{
  "best_metric": 0.6465878115264797,
  "best_model_checkpoint": "/xdisk/msurdeanu/enoriega/kw_pubmed_experiments/kw_pubmed_vanilla_sentence_10000_0.0003/checkpoint-100",
  "epoch": 1.3559903108724085,
  "global_step": 350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.00029998499999999995,
      "loss": 2.5567,
      "step": 5
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029997,
      "loss": 1.9715,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029995499999999997,
      "loss": 1.8692,
      "step": 15
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029994,
      "loss": 1.8491,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.000299925,
      "loss": 1.8861,
      "step": 25
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.6368282710280374,
      "eval_loss": 1.842035174369812,
      "eval_runtime": 60.5739,
      "eval_samples_per_second": 165.088,
      "eval_steps_per_second": 10.318,
      "step": 25
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029991299999999996,
      "loss": 1.8289,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.000299898,
      "loss": 1.813,
      "step": 35
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.000299883,
      "loss": 1.8096,
      "step": 40
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00029986799999999996,
      "loss": 1.8019,
      "step": 45
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000299853,
      "loss": 1.7913,
      "step": 50
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.6451718892607241,
      "eval_loss": 1.7883105278015137,
      "eval_runtime": 60.6012,
      "eval_samples_per_second": 165.013,
      "eval_steps_per_second": 10.313,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00029983799999999997,
      "loss": 1.7907,
      "step": 55
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.000299823,
      "loss": 1.7853,
      "step": 60
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000299808,
      "loss": 1.7915,
      "step": 65
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00029979299999999997,
      "loss": 1.7815,
      "step": 70
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000299778,
      "loss": 1.7876,
      "step": 75
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.6446729540614542,
      "eval_loss": 1.7794435024261475,
      "eval_runtime": 61.0108,
      "eval_samples_per_second": 163.905,
      "eval_steps_per_second": 10.244,
      "step": 75
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000299763,
      "loss": 1.7707,
      "step": 80
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00029974799999999996,
      "loss": 1.7773,
      "step": 85
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000299733,
      "loss": 1.7759,
      "step": 90
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000299718,
      "loss": 1.7662,
      "step": 95
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00029970299999999996,
      "loss": 1.7729,
      "step": 100
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.6465878115264797,
      "eval_loss": 1.7647390365600586,
      "eval_runtime": 60.7957,
      "eval_samples_per_second": 164.485,
      "eval_steps_per_second": 10.28,
      "step": 100
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00029968799999999994,
      "loss": 1.7748,
      "step": 105
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.000299673,
      "loss": 1.7706,
      "step": 110
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00029965799999999996,
      "loss": 1.785,
      "step": 115
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.000299643,
      "loss": 1.778,
      "step": 120
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029962799999999997,
      "loss": 1.7919,
      "step": 125
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.6427293636595962,
      "eval_loss": 1.7991459369659424,
      "eval_runtime": 60.6682,
      "eval_samples_per_second": 164.831,
      "eval_steps_per_second": 10.302,
      "step": 125
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00029961299999999995,
      "loss": 1.8059,
      "step": 130
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.000299598,
      "loss": 1.7882,
      "step": 135
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00029958299999999997,
      "loss": 2.0033,
      "step": 140
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00029956799999999995,
      "loss": 1.9014,
      "step": 145
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.000299553,
      "loss": 1.858,
      "step": 150
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.6386691654294546,
      "eval_loss": 1.8227039575576782,
      "eval_runtime": 60.7206,
      "eval_samples_per_second": 164.689,
      "eval_steps_per_second": 10.293,
      "step": 150
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00029953799999999996,
      "loss": 1.8575,
      "step": 155
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.000299523,
      "loss": 1.8191,
      "step": 160
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00029951099999999997,
      "loss": 1.8317,
      "step": 165
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00029949599999999995,
      "loss": 1.8389,
      "step": 170
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000299481,
      "loss": 1.8151,
      "step": 175
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.632689569566646,
      "eval_loss": 1.8751027584075928,
      "eval_runtime": 62.2872,
      "eval_samples_per_second": 160.547,
      "eval_steps_per_second": 10.034,
      "step": 175
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00029946599999999997,
      "loss": 1.8842,
      "step": 180
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00029945099999999995,
      "loss": 1.8344,
      "step": 185
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.000299436,
