{
  "best_metric": 1.0360217094421387,
  "best_model_checkpoint": "./res_2/checkpoint-12000",
  "epoch": 3.7685387794796985,
  "global_step": 15500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 2.5e-05,
      "loss": 4.4105,
      "step": 250
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.2648573215918387,
      "eval_f1_score": 0.1461888457719469,
      "eval_loss": 3.3562278747558594,
      "eval_runtime": 104.4517,
      "eval_samples_per_second": 198.953,
      "eval_steps_per_second": 2.078,
      "step": 250
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 2.7264,
      "step": 500
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.4484865983350176,
      "eval_f1_score": 0.36937033203465997,
      "eval_loss": 2.1725332736968994,
      "eval_runtime": 104.6188,
      "eval_samples_per_second": 198.635,
      "eval_steps_per_second": 2.074,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.9216399197592786e-05,
      "loss": 2.0303,
      "step": 750
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.5143640825754294,
      "eval_f1_score": 0.4607925258467065,
      "eval_loss": 1.784591555595398,
      "eval_runtime": 104.5717,
      "eval_samples_per_second": 198.725,
      "eval_steps_per_second": 2.075,
      "step": 750
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.8432798395185555e-05,
      "loss": 1.777,
      "step": 1000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.5497329291179442,
      "eval_f1_score": 0.5086586604390044,
      "eval_loss": 1.6212031841278076,
      "eval_runtime": 104.6796,
      "eval_samples_per_second": 198.52,
      "eval_steps_per_second": 2.073,
      "step": 1000
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.764919759277834e-05,
      "loss": 1.635,
      "step": 1250
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.57292719310909,
      "eval_f1_score": 0.5363731366139944,
      "eval_loss": 1.4894088506698608,
      "eval_runtime": 104.782,
      "eval_samples_per_second": 198.326,
      "eval_steps_per_second": 2.071,
      "step": 1250
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.6865596790371115e-05,
      "loss": 1.5492,
      "step": 1500
    },
    {
      "epoch": 0.36,
      "eval_accuracy": 0.5871228526057456,
      "eval_f1_score": 0.5565777587100311,
      "eval_loss": 1.4159808158874512,
      "eval_runtime": 104.9221,
      "eval_samples_per_second": 198.061,
      "eval_steps_per_second": 2.068,
      "step": 1500
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.608199598796389e-05,
      "loss": 1.4853,
      "step": 1750
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.6005004571483566,
      "eval_f1_score": 0.5743468856693701,
      "eval_loss": 1.3664453029632568,
      "eval_runtime": 105.1103,
      "eval_samples_per_second": 197.707,
      "eval_steps_per_second": 2.064,
      "step": 1750
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.5298395185556675e-05,
      "loss": 1.4178,
      "step": 2000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.6162841056734517,
      "eval_f1_score": 0.5926712498507521,
      "eval_loss": 1.3030893802642822,
      "eval_runtime": 104.7919,
      "eval_samples_per_second": 198.307,
      "eval_steps_per_second": 2.071,
      "step": 2000
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.451479438314945e-05,
      "loss": 1.4012,
      "step": 2250
    },
    {
      "epoch": 0.55,
      "eval_accuracy": 0.609932149559694,
      "eval_f1_score": 0.5892368884639717,
      "eval_loss": 1.3101539611816406,
      "eval_runtime": 105.0064,
      "eval_samples_per_second": 197.902,
