{
  "best_metric": 0.7994814174589455,
  "best_model_checkpoint": "../../checkpoints/baseline/default-baseline-uncleaned/lm_model/finetune/qnli/checkpoint-6000",
  "epoch": 10.0,
  "global_step": 6870,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "eval_accuracy": 0.7541557550430298,
      "eval_f1": 0.7606473594548551,
      "eval_loss": 0.513882577419281,
      "eval_runtime": 2.5883,
      "eval_samples_per_second": 883.22,
      "eval_steps_per_second": 110.499,
      "step": 200
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.7655293345451355,
      "eval_f1": 0.7691645133505599,
      "eval_loss": 0.49110308289527893,
      "eval_runtime": 2.5616,
      "eval_samples_per_second": 892.424,
      "eval_steps_per_second": 111.651,
      "step": 400
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.636098981077147e-05,
      "loss": 0.5823,
      "step": 500
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.7729659080505371,
      "eval_f1": 0.7746417716022579,
      "eval_loss": 0.4785449504852295,
      "eval_runtime": 2.5604,
      "eval_samples_per_second": 892.818,
      "eval_steps_per_second": 111.7,
      "step": 600
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.787401556968689,
      "eval_f1": 0.7940677966101695,
      "eval_loss": 0.4633069336414337,
      "eval_runtime": 2.5555,
      "eval_samples_per_second": 894.547,
      "eval_steps_per_second": 111.916,
      "step": 800
    },
    {
      "epoch": 1.46,
      "learning_rate": 4.272197962154294e-05,
      "loss": 0.5036,
      "step": 1000
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.7725284099578857,
      "eval_f1": 0.7653429602888085,
      "eval_loss": 0.48655858635902405,
      "eval_runtime": 2.5624,
      "eval_samples_per_second": 892.133,
      "eval_steps_per_second": 111.614,
      "step": 1000
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.7904636859893799,
      "eval_f1": 0.7993297025555091,
      "eval_loss": 0.45168524980545044,
      "eval_runtime": 2.5579,
      "eval_samples_per_second": 893.715,
      "eval_steps_per_second": 111.812,
      "step": 1200
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.7909011244773865,
      "eval_f1": 0.7941429801894918,
      "eval_loss": 0.45709288120269775,
      "eval_runtime": 2.5535,
      "eval_samples_per_second": 895.253,
      "eval_steps_per_second": 112.005,
      "step": 1400
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.9082969432314415e-05,
      "loss": 0.482,
      "step": 1500
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.7917760014533997,
      "eval_f1": 0.797274275979557,
      "eval_loss": 0.45153066515922546,
      "eval_runtime": 2.5517,
      "eval_samples_per_second": 895.864,
      "eval_steps_per_second": 112.081,
      "step": 1600
    },
    {
      "epoch": 2.62,
      "eval_accuracy": 0.7970253825187683,
      "eval_f1": 0.8028887000849616,
      "eval_loss": 0.45009395480155945,
      "eval_runtime": 2.5577,
      "eval_samples_per_second": 893.772,
      "eval_steps_per_second": 111.819,
      "step": 1800
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.544395924308588e-05,
      "loss": 0.461,
      "step": 2000
    },
    {
      "epoch": 2.91,
      "eval_accuracy": 0.7913385629653931,
      "eval_f1": 0.7899603698811096,
      "eval_loss": 0.4560691714286804,
      "eval_runtime": 2.5537,
      "eval_samples_per_second": 895.184,
      "eval_steps_per_second": 111.996,
      "step": 2000
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 0.8018372654914856,
      "eval_f1": 0.8119551681195518,
      "eval_loss": 0.4423236548900604,
      "eval_runtime": 2.5844,
      "eval_samples_per_second": 884.554,
      "eval_steps_per_second": 110.666,
      "step": 2200
    },
    {
      "epoch": 3.49,
      "eval_accuracy": 0.7939632534980774,
      "eval_f1": 0.8006771053745239,
      "eval_loss": 0.4459246098995209,
