{
  "best_metric": 1.3794686794281006,
  "best_model_checkpoint": "../saved_models_new/gptNEO_author_VJ_epochs15_lr5e-05/checkpoint-2000",
  "epoch": 14.37125748502994,
  "global_step": 24000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.2,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.5146,
      "step": 2000
    },
    {
      "epoch": 1.2,
      "eval_loss": 1.3794686794281006,
      "eval_runtime": 14.6483,
      "eval_samples_per_second": 25.395,
      "eval_steps_per_second": 12.698,
      "step": 2000
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.773242630385488e-05,
      "loss": 1.1782,
      "step": 4000
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3970938920974731,
      "eval_runtime": 14.6463,
      "eval_samples_per_second": 25.399,
      "eval_steps_per_second": 12.699,
      "step": 4000
    },
    {
      "epoch": 3.59,
      "learning_rate": 4.319727891156463e-05,
      "loss": 1.0049,
      "step": 6000
    },
    {
      "epoch": 3.59,
      "eval_loss": 1.4491934776306152,
      "eval_runtime": 14.6504,
      "eval_samples_per_second": 25.392,
      "eval_steps_per_second": 12.696,
      "step": 6000
    },
    {
      "epoch": 4.79,
      "learning_rate": 3.8662131519274384e-05,
      "loss": 0.8259,
      "step": 8000
    },
    {
      "epoch": 4.79,
      "eval_loss": 1.5493556261062622,
      "eval_runtime": 14.6505,
      "eval_samples_per_second": 25.392,
      "eval_steps_per_second": 12.696,
      "step": 8000
    },
    {
      "epoch": 5.99,
      "learning_rate": 3.412698412698413e-05,
      "loss": 0.6638,
      "step": 10000
    },
    {
      "epoch": 5.99,
      "eval_loss": 1.6542575359344482,
      "eval_runtime": 14.6609,
      "eval_samples_per_second": 25.374,
      "eval_steps_per_second": 12.687,
      "step": 10000
    },
    {
      "epoch": 7.19,
      "learning_rate": 2.959183673469388e-05,
      "loss": 0.4904,
      "step": 12000
    },
    {
      "epoch": 7.19,
      "eval_loss": 1.8380136489868164,
      "eval_runtime": 14.6527,
      "eval_samples_per_second": 25.388,
      "eval_steps_per_second": 12.694,
      "step": 12000
    },
    {
      "epoch": 8.38,
      "learning_rate": 2.5056689342403626e-05,
      "loss": 0.3432,
      "step": 14000
    },
    {
      "epoch": 8.38,
      "eval_loss": 1.9379925727844238,
      "eval_runtime": 14.6501,
      "eval_samples_per_second": 25.392,
      "eval_steps_per_second": 12.696,
      "step": 14000
    },
    {
      "epoch": 9.58,
      "learning_rate": 2.052154195011338e-05,
      "loss": 0.2499,
      "step": 16000
    },
    {
      "epoch": 9.58,
      "eval_loss": 2.04109787940979,
      "eval_runtime": 14.6527,
      "eval_samples_per_second": 25.388,
      "eval_steps_per_second": 12.694,
      "step": 16000
    },
    {
      "epoch": 10.78,
      "learning_rate": 1.5986394557823133e-05,
      "loss": 0.1809,
      "step": 18000
    },
    {
      "epoch": 10.78,
      "eval_loss": 2.1205220222473145,
      "eval_runtime": 14.6526,
      "eval_samples_per_second": 25.388,
      "eval_steps_per_second": 12.694,
      "step": 18000
    },
    {
      "epoch": 11.98,
      "learning_rate": 1.145124716553288e-05,
      "loss": 0.1246,
      "step": 20000
    },
    {
      "epoch": 11.98,
      "eval_loss": 2.1967058181762695,
      "eval_runtime": 14.6441,
      "eval_samples_per_second": 25.403,
      "eval_steps_per_second": 12.701,
      "step": 20000
    },
    {
      "epoch": 13.17,
      "learning_rate": 6.9160997732426305e-06,
      "loss": 0.0824,
      "step": 22000
    },
    {
      "epoch": 13.17,
      "eval_loss": 2.2765705585479736,
      "eval_runtime": 14.6502,
      "eval_samples_per_second": 25.392,
      "eval_steps_per_second": 12.696,
      "step": 22000
    },
    {
      "epoch": 14.37,
      "learning_rate": 2.3809523809523808e-06,
      "loss": 0.0615,
      "step": 24000
    },
    {
      "epoch": 14.37,
      "eval_loss": 2.3161871433258057,
      "eval_runtime": 14.6483,
      "eval_samples_per_second": 25.395,
      "eval_steps_per_second": 12.698,
      "step": 24000
    }
  ],
  "max_steps": 25050,
  "num_train_epochs": 15,
  "total_flos": 6268970336256000.0,
  "trial_name": null,
  "trial_params": null
}