{
  "best_metric": 0.735632183908046,
  "best_model_checkpoint": "/xdisk/msurdeanu/enoriega/kw_pubmed_experiments/kw_pubmed_keyword_sentence_10000_0.0003/checkpoint-25",
  "epoch": 1.06578081831338,
  "global_step": 275,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.000299991,
      "loss": 2.0241,
      "step": 5
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000299976,
      "loss": 1.3033,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000299961,
      "loss": 1.5029,
      "step": 15
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.000299946,
      "loss": 1.2483,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029993099999999997,
      "loss": 1.4697,
      "step": 25
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.735632183908046,
      "eval_loss": NaN,
      "eval_runtime": 59.6178,
      "eval_samples_per_second": 167.735,
      "eval_steps_per_second": 10.483,
      "step": 25
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.000299916,
      "loss": 1.2378,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029990099999999993,
      "loss": 1.7161,
      "step": 35
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029988599999999997,
      "loss": 2.0569,
      "step": 40
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00029987099999999995,
      "loss": 2.1263,
      "step": 45
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000299856,
      "loss": 2.5526,
      "step": 50
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.723092998955068,
      "eval_loss": NaN,
      "eval_runtime": 59.378,
      "eval_samples_per_second": 168.412,
      "eval_steps_per_second": 10.526,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00029984099999999996,
      "loss": 1.2865,
      "step": 55
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.000299829,
      "loss": 2.0071,
      "step": 60
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00029981399999999997,
      "loss": 2.3439,
      "step": 65
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000299799,
      "loss": 1.5348,
      "step": 70
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00029978399999999993,
      "loss": 2.3377,
      "step": 75
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.7082027168234065,
      "eval_loss": NaN,
      "eval_runtime": 59.6718,
      "eval_samples_per_second": 167.583,
      "eval_steps_per_second": 10.474,
      "step": 75
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00029976899999999997,
      "loss": 1.7853,
      "step": 80
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00029975399999999995,
      "loss": 1.9305,
      "step": 85
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000299739,
      "loss": 1.9862,
      "step": 90
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00029972399999999996,
      "loss": 3.8541,
      "step": 95
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00029970899999999995,
      "loss": 2.9479,
      "step": 100
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.6987983281086729,
      "eval_loss": NaN,
      "eval_runtime": 59.6999,
      "eval_samples_per_second": 167.505,
      "eval_steps_per_second": 10.469,
      "step": 100
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.000299694,
      "loss": 2.1467,
      "step": 105
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00029967899999999996,
      "loss": 2.8996,
      "step": 110
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.000299664,
      "loss": 3.0211,
      "step": 115
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.000299649,
      "loss": 2.2422,
      "step": 120
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029963399999999996,
      "loss": 1.4212,
      "step": 125
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.705067920585162,
      "eval_loss": NaN,
      "eval_runtime": 59.6589,
      "eval_samples_per_second": 167.62,
      "eval_steps_per_second": 10.476,
      "step": 125
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.000299619,
      "loss": 1.8562,
      "step": 130
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00029960399999999997,
      "loss": 2.0218,
      "step": 135
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00029958899999999995,
      "loss": 1.331,
      "step": 140
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.000299574,
      "loss": 1.9291,
      "step": 145
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00029955899999999997,
      "loss": 1.8373,
      "step": 150
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.7189132706374086,
      "eval_loss": NaN,
      "eval_runtime": 59.6305,
      "eval_samples_per_second": 167.699,
      "eval_steps_per_second": 10.481,
      "step": 150
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.000299544,
      "loss": 1.9669,
      "step": 155
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.000299529,
      "loss": 1.8719,
      "step": 160
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00029951399999999997,
      "loss": 2.4312,
      "step": 165
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.000299499,
      "loss": 3.1182,
      "step": 170
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000299484,
      "loss": 1.5001,
      "step": 175
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.6812957157784744,
      "eval_loss": NaN,
      "eval_runtime": 59.7742,
      "eval_samples_per_second": 167.296,
      "eval_steps_per_second": 10.456,
      "step": 175
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00029946899999999996,
      "loss": 1.493,
      "step": 180
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00029945399999999994,
      "loss": 1.5281,
      "step": 185
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.000299439,
      "loss": 1.6081,
      "step": 190
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00029942399999999996,
      "loss": 1.4183,
      "step": 195
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.000299409,
      "loss": 1.9615,
      "step": 200
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.7055903866248694,
      "eval_loss": NaN,
      "eval_runtime": 59.6776,
      "eval_samples_per_second": 167.567,
      "eval_steps_per_second": 10.473,
      "step": 200
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000299394,
      "loss": 1.7213,
      "step": 205
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00029937899999999995,
      "loss": 1.5657,
      "step": 210
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.000299364,
      "loss": 1.6782,
      "step": 215
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00029934899999999997,
      "loss": 2.0081,
      "step": 220
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00029933399999999995,
      "loss": 1.4458,
      "step": 225
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.6974921630094044,
      "eval_loss": NaN,
      "eval_runtime": 59.7906,
      "eval_samples_per_second": 167.25,
      "eval_steps_per_second": 10.453,
      "step": 225
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.000299322,
      "loss": 1.7357,
      "step": 230
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00029930699999999996,
      "loss": 4.7272,
      "step": 235
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00029929199999999994,
      "loss": 1.6994,
      "step": 240
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.000299277,
      "loss": 1.6879,
      "step": 245
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00029926199999999995,
      "loss": 1.4237,
      "step": 250
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.7087251828631139,
      "eval_loss": NaN,
      "eval_runtime": 59.7381,
      "eval_samples_per_second": 167.397,
      "eval_steps_per_second": 10.462,
      "step": 250
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.000299247,
      "loss": 1.5802,
      "step": 255
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00029923199999999997,
      "loss": 2.2637,
      "step": 260
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00029921699999999995,
      "loss": 1.4615,
      "step": 265
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.000299202,
      "loss": 1.6544,
      "step": 270
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00029918699999999997,
      "loss": 1.6758,
      "step": 275
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.6922675026123302,
      "eval_loss": NaN,
      "eval_runtime": 59.6857,
      "eval_samples_per_second": 167.544,
      "eval_steps_per_second": 10.472,
      "step": 275
    },
    {
      "epoch": 1.07,
      "step": 275,
      "total_flos": 1.493606932357716e+17,
      "train_loss": 1.9488013319535689,
      "train_runtime": 36132.536,
      "train_samples_per_second": 22140.71,
      "train_steps_per_second": 2.768
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.7343260188087775,
      "eval_loss": NaN,
      "eval_runtime": 59.7956,
      "eval_samples_per_second": 167.236,
      "eval_steps_per_second": 10.452,
      "step": 275
    }
  ],
  "max_steps": 100000,
  "num_train_epochs": 388,
  "total_flos": 1.493606932357716e+17,
  "trial_name": null,
  "trial_params": null
}