      "eval_steps_per_second": 2.067,
      "step": 2250
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.373119358074223e-05,
      "loss": 1.363,
      "step": 2500
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.6175833694239931,
      "eval_f1_score": 0.596718851218216,
      "eval_loss": 1.2853854894638062,
      "eval_runtime": 104.962,
      "eval_samples_per_second": 197.986,
      "eval_steps_per_second": 2.067,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.2947592778335004e-05,
      "loss": 1.3349,
      "step": 2750
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.6279293585486743,
      "eval_f1_score": 0.6094187627596529,
      "eval_loss": 1.230573058128357,
      "eval_runtime": 104.847,
      "eval_samples_per_second": 198.203,
      "eval_steps_per_second": 2.07,
      "step": 2750
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.216399197592779e-05,
      "loss": 1.3324,
      "step": 3000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.6324046003560945,
      "eval_f1_score": 0.6161864856783154,
      "eval_loss": 1.2122113704681396,
      "eval_runtime": 105.0625,
      "eval_samples_per_second": 197.796,
      "eval_steps_per_second": 2.065,
      "step": 3000
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.1380391173520564e-05,
      "loss": 1.292,
      "step": 3250
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.6309128530869544,
      "eval_f1_score": 0.6119500434777825,
      "eval_loss": 1.20195472240448,
      "eval_runtime": 105.1108,
      "eval_samples_per_second": 197.706,
      "eval_steps_per_second": 2.064,
      "step": 3250
    },
    {
      "epoch": 0.85,
      "learning_rate": 4.059679037111334e-05,
      "loss": 1.2693,
      "step": 3500
    },
    {
      "epoch": 0.85,
      "eval_accuracy": 0.6373129300803618,
      "eval_f1_score": 0.624172345328055,
      "eval_loss": 1.1805213689804077,
      "eval_runtime": 104.8144,
      "eval_samples_per_second": 198.265,
      "eval_steps_per_second": 2.07,
      "step": 3500
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.9813189568706124e-05,
      "loss": 1.2729,
      "step": 3750
    },
    {
      "epoch": 0.91,
      "eval_accuracy": 0.6417881718877821,
      "eval_f1_score": 0.6227824246172433,
      "eval_loss": 1.1760473251342773,
      "eval_runtime": 104.7877,
      "eval_samples_per_second": 198.315,
      "eval_steps_per_second": 2.071,
      "step": 3750
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.90295887662989e-05,
      "loss": 1.2382,
      "step": 4000
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.6395746114238968,
      "eval_f1_score": 0.6232000558005116,
      "eval_loss": 1.1657843589782715,
      "eval_runtime": 104.9017,
      "eval_samples_per_second": 198.1,
      "eval_steps_per_second": 2.069,
      "step": 4000
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.8245987963891677e-05,
      "loss": 1.175,
      "step": 4250
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.6491025455945335,
      "eval_f1_score": 0.6333235301999404,
      "eval_loss": 1.1377447843551636,
      "eval_runtime": 104.8132,
      "eval_samples_per_second": 198.267,
      "eval_steps_per_second": 2.07,
      "step": 4250
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.746238716148446e-05,
      "loss": 1.1071,
      "step": 4500
    },
    {
      "epoch": 1.09,
      "eval_accuracy": 0.6498243587892787,
      "eval_f1_score": 0.6338669978734944,
      "eval_loss": 1.134881854057312,
      "eval_runtime": 104.9571,
      "eval_samples_per_second": 197.995,
      "eval_steps_per_second": 2.068,
