{
  "best_metric": 6.402657508850098,
  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/lm-training/src/outputs/2023-11-15/09-23-00/checkpoints/checkpoint-250",
  "epoch": 2.0,
  "eval_steps": 10,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 4.866666666666667e-05,
      "loss": 9.9717,
      "step": 10
    },
    {
      "epoch": 0.08,
      "eval_loss": 9.324381828308105,
      "eval_runtime": 2.0042,
      "eval_samples_per_second": 498.949,
      "eval_steps_per_second": 62.369,
      "step": 10
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.7333333333333336e-05,
      "loss": 9.1002,
      "step": 20
    },
    {
      "epoch": 0.16,
      "eval_loss": 8.892333984375,
      "eval_runtime": 1.9424,
      "eval_samples_per_second": 514.829,
      "eval_steps_per_second": 64.354,
      "step": 20
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.600000000000001e-05,
      "loss": 8.7625,
      "step": 30
    },
    {
      "epoch": 0.24,
      "eval_loss": 8.54922103881836,
      "eval_runtime": 1.9052,
      "eval_samples_per_second": 524.878,
      "eval_steps_per_second": 65.61,
      "step": 30
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.466666666666667e-05,
      "loss": 8.4987,
      "step": 40
    },
    {
      "epoch": 0.32,
      "eval_loss": 8.234408378601074,
      "eval_runtime": 2.0895,
      "eval_samples_per_second": 478.585,
      "eval_steps_per_second": 59.823,
      "step": 40
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 8.2485,
      "step": 50
    },
    {
      "epoch": 0.4,
      "eval_loss": 7.941398620605469,
      "eval_runtime": 2.1108,
      "eval_samples_per_second": 473.76,
      "eval_steps_per_second": 59.22,
      "step": 50
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.2e-05,
      "loss": 8.0802,
      "step": 60
    },
    {
      "epoch": 0.48,
      "eval_loss": 7.663424968719482,
      "eval_runtime": 1.991,
      "eval_samples_per_second": 502.251,
      "eval_steps_per_second": 62.781,
      "step": 60
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.066666666666667e-05,
      "loss": 7.7924,
      "step": 70
    },
    {
      "epoch": 0.56,
      "eval_loss": 7.419986724853516,
      "eval_runtime": 1.9753,
      "eval_samples_per_second": 506.241,
      "eval_steps_per_second": 63.28,
      "step": 70
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.933333333333333e-05,
      "loss": 7.5139,
      "step": 80
    },
    {
      "epoch": 0.64,
      "eval_loss": 7.188984394073486,
      "eval_runtime": 1.9721,
      "eval_samples_per_second": 507.08,
      "eval_steps_per_second": 63.385,
      "step": 80
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.8e-05,
      "loss": 7.3932,
      "step": 90
    },
    {
      "epoch": 0.72,
      "eval_loss": 7.042924880981445,
      "eval_runtime": 2.0739,
      "eval_samples_per_second": 482.185,
      "eval_steps_per_second": 60.273,
      "step": 90
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 7.334,
      "step": 100
    },
    {
      "epoch": 0.8,
      "eval_loss": 6.937138557434082,
      "eval_runtime": 2.0003,
      "eval_samples_per_second": 499.914,
      "eval_steps_per_second": 62.489,
      "step": 100
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.5333333333333336e-05,
      "loss": 7.1699,
      "step": 110
    },
    {
      "epoch": 0.88,
      "eval_loss": 6.849837779998779,
      "eval_runtime": 2.0147,
      "eval_samples_per_second": 496.356,
      "eval_steps_per_second": 62.045,
      "step": 110
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 7.2476,
      "step": 120
    },
    {
      "epoch": 0.96,
      "eval_loss": 6.779170036315918,
      "eval_runtime": 2.0584,
      "eval_samples_per_second": 485.82,
      "eval_steps_per_second": 60.727,
      "step": 120
    },
    {
      "epoch": 1.04,
      "learning_rate": 3.266666666666667e-05,
      "loss": 6.8774,
      "step": 130
    },
    {
