{
  "best_metric": 0.025097770616412163,
  "best_model_checkpoint": "./results_pnum/results_cvrev_pnum_f4_large_b4e15_5000/checkpoint-10000",
  "epoch": 10.26694045174538,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "learning_rate": 4.914946777561009e-05,
      "loss": 0.5832,
      "step": 500
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.829380155388986e-05,
      "loss": 0.0339,
      "step": 1000
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.743813533216963e-05,
      "loss": 0.0314,
      "step": 1500
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.65824691104494e-05,
      "loss": 0.0301,
      "step": 2000
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.572680288872916e-05,
      "loss": 0.0292,
      "step": 2500
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.487113666700893e-05,
      "loss": 0.029,
      "step": 3000
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.40154704452887e-05,
      "loss": 0.0278,
      "step": 3500
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.3159804223568475e-05,
      "loss": 0.0273,
      "step": 4000
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.2304138001848244e-05,
      "loss": 0.0261,
      "step": 4500
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.144847178012801e-05,
      "loss": 0.0266,
      "step": 5000
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.025240449234843254,
      "eval_runtime": 181.3134,
      "eval_samples_per_second": 14.307,
      "eval_steps_per_second": 3.579,
      "step": 5000
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.0592805558407774e-05,
      "loss": 0.026,
      "step": 5500
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.973713933668754e-05,
      "loss": 0.0265,
      "step": 6000
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.888147311496732e-05,
      "loss": 0.0248,
      "step": 6500
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.8025806893247086e-05,
      "loss": 0.0247,
      "step": 7000
    },
    {
      "epoch": 3.85,
      "learning_rate": 3.7170140671526854e-05,
      "loss": 0.0245,
      "step": 7500
    },
    {
      "epoch": 4.11,
      "learning_rate": 3.631447444980662e-05,
      "loss": 0.0241,
      "step": 8000
    },
    {
      "epoch": 4.36,
      "learning_rate": 3.5458808228086384e-05,
      "loss": 0.0226,
      "step": 8500
    },
    {
      "epoch": 4.62,
      "learning_rate": 3.460314200636616e-05,
      "loss": 0.0232,
      "step": 9000
    },
    {
      "epoch": 4.88,
      "learning_rate": 3.374747578464593e-05,
      "loss": 0.0232,
      "step": 9500
    },
    {
      "epoch": 5.13,
      "learning_rate": 3.2891809562925697e-05,
      "loss": 0.0215,
      "step": 10000
    },
    {
      "epoch": 5.13,
      "eval_loss": 0.025097770616412163,
      "eval_runtime": 120.6593,
      "eval_samples_per_second": 21.499,
      "eval_steps_per_second": 5.379,
      "step": 10000
    },
    {
      "epoch": 5.39,
      "learning_rate": 3.2036143341205465e-05,
      "loss": 0.0211,
      "step": 10500
    },
    {
      "epoch": 5.65,
      "learning_rate": 3.1180477119485233e-05,
      "loss": 0.0211,
      "step": 11000
    },
    {
      "epoch": 5.9,
      "learning_rate": 3.0324810897765e-05,
      "loss": 0.0216,
      "step": 11500
    },
    {
      "epoch": 6.16,
      "learning_rate": 2.946914467604477e-05,
      "loss": 0.0201,
      "step": 12000
    },
    {
      "epoch": 6.42,
      "learning_rate": 2.861347845432454e-05,
      "loss": 0.0197,
      "step": 12500
    },
    {
      "epoch": 6.67,
      "learning_rate": 2.7757812232604307e-05,
      "loss": 0.0196,
      "step": 13000
    },
    {
      "epoch": 6.93,
      "learning_rate": 2.690214601088408e-05,
      "loss": 0.0186,
      "step": 13500
    },
    {
      "epoch": 7.19,
      "learning_rate": 2.6046479789163847e-05,
      "loss": 0.0179,
      "step": 14000
    },
    {
      "epoch": 7.44,
      "learning_rate": 2.5190813567443612e-05,
      "loss": 0.0178,
      "step": 14500
    },
    {
      "epoch": 7.7,
      "learning_rate": 2.433514734572338e-05,
      "loss": 0.0178,
      "step": 15000
    },
    {
      "epoch": 7.7,
      "eval_loss": 0.02974303811788559,
      "eval_runtime": 120.9977,
      "eval_samples_per_second": 21.438,
      "eval_steps_per_second": 5.364,
      "step": 15000
    },
    {
      "epoch": 7.96,
      "learning_rate": 2.347948112400315e-05,
      "loss": 0.0174,
      "step": 15500
    },
    {
      "epoch": 8.21,
      "learning_rate": 2.2623814902282918e-05,
      "loss": 0.0159,
      "step": 16000
    },
    {
      "epoch": 8.47,
      "learning_rate": 2.1768148680562686e-05,
      "loss": 0.016,
      "step": 16500
    },
    {
      "epoch": 8.73,
      "learning_rate": 2.0912482458842455e-05,
      "loss": 0.0158,
      "step": 17000
    },
    {
      "epoch": 8.98,
      "learning_rate": 2.0056816237122226e-05,
      "loss": 0.0163,
      "step": 17500
    },
    {
      "epoch": 9.24,
      "learning_rate": 1.920115001540199e-05,
      "loss": 0.014,
      "step": 18000
    },
    {
      "epoch": 9.5,
      "learning_rate": 1.834548379368176e-05,
      "loss": 0.0143,
      "step": 18500
    },
    {
      "epoch": 9.75,
      "learning_rate": 1.748981757196153e-05,
      "loss": 0.0149,
      "step": 19000
    },
    {
      "epoch": 10.01,
      "learning_rate": 1.6634151350241297e-05,
      "loss": 0.0142,
      "step": 19500
    },
    {
      "epoch": 10.27,
      "learning_rate": 1.577848512852107e-05,
      "loss": 0.0128,
      "step": 20000
    },
    {
      "epoch": 10.27,
      "eval_loss": 0.03699657693505287,
      "eval_runtime": 120.8215,
      "eval_samples_per_second": 21.47,
      "eval_steps_per_second": 5.372,
      "step": 20000
    }
  ],
  "max_steps": 29220,
  "num_train_epochs": 15,
  "total_flos": 1.7313883226112e+17,
  "trial_name": null,
  "trial_params": null
}