,
      "loss": 0.6229,
      "step": 2160
    },
    {
      "epoch": 11.19,
      "learning_rate": 4.8969072164948455e-06,
      "loss": 0.6611,
      "step": 2170
    },
    {
      "epoch": 11.24,
      "learning_rate": 4.868270332187858e-06,
      "loss": 0.6164,
      "step": 2180
    },
    {
      "epoch": 11.29,
      "learning_rate": 4.839633447880871e-06,
      "loss": 0.6623,
      "step": 2190
    },
    {
      "epoch": 11.34,
      "learning_rate": 4.810996563573884e-06,
      "loss": 0.6415,
      "step": 2200
    },
    {
      "epoch": 11.39,
      "learning_rate": 4.782359679266896e-06,
      "loss": 0.6349,
      "step": 2210
    },
    {
      "epoch": 11.44,
      "learning_rate": 4.753722794959909e-06,
      "loss": 0.6506,
      "step": 2220
    },
    {
      "epoch": 11.49,
      "learning_rate": 4.725085910652922e-06,
      "loss": 0.6366,
      "step": 2230
    },
    {
      "epoch": 11.55,
      "learning_rate": 4.6964490263459336e-06,
      "loss": 0.6533,
      "step": 2240
    },
    {
      "epoch": 11.6,
      "learning_rate": 4.667812142038946e-06,
      "loss": 0.6704,
      "step": 2250
    },
    {
      "epoch": 11.65,
      "learning_rate": 4.639175257731959e-06,
      "loss": 0.6482,
      "step": 2260
    },
    {
      "epoch": 11.7,
      "learning_rate": 4.610538373424972e-06,
      "loss": 0.5899,
      "step": 2270
    },
    {
      "epoch": 11.75,
      "learning_rate": 4.581901489117984e-06,
      "loss": 0.6463,
      "step": 2280
    },
    {
      "epoch": 11.8,
      "learning_rate": 4.553264604810997e-06,
      "loss": 0.6575,
      "step": 2290
    },
    {
      "epoch": 11.86,
      "learning_rate": 4.52462772050401e-06,
      "loss": 0.6422,
      "step": 2300
    },
    {
      "epoch": 11.91,
      "learning_rate": 4.4959908361970224e-06,
      "loss": 0.6736,
      "step": 2310
    },
    {
      "epoch": 11.96,
      "learning_rate": 4.467353951890035e-06,
      "loss": 0.6549,
      "step": 2320
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7133182844243793,
      "eval_f1": 0.711971866511754,
      "eval_loss": 0.8296553492546082,
      "eval_precision": 0.7134785190322149,
      "eval_recall": 0.7133182844243793,
      "eval_runtime": 142.0833,
      "eval_samples_per_second": 21.825,
      "eval_steps_per_second": 0.683,
      "step": 2328
    },
    {
      "epoch": 12.01,
      "learning_rate": 4.438717067583047e-06,
      "loss": 0.5999,
      "step": 2330
    },
    {
      "epoch": 12.06,
      "learning_rate": 4.41008018327606e-06,
      "loss": 0.6216,
      "step": 2340
    },
    {
      "epoch": 12.11,
      "learning_rate": 4.381443298969073e-06,
      "loss": 0.6263,
      "step": 2350
    },
    {
      "epoch": 12.16,
      "learning_rate": 4.352806414662085e-06,
      "loss": 0.6119,
      "step": 2360
    },
    {
      "epoch": 12.22,
      "learning_rate": 4.324169530355098e-06,
      "loss": 0.6132,
      "step": 2370
    },
    {
      "epoch": 12.27,
      "learning_rate": 4.2955326460481105e-06,
      "loss": 0.6269,
      "step": 2380
    },
    {
      "epoch": 12.32,
      "learning_rate": 4.2668957617411224e-06,
      "loss": 0.6457,
      "step": 2390
    },
    {
      "epoch": 12.37,
      "learning_rate": 4.238258877434135e-06,
      "loss": 0.6278,
      "step": 2400
    },
    {
      "epoch": 12.42,
      "learning_rate": 4.209621993127148e-06,
      "loss": 0.6119,
      "step": 2410
    },
    {
      "epoch": 12.47,
      "learning_rate": 4.180985108820161e-06,
