{
  "best_metric": 0.17471107840538025,
  "best_model_checkpoint": "xlnet-base-cased/subtaskA/0/checkpoint-23952",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 71856,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.9860832776664444e-05,
      "loss": 0.5163,
      "step": 500
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9721665553328883e-05,
      "loss": 0.3694,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9582498329993322e-05,
      "loss": 0.3342,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.944333110665776e-05,
      "loss": 0.3631,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9304163883322203e-05,
      "loss": 0.2701,
      "step": 2500
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9164996659986642e-05,
      "loss": 0.2449,
      "step": 3000
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.902582943665108e-05,
      "loss": 0.2478,
      "step": 3500
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.888666221331552e-05,
      "loss": 0.2499,
      "step": 4000
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8747494989979962e-05,
      "loss": 0.2542,
      "step": 4500
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.86083277666444e-05,
      "loss": 0.2503,
      "step": 5000
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8469160543308843e-05,
      "loss": 0.2436,
      "step": 5500
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.832999331997328e-05,
      "loss": 0.2492,
      "step": 6000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.819082609663772e-05,
      "loss": 0.228,
      "step": 6500
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.805165887330216e-05,
      "loss": 0.2293,
      "step": 7000
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.7912491649966602e-05,
      "loss": 0.2179,
      "step": 7500
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.777332442663104e-05,
      "loss": 0.2044,
      "step": 8000
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.763415720329548e-05,
      "loss": 0.1854,
      "step": 8500
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7494989979959922e-05,
      "loss": 0.1957,
      "step": 9000
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.735582275662436e-05,
      "loss": 0.2028,
      "step": 9500
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.72166555332888e-05,
      "loss": 0.1937,
      "step": 10000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.707748830995324e-05,
      "loss": 0.1722,
      "step": 10500
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.693832108661768e-05,
      "loss": 0.1461,
      "step": 11000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.679915386328212e-05,
      "loss": 0.1734,
      "step": 11500
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6659986639946563e-05,
      "loss": 0.1978,
      "step": 12000
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6520819416611002e-05,
      "loss": 0.1648,
      "step": 12500
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.638165219327544e-05,
      "loss": 0.1769,
      "step": 13000
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.624248496993988e-05,
      "loss": 0.1984,
      "step": 13500
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6103317746604322e-05,
      "loss": 0.1515,
      "step": 14000
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.596415052326876e-05,
      "loss": 0.1543,
      "step": 14500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.58249832999332e-05,
      "loss": 0.1974,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5685816076597642e-05,
      "loss": 0.162,
      "step": 15500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.554664885326208e-05,
      "loss": 0.1493,
      "step": 16000
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5407481629926523e-05,
      "loss": 0.1776,
      "step": 16500
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5268314406590962e-05,
      "loss": 0.1675,
      "step": 17000
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5129147183255401e-05,
      "loss": 0.1587,
      "step": 17500
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.498997995991984e-05,
      "loss": 0.1289,
      "step": 18000
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.485081273658428e-05,
      "loss": 0.1719,
      "step": 18500
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.471164551324872e-05,
      "loss": 0.1658,
      "step": 19000
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4572478289913162e-05,
      "loss": 0.1505,
      "step": 19500
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.44333110665776e-05,
      "loss": 0.1898,
      "step": 20000
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4294143843242041e-05,
      "loss": 0.1813,
      "step": 20500
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.415497661990648e-05,
      "loss": 0.1528,
      "step": 21000
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4015809396570921e-05,
      "loss": 0.1635,
      "step": 21500
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3876642173235361e-05,
      "loss": 0.1543,
      "step": 22000
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.37374749498998e-05,
      "loss": 0.1539,
      "step": 22500
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3598307726564241e-05,
      "loss": 0.1421,
      "step": 23000
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.345914050322868e-05,
      "loss": 0.1217,
      "step": 23500
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.9711923847695391,
      "eval_loss": 0.17471107840538025,
      "eval_runtime": 440.9341,
      "eval_samples_per_second": 54.321,
      "eval_steps_per_second": 13.58,
      "step": 23952
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.331997327989312e-05,
      "loss": 0.1296,
