{
  "best_metric": 3.6031816005706787,
  "best_model_checkpoint": "models/GPT2_five_function_67/checkpoint-64080",
  "epoch": 10.0,
  "global_step": 64080,
  "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": 6.9773,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "learning_rate": 2e-05,
      "loss": 5.7479,
      "step": 2000
    },
    {
      "epoch": 0.47,
      "learning_rate": 3e-05,
      "loss": 5.4062,
      "step": 3000
    },
    {
      "epoch": 0.62,
      "learning_rate": 4e-05,
      "loss": 5.1414,
      "step": 4000
    },
    {
      "epoch": 0.78,
      "learning_rate": 5e-05,
      "loss": 4.9261,
      "step": 5000
    },
    {
      "epoch": 0.94,
      "learning_rate": 6e-05,
      "loss": 4.7546,
      "step": 6000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3263839685574108,
      "eval_loss": 4.561105728149414,
      "eval_runtime": 1.9761,
      "eval_samples_per_second": 596.134,
      "eval_steps_per_second": 5.061,
      "step": 6408
    },
    {
      "epoch": 1.09,
      "learning_rate": 7e-05,
      "loss": 4.6053,
      "step": 7000
    },
    {
      "epoch": 1.25,
      "learning_rate": 8e-05,
      "loss": 4.4838,
      "step": 8000
    },
    {
      "epoch": 1.4,
      "learning_rate": 9e-05,
      "loss": 4.3905,
      "step": 9000
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 4.3098,
      "step": 10000
    },
    {
      "epoch": 1.72,
      "learning_rate": 9.815273668639054e-05,
      "loss": 4.2407,
      "step": 11000
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.630362426035503e-05,
      "loss": 4.1827,
      "step": 12000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3615429862438672,
      "eval_loss": 4.056678771972656,
      "eval_runtime": 1.9627,
      "eval_samples_per_second": 600.185,
      "eval_steps_per_second": 5.095,
      "step": 12816
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.445636094674557e-05,
      "loss": 4.125,
      "step": 13000
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.260724852071006e-05,
      "loss": 4.0549,
      "step": 14000
    },
    {
      "epoch": 2.34,
      "learning_rate": 9.07599852071006e-05,
      "loss": 4.0283,
      "step": 15000
    },
    {
      "epoch": 2.5,
      "learning_rate": 8.891087278106509e-05,
      "loss": 3.9989,
      "step": 16000
    },
    {
      "epoch": 2.65,
      "learning_rate": 8.706360946745562e-05,
      "loss": 3.9761,
      "step": 17000
    },
    {
      "epoch": 2.81,
      "learning_rate": 8.521449704142012e-05,
      "loss": 3.954,
      "step": 18000
    },
    {
      "epoch": 2.97,
      "learning_rate": 8.336723372781065e-05,
      "loss": 3.9339,
      "step": 19000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3764554897530848,
      "eval_loss": 3.8715641498565674,
      "eval_runtime": 1.9702,
      "eval_samples_per_second": 597.906,
      "eval_steps_per_second": 5.076,
      "step": 19224
    },
    {
      "epoch": 3.12,
      "learning_rate": 8.151812130177516e-05,
      "loss": 3.8682,
      "step": 20000
    },
    {
      "epoch": 3.28,
      "learning_rate": 7.967085798816568e-05,
      "loss": 3.8504,
      "step": 21000
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.782174556213018e-05,
      "loss": 3.8432,
      "step": 22000
    },
    {
      "epoch": 3.59,
      "learning_rate": 7.597448224852071e-05,
      "loss": 3.8319,
      "step": 23000
    },
    {
      "epoch": 3.75,
      "learning_rate": 7.412721893491125e-05,
      "loss": 3.8222,
      "step": 24000
    },
    {
      "epoch": 3.9,
      "learning_rate": 7.227810650887574e-05,
      "loss": 3.8136,
      "step": 25000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.38571982407122707,
      "eval_loss": 3.777761459350586,
      "eval_runtime": 1.9592,
      "eval_samples_per_second": 601.256,
      "eval_steps_per_second": 5.104,
      "step": 25632
    },
    {
      "epoch": 4.06,
      "learning_rate": 7.042899408284024e-05,
      "loss": 3.7771,
      "step": 26000
    },
    {
      "epoch": 4.21,
      "learning_rate": 6.858173076923077e-05,
      "loss": 3.7353,
      "step": 27000
    },
    {
      "epoch": 4.37,
      "learning_rate": 6.673261834319527e-05,
      "loss": 3.7357,
      "step": 28000
    },
    {
      "epoch": 4.53,
      "learning_rate": 6.488350591715976e-05,
      "loss": 3.7324,
      "step": 29000
    },
    {
      "epoch": 4.68,
      "learning_rate": 6.30362426035503e-05,
      "loss": 3.7305,
      "step": 30000
    },
    {
      "epoch": 4.84,
      "learning_rate": 6.11871301775148e-05,
      "loss": 3.7248,
      "step": 31000
    },
    {
      "epoch": 4.99,
      "learning_rate": 5.933801775147929e-05,
      "loss": 3.72,
      "step": 32000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3916286779941981,
      "eval_loss": 3.7189266681671143,
      "eval_runtime": 1.9479,
      "eval_samples_per_second": 604.755,
      "eval_steps_per_second": 5.134,
      "step": 32040
    },
    {
      "epoch": 5.15,
      "learning_rate": 5.7488905325443785e-05,
      "loss": 3.6509,
      "step": 33000
    },
    {
      "epoch": 5.31,
      "learning_rate": 5.564349112426036e-05,
      "loss": 3.6553,
      "step": 34000