      "eval_runtime": 2.5753,
      "eval_samples_per_second": 887.668,
      "eval_steps_per_second": 111.056,
      "step": 2400
    },
    {
      "epoch": 3.64,
      "learning_rate": 3.1804949053857355e-05,
      "loss": 0.4398,
      "step": 2500
    },
    {
      "epoch": 3.78,
      "eval_accuracy": 0.8027121424674988,
      "eval_f1": 0.8112180828798661,
      "eval_loss": 0.4345141649246216,
      "eval_runtime": 2.5691,
      "eval_samples_per_second": 889.815,
      "eval_steps_per_second": 111.324,
      "step": 2600
    },
    {
      "epoch": 4.08,
      "eval_accuracy": 0.8009623885154724,
      "eval_f1": 0.8151158065826899,
      "eval_loss": 0.4380125403404236,
      "eval_runtime": 2.563,
      "eval_samples_per_second": 891.909,
      "eval_steps_per_second": 111.586,
      "step": 2800
    },
    {
      "epoch": 4.37,
      "learning_rate": 2.816593886462882e-05,
      "loss": 0.4272,
      "step": 3000
    },
    {
      "epoch": 4.37,
      "eval_accuracy": 0.7913385629653931,
      "eval_f1": 0.7934170636639237,
      "eval_loss": 0.45032593607902527,
      "eval_runtime": 2.5682,
      "eval_samples_per_second": 890.103,
      "eval_steps_per_second": 111.36,
      "step": 3000
    },
    {
      "epoch": 4.66,
      "eval_accuracy": 0.8005249500274658,
      "eval_f1": 0.8059574468085107,
      "eval_loss": 0.4483606219291687,
      "eval_runtime": 2.5703,
      "eval_samples_per_second": 889.378,
      "eval_steps_per_second": 111.269,
      "step": 3200
    },
    {
      "epoch": 4.95,
      "eval_accuracy": 0.7948381304740906,
      "eval_f1": 0.7934830471158081,
      "eval_loss": 0.456741064786911,
      "eval_runtime": 2.5765,
      "eval_samples_per_second": 887.256,
      "eval_steps_per_second": 111.004,
      "step": 3400
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.452692867540029e-05,
      "loss": 0.4153,
      "step": 3500
    },
    {
      "epoch": 5.24,
      "eval_accuracy": 0.7970253825187683,
      "eval_f1": 0.8008583690987126,
      "eval_loss": 0.4510791301727295,
      "eval_runtime": 2.5665,
      "eval_samples_per_second": 890.719,
      "eval_steps_per_second": 111.437,
      "step": 3600
    },
    {
      "epoch": 5.53,
      "eval_accuracy": 0.7957130074501038,
      "eval_f1": 0.797044763146458,
      "eval_loss": 0.44894999265670776,
      "eval_runtime": 2.5695,
      "eval_samples_per_second": 889.652,
      "eval_steps_per_second": 111.304,
      "step": 3800
    },
    {
      "epoch": 5.82,
      "learning_rate": 2.088791848617176e-05,
      "loss": 0.4055,
      "step": 4000
    },
    {
      "epoch": 5.82,
      "eval_accuracy": 0.7939632534980774,
      "eval_f1": 0.7954841511072515,
      "eval_loss": 0.45642799139022827,
      "eval_runtime": 2.5702,
      "eval_samples_per_second": 889.408,
      "eval_steps_per_second": 111.273,
      "step": 4000
    },
    {
      "epoch": 6.11,
      "eval_accuracy": 0.7992125749588013,
      "eval_f1": 0.8040973111395647,
      "eval_loss": 0.4486392140388489,
      "eval_runtime": 2.5739,
      "eval_samples_per_second": 888.144,
      "eval_steps_per_second": 111.115,
      "step": 4200
    },
    {
      "epoch": 6.4,
      "eval_accuracy": 0.8018372654914856,
      "eval_f1": 0.8074798130046749,
      "eval_loss": 0.44145750999450684,
      "eval_runtime": 2.5725,
      "eval_samples_per_second": 888.634,
      "eval_steps_per_second": 111.176,
      "step": 4400
    },
    {
      "epoch": 6.55,
      "learning_rate": 1.7248908296943234e-05,
      "loss": 0.3918,
      "step": 4500
    },
    {
      "epoch": 6.7,
      "eval_accuracy": 0.7970253825187683,
      "eval_f1": 0.7980852915578764,
      "eval_loss": 0.4575858414173126,
      "eval_runtime": 2.5712,
      "eval_samples_per_second": 889.092,
      "eval_steps_per_second": 111.234,
      "step": 4600
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8018372654914856,