      "step": 24000
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.318080605655756e-05,
      "loss": 0.0915,
      "step": 24500
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3041638833222002e-05,
      "loss": 0.1019,
      "step": 25000
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.290247160988644e-05,
      "loss": 0.0997,
      "step": 25500
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2763304386550881e-05,
      "loss": 0.2252,
      "step": 26000
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.262413716321532e-05,
      "loss": 0.1427,
      "step": 26500
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.248496993987976e-05,
      "loss": 0.1348,
      "step": 27000
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.23458027165442e-05,
      "loss": 0.1055,
      "step": 27500
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.220663549320864e-05,
      "loss": 0.1216,
      "step": 28000
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.206746826987308e-05,
      "loss": 0.1,
      "step": 28500
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.192830104653752e-05,
      "loss": 0.1036,
      "step": 29000
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1789133823201962e-05,
      "loss": 0.0946,
      "step": 29500
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.16499665998664e-05,
      "loss": 0.1058,
      "step": 30000
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1510799376530842e-05,
      "loss": 0.0957,
      "step": 30500
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.137163215319528e-05,
      "loss": 0.0952,
      "step": 31000
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1232464929859721e-05,
      "loss": 0.127,
      "step": 31500
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.109329770652416e-05,
      "loss": 0.0919,
      "step": 32000
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.09541304831886e-05,
      "loss": 0.1056,
      "step": 32500
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.081496325985304e-05,
      "loss": 0.1025,
      "step": 33000
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.067579603651748e-05,
      "loss": 0.0879,
      "step": 33500
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.053662881318192e-05,
      "loss": 0.0847,
      "step": 34000
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.039746158984636e-05,
      "loss": 0.088,
      "step": 34500
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0258294366510799e-05,
      "loss": 0.1225,
      "step": 35000
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0119127143175241e-05,
      "loss": 0.1046,
      "step": 35500
    },
    {
      "epoch": 1.5,
      "learning_rate": 9.97995991983968e-06,
      "loss": 0.0962,
      "step": 36000
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.84079269650412e-06,
      "loss": 0.0994,
      "step": 36500
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.70162547316856e-06,
      "loss": 0.1223,
      "step": 37000
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.562458249833e-06,
      "loss": 0.1025,
      "step": 37500
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.423291026497439e-06,
      "loss": 0.1094,
      "step": 38000
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.28412380316188e-06,
      "loss": 0.1072,
      "step": 38500
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.14495657982632e-06,
      "loss": 0.0931,
      "step": 39000
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.005789356490761e-06,
      "loss": 0.0762,
      "step": 39500
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.8666221331552e-06,
      "loss": 0.0909,
      "step": 40000
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.72745490981964e-06,
      "loss": 0.087,
      "step": 40500
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.588287686484081e-06,
      "loss": 0.0679,
      "step": 41000
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.44912046314852e-06,
      "loss": 0.0698,
      "step": 41500
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.30995323981296e-06,
      "loss": 0.0813,
      "step": 42000
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.1707860164774e-06,
      "loss": 0.0868,
      "step": 42500
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.03161879314184e-06,
      "loss": 0.0939,
      "step": 43000
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.892451569806279e-06,
      "loss": 0.0606,
      "step": 43500
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.75328434647072e-06,
      "loss": 0.1026,
      "step": 44000
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.614117123135159e-06,
      "loss": 0.0757,
      "step": 44500
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.474949899799599e-06,
      "loss": 0.0845,
      "step": 45000
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.33578267646404e-06,
      "loss": 0.0753,
      "step": 45500
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.1966154531284795e-06,
      "loss": 0.0633,
      "step": 46000
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.057448229792919e-06,
      "loss": 0.0847,
      "step": 46500
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.91828100645736e-06,
      "loss": 0.0602,
      "step": 47000
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.7791137831218005e-06,
      "loss": 0.0683,
      "step": 47500
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9619238476953907,
      "eval_loss": 0.20673486590385437,
      "eval_runtime": 441.9436,
      "eval_samples_per_second": 54.197,
      "eval_steps_per_second": 13.549,
      "step": 47904
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.63994655978624e-06,
      "loss": 0.0522,
      "step": 48000
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.50077933645068e-06,
