{
  "best_metric": 0.024844545871019363,
  "best_model_checkpoint": "./results_pnum/results_cvrev_pnum_f0_large_b4e15_5000/checkpoint-10000",
  "epoch": 10.277492291880781,
  "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": 4.91485935519238e-05,
      "loss": 0.5437,
      "step": 500
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.829204782951314e-05,
      "loss": 0.0334,
      "step": 1000
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.743550210710248e-05,
      "loss": 0.0315,
      "step": 1500
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.657895638469182e-05,
      "loss": 0.0313,
      "step": 2000
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.572241066228116e-05,
      "loss": 0.0295,
      "step": 2500
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.48658649398705e-05,
      "loss": 0.0284,
      "step": 3000
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.400931921745983e-05,
      "loss": 0.0283,
      "step": 3500
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.315277349504917e-05,
      "loss": 0.0268,
      "step": 4000
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.229622777263851e-05,
      "loss": 0.0267,
      "step": 4500
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.143968205022784e-05,
      "loss": 0.0263,
      "step": 5000
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.02492157183587551,
      "eval_runtime": 103.3756,
      "eval_samples_per_second": 25.083,
      "eval_steps_per_second": 6.278,
      "step": 5000
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.058313632781718e-05,
      "loss": 0.0262,
      "step": 5500
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.972659060540652e-05,
      "loss": 0.0252,
      "step": 6000
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.887004488299585e-05,
      "loss": 0.0247,
      "step": 6500
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.801349916058519e-05,
      "loss": 0.0251,
      "step": 7000
    },
    {
      "epoch": 3.85,
      "learning_rate": 3.715695343817453e-05,
      "loss": 0.0244,
      "step": 7500
    },
    {
      "epoch": 4.11,
      "learning_rate": 3.630040771576387e-05,
      "loss": 0.0239,
      "step": 8000
    },
    {
      "epoch": 4.37,
      "learning_rate": 3.544386199335321e-05,
      "loss": 0.0233,
      "step": 8500
    },
    {
      "epoch": 4.62,
      "learning_rate": 3.458731627094255e-05,
      "loss": 0.0229,
      "step": 9000
    },
    {
      "epoch": 4.88,
      "learning_rate": 3.373077054853188e-05,
      "loss": 0.0232,
      "step": 9500
    },
    {
      "epoch": 5.14,
      "learning_rate": 3.287422482612122e-05,
      "loss": 0.0226,
      "step": 10000
    },
    {
      "epoch": 5.14,
      "eval_loss": 0.024844545871019363,
      "eval_runtime": 103.4165,
      "eval_samples_per_second": 25.073,
      "eval_steps_per_second": 6.276,
      "step": 10000
    },
    {
      "epoch": 5.4,
      "learning_rate": 3.201767910371056e-05,
      "loss": 0.0216,
      "step": 10500
    },
    {
      "epoch": 5.65,
      "learning_rate": 3.11611333812999e-05,
      "loss": 0.0219,
      "step": 11000
    },
    {
      "epoch": 5.91,
      "learning_rate": 3.0304587658889232e-05,
      "loss": 0.0213,
      "step": 11500
    },
    {
      "epoch": 6.17,
      "learning_rate": 2.944804193647857e-05,
      "loss": 0.0203,
      "step": 12000
    },
    {
      "epoch": 6.42,
      "learning_rate": 2.8591496214067907e-05,
      "loss": 0.0204,
      "step": 12500
    },
    {
      "epoch": 6.68,
      "learning_rate": 2.7734950491657246e-05,
      "loss": 0.0205,
      "step": 13000
    },
    {
      "epoch": 6.94,
      "learning_rate": 2.6878404769246586e-05,
      "loss": 0.02,
      "step": 13500
    },
    {
      "epoch": 7.19,
      "learning_rate": 2.602185904683592e-05,
      "loss": 0.0186,
      "step": 14000
    },
    {
      "epoch": 7.45,
      "learning_rate": 2.516531332442526e-05,
      "loss": 0.0181,
      "step": 14500
    },
    {
      "epoch": 7.71,
      "learning_rate": 2.4308767602014596e-05,
      "loss": 0.0181,
      "step": 15000
    },
    {
      "epoch": 7.71,
      "eval_loss": 0.026644999161362648,
      "eval_runtime": 103.372,
      "eval_samples_per_second": 25.084,
      "eval_steps_per_second": 6.278,
      "step": 15000
    },
    {
      "epoch": 7.97,
      "learning_rate": 2.3452221879603932e-05,
      "loss": 0.0187,
      "step": 15500
    },
    {
      "epoch": 8.22,
      "learning_rate": 2.259567615719327e-05,
      "loss": 0.0169,
      "step": 16000
    },
    {
      "epoch": 8.48,
      "learning_rate": 2.173913043478261e-05,
      "loss": 0.0168,
      "step": 16500
    },
    {
      "epoch": 8.74,
      "learning_rate": 2.0882584712371946e-05,
      "loss": 0.0164,
      "step": 17000
    },
    {
      "epoch": 8.99,
      "learning_rate": 2.0026038989961285e-05,
      "loss": 0.0167,
      "step": 17500
    },
    {
      "epoch": 9.25,
      "learning_rate": 1.9169493267550624e-05,
      "loss": 0.0152,
      "step": 18000
    },
    {
      "epoch": 9.51,
      "learning_rate": 1.831294754513996e-05,
      "loss": 0.0154,
      "step": 18500
    },
    {
      "epoch": 9.76,
      "learning_rate": 1.74564018227293e-05,
      "loss": 0.0148,
      "step": 19000
    },
    {
      "epoch": 10.02,
      "learning_rate": 1.6599856100318635e-05,
      "loss": 0.0147,
      "step": 19500
    },
    {
      "epoch": 10.28,
      "learning_rate": 1.5743310377907974e-05,
      "loss": 0.0137,
      "step": 20000
    },
    {
      "epoch": 10.28,
      "eval_loss": 0.03126714006066322,
      "eval_runtime": 103.3509,
      "eval_samples_per_second": 25.089,
      "eval_steps_per_second": 6.28,
      "step": 20000
    }
  ],
  "max_steps": 29190,
  "num_train_epochs": 15,
  "total_flos": 1.7320378368e+17,
  "trial_name": null,
  "trial_params": null
}