{
  "best_metric": 0.8361828808652427,
  "best_model_checkpoint": "deberta-stance-transcripts/checkpoint-24656",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 61640,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 9.91888384166126e-06,
      "loss": 0.4064,
      "step": 500
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.83776768332252e-06,
      "loss": 0.3183,
      "step": 1000
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.756651524983777e-06,
      "loss": 0.2936,
      "step": 1500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.675535366645036e-06,
      "loss": 0.2957,
      "step": 2000
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.594419208306296e-06,
      "loss": 0.2816,
      "step": 2500
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.513303049967553e-06,
      "loss": 0.2713,
      "step": 3000
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.432186891628813e-06,
      "loss": 0.2662,
      "step": 3500
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.351070733290072e-06,
      "loss": 0.259,
      "step": 4000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.269954574951331e-06,
      "loss": 0.2549,
      "step": 4500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.188838416612589e-06,
      "loss": 0.2554,
      "step": 5000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.107722258273848e-06,
      "loss": 0.2513,
      "step": 5500
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.026606099935108e-06,
      "loss": 0.262,
      "step": 6000
    },
    {
      "epoch": 0.53,
      "learning_rate": 8.945489941596367e-06,
      "loss": 0.2374,
      "step": 6500
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.864373783257626e-06,
      "loss": 0.2529,
      "step": 7000
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.783257624918886e-06,
      "loss": 0.2496,
      "step": 7500
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.702141466580143e-06,
      "loss": 0.2503,
      "step": 8000
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.621025308241403e-06,
      "loss": 0.2401,
      "step": 8500
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.53990914990266e-06,
      "loss": 0.2376,
      "step": 9000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.45879299156392e-06,
      "loss": 0.249,
      "step": 9500
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.377676833225179e-06,
      "loss": 0.2358,
      "step": 10000
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.296560674886438e-06,
      "loss": 0.2262,
      "step": 10500
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.215444516547698e-06,
      "loss": 0.2389,
      "step": 11000
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.134328358208955e-06,
      "loss": 0.2414,
      "step": 11500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.053212199870215e-06,
      "loss": 0.232,
      "step": 12000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7983222247759136,
      "eval_f1": 0.8218727740945702,
      "eval_loss": 0.24177740514278412,
      "eval_roc_auc": 0.874017912487587,
      "eval_runtime": 131.4468,
      "eval_samples_per_second": 66.202,
      "eval_steps_per_second": 8.277,
      "step": 12328
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.972096041531474e-06,
      "loss": 0.2243,
      "step": 12500
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.890979883192733e-06,
      "loss": 0.2002,
      "step": 13000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.809863724853991e-06,
      "loss": 0.2076,
      "step": 13500
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.72874756651525e-06,
      "loss": 0.2026,
      "step": 14000
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.64763140817651e-06,
      "loss": 0.2077,
      "step": 14500
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.566515249837768e-06,
      "loss": 0.2021,
      "step": 15000
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.4853990914990275e-06,
      "loss": 0.204,
      "step": 15500
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.404282933160286e-06,
      "loss": 0.2089,
      "step": 16000
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.323166774821545e-06,
      "loss": 0.2095,
      "step": 16500
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.242050616482804e-06,
      "loss": 0.2004,
      "step": 17000
    },
    {
      "epoch": 1.42,
      "learning_rate": 7.160934458144063e-06,
      "loss": 0.2102,
      "step": 17500
    },
    {
      "epoch": 1.46,
      "learning_rate": 7.0798182998053225e-06,
      "loss": 0.1962,
      "step": 18000
    },
    {
      "epoch": 1.5,
      "learning_rate": 6.998702141466581e-06,
      "loss": 0.2009,
      "step": 18500
    },
    {
      "epoch": 1.54,
      "learning_rate": 6.9175859831278394e-06,
      "loss": 0.2,
      "step": 19000
    },
    {
      "epoch": 1.58,
      "learning_rate": 6.836469824789098e-06,
      "loss": 0.2032,
      "step": 19500
    },
    {
      "epoch": 1.62,
      "learning_rate": 6.755353666450357e-06,
      "loss": 0.2029,
      "step": 20000
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.674237508111617e-06,
      "loss": 0.2098,
      "step": 20500
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.593121349772875e-06,
      "loss": 0.1985,
      "step": 21000
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.512005191434134e-06,
      "loss": 0.214,