      "loss": 0.6198,
      "step": 2420
    },
    {
      "epoch": 12.53,
      "learning_rate": 4.152348224513174e-06,
      "loss": 0.6232,
      "step": 2430
    },
    {
      "epoch": 12.58,
      "learning_rate": 4.123711340206186e-06,
      "loss": 0.6215,
      "step": 2440
    },
    {
      "epoch": 12.63,
      "learning_rate": 4.0950744558991985e-06,
      "loss": 0.644,
      "step": 2450
    },
    {
      "epoch": 12.68,
      "learning_rate": 4.066437571592211e-06,
      "loss": 0.6071,
      "step": 2460
    },
    {
      "epoch": 12.73,
      "learning_rate": 4.037800687285224e-06,
      "loss": 0.6035,
      "step": 2470
    },
    {
      "epoch": 12.78,
      "learning_rate": 4.009163802978236e-06,
      "loss": 0.6482,
      "step": 2480
    },
    {
      "epoch": 12.84,
      "learning_rate": 3.980526918671249e-06,
      "loss": 0.6152,
      "step": 2490
    },
    {
      "epoch": 12.89,
      "learning_rate": 3.951890034364262e-06,
      "loss": 0.6212,
      "step": 2500
    },
    {
      "epoch": 12.94,
      "learning_rate": 3.923253150057274e-06,
      "loss": 0.6033,
      "step": 2510
    },
    {
      "epoch": 12.99,
      "learning_rate": 3.8946162657502865e-06,
      "loss": 0.6762,
      "step": 2520
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7155756207674944,
      "eval_f1": 0.7152962310825538,
      "eval_loss": 0.8179510235786438,
      "eval_precision": 0.7161790866733511,
      "eval_recall": 0.7155756207674944,
      "eval_runtime": 138.6525,
      "eval_samples_per_second": 22.365,
      "eval_steps_per_second": 0.7,
      "step": 2522
    },
    {
      "epoch": 13.04,
      "learning_rate": 3.865979381443299e-06,
      "loss": 0.5963,
      "step": 2530
    },
    {
      "epoch": 13.09,
      "learning_rate": 3.837342497136312e-06,
      "loss": 0.6163,
      "step": 2540
    },
    {
      "epoch": 13.14,
      "learning_rate": 3.808705612829324e-06,
      "loss": 0.6262,
      "step": 2550
    },
    {
      "epoch": 13.2,
      "learning_rate": 3.780068728522337e-06,
      "loss": 0.6166,
      "step": 2560
    },
    {
      "epoch": 13.25,
      "learning_rate": 3.7514318442153498e-06,
      "loss": 0.62,
      "step": 2570
    },
    {
      "epoch": 13.3,
      "learning_rate": 3.7227949599083626e-06,
      "loss": 0.5676,
      "step": 2580
    },
    {
      "epoch": 13.35,
      "learning_rate": 3.694158075601375e-06,
      "loss": 0.6093,
      "step": 2590
    },
    {
      "epoch": 13.4,
      "learning_rate": 3.6655211912943874e-06,
      "loss": 0.5997,
      "step": 2600
    },
    {
      "epoch": 13.45,
      "learning_rate": 3.6368843069873997e-06,
      "loss": 0.619,
      "step": 2610
    },
    {
      "epoch": 13.51,
      "learning_rate": 3.6082474226804126e-06,
      "loss": 0.6019,
      "step": 2620
    },
    {
      "epoch": 13.56,
      "learning_rate": 3.5796105383734254e-06,
      "loss": 0.6117,
      "step": 2630
    },
    {
      "epoch": 13.61,
      "learning_rate": 3.550973654066438e-06,
      "loss": 0.5879,
      "step": 2640
    },
    {
      "epoch": 13.66,
      "learning_rate": 3.5223367697594506e-06,
      "loss": 0.581,
      "step": 2650
    },
    {
      "epoch": 13.71,
      "learning_rate": 3.493699885452463e-06,
      "loss": 0.633,
      "step": 2660
    },
    {
      "epoch": 13.76,
      "learning_rate": 3.4650630011454754e-06,
      "loss": 0.6022,
      "step": 2670
    },
    {
      "epoch": 13.81,