      "step": 4500
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.6678786359077236e-05,
      "loss": 1.1257,
      "step": 4750
    },
    {
      "epoch": 1.15,
      "eval_accuracy": 0.6517491939752659,
      "eval_f1_score": 0.6390275803026418,
      "eval_loss": 1.1246434450149536,
      "eval_runtime": 118.4939,
      "eval_samples_per_second": 175.376,
      "eval_steps_per_second": 1.831,
      "step": 4750
    },
    {
      "epoch": 1.22,
      "learning_rate": 3.589518555667001e-05,
      "loss": 1.0846,
      "step": 5000
    },
    {
      "epoch": 1.22,
      "eval_accuracy": 0.6530484577258072,
      "eval_f1_score": 0.6374295817702883,
      "eval_loss": 1.134043574333191,
      "eval_runtime": 104.8506,
      "eval_samples_per_second": 198.196,
      "eval_steps_per_second": 2.07,
      "step": 5000
    },
    {
      "epoch": 1.28,
      "learning_rate": 3.511158475426279e-05,
      "loss": 1.1093,
      "step": 5250
    },
    {
      "epoch": 1.28,
      "eval_accuracy": 0.6527116115682595,
      "eval_f1_score": 0.6433227186941691,
      "eval_loss": 1.1218096017837524,
      "eval_runtime": 105.0759,
      "eval_samples_per_second": 197.771,
      "eval_steps_per_second": 2.065,
      "step": 5250
    },
    {
      "epoch": 1.34,
      "learning_rate": 3.4327983951855566e-05,
      "loss": 1.0895,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.6540108753188009,
      "eval_f1_score": 0.640048503150472,
      "eval_loss": 1.120483160018921,
      "eval_runtime": 105.0798,
      "eval_samples_per_second": 197.764,
      "eval_steps_per_second": 2.065,
      "step": 5500
    },
    {
      "epoch": 1.4,
      "learning_rate": 3.354438314944834e-05,
      "loss": 1.0947,
      "step": 5750
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.6575236995332274,
      "eval_f1_score": 0.6418486412023349,
      "eval_loss": 1.1041547060012817,
      "eval_runtime": 105.0665,
      "eval_samples_per_second": 197.789,
      "eval_steps_per_second": 2.065,
      "step": 5750
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.2760782347041125e-05,
      "loss": 1.0784,
      "step": 6000
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.6592079303209663,
      "eval_f1_score": 0.645978450692562,
      "eval_loss": 1.0955352783203125,
      "eval_runtime": 104.9867,
      "eval_samples_per_second": 197.939,
      "eval_steps_per_second": 2.067,
      "step": 6000
    },
    {
      "epoch": 1.52,
      "learning_rate": 3.19771815446339e-05,
      "loss": 1.0838,
      "step": 6250
    },
    {
      "epoch": 1.52,
      "eval_accuracy": 0.6625763918964439,
      "eval_f1_score": 0.6483699052640116,
      "eval_loss": 1.095639944076538,
      "eval_runtime": 104.9633,
      "eval_samples_per_second": 197.983,
      "eval_steps_per_second": 2.067,
      "step": 6250
    },
    {
      "epoch": 1.58,
      "learning_rate": 3.119358074222668e-05,
      "loss": 1.0764,
      "step": 6500
    },
    {
      "epoch": 1.58,
      "eval_accuracy": 0.6629132380539916,
      "eval_f1_score": 0.6497593784369542,
      "eval_loss": 1.0892637968063354,
      "eval_runtime": 105.0455,
      "eval_samples_per_second": 197.829,
      "eval_steps_per_second": 2.066,
      "step": 6500
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.040997993981946e-05,
      "loss": 1.0563,
      "step": 6750
    },
    {
      "epoch": 1.64,
      "eval_accuracy": 0.6619508204609981,
      "eval_f1_score": 0.6484351513187849,
      "eval_loss": 1.089970350265503,
      "eval_runtime": 105.3681,
      "eval_samples_per_second": 197.223,