      "eval_f1": 0.8084566596194505,
      "eval_loss": 0.4492335319519043,
      "eval_runtime": 2.5716,
      "eval_samples_per_second": 888.937,
      "eval_steps_per_second": 111.214,
      "step": 4800
    },
    {
      "epoch": 7.28,
      "learning_rate": 1.3609898107714703e-05,
      "loss": 0.3772,
      "step": 5000
    },
    {
      "epoch": 7.28,
      "eval_accuracy": 0.7961505055427551,
      "eval_f1": 0.7963286713286714,
      "eval_loss": 0.4673013389110565,
      "eval_runtime": 2.5753,
      "eval_samples_per_second": 887.674,
      "eval_steps_per_second": 111.056,
      "step": 5000
    },
    {
      "epoch": 7.57,
      "eval_accuracy": 0.7996500730514526,
      "eval_f1": 0.7994746059544658,
      "eval_loss": 0.4637942910194397,
      "eval_runtime": 2.5678,
      "eval_samples_per_second": 890.24,
      "eval_steps_per_second": 111.377,
      "step": 5200
    },
    {
      "epoch": 7.86,
      "eval_accuracy": 0.7974628210067749,
      "eval_f1": 0.8013728013728013,
      "eval_loss": 0.45550382137298584,
      "eval_runtime": 2.5738,
      "eval_samples_per_second": 888.174,
      "eval_steps_per_second": 111.119,
      "step": 5400
    },
    {
      "epoch": 8.01,
      "learning_rate": 9.970887918486172e-06,
      "loss": 0.377,
      "step": 5500
    },
    {
      "epoch": 8.15,
      "eval_accuracy": 0.7961505055427551,
      "eval_f1": 0.7991379310344828,
      "eval_loss": 0.4626655578613281,
      "eval_runtime": 2.5693,
      "eval_samples_per_second": 889.72,
      "eval_steps_per_second": 111.312,
      "step": 5600
    },
    {
      "epoch": 8.44,
      "eval_accuracy": 0.7970253825187683,
      "eval_f1": 0.8017094017094017,
      "eval_loss": 0.4536679685115814,
      "eval_runtime": 2.5725,
      "eval_samples_per_second": 888.631,
      "eval_steps_per_second": 111.176,
      "step": 5800
    },
    {
      "epoch": 8.73,
      "learning_rate": 6.3318777292576415e-06,
      "loss": 0.3707,
      "step": 6000
    },
    {
      "epoch": 8.73,
      "eval_accuracy": 0.7970253825187683,
      "eval_f1": 0.7994814174589455,
      "eval_loss": 0.45931074023246765,
      "eval_runtime": 2.5683,
      "eval_samples_per_second": 890.099,
      "eval_steps_per_second": 111.36,
      "step": 6000
    },
    {
      "epoch": 9.02,
      "eval_accuracy": 0.7965879440307617,
      "eval_f1": 0.7975620374401393,
      "eval_loss": 0.46558013558387756,
      "eval_runtime": 2.5731,
      "eval_samples_per_second": 888.418,
      "eval_steps_per_second": 111.149,
      "step": 6200
    },
    {
      "epoch": 9.32,
      "eval_accuracy": 0.7965879440307617,
      "eval_f1": 0.8,
      "eval_loss": 0.4601030945777893,
      "eval_runtime": 2.5774,
      "eval_samples_per_second": 886.945,
      "eval_steps_per_second": 110.965,
      "step": 6400
    },
    {
      "epoch": 9.46,
      "learning_rate": 2.692867540029112e-06,
      "loss": 0.366,
      "step": 6500
    },
    {
      "epoch": 9.61,
      "eval_accuracy": 0.7961505055427551,
      "eval_f1": 0.7989646246764452,
      "eval_loss": 0.45923250913619995,
      "eval_runtime": 2.5827,
      "eval_samples_per_second": 885.108,
      "eval_steps_per_second": 110.735,
      "step": 6600
    },
    {
      "epoch": 9.9,
      "eval_accuracy": 0.7974628210067749,
      "eval_f1": 0.8015430775825118,
      "eval_loss": 0.4570430517196655,
      "eval_runtime": 2.5678,
      "eval_samples_per_second": 890.266,
      "eval_steps_per_second": 111.381,
      "step": 6800
    },
    {
      "epoch": 10.0,
      "step": 6870,
      "total_flos": 1.022590196992512e+16,
      "train_loss": 0.4268792453638157,
      "train_runtime": 1110.0444,
      "train_samples_per_second": 395.633,
      "train_steps_per_second": 6.189
    }
  ],
  "max_steps": 6870,
  "num_train_epochs": 10,
  "total_flos": 1.022590196992512e+16,
  "trial_name": null,
  "trial_params": null
}