      "loss": 0.0483,
      "step": 48500
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.36161211311512e-06,
      "loss": 0.0578,
      "step": 49000
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.22244488977956e-06,
      "loss": 0.0595,
      "step": 49500
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.083277666443999e-06,
      "loss": 0.0485,
      "step": 50000
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.944110443108439e-06,
      "loss": 0.0639,
      "step": 50500
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.80494321977288e-06,
      "loss": 0.0564,
      "step": 51000
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.6657759964373195e-06,
      "loss": 0.0368,
      "step": 51500
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.526608773101759e-06,
      "loss": 0.0504,
      "step": 52000
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.387441549766199e-06,
      "loss": 0.0539,
      "step": 52500
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.248274326430639e-06,
      "loss": 0.0341,
      "step": 53000
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1091071030950786e-06,
      "loss": 0.0956,
      "step": 53500
    },
    {
      "epoch": 2.25,
      "learning_rate": 4.969939879759519e-06,
      "loss": 0.0393,
      "step": 54000
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.830772656423959e-06,
      "loss": 0.0383,
      "step": 54500
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6916054330883996e-06,
      "loss": 0.0617,
      "step": 55000
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.552438209752839e-06,
      "loss": 0.0715,
      "step": 55500
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.41327098641728e-06,
      "loss": 0.0499,
      "step": 56000
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.27410376308172e-06,
      "loss": 0.0286,
      "step": 56500
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1349365397461595e-06,
      "loss": 0.0633,
      "step": 57000
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.995769316410599e-06,
      "loss": 0.0429,
      "step": 57500
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.856602093075039e-06,
      "loss": 0.0519,
      "step": 58000
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.717434869739479e-06,
      "loss": 0.0336,
      "step": 58500
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.578267646403919e-06,
      "loss": 0.0334,
      "step": 59000
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4391004230683596e-06,
      "loss": 0.0443,
      "step": 59500
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2999331997327993e-06,
      "loss": 0.0382,
      "step": 60000
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.160765976397239e-06,
      "loss": 0.0227,
      "step": 60500
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0215987530616793e-06,
      "loss": 0.0484,
      "step": 61000
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.882431529726119e-06,
      "loss": 0.0421,
      "step": 61500
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.743264306390559e-06,
      "loss": 0.0572,
      "step": 62000
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.604097083054999e-06,
      "loss": 0.0373,
      "step": 62500
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.464929859719439e-06,
      "loss": 0.0668,
      "step": 63000
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.325762636383879e-06,
      "loss": 0.0406,
      "step": 63500
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.186595413048319e-06,
      "loss": 0.0477,
      "step": 64000
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.047428189712759e-06,
      "loss": 0.0292,
      "step": 64500
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.908260966377199e-06,
      "loss": 0.0546,
      "step": 65000
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.769093743041639e-06,
      "loss": 0.049,
      "step": 65500
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.629926519706079e-06,
      "loss": 0.0295,
      "step": 66000
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4907592963705188e-06,
      "loss": 0.0393,
      "step": 66500
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3515920730349588e-06,
      "loss": 0.029,
      "step": 67000
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2124248496993988e-06,
      "loss": 0.0295,
      "step": 67500
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.073257626363839e-06,
      "loss": 0.044,
      "step": 68000
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.340904030282788e-07,
      "loss": 0.0248,
      "step": 68500
    },
    {
      "epoch": 2.88,
      "learning_rate": 7.949231796927188e-07,
      "loss": 0.0589,
      "step": 69000
    },
    {
      "epoch": 2.9,
      "learning_rate": 6.557559563571589e-07,
      "loss": 0.0329,
      "step": 69500
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.165887330215988e-07,
      "loss": 0.0325,
      "step": 70000
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.7742150968603874e-07,
      "loss": 0.0306,
      "step": 70500
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.3825428635047877e-07,
      "loss": 0.0554,
      "step": 71000
    },
    {
      "epoch": 2.99,
      "learning_rate": 9.908706301491873e-08,
      "loss": 0.0345,
      "step": 71500
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9544088176352705,
      "eval_loss": 0.2647545337677002,
      "eval_runtime": 440.9856,
      "eval_samples_per_second": 54.315,
      "eval_steps_per_second": 13.579,
      "step": 71856
    }
  ],
  "logging_steps": 500,
  "max_steps": 71856,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 8.187886554743808e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}