      "step": 21500
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.430889033095393e-06,
      "loss": 0.2013,
      "step": 22000
    },
    {
      "epoch": 1.83,
      "learning_rate": 6.349772874756652e-06,
      "loss": 0.2039,
      "step": 22500
    },
    {
      "epoch": 1.87,
      "learning_rate": 6.2686567164179116e-06,
      "loss": 0.2036,
      "step": 23000
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.18754055807917e-06,
      "loss": 0.1989,
      "step": 23500
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.106424399740429e-06,
      "loss": 0.2012,
      "step": 24000
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.025308241401687e-06,
      "loss": 0.2023,
      "step": 24500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8142955642381062,
      "eval_f1": 0.8361828808652427,
      "eval_loss": 0.24095113575458527,
      "eval_roc_auc": 0.8841932788750095,
      "eval_runtime": 131.1521,
      "eval_samples_per_second": 66.35,
      "eval_steps_per_second": 8.296,
      "step": 24656
    },
    {
      "epoch": 2.03,
      "learning_rate": 5.944192083062946e-06,
      "loss": 0.1802,
      "step": 25000
    },
    {
      "epoch": 2.07,
      "learning_rate": 5.863075924724206e-06,
      "loss": 0.1702,
      "step": 25500
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.781959766385464e-06,
      "loss": 0.1705,
      "step": 26000
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7008436080467235e-06,
      "loss": 0.1816,
      "step": 26500
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.619727449707982e-06,
      "loss": 0.1741,
      "step": 27000
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.538611291369241e-06,
      "loss": 0.1731,
      "step": 27500
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.457495133030501e-06,
      "loss": 0.1702,
      "step": 28000
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.376378974691759e-06,
      "loss": 0.1766,
      "step": 28500
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.2952628163530185e-06,
      "loss": 0.1648,
      "step": 29000
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.214146658014276e-06,
      "loss": 0.1796,
      "step": 29500
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.1330304996755355e-06,
      "loss": 0.1795,
      "step": 30000
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.051914341336794e-06,
      "loss": 0.1745,
      "step": 30500
    },
    {
      "epoch": 2.51,
      "learning_rate": 4.970798182998053e-06,
      "loss": 0.173,
      "step": 31000
    },
    {
      "epoch": 2.56,
      "learning_rate": 4.889682024659313e-06,
      "loss": 0.1684,
      "step": 31500
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.808565866320571e-06,
      "loss": 0.1753,
      "step": 32000
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.7274497079818305e-06,
      "loss": 0.1735,
      "step": 32500
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.64633354964309e-06,
      "loss": 0.1792,
      "step": 33000
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.565217391304348e-06,
      "loss": 0.1765,
      "step": 33500
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.484101232965607e-06,
      "loss": 0.1812,
      "step": 34000
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.402985074626866e-06,
      "loss": 0.1705,
      "step": 34500
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.321868916288125e-06,
      "loss": 0.1783,
      "step": 35000
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.240752757949384e-06,
      "loss": 0.1818,
      "step": 35500
    },
    {
      "epoch": 2.92,
      "learning_rate": 4.159636599610643e-06,
      "loss": 0.1761,
      "step": 36000
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.078520441271902e-06,
      "loss": 0.1744,
      "step": 36500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.809928752011032,
      "eval_f1": 0.8312578171733102,
      "eval_loss": 0.26303380727767944,
      "eval_roc_auc": 0.8817786701538314,
      "eval_runtime": 131.6921,
      "eval_samples_per_second": 66.078,
      "eval_steps_per_second": 8.262,
      "step": 36984
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.99740428293316e-06,
      "loss": 0.1749,
      "step": 37000
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.9162881245944196e-06,
      "loss": 0.1478,
      "step": 37500
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.835171966255679e-06,
      "loss": 0.1498,
      "step": 38000
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.7540558079169374e-06,
      "loss": 0.1447,
      "step": 38500
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.672939649578196e-06,
      "loss": 0.1587,
      "step": 39000
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.591823491239455e-06,
      "loss": 0.1558,
      "step": 39500
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.510707332900714e-06,
      "loss": 0.1459,
      "step": 40000
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.429591174561973e-06,
      "loss": 0.1538,
      "step": 40500
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.348475016223232e-06,
      "loss": 0.1487,
      "step": 41000
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.2673588578844904e-06,
      "loss": 0.1492,
      "step": 41500
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.1862426995457498e-06,
      "loss": 0.1441,
      "step": 42000
    },
    {
