{
  "best_metric": 4.350083351135254,
  "best_model_checkpoint": "models/GPT2_more_function_67/checkpoint-64170",
  "epoch": 10.0,
  "global_step": 64170,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 1e-05,
      "loss": 8.2024,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "learning_rate": 2e-05,
      "loss": 6.8723,
      "step": 2000
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 6.4179,
      "step": 3000
    },
    {
      "epoch": 0.62,
      "learning_rate": 4e-05,
      "loss": 6.0889,
      "step": 4000
    },
    {
      "epoch": 0.78,
      "learning_rate": 5e-05,
      "loss": 5.8357,
      "step": 5000
    },
    {
      "epoch": 0.94,
      "learning_rate": 6e-05,
      "loss": 5.634,
      "step": 6000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.19016516065262834,
      "eval_loss": 5.403344631195068,
      "eval_runtime": 1.9384,
      "eval_samples_per_second": 608.231,
      "eval_steps_per_second": 5.159,
      "step": 6417
    },
    {
      "epoch": 1.09,
      "learning_rate": 7e-05,
      "loss": 5.4636,
      "step": 7000
    },
    {
      "epoch": 1.25,
      "learning_rate": 8e-05,
      "loss": 5.3276,
      "step": 8000
    },
    {
      "epoch": 1.4,
      "learning_rate": 9e-05,
      "loss": 5.2192,
      "step": 9000
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 5.1306,
      "step": 10000
    },
    {
      "epoch": 1.71,
      "learning_rate": 9.815580579656637e-05,
      "loss": 5.0532,
      "step": 11000
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.630976555288906e-05,
      "loss": 4.9862,
      "step": 12000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2242725384517775,
      "eval_loss": 4.8415141105651855,
      "eval_runtime": 1.9614,
      "eval_samples_per_second": 601.103,
      "eval_steps_per_second": 5.098,
      "step": 12834
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.446557134945542e-05,
      "loss": 4.9227,
      "step": 13000
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.26195311057781e-05,
      "loss": 4.844,
      "step": 14000
    },
    {
      "epoch": 2.34,
      "learning_rate": 9.077533690234448e-05,
      "loss": 4.8161,
      "step": 15000
    },
    {
      "epoch": 2.49,
      "learning_rate": 8.892929665866716e-05,
      "loss": 4.7876,
      "step": 16000
    },
    {
      "epoch": 2.65,
      "learning_rate": 8.708510245523353e-05,
      "loss": 4.7584,
      "step": 17000
    },
    {
      "epoch": 2.81,
      "learning_rate": 8.523906221155622e-05,
      "loss": 4.7349,
      "step": 18000
    },
    {
      "epoch": 2.96,
      "learning_rate": 8.339486800812258e-05,
      "loss": 4.7151,
      "step": 19000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.24096892468594097,
      "eval_loss": 4.639059543609619,
      "eval_runtime": 1.9547,
      "eval_samples_per_second": 603.16,
      "eval_steps_per_second": 5.116,
      "step": 19251
    },
    {
      "epoch": 3.12,
      "learning_rate": 8.154882776444527e-05,
      "loss": 4.6475,
      "step": 20000
    },
    {
      "epoch": 3.27,
      "learning_rate": 7.970463356101163e-05,
      "loss": 4.623,
      "step": 21000
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.785859331733432e-05,
      "loss": 4.6159,
      "step": 22000
    },
    {
      "epoch": 3.58,
      "learning_rate": 7.60143991139007e-05,
      "loss": 4.6039,
      "step": 23000
    },
    {
      "epoch": 3.74,
      "learning_rate": 7.416835887022337e-05,
      "loss": 4.5934,
      "step": 24000
    },
    {
      "epoch": 3.9,
      "learning_rate": 7.232416466678975e-05,
      "loss": 4.5832,
      "step": 25000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.24864926235365617,
      "eval_loss": 4.5397257804870605,
      "eval_runtime": 1.9754,
      "eval_samples_per_second": 596.829,
      "eval_steps_per_second": 5.062,
      "step": 25668
    },
    {
      "epoch": 4.05,
      "learning_rate": 7.047812442311242e-05,
      "loss": 4.5467,
      "step": 26000
    },
    {
      "epoch": 4.21,
      "learning_rate": 6.863393021967878e-05,
      "loss": 4.5013,
      "step": 27000
    },
    {
      "epoch": 4.36,
      "learning_rate": 6.678788997600149e-05,
      "loss": 4.4996,
      "step": 28000
    },
    {
      "epoch": 4.52,
      "learning_rate": 6.494369577256785e-05,
      "loss": 4.4983,
      "step": 29000
    },
    {
      "epoch": 4.68,
      "learning_rate": 6.309765552889054e-05,
      "loss": 4.4935,
      "step": 30000
    },
    {
      "epoch": 4.83,
      "learning_rate": 6.12534613254569e-05,
      "loss": 4.4882,
      "step": 31000
    },
    {
      "epoch": 4.99,
      "learning_rate": 5.940742108177959e-05,
      "loss": 4.4844,
      "step": 32000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2552142814209293,
      "eval_loss": 4.475869178771973,
      "eval_runtime": 1.9612,
      "eval_samples_per_second": 601.15,
      "eval_steps_per_second": 5.099,
      "step": 32085
    },
    {
      "epoch": 5.14,
      "learning_rate": 5.7563226878345955e-05,
      "loss": 4.4145,
      "step": 33000
    },
    {
      "epoch": 5.3,
      "learning_rate": 5.571718663466864e-05,
      "loss": 4.4136,
      "step": 34000
    },
