{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.18099547511312217,
  "eval_steps": 20,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 2.4375,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 2.5231,
      "step": 20
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.4237778186798096,
      "eval_runtime": 171.986,
      "eval_samples_per_second": 45.69,
      "eval_steps_per_second": 5.716,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.999999999999999e-05,
      "loss": 2.3284,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00011999999999999999,
      "loss": 1.8222,
      "step": 40
    },
    {
      "epoch": 0.02,
      "eval_loss": 1.525316596031189,
      "eval_runtime": 172.7335,
      "eval_samples_per_second": 45.492,
      "eval_steps_per_second": 5.691,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00015,
      "loss": 1.025,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00017999999999999998,
      "loss": 1.0834,
      "step": 60
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.8071926832199097,
      "eval_runtime": 172.0838,
      "eval_samples_per_second": 45.664,
      "eval_steps_per_second": 5.712,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.8541,
      "step": 70
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.7104,
      "step": 80
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.7147021889686584,
      "eval_runtime": 172.8542,
      "eval_samples_per_second": 45.46,
      "eval_steps_per_second": 5.687,
      "step": 80
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00027,
      "loss": 0.6001,
      "step": 90
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0003,
      "loss": 0.4885,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.70524662733078,
      "eval_runtime": 172.2659,
      "eval_samples_per_second": 45.616,
      "eval_steps_per_second": 5.706,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029,
      "loss": 0.9048,
      "step": 110
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00028,
      "loss": 0.7366,
      "step": 120
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.6695398688316345,
      "eval_runtime": 172.5803,
      "eval_samples_per_second": 45.532,
      "eval_steps_per_second": 5.696,
      "step": 120
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00027,
      "loss": 0.6551,
      "step": 130
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00026,
      "loss": 0.5573,
      "step": 140
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.6578989624977112,
      "eval_runtime": 172.8835,
      "eval_samples_per_second": 45.453,
      "eval_steps_per_second": 5.686,
      "step": 140
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00025,
      "loss": 0.4508,
      "step": 150
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.8456,
      "step": 160
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.6626420021057129,
      "eval_runtime": 172.8197,
      "eval_samples_per_second": 45.469,
      "eval_steps_per_second": 5.688,
      "step": 160
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00023,
      "loss": 0.7237,
      "step": 170
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00021999999999999995,
      "loss": 0.6205,
      "step": 180
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.63532555103302,
      "eval_runtime": 172.3005,
      "eval_samples_per_second": 45.606,
      "eval_steps_per_second": 5.705,
      "step": 180
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.5346,
      "step": 190
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019999999999999998,
      "loss": 0.4467,
      "step": 200
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.6415661573410034,
      "eval_runtime": 172.5689,
      "eval_samples_per_second": 45.535,
      "eval_steps_per_second": 5.696,
      "step": 200
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00018999999999999998,
      "loss": 0.8274,
      "step": 210
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.689,
      "step": 220
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.6247297525405884,
      "eval_runtime": 172.7326,
      "eval_samples_per_second": 45.492,
      "eval_steps_per_second": 5.691,
      "step": 220
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00016999999999999999,
      "loss": 0.6195,
      "step": 230
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00015999999999999999,
      "loss": 0.5187,
      "step": 240
    },
    {
      "epoch": 0.11,
      "eval_loss": 0.6187988519668579,
      "eval_runtime": 172.5129,
      "eval_samples_per_second": 45.55,
      "eval_steps_per_second": 5.698,
      "step": 240
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00015,
      "loss": 0.4326,
      "step": 250
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00014,
      "loss": 0.8144,
      "step": 260
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.6190418004989624,
      "eval_runtime": 173.1094,
      "eval_samples_per_second": 45.393,
      "eval_steps_per_second": 5.678,
      "step": 260
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00013,
      "loss": 0.6642,
      "step": 270
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.5946,
      "step": 280
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.6081172227859497,
      "eval_runtime": 172.463,
      "eval_samples_per_second": 45.563,
      "eval_steps_per_second": 5.7,
      "step": 280
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00010999999999999998,
      "loss": 0.515,
      "step": 290
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.999999999999999e-05,
      "loss": 0.4127,
      "step": 300
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.622405469417572,
      "eval_runtime": 172.9603,
      "eval_samples_per_second": 45.432,
      "eval_steps_per_second": 5.683,
      "step": 300
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.8053,
      "step": 310
    },
    {
      "epoch": 0.14,
      "learning_rate": 7.999999999999999e-05,
      "loss": 0.6842,
      "step": 320
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.6062848567962646,
      "eval_runtime": 172.4692,
      "eval_samples_per_second": 45.562,
      "eval_steps_per_second": 5.7,
      "step": 320
    },
    {
      "epoch": 0.15,
      "learning_rate": 7e-05,
      "loss": 0.5868,
      "step": 330
    },
    {
      "epoch": 0.15,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.5091,
      "step": 340
    },
    {
      "epoch": 0.15,
      "eval_loss": 0.6016470193862915,
      "eval_runtime": 172.4602,
      "eval_samples_per_second": 45.564,
      "eval_steps_per_second": 5.7,
      "step": 340
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 0.4201,
      "step": 350
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.9999999999999996e-05,
      "loss": 0.7834,
      "step": 360
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.6014373302459717,
      "eval_runtime": 172.2045,
      "eval_samples_per_second": 45.632,
      "eval_steps_per_second": 5.708,
      "step": 360
    },
    {
      "epoch": 0.17,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.6635,
      "step": 370
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 0.5875,
      "step": 380
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.5969696044921875,
      "eval_runtime": 172.1021,
      "eval_samples_per_second": 45.659,
      "eval_steps_per_second": 5.712,
      "step": 380
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.999999999999999e-06,
      "loss": 0.4966,
      "step": 390
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0,
      "loss": 0.4273,
      "step": 400
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.5965555906295776,
      "eval_runtime": 172.4595,
      "eval_samples_per_second": 45.564,
      "eval_steps_per_second": 5.7,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_train_epochs": 1,
  "save_steps": 20,
  "total_flos": 1.346139374616576e+16,
  "trial_name": null,
  "trial_params": null
}