      "epoch": 3.45,
      "learning_rate": 3.1051265412070087e-06,
      "loss": 0.1675,
      "step": 42500
    },
    {
      "epoch": 3.49,
      "learning_rate": 3.0240103828682676e-06,
      "loss": 0.1448,
      "step": 43000
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.9428942245295265e-06,
      "loss": 0.1402,
      "step": 43500
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.861778066190786e-06,
      "loss": 0.1676,
      "step": 44000
    },
    {
      "epoch": 3.61,
      "learning_rate": 2.7806619078520443e-06,
      "loss": 0.1629,
      "step": 44500
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.6995457495133032e-06,
      "loss": 0.1549,
      "step": 45000
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.618429591174562e-06,
      "loss": 0.1649,
      "step": 45500
    },
    {
      "epoch": 3.73,
      "learning_rate": 2.537313432835821e-06,
      "loss": 0.1475,
      "step": 46000
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.45619727449708e-06,
      "loss": 0.1514,
      "step": 46500
    },
    {
      "epoch": 3.81,
      "learning_rate": 2.375081116158339e-06,
      "loss": 0.165,
      "step": 47000
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.2939649578195978e-06,
      "loss": 0.1619,
      "step": 47500
    },
    {
      "epoch": 3.89,
      "learning_rate": 2.2128487994808567e-06,
      "loss": 0.1482,
      "step": 48000
    },
    {
      "epoch": 3.93,
      "learning_rate": 2.1317326411421156e-06,
      "loss": 0.1566,
      "step": 48500
    },
    {
      "epoch": 3.97,
      "learning_rate": 2.0506164828033745e-06,
      "loss": 0.1427,
      "step": 49000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8131464031257182,
      "eval_f1": 0.8350062394878195,
      "eval_loss": 0.2954484820365906,
      "eval_roc_auc": 0.8847613364883563,
      "eval_runtime": 131.8182,
      "eval_samples_per_second": 66.015,
      "eval_steps_per_second": 8.254,
      "step": 49312
    },
    {
      "epoch": 4.02,
      "learning_rate": 1.9695003244646334e-06,
      "loss": 0.1421,
      "step": 49500
    },
    {
      "epoch": 4.06,
      "learning_rate": 1.8883841661258925e-06,
      "loss": 0.1246,
      "step": 50000
    },
    {
      "epoch": 4.1,
      "learning_rate": 1.8072680077871512e-06,
      "loss": 0.1298,
      "step": 50500
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.7261518494484104e-06,
      "loss": 0.1493,
      "step": 51000
    },
    {
      "epoch": 4.18,
      "learning_rate": 1.6450356911096693e-06,
      "loss": 0.1324,
      "step": 51500
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.563919532770928e-06,
      "loss": 0.1308,
      "step": 52000
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.482803374432187e-06,
      "loss": 0.1248,
      "step": 52500
    },
    {
      "epoch": 4.3,
      "learning_rate": 1.4016872160934458e-06,
      "loss": 0.1403,
      "step": 53000
    },
    {
      "epoch": 4.34,
      "learning_rate": 1.320571057754705e-06,
      "loss": 0.1345,
      "step": 53500
    },
    {
      "epoch": 4.38,
      "learning_rate": 1.2394548994159636e-06,
      "loss": 0.1383,
      "step": 54000
    },
    {
      "epoch": 4.42,
      "learning_rate": 1.1583387410772227e-06,
      "loss": 0.1463,
      "step": 54500
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.0772225827384817e-06,
      "loss": 0.1304,
      "step": 55000
    },
    {
      "epoch": 4.5,
      "learning_rate": 9.961064243997406e-07,
      "loss": 0.1421,
      "step": 55500
    },
    {
      "epoch": 4.54,
      "learning_rate": 9.149902660609994e-07,
      "loss": 0.1299,
      "step": 56000
    },
    {
      "epoch": 4.58,
      "learning_rate": 8.338741077222584e-07,
      "loss": 0.1472,
      "step": 56500
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.527579493835173e-07,
      "loss": 0.1444,
      "step": 57000
    },
    {
      "epoch": 4.66,
      "learning_rate": 6.716417910447762e-07,
      "loss": 0.1395,
      "step": 57500
    },
    {
      "epoch": 4.7,
      "learning_rate": 5.905256327060351e-07,
      "loss": 0.1212,
      "step": 58000
    },
    {
      "epoch": 4.75,
      "learning_rate": 5.09409474367294e-07,
      "loss": 0.1302,
      "step": 58500
    },
    {
      "epoch": 4.79,
      "learning_rate": 4.2829331602855294e-07,
      "loss": 0.1383,
      "step": 59000
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.4717715768981185e-07,
      "loss": 0.1318,
      "step": 59500
    },
    {
      "epoch": 4.87,
      "learning_rate": 2.6606099935107076e-07,
      "loss": 0.1339,
      "step": 60000
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.8494484101232967e-07,
      "loss": 0.1361,
      "step": 60500
    },
    {
      "epoch": 4.95,
      "learning_rate": 1.0382868267358859e-07,
      "loss": 0.1368,
      "step": 61000
    },
    {
      "epoch": 4.99,
      "learning_rate": 2.2712524334847506e-08,
      "loss": 0.1311,
      "step": 61500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8113077453458974,
      "eval_f1": 0.8333153055705786,
      "eval_loss": 0.3088623583316803,
      "eval_roc_auc": 0.8842623226809119,
      "eval_runtime": 131.6271,
      "eval_samples_per_second": 66.111,
      "eval_steps_per_second": 8.266,
      "step": 61640
    }
  ],
  "logging_steps": 500,
  "max_steps": 61640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.2974734910803968e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}