{
  "best_metric": 4.725980758666992,
  "best_model_checkpoint": "models/GPT2_no_function_67/checkpoint-47020",
  "epoch": 10.0,
  "global_step": 47020,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21,
      "learning_rate": 1e-05,
      "loss": 8.0666,
      "step": 1000
    },
    {
      "epoch": 0.43,
      "learning_rate": 2e-05,
      "loss": 7.0832,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 6.6789,
      "step": 3000
    },
    {
      "epoch": 0.85,
      "learning_rate": 4e-05,
      "loss": 6.3673,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22100889236420504,
      "eval_loss": 6.009326934814453,
      "eval_runtime": 1.4336,
      "eval_samples_per_second": 604.067,
      "eval_steps_per_second": 4.883,
      "step": 4702
    },
    {
      "epoch": 1.06,
      "learning_rate": 5e-05,
      "loss": 6.1194,
      "step": 5000
    },
    {
      "epoch": 1.28,
      "learning_rate": 6e-05,
      "loss": 5.9118,
      "step": 6000
    },
    {
      "epoch": 1.49,
      "learning_rate": 7e-05,
      "loss": 5.7482,
      "step": 7000
    },
    {
      "epoch": 1.7,
      "learning_rate": 8e-05,
      "loss": 5.6171,
      "step": 8000
    },
    {
      "epoch": 1.91,
      "learning_rate": 9e-05,
      "loss": 5.5122,
      "step": 9000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2530959611572803,
      "eval_loss": 5.362746238708496,
      "eval_runtime": 1.4445,
      "eval_samples_per_second": 599.499,
      "eval_steps_per_second": 4.846,
      "step": 9404
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0001,
      "loss": 5.3993,
      "step": 10000
    },
    {
      "epoch": 2.34,
      "learning_rate": 9.730145867098865e-05,
      "loss": 5.3202,
      "step": 11000
    },
    {
      "epoch": 2.55,
      "learning_rate": 9.460021609940574e-05,
      "loss": 5.2575,
      "step": 12000
    },
    {
      "epoch": 2.76,
      "learning_rate": 9.190167477039439e-05,
      "loss": 5.2089,
      "step": 13000
    },
    {
      "epoch": 2.98,
      "learning_rate": 8.920043219881146e-05,
      "loss": 5.1637,
      "step": 14000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2754268880362241,
      "eval_loss": 5.0802507400512695,
      "eval_runtime": 1.4537,
      "eval_samples_per_second": 595.734,
      "eval_steps_per_second": 4.815,
      "step": 14106
    },
    {
      "epoch": 3.19,
      "learning_rate": 8.650189086980012e-05,
      "loss": 5.0658,
      "step": 15000
    },
    {
      "epoch": 3.4,
      "learning_rate": 8.380064829821718e-05,
      "loss": 5.0423,
      "step": 16000
    },
    {
      "epoch": 3.62,
      "learning_rate": 8.110210696920584e-05,
      "loss": 5.0206,
      "step": 17000
    },
    {
      "epoch": 3.83,
      "learning_rate": 7.84008643976229e-05,
      "loss": 5.0001,
      "step": 18000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2882471677183539,
      "eval_loss": 4.943727970123291,
      "eval_runtime": 1.5715,
      "eval_samples_per_second": 551.081,
      "eval_steps_per_second": 4.454,
      "step": 18808
    },
    {
      "epoch": 4.04,
      "learning_rate": 7.570232306861156e-05,
      "loss": 4.9618,
      "step": 19000
    },
    {
      "epoch": 4.25,
      "learning_rate": 7.300108049702864e-05,
      "loss": 4.8897,
      "step": 20000
    },
    {
      "epoch": 4.47,
      "learning_rate": 7.030253916801729e-05,
      "loss": 4.8867,
      "step": 21000
    },
    {
      "epoch": 4.68,
      "learning_rate": 6.760129659643436e-05,
      "loss": 4.8784,
      "step": 22000
    },
    {
      "epoch": 4.89,
      "learning_rate": 6.490275526742302e-05,
      "loss": 4.8699,
      "step": 23000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.29503009583386375,
      "eval_loss": 4.865215301513672,
      "eval_runtime": 1.4538,
      "eval_samples_per_second": 595.674,
      "eval_steps_per_second": 4.815,
      "step": 23510
    },
    {
      "epoch": 5.1,
      "learning_rate": 6.22015126958401e-05,
      "loss": 4.8168,
      "step": 24000
    },
    {
      "epoch": 5.32,
      "learning_rate": 5.950297136682874e-05,
      "loss": 4.7782,
      "step": 25000
    },
    {
      "epoch": 5.53,