      "learning_rate": 3.436426116838488e-06,
      "loss": 0.602,
      "step": 2680
    },
    {
      "epoch": 13.87,
      "learning_rate": 3.407789232531501e-06,
      "loss": 0.5738,
      "step": 2690
    },
    {
      "epoch": 13.92,
      "learning_rate": 3.379152348224514e-06,
      "loss": 0.6074,
      "step": 2700
    },
    {
      "epoch": 13.97,
      "learning_rate": 3.350515463917526e-06,
      "loss": 0.5937,
      "step": 2710
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7188003869719445,
      "eval_f1": 0.718989971086903,
      "eval_loss": 0.8270788192749023,
      "eval_precision": 0.7219837313936599,
      "eval_recall": 0.7188003869719445,
      "eval_runtime": 137.3251,
      "eval_samples_per_second": 22.581,
      "eval_steps_per_second": 0.706,
      "step": 2716
    },
    {
      "epoch": 14.02,
      "learning_rate": 3.3218785796105386e-06,
      "loss": 0.5785,
      "step": 2720
    },
    {
      "epoch": 14.07,
      "learning_rate": 3.293241695303551e-06,
      "loss": 0.5945,
      "step": 2730
    },
    {
      "epoch": 14.12,
      "learning_rate": 3.264604810996564e-06,
      "loss": 0.5927,
      "step": 2740
    },
    {
      "epoch": 14.18,
      "learning_rate": 3.2359679266895766e-06,
      "loss": 0.6038,
      "step": 2750
    },
    {
      "epoch": 14.23,
      "learning_rate": 3.2073310423825895e-06,
      "loss": 0.5904,
      "step": 2760
    },
    {
      "epoch": 14.28,
      "learning_rate": 3.1786941580756014e-06,
      "loss": 0.5885,
      "step": 2770
    },
    {
      "epoch": 14.33,
      "learning_rate": 3.1500572737686143e-06,
      "loss": 0.5421,
      "step": 2780
    },
    {
      "epoch": 14.38,
      "learning_rate": 3.1214203894616266e-06,
      "loss": 0.5882,
      "step": 2790
    },
    {
      "epoch": 14.43,
      "learning_rate": 3.0927835051546395e-06,
      "loss": 0.6144,
      "step": 2800
    },
    {
      "epoch": 14.48,
      "learning_rate": 3.0641466208476523e-06,
      "loss": 0.6108,
      "step": 2810
    },
    {
      "epoch": 14.54,
      "learning_rate": 3.0355097365406643e-06,
      "loss": 0.5852,
      "step": 2820
    },
    {
      "epoch": 14.59,
      "learning_rate": 3.006872852233677e-06,
      "loss": 0.5534,
      "step": 2830
    },
    {
      "epoch": 14.64,
      "learning_rate": 2.97823596792669e-06,
      "loss": 0.583,
      "step": 2840
    },
    {
      "epoch": 14.69,
      "learning_rate": 2.9495990836197023e-06,
      "loss": 0.5909,
      "step": 2850
    },
    {
      "epoch": 14.74,
      "learning_rate": 2.920962199312715e-06,
      "loss": 0.5972,
      "step": 2860
    },
    {
      "epoch": 14.79,
      "learning_rate": 2.892325315005728e-06,
      "loss": 0.5697,
      "step": 2870
    },
    {
      "epoch": 14.85,
      "learning_rate": 2.86368843069874e-06,
      "loss": 0.5914,
      "step": 2880
    },
    {
      "epoch": 14.9,
      "learning_rate": 2.8350515463917527e-06,
      "loss": 0.6051,
      "step": 2890
    },
    {
      "epoch": 14.95,
      "learning_rate": 2.8064146620847655e-06,
      "loss": 0.566,
      "step": 2900
    },
    {
      "epoch": 15.0,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.569,
      "step": 2910
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7178329571106095,
      "eval_f1": 0.7165091349937845,
      "eval_loss": 0.8245362043380737,
      "eval_precision": 0.7174763240731499,
      "eval_recall": 0.7178329571106095,
      "eval_runtime": 137.9864,
      "eval_samples_per_second": 22.473,