      "eval_steps_per_second": 2.059,
      "step": 6750
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.9626379137412235e-05,
      "loss": 1.066,
      "step": 7000
    },
    {
      "epoch": 1.7,
      "eval_accuracy": 0.6659929743515711,
      "eval_f1_score": 0.6533335696048064,
      "eval_loss": 1.0745749473571777,
      "eval_runtime": 105.1237,
      "eval_samples_per_second": 197.681,
      "eval_steps_per_second": 2.064,
      "step": 7000
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.8842778335005018e-05,
      "loss": 1.0603,
      "step": 7250
    },
    {
      "epoch": 1.76,
      "eval_accuracy": 0.6680140512968578,
      "eval_f1_score": 0.657494747880274,
      "eval_loss": 1.0701572895050049,
      "eval_runtime": 105.023,
      "eval_samples_per_second": 197.871,
      "eval_steps_per_second": 2.066,
      "step": 7250
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.8059177532597798e-05,
      "loss": 1.0624,
      "step": 7500
    },
    {
      "epoch": 1.82,
      "eval_accuracy": 0.6685915018526538,
      "eval_f1_score": 0.6554959146242652,
      "eval_loss": 1.0651373863220215,
      "eval_runtime": 105.2005,
      "eval_samples_per_second": 197.537,
      "eval_steps_per_second": 2.063,
      "step": 7500
    },
    {
      "epoch": 1.88,
      "learning_rate": 2.727557673019057e-05,
      "loss": 1.0477,
      "step": 7750
    },
    {
      "epoch": 1.88,
      "eval_accuracy": 0.6685915018526538,
      "eval_f1_score": 0.655750687605972,
      "eval_loss": 1.059809684753418,
      "eval_runtime": 105.0102,
      "eval_samples_per_second": 197.895,
      "eval_steps_per_second": 2.066,
      "step": 7750
    },
    {
      "epoch": 1.95,
      "learning_rate": 2.649197592778335e-05,
      "loss": 1.0455,
      "step": 8000
    },
    {
      "epoch": 1.95,
      "eval_accuracy": 0.6643087435638323,
      "eval_f1_score": 0.6529621516807526,
      "eval_loss": 1.0628999471664429,
      "eval_runtime": 104.8979,
      "eval_samples_per_second": 198.107,
      "eval_steps_per_second": 2.069,
      "step": 8000
    },
    {
      "epoch": 2.01,
      "learning_rate": 2.5708375125376127e-05,
      "loss": 1.0403,
      "step": 8250
    },
    {
      "epoch": 2.01,
      "eval_accuracy": 0.6683990183340551,
      "eval_f1_score": 0.655431489724863,
      "eval_loss": 1.0640465021133423,
      "eval_runtime": 104.8293,
      "eval_samples_per_second": 198.237,
      "eval_steps_per_second": 2.07,
      "step": 8250
    },
    {
      "epoch": 2.07,
      "learning_rate": 2.4924774322968907e-05,
      "loss": 0.9083,
      "step": 8500
    },
    {
      "epoch": 2.07,
      "eval_accuracy": 0.6707569414368895,
      "eval_f1_score": 0.6605996988642956,
      "eval_loss": 1.0650361776351929,
      "eval_runtime": 104.9496,
      "eval_samples_per_second": 198.009,
      "eval_steps_per_second": 2.068,
      "step": 8500
    },
    {
      "epoch": 2.13,
      "learning_rate": 2.4141173520561687e-05,
      "loss": 0.9045,
      "step": 8750
    },
    {
      "epoch": 2.13,
      "eval_accuracy": 0.6696501612049468,
      "eval_f1_score": 0.6590246628667181,
      "eval_loss": 1.0609605312347412,
      "eval_runtime": 105.2135,
      "eval_samples_per_second": 197.513,
      "eval_steps_per_second": 2.062,
      "step": 8750
    },
    {
      "epoch": 2.19,
      "learning_rate": 2.3357572718154463e-05,
      "loss": 0.8998,
      "step": 9000
    },
    {
      "epoch": 2.19,
      "eval_accuracy": 0.6709975458351378,
      "eval_f1_score": 0.6609721373546371,
      "eval_loss": 1.0593194961547852,
      "eval_runtime": 104.8728,