      "learning_rate": 5.680172879524581e-05,
      "loss": 4.7791,
      "step": 26000
    },
    {
      "epoch": 5.74,
      "learning_rate": 5.410318746623447e-05,
      "loss": 4.7784,
      "step": 27000
    },
    {
      "epoch": 5.95,
      "learning_rate": 5.140194489465154e-05,
      "loss": 4.7738,
      "step": 28000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.30020366969140405,
      "eval_loss": 4.812718391418457,
      "eval_runtime": 1.4498,
      "eval_samples_per_second": 597.338,
      "eval_steps_per_second": 4.828,
      "step": 28212
    },
    {
      "epoch": 6.17,
      "learning_rate": 4.87034035656402e-05,
      "loss": 4.7033,
      "step": 29000
    },
    {
      "epoch": 6.38,
      "learning_rate": 4.600216099405727e-05,
      "loss": 4.6956,
      "step": 30000
    },
    {
      "epoch": 6.59,
      "learning_rate": 4.330361966504592e-05,
      "loss": 4.6959,
      "step": 31000
    },
    {
      "epoch": 6.81,
      "learning_rate": 4.0602377093463e-05,
      "loss": 4.6966,
      "step": 32000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3044225418704879,
      "eval_loss": 4.7761945724487305,
      "eval_runtime": 1.4505,
      "eval_samples_per_second": 597.024,
      "eval_steps_per_second": 4.826,
      "step": 32914
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.790383576445165e-05,
      "loss": 4.6884,
      "step": 33000
    },
    {
      "epoch": 7.23,
      "learning_rate": 3.520259319286872e-05,
      "loss": 4.6182,
      "step": 34000
    },
    {
      "epoch": 7.44,
      "learning_rate": 3.250405186385738e-05,
      "loss": 4.6246,
      "step": 35000
    },
    {
      "epoch": 7.66,
      "learning_rate": 2.9802809292274448e-05,
      "loss": 4.6318,
      "step": 36000
    },
    {
      "epoch": 7.87,
      "learning_rate": 2.7104267963263103e-05,
      "loss": 4.6296,
      "step": 37000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.30685021185284866,
      "eval_loss": 4.7499237060546875,
      "eval_runtime": 1.4456,
      "eval_samples_per_second": 599.074,
      "eval_steps_per_second": 4.842,
      "step": 37616
    },
    {
      "epoch": 8.08,
      "learning_rate": 2.4403025391680173e-05,
      "loss": 4.6027,
      "step": 38000
    },
    {
      "epoch": 8.29,
      "learning_rate": 2.170448406266883e-05,
      "loss": 4.5656,
      "step": 39000
    },
    {
      "epoch": 8.51,
      "learning_rate": 1.9003241491085898e-05,
      "loss": 4.5703,
      "step": 40000
    },
    {
      "epoch": 8.72,
      "learning_rate": 1.6304700162074554e-05,
      "loss": 4.5718,
      "step": 41000
    },
    {
      "epoch": 8.93,
      "learning_rate": 1.3603457590491628e-05,
      "loss": 4.5712,
      "step": 42000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.30869596843119784,
      "eval_loss": 4.733290195465088,
      "eval_runtime": 1.4519,
      "eval_samples_per_second": 596.477,
      "eval_steps_per_second": 4.821,
      "step": 42318
    },
    {
      "epoch": 9.15,
      "learning_rate": 1.0904916261480282e-05,
      "loss": 4.5367,
      "step": 43000
    },
    {
      "epoch": 9.36,
      "learning_rate": 8.203673689897353e-06,
      "loss": 4.5226,
      "step": 44000
    },
    {
      "epoch": 9.57,
      "learning_rate": 5.505132360886007e-06,
      "loss": 4.5228,
      "step": 45000
    },
    {
      "epoch": 9.78,
      "learning_rate": 2.8038897893030795e-06,
      "loss": 4.5214,
      "step": 46000
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.0534846029173421e-07,
      "loss": 4.5227,
      "step": 47000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.31029623029222964,
      "eval_loss": 4.725980758666992,
      "eval_runtime": 1.4835,
      "eval_samples_per_second": 583.771,
      "eval_steps_per_second": 4.719,
      "step": 47020
    },
    {
      "epoch": 10.0,
      "step": 47020,
      "total_flos": 3.930942652416e+17,
      "train_loss": 5.095248515795566,
      "train_runtime": 20611.3432,
      "train_samples_per_second": 291.961,
      "train_steps_per_second": 2.281
    }
  ],
  "max_steps": 47020,
  "num_train_epochs": 10,
  "total_flos": 3.930942652416e+17,
  "trial_name": null,
  "trial_params": null
}