      "eval_steps_per_second": 0.703,
      "step": 2910
    },
    {
      "epoch": 15.05,
      "learning_rate": 2.7491408934707907e-06,
      "loss": 0.5732,
      "step": 2920
    },
    {
      "epoch": 15.1,
      "learning_rate": 2.720504009163803e-06,
      "loss": 0.6155,
      "step": 2930
    },
    {
      "epoch": 15.15,
      "learning_rate": 2.6918671248568155e-06,
      "loss": 0.5725,
      "step": 2940
    },
    {
      "epoch": 15.21,
      "learning_rate": 2.6632302405498283e-06,
      "loss": 0.57,
      "step": 2950
    },
    {
      "epoch": 15.26,
      "learning_rate": 2.634593356242841e-06,
      "loss": 0.5978,
      "step": 2960
    },
    {
      "epoch": 15.31,
      "learning_rate": 2.605956471935854e-06,
      "loss": 0.5634,
      "step": 2970
    },
    {
      "epoch": 15.36,
      "learning_rate": 2.577319587628866e-06,
      "loss": 0.5842,
      "step": 2980
    },
    {
      "epoch": 15.41,
      "learning_rate": 2.5486827033218788e-06,
      "loss": 0.5699,
      "step": 2990
    },
    {
      "epoch": 15.46,
      "learning_rate": 2.520045819014891e-06,
      "loss": 0.576,
      "step": 3000
    },
    {
      "epoch": 15.52,
      "learning_rate": 2.491408934707904e-06,
      "loss": 0.5578,
      "step": 3010
    },
    {
      "epoch": 15.57,
      "learning_rate": 2.4627720504009168e-06,
      "loss": 0.5726,
      "step": 3020
    },
    {
      "epoch": 15.62,
      "learning_rate": 2.434135166093929e-06,
      "loss": 0.5669,
      "step": 3030
    },
    {
      "epoch": 15.67,
      "learning_rate": 2.405498281786942e-06,
      "loss": 0.5945,
      "step": 3040
    },
    {
      "epoch": 15.72,
      "learning_rate": 2.3768613974799544e-06,
      "loss": 0.5548,
      "step": 3050
    },
    {
      "epoch": 15.77,
      "learning_rate": 2.3482245131729668e-06,
      "loss": 0.5442,
      "step": 3060
    },
    {
      "epoch": 15.82,
      "learning_rate": 2.3195876288659796e-06,
      "loss": 0.5575,
      "step": 3070
    },
    {
      "epoch": 15.88,
      "learning_rate": 2.290950744558992e-06,
      "loss": 0.5429,
      "step": 3080
    },
    {
      "epoch": 15.93,
      "learning_rate": 2.262313860252005e-06,
      "loss": 0.5476,
      "step": 3090
    },
    {
      "epoch": 15.98,
      "learning_rate": 2.2336769759450176e-06,
      "loss": 0.5623,
      "step": 3100
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7165430506288294,
      "eval_f1": 0.7156652378724117,
      "eval_loss": 0.8228491544723511,
      "eval_precision": 0.7153021143244204,
      "eval_recall": 0.7165430506288294,
      "eval_runtime": 137.8917,
      "eval_samples_per_second": 22.489,
      "eval_steps_per_second": 0.703,
      "step": 3104
    },
    {
      "epoch": 16.03,
      "learning_rate": 2.20504009163803e-06,
      "loss": 0.5951,
      "step": 3110
    },
    {
      "epoch": 16.08,
      "learning_rate": 2.1764032073310424e-06,
      "loss": 0.5594,
      "step": 3120
    },
    {
      "epoch": 16.13,
      "learning_rate": 2.1477663230240552e-06,
      "loss": 0.6058,
      "step": 3130
    },
    {
      "epoch": 16.19,
      "learning_rate": 2.1191294387170676e-06,
      "loss": 0.5372,
      "step": 3140
    },
    {
      "epoch": 16.24,
      "learning_rate": 2.0904925544100804e-06,
      "loss": 0.566,
      "step": 3150
    },
    {
      "epoch": 16.29,
      "learning_rate": 2.061855670103093e-06,
      "loss": 0.5524,
      "step": 3160
    },
    {
      "epoch": 16.34,