      "loss": 1.8594,
      "step": 190
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00029942099999999996,
      "loss": 1.9278,
      "step": 195
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.000299406,
      "loss": 1.8539,
      "step": 200
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.6372984484332218,
      "eval_loss": 1.8315496444702148,
      "eval_runtime": 61.0171,
      "eval_samples_per_second": 163.888,
      "eval_steps_per_second": 10.243,
      "step": 200
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000299391,
      "loss": 1.8632,
      "step": 205
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00029937599999999996,
      "loss": 1.907,
      "step": 210
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.000299361,
      "loss": 1.8705,
      "step": 215
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.000299346,
      "loss": 1.8825,
      "step": 220
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.000299331,
      "loss": 1.8623,
      "step": 225
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.6348281107392759,
      "eval_loss": 1.846932053565979,
      "eval_runtime": 60.9038,
      "eval_samples_per_second": 164.193,
      "eval_steps_per_second": 10.262,
      "step": 225
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.000299316,
      "loss": 1.8923,
      "step": 230
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00029930099999999997,
      "loss": 1.852,
      "step": 235
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00029928599999999995,
      "loss": 1.8754,
      "step": 240
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00029927099999999993,
      "loss": 1.9014,
      "step": 245
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00029925599999999997,
      "loss": 1.8992,
      "step": 250
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.6353137321715426,
      "eval_loss": 1.846245288848877,
      "eval_runtime": 60.8105,
      "eval_samples_per_second": 164.445,
      "eval_steps_per_second": 10.278,
      "step": 250
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00029924099999999995,
      "loss": 1.8759,
      "step": 255
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.000299226,
      "loss": 2.0852,
      "step": 260
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00029921099999999996,
      "loss": 1.8663,
      "step": 265
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00029919599999999995,
      "loss": 1.8555,
      "step": 270
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.000299181,
      "loss": 1.8717,
      "step": 275
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.6298189252336449,
      "eval_loss": 1.8705145120620728,
      "eval_runtime": 60.9334,
      "eval_samples_per_second": 164.114,
      "eval_steps_per_second": 10.257,
      "step": 275
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00029916599999999996,
      "loss": 1.9226,
      "step": 280
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.000299151,
      "loss": 2.0094,
      "step": 285
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.000299136,
      "loss": 1.9307,
      "step": 290
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00029912099999999996,
      "loss": 1.9055,
      "step": 295
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.000299106,
      "loss": 1.9057,
      "step": 300
    },
    {
      "epoch": 1.16,
      "eval_accuracy": 0.6247444509345794,
      "eval_loss": 1.913194179534912,
      "eval_runtime": 60.8541,
      "eval_samples_per_second": 164.327,
      "eval_steps_per_second": 10.27,
      "step": 300
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00029909099999999997,
      "loss": 1.8821,
      "step": 305
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.000299076,
      "loss": 1.8635,
      "step": 310
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.000299061,
      "loss": 1.8655,
      "step": 315
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00029904599999999997,
      "loss": 1.8642,
      "step": 320
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.000299031,
      "loss": 1.8637,
      "step": 325
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.6270322235202492,
      "eval_loss": 1.900992751121521,
      "eval_runtime": 60.9918,
      "eval_samples_per_second": 163.957,
      "eval_steps_per_second": 10.247,
      "step": 325
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.000299016,
      "loss": 1.9827,
      "step": 330
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00029900099999999997,
      "loss": 1.95,
      "step": 335
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00029898599999999995,
      "loss": 1.9358,
      "step": 340
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.000298971,
      "loss": 1.9403,
      "step": 345
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00029895599999999996,
      "loss": 1.947,
      "step": 350
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.6236979166666666,
      "eval_loss": 1.903308629989624,
      "eval_runtime": 60.9439,
      "eval_samples_per_second": 164.085,
      "eval_steps_per_second": 10.255,
      "step": 350
    },
    {
      "epoch": 1.36,
      "step": 350,
      "total_flos": 1.899141968469924e+17,
      "train_loss": 1.8697899763924735,
      "train_runtime": 46350.9183,
      "train_samples_per_second": 17259.636,
      "train_steps_per_second": 2.157
    },
    {
      "epoch": 1.36,
      "eval_accuracy": 0.6470223174085522,
      "eval_loss": 1.767507791519165,
      "eval_runtime": 61.0991,
      "eval_samples_per_second": 163.669,
      "eval_steps_per_second": 10.229,
      "step": 350
    }
  ],
  "max_steps": 100000,
  "num_train_epochs": 388,
  "total_flos": 1.899141968469924e+17,
  "trial_name": null,
  "trial_params": null
}