      "eval_samples_per_second": 198.154,
      "eval_steps_per_second": 2.069,
      "step": 9000
    },
    {
      "epoch": 2.25,
      "learning_rate": 2.2573971915747243e-05,
      "loss": 0.8994,
      "step": 9250
    },
    {
      "epoch": 2.25,
      "eval_accuracy": 0.6718156007891825,
      "eval_f1_score": 0.6593984410226574,
      "eval_loss": 1.064655065536499,
      "eval_runtime": 104.9299,
      "eval_samples_per_second": 198.046,
      "eval_steps_per_second": 2.068,
      "step": 9250
    },
    {
      "epoch": 2.31,
      "learning_rate": 2.1790371113340023e-05,
      "loss": 0.9093,
      "step": 9500
    },
    {
      "epoch": 2.31,
      "eval_accuracy": 0.668880227130552,
      "eval_f1_score": 0.6589669423489957,
      "eval_loss": 1.0653289556503296,
      "eval_runtime": 105.0839,
      "eval_samples_per_second": 197.756,
      "eval_steps_per_second": 2.065,
      "step": 9500
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.10067703109328e-05,
      "loss": 0.8864,
      "step": 9750
    },
    {
      "epoch": 2.37,
      "eval_accuracy": 0.674895337086762,
      "eval_f1_score": 0.664902640512849,
      "eval_loss": 1.050571322441101,
      "eval_runtime": 105.0155,
      "eval_samples_per_second": 197.885,
      "eval_steps_per_second": 2.066,
      "step": 9750
    },
    {
      "epoch": 2.43,
      "learning_rate": 2.0223169508525576e-05,
      "loss": 0.8905,
      "step": 10000
    },
    {
      "epoch": 2.43,
      "eval_accuracy": 0.6742697656513161,
      "eval_f1_score": 0.6644634718259311,
      "eval_loss": 1.0541752576828003,
      "eval_runtime": 104.7466,
      "eval_samples_per_second": 198.393,
      "eval_steps_per_second": 2.072,
      "step": 10000
    },
    {
      "epoch": 2.49,
      "learning_rate": 1.9439568706118356e-05,
      "loss": 0.8981,
      "step": 10250
    },
    {
      "epoch": 2.49,
      "eval_accuracy": 0.6703238535200423,
      "eval_f1_score": 0.6590618499027966,
      "eval_loss": 1.0549697875976562,
      "eval_runtime": 105.0016,
      "eval_samples_per_second": 197.911,
      "eval_steps_per_second": 2.067,
      "step": 10250
    },
    {
      "epoch": 2.55,
      "learning_rate": 1.8655967903711136e-05,
      "loss": 0.9038,
      "step": 10500
    },
    {
      "epoch": 2.55,
      "eval_accuracy": 0.671286271113036,
      "eval_f1_score": 0.6622994563067591,
      "eval_loss": 1.047540307044983,
      "eval_runtime": 105.0268,
      "eval_samples_per_second": 197.864,
      "eval_steps_per_second": 2.066,
      "step": 10500
    },
    {
      "epoch": 2.61,
      "learning_rate": 1.7872367101303912e-05,
      "loss": 0.8884,
      "step": 10750
    },
    {
      "epoch": 2.61,
      "eval_accuracy": 0.6739329194937683,
      "eval_f1_score": 0.6640063633487168,
      "eval_loss": 1.0482571125030518,
      "eval_runtime": 104.9667,
      "eval_samples_per_second": 197.977,
      "eval_steps_per_second": 2.067,
      "step": 10750
    },
    {
      "epoch": 2.67,
      "learning_rate": 1.7088766298896692e-05,
      "loss": 0.8993,
      "step": 11000
    },
    {
      "epoch": 2.67,
      "eval_accuracy": 0.6732592271786728,
      "eval_f1_score": 0.6644247957468762,
      "eval_loss": 1.0455670356750488,
      "eval_runtime": 104.9075,
      "eval_samples_per_second": 198.089,
      "eval_steps_per_second": 2.068,
      "step": 11000
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.630516549648947e-05,
      "loss": 0.8881,
      "step": 11250
    },
    {
      "epoch": 2.74,
      "eval_accuracy": 0.6758577546797555,
      "eval_f1_score": 0.6663120997306459,