      "learning_rate": 2.0332187857961057e-06,
      "loss": 0.527,
      "step": 3170
    },
    {
      "epoch": 16.39,
      "learning_rate": 2.004581901489118e-06,
      "loss": 0.5995,
      "step": 3180
    },
    {
      "epoch": 16.44,
      "learning_rate": 1.975945017182131e-06,
      "loss": 0.5271,
      "step": 3190
    },
    {
      "epoch": 16.49,
      "learning_rate": 1.9473081328751433e-06,
      "loss": 0.5892,
      "step": 3200
    },
    {
      "epoch": 16.55,
      "learning_rate": 1.918671248568156e-06,
      "loss": 0.5539,
      "step": 3210
    },
    {
      "epoch": 16.6,
      "learning_rate": 1.8900343642611685e-06,
      "loss": 0.551,
      "step": 3220
    },
    {
      "epoch": 16.65,
      "learning_rate": 1.8613974799541813e-06,
      "loss": 0.5638,
      "step": 3230
    },
    {
      "epoch": 16.7,
      "learning_rate": 1.8327605956471937e-06,
      "loss": 0.5533,
      "step": 3240
    },
    {
      "epoch": 16.75,
      "learning_rate": 1.8041237113402063e-06,
      "loss": 0.5886,
      "step": 3250
    },
    {
      "epoch": 16.8,
      "learning_rate": 1.775486827033219e-06,
      "loss": 0.5325,
      "step": 3260
    },
    {
      "epoch": 16.86,
      "learning_rate": 1.7468499427262315e-06,
      "loss": 0.5441,
      "step": 3270
    },
    {
      "epoch": 16.91,
      "learning_rate": 1.718213058419244e-06,
      "loss": 0.5422,
      "step": 3280
    },
    {
      "epoch": 16.96,
      "learning_rate": 1.689576174112257e-06,
      "loss": 0.5291,
      "step": 3290
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.7162205740083843,
      "eval_f1": 0.7155771742857344,
      "eval_loss": 0.8238459825515747,
      "eval_precision": 0.7165407692005947,
      "eval_recall": 0.7162205740083843,
      "eval_runtime": 138.7789,
      "eval_samples_per_second": 22.345,
      "eval_steps_per_second": 0.699,
      "step": 3298
    },
    {
      "epoch": 17.01,
      "learning_rate": 1.6609392898052693e-06,
      "loss": 0.544,
      "step": 3300
    },
    {
      "epoch": 17.06,
      "learning_rate": 1.632302405498282e-06,
      "loss": 0.5365,
      "step": 3310
    },
    {
      "epoch": 17.11,
      "learning_rate": 1.6036655211912947e-06,
      "loss": 0.5507,
      "step": 3320
    },
    {
      "epoch": 17.16,
      "learning_rate": 1.5750286368843071e-06,
      "loss": 0.5537,
      "step": 3330
    },
    {
      "epoch": 17.22,
      "learning_rate": 1.5463917525773197e-06,
      "loss": 0.5241,
      "step": 3340
    },
    {
      "epoch": 17.27,
      "learning_rate": 1.5177548682703321e-06,
      "loss": 0.5106,
      "step": 3350
    },
    {
      "epoch": 17.32,
      "learning_rate": 1.489117983963345e-06,
      "loss": 0.5773,
      "step": 3360
    },
    {
      "epoch": 17.37,
      "learning_rate": 1.4604810996563575e-06,
      "loss": 0.5466,
      "step": 3370
    },
    {
      "epoch": 17.42,
      "learning_rate": 1.43184421534937e-06,
      "loss": 0.538,
      "step": 3380
    },
    {
      "epoch": 17.47,
      "learning_rate": 1.4032073310423828e-06,
      "loss": 0.5373,
      "step": 3390
    },
    {
      "epoch": 17.53,
      "learning_rate": 1.3745704467353954e-06,
      "loss": 0.5416,
      "step": 3400
    },
    {
      "epoch": 17.58,
      "learning_rate": 1.3459335624284078e-06,
      "loss": 0.5536,
      "step": 3410
    },
    {
      "epoch": 17.63,
      "learning_rate": 1.3172966781214206e-06,
      "loss": 0.5853,
      "step": 3420
    },
    {