      "epoch": 1.04,
      "eval_loss": 6.709743022918701,
      "eval_runtime": 2.4463,
      "eval_samples_per_second": 408.786,
      "eval_steps_per_second": 51.098,
      "step": 130
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.1333333333333334e-05,
      "loss": 6.7219,
      "step": 140
    },
    {
      "epoch": 1.12,
      "eval_loss": 6.656557559967041,
      "eval_runtime": 2.0119,
      "eval_samples_per_second": 497.045,
      "eval_steps_per_second": 62.131,
      "step": 140
    },
    {
      "epoch": 1.2,
      "learning_rate": 3e-05,
      "loss": 6.6353,
      "step": 150
    },
    {
      "epoch": 1.2,
      "eval_loss": 6.612905502319336,
      "eval_runtime": 2.0567,
      "eval_samples_per_second": 486.217,
      "eval_steps_per_second": 60.777,
      "step": 150
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.8666666666666668e-05,
      "loss": 6.6524,
      "step": 160
    },
    {
      "epoch": 1.28,
      "eval_loss": 6.580767631530762,
      "eval_runtime": 1.9849,
      "eval_samples_per_second": 503.794,
      "eval_steps_per_second": 62.974,
      "step": 160
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.733333333333333e-05,
      "loss": 6.484,
      "step": 170
    },
    {
      "epoch": 1.36,
      "eval_loss": 6.5504679679870605,
      "eval_runtime": 2.1014,
      "eval_samples_per_second": 475.877,
      "eval_steps_per_second": 59.485,
      "step": 170
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 6.5846,
      "step": 180
    },
    {
      "epoch": 1.44,
      "eval_loss": 6.531940460205078,
      "eval_runtime": 3.1229,
      "eval_samples_per_second": 320.22,
      "eval_steps_per_second": 40.027,
      "step": 180
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.466666666666667e-05,
      "loss": 6.558,
      "step": 190
    },
    {
      "epoch": 1.52,
      "eval_loss": 6.510298728942871,
      "eval_runtime": 1.9686,
      "eval_samples_per_second": 507.978,
      "eval_steps_per_second": 63.497,
      "step": 190
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 6.4792,
      "step": 200
    },
    {
      "epoch": 1.6,
      "eval_loss": 6.493106365203857,
      "eval_runtime": 1.8922,
      "eval_samples_per_second": 528.475,
      "eval_steps_per_second": 66.059,
      "step": 200
    },
    {
      "epoch": 1.68,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 6.5203,
      "step": 210
    },
    {
      "epoch": 1.68,
      "eval_loss": 6.469159126281738,
      "eval_runtime": 2.0323,
      "eval_samples_per_second": 492.062,
      "eval_steps_per_second": 61.508,
      "step": 210
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.0666666666666666e-05,
      "loss": 6.6041,
      "step": 220
    },
    {
      "epoch": 1.76,
      "eval_loss": 6.450395107269287,
      "eval_runtime": 3.3679,
      "eval_samples_per_second": 296.917,
      "eval_steps_per_second": 37.115,
      "step": 220
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 6.646,
      "step": 230
    },
    {
      "epoch": 1.84,
      "eval_loss": 6.437198638916016,
      "eval_runtime": 2.0292,
      "eval_samples_per_second": 492.807,
      "eval_steps_per_second": 61.601,
      "step": 230
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.8e-05,
      "loss": 6.5665,
      "step": 240
    },
    {
      "epoch": 1.92,
      "eval_loss": 6.420088768005371,
      "eval_runtime": 2.0424,
      "eval_samples_per_second": 489.62,
      "eval_steps_per_second": 61.202,
      "step": 240
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 6.6006,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_loss": 6.402657508850098,
      "eval_runtime": 1.8829,
      "eval_samples_per_second": 531.107,
      "eval_steps_per_second": 66.388,
      "step": 250
    }
  ],
  "logging_steps": 10,
  "max_steps": 375,
  "num_train_epochs": 3,
  "save_steps": 10,
  "total_flos": 29623007232000.0,
  "trial_name": null,
  "trial_params": null
}