      "eval_loss": 1.0385903120040894,
      "eval_runtime": 104.9411,
      "eval_samples_per_second": 198.025,
      "eval_steps_per_second": 2.068,
      "step": 11250
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.552156469408225e-05,
      "loss": 0.8774,
      "step": 11500
    },
    {
      "epoch": 2.8,
      "eval_accuracy": 0.677590106347144,
      "eval_f1_score": 0.6679638313459818,
      "eval_loss": 1.0412800312042236,
      "eval_runtime": 104.9104,
      "eval_samples_per_second": 198.083,
      "eval_steps_per_second": 2.068,
      "step": 11500
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.4737963891675025e-05,
      "loss": 0.8725,
      "step": 11750
    },
    {
      "epoch": 2.86,
      "eval_accuracy": 0.6772051393099466,
      "eval_f1_score": 0.6684538458032429,
      "eval_loss": 1.041609287261963,
      "eval_runtime": 104.9692,
      "eval_samples_per_second": 197.972,
      "eval_steps_per_second": 2.067,
      "step": 11750
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.3954363089267805e-05,
      "loss": 0.8814,
      "step": 12000
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.6780231942639912,
      "eval_f1_score": 0.668403494357387,
      "eval_loss": 1.0360217094421387,
      "eval_runtime": 104.9925,
      "eval_samples_per_second": 197.929,
      "eval_steps_per_second": 2.067,
      "step": 12000
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.3170762286860583e-05,
      "loss": 0.8707,
      "step": 12250
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.6776382272267937,
      "eval_f1_score": 0.6678552195861074,
      "eval_loss": 1.036238193511963,
      "eval_runtime": 104.9928,
      "eval_samples_per_second": 197.928,
      "eval_steps_per_second": 2.067,
      "step": 12250
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.2387161484453361e-05,
      "loss": 0.803,
      "step": 12500
    },
    {
      "epoch": 3.04,
      "eval_accuracy": 0.678504403060488,
      "eval_f1_score": 0.6693071368875216,
      "eval_loss": 1.048478364944458,
      "eval_runtime": 104.9691,
      "eval_samples_per_second": 197.973,
      "eval_steps_per_second": 2.067,
      "step": 12500
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.160356068204614e-05,
      "loss": 0.7647,
      "step": 12750
    },
    {
      "epoch": 3.1,
      "eval_accuracy": 0.6773976228285453,
      "eval_f1_score": 0.667836848673267,
      "eval_loss": 1.0551421642303467,
      "eval_runtime": 104.9474,
      "eval_samples_per_second": 198.013,
      "eval_steps_per_second": 2.068,
      "step": 12750
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.0819959879638917e-05,
      "loss": 0.7595,
      "step": 13000
    },
    {
      "epoch": 3.16,
      "eval_accuracy": 0.6766758096338001,
      "eval_f1_score": 0.6681691840022692,
      "eval_loss": 1.0496253967285156,
      "eval_runtime": 104.8372,
      "eval_samples_per_second": 198.222,
      "eval_steps_per_second": 2.07,
      "step": 13000
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.0036359077231696e-05,
      "loss": 0.7628,
      "step": 13250
    },
    {
      "epoch": 3.22,
      "eval_accuracy": 0.6778788316250421,
      "eval_f1_score": 0.6698784049603582,
      "eval_loss": 1.0489078760147095,
      "eval_runtime": 104.8411,
      "eval_samples_per_second": 198.214,
      "eval_steps_per_second": 2.07,
      "step": 13250
    },
    {
      "epoch": 3.28,
      "learning_rate": 9.252758274824474e-06,
      "loss": 0.7711,
      "step": 13500
    },
    {
      "epoch": 3.28,
      "eval_accuracy": 0.6778307107453925,