      "epoch": 17.68,
      "learning_rate": 1.288659793814433e-06,
      "loss": 0.5522,
      "step": 3430
    },
    {
      "epoch": 17.73,
      "learning_rate": 1.2600229095074456e-06,
      "loss": 0.5318,
      "step": 3440
    },
    {
      "epoch": 17.78,
      "learning_rate": 1.2313860252004584e-06,
      "loss": 0.553,
      "step": 3450
    },
    {
      "epoch": 17.84,
      "learning_rate": 1.202749140893471e-06,
      "loss": 0.5386,
      "step": 3460
    },
    {
      "epoch": 17.89,
      "learning_rate": 1.1741122565864834e-06,
      "loss": 0.5333,
      "step": 3470
    },
    {
      "epoch": 17.94,
      "learning_rate": 1.145475372279496e-06,
      "loss": 0.568,
      "step": 3480
    },
    {
      "epoch": 17.99,
      "learning_rate": 1.1168384879725088e-06,
      "loss": 0.5775,
      "step": 3490
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7152531441470493,
      "eval_f1": 0.7151086078956284,
      "eval_loss": 0.8245630860328674,
      "eval_precision": 0.7161566827277972,
      "eval_recall": 0.7152531441470493,
      "eval_runtime": 137.143,
      "eval_samples_per_second": 22.611,
      "eval_steps_per_second": 0.707,
      "step": 3492
    },
    {
      "epoch": 18.04,
      "learning_rate": 1.0882016036655212e-06,
      "loss": 0.5582,
      "step": 3500
    },
    {
      "epoch": 18.09,
      "learning_rate": 1.0595647193585338e-06,
      "loss": 0.5404,
      "step": 3510
    },
    {
      "epoch": 18.14,
      "learning_rate": 1.0309278350515464e-06,
      "loss": 0.57,
      "step": 3520
    },
    {
      "epoch": 18.2,
      "learning_rate": 1.002290950744559e-06,
      "loss": 0.5514,
      "step": 3530
    },
    {
      "epoch": 18.25,
      "learning_rate": 9.736540664375716e-07,
      "loss": 0.48,
      "step": 3540
    },
    {
      "epoch": 18.3,
      "learning_rate": 9.450171821305842e-07,
      "loss": 0.4917,
      "step": 3550
    },
    {
      "epoch": 18.35,
      "learning_rate": 9.163802978235968e-07,
      "loss": 0.5253,
      "step": 3560
    },
    {
      "epoch": 18.4,
      "learning_rate": 8.877434135166096e-07,
      "loss": 0.5138,
      "step": 3570
    },
    {
      "epoch": 18.45,
      "learning_rate": 8.59106529209622e-07,
      "loss": 0.5686,
      "step": 3580
    },
    {
      "epoch": 18.51,
      "learning_rate": 8.304696449026347e-07,
      "loss": 0.5669,
      "step": 3590
    },
    {
      "epoch": 18.56,
      "learning_rate": 8.018327605956474e-07,
      "loss": 0.5286,
      "step": 3600
    },
    {
      "epoch": 18.61,
      "learning_rate": 7.731958762886599e-07,
      "loss": 0.5736,
      "step": 3610
    },
    {
      "epoch": 18.66,
      "learning_rate": 7.445589919816725e-07,
      "loss": 0.5202,
      "step": 3620
    },
    {
      "epoch": 18.71,
      "learning_rate": 7.15922107674685e-07,
      "loss": 0.5417,
      "step": 3630
    },
    {
      "epoch": 18.76,
      "learning_rate": 6.872852233676977e-07,
      "loss": 0.5409,
      "step": 3640
    },
    {
      "epoch": 18.81,
      "learning_rate": 6.586483390607103e-07,
      "loss": 0.5737,
      "step": 3650
    },
    {
      "epoch": 18.87,
      "learning_rate": 6.300114547537228e-07,
      "loss": 0.5383,
      "step": 3660
    },
    {
      "epoch": 18.92,
      "learning_rate": 6.013745704467355e-07,
      "loss": 0.5322,
      "step": 3670
    },
    {
      "epoch": 18.97,
      "learning_rate": 5.72737686139748e-07,
      "loss": 0.545,
      "step": 3680