    },
    {
      "epoch": 5.46,
      "learning_rate": 5.379437869822486e-05,
      "loss": 3.6559,
      "step": 35000
    },
    {
      "epoch": 5.62,
      "learning_rate": 5.1945266272189354e-05,
      "loss": 3.658,
      "step": 36000
    },
    {
      "epoch": 5.77,
      "learning_rate": 5.009615384615385e-05,
      "loss": 3.655,
      "step": 37000
    },
    {
      "epoch": 5.93,
      "learning_rate": 4.8248890532544384e-05,
      "loss": 3.6514,
      "step": 38000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3962541609293745,
      "eval_loss": 3.6761600971221924,
      "eval_runtime": 1.9449,
      "eval_samples_per_second": 605.676,
      "eval_steps_per_second": 5.142,
      "step": 38448
    },
    {
      "epoch": 6.09,
      "learning_rate": 4.639977810650888e-05,
      "loss": 3.6115,
      "step": 39000
    },
    {
      "epoch": 6.24,
      "learning_rate": 4.4552514792899406e-05,
      "loss": 3.5883,
      "step": 40000
    },
    {
      "epoch": 6.4,
      "learning_rate": 4.2705251479289946e-05,
      "loss": 3.5928,
      "step": 41000
    },
    {
      "epoch": 6.55,
      "learning_rate": 4.0856139053254435e-05,
      "loss": 3.5949,
      "step": 42000
    },
    {
      "epoch": 6.71,
      "learning_rate": 3.900702662721893e-05,
      "loss": 3.5947,
      "step": 43000
    },
    {
      "epoch": 6.87,
      "learning_rate": 3.715976331360947e-05,
      "loss": 3.5942,
      "step": 44000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.39874069221822656,
      "eval_loss": 3.6474609375,
      "eval_runtime": 1.9407,
      "eval_samples_per_second": 607.008,
      "eval_steps_per_second": 5.153,
      "step": 44856
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.531065088757397e-05,
      "loss": 3.5838,
      "step": 45000
    },
    {
      "epoch": 7.18,
      "learning_rate": 3.346153846153846e-05,
      "loss": 3.533,
      "step": 46000
    },
    {
      "epoch": 7.33,
      "learning_rate": 3.161242603550296e-05,
      "loss": 3.5389,
      "step": 47000
    },
    {
      "epoch": 7.49,
      "learning_rate": 2.9765162721893493e-05,
      "loss": 3.5412,
      "step": 48000
    },
    {
      "epoch": 7.65,
      "learning_rate": 2.791605029585799e-05,
      "loss": 3.5426,
      "step": 49000
    },
    {
      "epoch": 7.8,
      "learning_rate": 2.6068786982248522e-05,
      "loss": 3.5431,
      "step": 50000
    },
    {
      "epoch": 7.96,
      "learning_rate": 2.4219674556213018e-05,
      "loss": 3.5418,
      "step": 51000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4025039102709784,
      "eval_loss": 3.625131845474243,
      "eval_runtime": 1.961,
      "eval_samples_per_second": 600.704,
      "eval_steps_per_second": 5.099,
      "step": 51264
    },
    {
      "epoch": 8.11,
      "learning_rate": 2.2374260355029585e-05,
      "loss": 3.5,
      "step": 52000
    },
    {
      "epoch": 8.27,
      "learning_rate": 2.0525147928994084e-05,
      "loss": 3.4933,
      "step": 53000
    },
    {
      "epoch": 8.43,
      "learning_rate": 1.867603550295858e-05,
      "loss": 3.4939,
      "step": 54000
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.682692307692308e-05,
      "loss": 3.4967,
      "step": 55000
    },
    {
      "epoch": 8.74,
      "learning_rate": 1.4979659763313611e-05,
      "loss": 3.4965,
      "step": 56000
    },
    {
      "epoch": 8.9,
      "learning_rate": 1.3130547337278107e-05,
      "loss": 3.4961,
      "step": 57000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.40393433418445784,
      "eval_loss": 3.610266923904419,
      "eval_runtime": 1.9778,
      "eval_samples_per_second": 595.607,
      "eval_steps_per_second": 5.056,
      "step": 57672
    },
    {
      "epoch": 9.05,
      "learning_rate": 1.128328402366864e-05,
      "loss": 3.4838,
      "step": 58000
    },
    {
      "epoch": 9.21,
      "learning_rate": 9.434171597633136e-06,
      "loss": 3.4559,
      "step": 59000
    },
    {
      "epoch": 9.36,
      "learning_rate": 7.58690828402367e-06,
      "loss": 3.4569,
      "step": 60000
    },
    {
      "epoch": 9.52,
      "learning_rate": 5.739644970414201e-06,
      "loss": 3.4573,
      "step": 61000
    },
    {
      "epoch": 9.68,
      "learning_rate": 3.890532544378699e-06,
      "loss": 3.4569,
      "step": 62000
    },
    {
      "epoch": 9.83,
      "learning_rate": 2.0414201183431953e-06,
      "loss": 3.4574,
      "step": 63000
    },
    {
      "epoch": 9.99,
      "learning_rate": 1.9230769230769234e-07,
      "loss": 3.4556,
      "step": 64000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.4048032832907771,
      "eval_loss": 3.6031816005706787,
      "eval_runtime": 1.9577,
      "eval_samples_per_second": 601.715,
      "eval_steps_per_second": 5.108,
      "step": 64080
    },
    {
      "epoch": 10.0,
      "step": 64080,
      "total_flos": 5.3573097259008e+17,
      "train_loss": 3.9049099916227155,
      "train_runtime": 28071.2959,
      "train_samples_per_second": 292.158,
      "train_steps_per_second": 2.283
    }
  ],
  "max_steps": 64080,
  "num_train_epochs": 10,
  "total_flos": 5.3573097259008e+17,
  "trial_name": null,
  "trial_params": null
}