      "eval_f1_score": 0.670537104393122,
      "eval_loss": 1.0493261814117432,
      "eval_runtime": 104.9371,
      "eval_samples_per_second": 198.033,
      "eval_steps_per_second": 2.068,
      "step": 13500
    },
    {
      "epoch": 3.34,
      "learning_rate": 8.469157472417252e-06,
      "loss": 0.7659,
      "step": 13750
    },
    {
      "epoch": 3.34,
      "eval_accuracy": 0.677445743708195,
      "eval_f1_score": 0.6690005925034824,
      "eval_loss": 1.0476195812225342,
      "eval_runtime": 104.8587,
      "eval_samples_per_second": 198.181,
      "eval_steps_per_second": 2.069,
      "step": 13750
    },
    {
      "epoch": 3.4,
      "learning_rate": 7.68555667001003e-06,
      "loss": 0.7466,
      "step": 14000
    },
    {
      "epoch": 3.4,
      "eval_accuracy": 0.6767720513930995,
      "eval_f1_score": 0.6682514083974397,
      "eval_loss": 1.0503697395324707,
      "eval_runtime": 104.8988,
      "eval_samples_per_second": 198.105,
      "eval_steps_per_second": 2.069,
      "step": 14000
    },
    {
      "epoch": 3.46,
      "learning_rate": 6.901955867602809e-06,
      "loss": 0.7438,
      "step": 14250
    },
    {
      "epoch": 3.46,
      "eval_accuracy": 0.6778788316250421,
      "eval_f1_score": 0.6704394720403158,
      "eval_loss": 1.0488700866699219,
      "eval_runtime": 104.8654,
      "eval_samples_per_second": 198.168,
      "eval_steps_per_second": 2.069,
      "step": 14250
    },
    {
      "epoch": 3.53,
      "learning_rate": 6.118355065195587e-06,
      "loss": 0.7492,
      "step": 14500
    },
    {
      "epoch": 3.53,
      "eval_accuracy": 0.6762908425966027,
      "eval_f1_score": 0.668875699854832,
      "eval_loss": 1.0487704277038574,
      "eval_runtime": 105.1614,
      "eval_samples_per_second": 197.611,
      "eval_steps_per_second": 2.063,
      "step": 14500
    },
    {
      "epoch": 3.59,
      "learning_rate": 5.334754262788365e-06,
      "loss": 0.7534,
      "step": 14750
    },
    {
      "epoch": 3.59,
      "eval_accuracy": 0.6767239305134498,
      "eval_f1_score": 0.6693352825093323,
      "eval_loss": 1.0485919713974,
      "eval_runtime": 105.4042,
      "eval_samples_per_second": 197.155,
      "eval_steps_per_second": 2.059,
      "step": 14750
    },
    {
      "epoch": 3.65,
      "learning_rate": 4.551153460381144e-06,
      "loss": 0.7391,
      "step": 15000
    },
    {
      "epoch": 3.65,
      "eval_accuracy": 0.6791299744959338,
      "eval_f1_score": 0.6716788527298224,
      "eval_loss": 1.0460212230682373,
      "eval_runtime": 105.024,
      "eval_samples_per_second": 197.869,
      "eval_steps_per_second": 2.066,
      "step": 15000
    },
    {
      "epoch": 3.71,
      "learning_rate": 3.7675526579739217e-06,
      "loss": 0.7543,
      "step": 15250
    },
    {
      "epoch": 3.71,
      "eval_accuracy": 0.6804773591261248,
      "eval_f1_score": 0.6726007691455318,
      "eval_loss": 1.045469880104065,
      "eval_runtime": 104.6353,
      "eval_samples_per_second": 198.604,
      "eval_steps_per_second": 2.074,
      "step": 15250
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.9839518555667003e-06,
      "loss": 0.7382,
      "step": 15500
    },
    {
      "epoch": 3.77,
      "eval_accuracy": 0.6802848756075262,
      "eval_f1_score": 0.6728396464392151,
      "eval_loss": 1.0436148643493652,
      "eval_runtime": 105.0033,
      "eval_samples_per_second": 197.908,
      "eval_steps_per_second": 2.067,
      "step": 15500
    }
  ],
  "max_steps": 16452,
  "num_train_epochs": 4,
  "total_flos": 1.9763710480077926e+17,
  "trial_name": null,
  "trial_params": null
}