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7178329571106095,
      "eval_f1": 0.7173917212076267,
      "eval_loss": 0.825716495513916,
      "eval_precision": 0.7191676309063934,
      "eval_recall": 0.7178329571106095,
      "eval_runtime": 137.8289,
      "eval_samples_per_second": 22.499,
      "eval_steps_per_second": 0.704,
      "step": 3686
    },
    {
      "epoch": 19.02,
      "learning_rate": 5.441008018327606e-07,
      "loss": 0.567,
      "step": 3690
    },
    {
      "epoch": 19.07,
      "learning_rate": 5.154639175257732e-07,
      "loss": 0.5309,
      "step": 3700
    },
    {
      "epoch": 19.12,
      "learning_rate": 4.868270332187858e-07,
      "loss": 0.548,
      "step": 3710
    },
    {
      "epoch": 19.18,
      "learning_rate": 4.581901489117984e-07,
      "loss": 0.5147,
      "step": 3720
    },
    {
      "epoch": 19.23,
      "learning_rate": 4.29553264604811e-07,
      "loss": 0.4884,
      "step": 3730
    },
    {
      "epoch": 19.28,
      "learning_rate": 4.009163802978237e-07,
      "loss": 0.5347,
      "step": 3740
    },
    {
      "epoch": 19.33,
      "learning_rate": 3.7227949599083624e-07,
      "loss": 0.5193,
      "step": 3750
    },
    {
      "epoch": 19.38,
      "learning_rate": 3.4364261168384884e-07,
      "loss": 0.5337,
      "step": 3760
    },
    {
      "epoch": 19.43,
      "learning_rate": 3.150057273768614e-07,
      "loss": 0.5594,
      "step": 3770
    },
    {
      "epoch": 19.48,
      "learning_rate": 2.86368843069874e-07,
      "loss": 0.531,
      "step": 3780
    },
    {
      "epoch": 19.54,
      "learning_rate": 2.577319587628866e-07,
      "loss": 0.5651,
      "step": 3790
    },
    {
      "epoch": 19.59,
      "learning_rate": 2.290950744558992e-07,
      "loss": 0.4949,
      "step": 3800
    },
    {
      "epoch": 19.64,
      "learning_rate": 2.0045819014891184e-07,
      "loss": 0.5284,
      "step": 3810
    },
    {
      "epoch": 19.69,
      "learning_rate": 1.7182130584192442e-07,
      "loss": 0.5183,
      "step": 3820
    },
    {
      "epoch": 19.74,
      "learning_rate": 1.43184421534937e-07,
      "loss": 0.5412,
      "step": 3830
    },
    {
      "epoch": 19.79,
      "learning_rate": 1.145475372279496e-07,
      "loss": 0.547,
      "step": 3840
    },
    {
      "epoch": 19.85,
      "learning_rate": 8.591065292096221e-08,
      "loss": 0.5675,
      "step": 3850
    },
    {
      "epoch": 19.9,
      "learning_rate": 5.72737686139748e-08,
      "loss": 0.5541,
      "step": 3860
    },
    {
      "epoch": 19.95,
      "learning_rate": 2.86368843069874e-08,
      "loss": 0.5541,
      "step": 3870
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0,
      "loss": 0.5409,
      "step": 3880
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7178329571106095,
      "eval_f1": 0.717743945710896,
      "eval_loss": 0.8245128989219666,
      "eval_precision": 0.7187025517730355,
      "eval_recall": 0.7178329571106095,
      "eval_runtime": 138.2737,
      "eval_samples_per_second": 22.427,
      "eval_steps_per_second": 0.702,
      "step": 3880
    },
    {
      "epoch": 20.0,
      "step": 3880,
      "total_flos": 3.843964371168166e+19,
      "train_loss": 0.8177650722031741,
      "train_runtime": 43638.4535,
      "train_samples_per_second": 11.367,
      "train_steps_per_second": 0.089
    }
  ],
  "max_steps": 3880,
  "num_train_epochs": 20,
  "total_flos": 3.843964371168166e+19,
  "trial_name": null,
  "trial_params": null
}