    {
      "epoch": 5.45,
      "learning_rate": 5.3872992431235006e-05,
      "loss": 4.4166,
      "step": 35000
    },
    {
      "epoch": 5.61,
      "learning_rate": 5.202695218755769e-05,
      "loss": 4.4143,
      "step": 36000
    },
    {
      "epoch": 5.77,
      "learning_rate": 5.018275798412405e-05,
      "loss": 4.4144,
      "step": 37000
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.833671774044675e-05,
      "loss": 4.4117,
      "step": 38000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.25894091482839454,
      "eval_loss": 4.4317240715026855,
      "eval_runtime": 1.9658,
      "eval_samples_per_second": 599.759,
      "eval_steps_per_second": 5.087,
      "step": 38502
    },
    {
      "epoch": 6.08,
      "learning_rate": 4.6492523537013114e-05,
      "loss": 4.3734,
      "step": 39000
    },
    {
      "epoch": 6.23,
      "learning_rate": 4.46464832933358e-05,
      "loss": 4.3426,
      "step": 40000
    },
    {
      "epoch": 6.39,
      "learning_rate": 4.2802289089902164e-05,
      "loss": 4.3469,
      "step": 41000
    },
    {
      "epoch": 6.55,
      "learning_rate": 4.095624884622485e-05,
      "loss": 4.3509,
      "step": 42000
    },
    {
      "epoch": 6.7,
      "learning_rate": 3.9112054642791215e-05,
      "loss": 4.352,
      "step": 43000
    },
    {
      "epoch": 6.86,
      "learning_rate": 3.7266014399113905e-05,
      "loss": 4.3505,
      "step": 44000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.2628077978802268,
      "eval_loss": 4.400611400604248,
      "eval_runtime": 1.9666,
      "eval_samples_per_second": 599.513,
      "eval_steps_per_second": 5.085,
      "step": 44919
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.5421820195680266e-05,
      "loss": 4.3429,
      "step": 45000
    },
    {
      "epoch": 7.17,
      "learning_rate": 3.3575779952002956e-05,
      "loss": 4.2836,
      "step": 46000
    },
    {
      "epoch": 7.32,
      "learning_rate": 3.173158574856932e-05,
      "loss": 4.2905,
      "step": 47000
    },
    {
      "epoch": 7.48,
      "learning_rate": 2.988554550489201e-05,
      "loss": 4.2925,
      "step": 48000
    },
    {
      "epoch": 7.64,
      "learning_rate": 2.8041351301458374e-05,
      "loss": 4.2935,
      "step": 49000
    },
    {
      "epoch": 7.79,
      "learning_rate": 2.6195311057781064e-05,
      "loss": 4.2968,
      "step": 50000
    },
    {
      "epoch": 7.95,
      "learning_rate": 2.4351116854347424e-05,
      "loss": 4.2941,
      "step": 51000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.26521207749794634,
      "eval_loss": 4.375330448150635,
      "eval_runtime": 1.9553,
      "eval_samples_per_second": 602.989,
      "eval_steps_per_second": 5.114,
      "step": 51336
    },
    {
      "epoch": 8.1,
      "learning_rate": 2.2505076610670114e-05,
      "loss": 4.2558,
      "step": 52000
    },
    {
      "epoch": 8.26,
      "learning_rate": 2.0660882407236478e-05,
      "loss": 4.2412,
      "step": 53000
    },
    {
      "epoch": 8.42,
      "learning_rate": 1.8814842163559165e-05,
      "loss": 4.2469,
      "step": 54000
    },
    {
      "epoch": 8.57,
      "learning_rate": 1.6970647960125532e-05,
      "loss": 4.2463,
      "step": 55000
    },
    {
      "epoch": 8.73,
      "learning_rate": 1.5124607716448219e-05,
      "loss": 4.2458,
      "step": 56000
    },
    {
      "epoch": 8.88,
      "learning_rate": 1.3280413513014583e-05,
      "loss": 4.2479,
      "step": 57000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.2669418231118057,
      "eval_loss": 4.358160495758057,
      "eval_runtime": 1.9826,
      "eval_samples_per_second": 594.673,
      "eval_steps_per_second": 5.044,
      "step": 57753
    },
    {
      "epoch": 9.04,
      "learning_rate": 1.1434373269337271e-05,
      "loss": 4.2333,
      "step": 58000
    },
    {
      "epoch": 9.19,
      "learning_rate": 9.590179065903637e-06,
      "loss": 4.2031,
      "step": 59000
    },
    {
      "epoch": 9.35,
      "learning_rate": 7.744138822226325e-06,
      "loss": 4.2029,
      "step": 60000
    },
    {
      "epoch": 9.51,
      "learning_rate": 5.899944618792691e-06,
      "loss": 4.2072,
      "step": 61000
    },
    {
      "epoch": 9.66,
      "learning_rate": 4.0539043751153775e-06,
      "loss": 4.2067,
      "step": 62000
    },
    {
      "epoch": 9.82,
      "learning_rate": 2.209710171681743e-06,
      "loss": 4.2061,
      "step": 63000
    },
    {
      "epoch": 9.97,
      "learning_rate": 3.636699280044305e-07,
      "loss": 4.2036,
      "step": 64000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.2679636419493365,
      "eval_loss": 4.350083351135254,
      "eval_runtime": 1.9519,
      "eval_samples_per_second": 604.023,
      "eval_steps_per_second": 5.123,
      "step": 64170
    },
    {
      "epoch": 10.0,
      "step": 64170,
      "total_flos": 5.36497864704e+17,
      "train_loss": 4.696056910142249,
      "train_runtime": 28033.3871,
      "train_samples_per_second": 292.972,
      "train_steps_per_second": 2.289
    }
  ],
  "max_steps": 64170,
  "num_train_epochs": 10,
  "total_flos": 5.36497864704e+17,
  "trial_name": null,
  "trial_params": null
}