{
  "best_metric": 0.9415691141448361,
  "best_model_checkpoint": "model_saves/deberta-large_spell_10k_1_p3/checkpoint-536",
  "epoch": 8.0,
  "global_step": 2144,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9410715683314885,
      "eval_loss": 0.4239862263202667,
      "eval_runtime": 4.6841,
      "eval_samples_per_second": 935.507,
      "eval_steps_per_second": 7.472,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9415691141448361,
      "eval_loss": 0.4533962905406952,
      "eval_runtime": 4.6827,
      "eval_samples_per_second": 935.78,
      "eval_steps_per_second": 7.474,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9409429890763538,
      "eval_loss": 0.47931021451950073,
      "eval_runtime": 4.6901,
      "eval_samples_per_second": 934.314,
      "eval_steps_per_second": 7.463,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2492,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9402945023982826,
      "eval_loss": 0.5379701852798462,
      "eval_runtime": 4.6897,
      "eval_samples_per_second": 934.389,
      "eval_steps_per_second": 7.463,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9399199454376727,
      "eval_loss": 0.5923169851303101,
      "eval_runtime": 4.6909,
      "eval_samples_per_second": 934.151,
      "eval_steps_per_second": 7.461,
      "step": 1340
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.939796956584935,
      "eval_loss": 0.6551798582077026,
      "eval_runtime": 4.6848,
      "eval_samples_per_second": 935.373,
      "eval_steps_per_second": 7.471,
      "step": 1608
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9385950200695446,
      "eval_loss": 0.7205362319946289,
      "eval_runtime": 4.6965,
      "eval_samples_per_second": 933.025,
      "eval_steps_per_second": 7.452,
      "step": 1876
    },
    {
      "epoch": 7.46,
      "learning_rate": 1e-05,
      "loss": 0.0701,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9394950748554881,
      "eval_loss": 0.7645609378814697,
      "eval_runtime": 4.6967,
      "eval_samples_per_second": 932.997,
      "eval_steps_per_second": 7.452,
      "step": 2144
    },
    {
      "epoch": 8.0,
      "step": 2144,
      "total_flos": 3.262047243678515e+16,
      "train_loss": 0.025062660005555223,
      "train_runtime": 358.6275,
      "train_samples_per_second": 1434.804,
      "train_steps_per_second": 11.209
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 3.262047243678515e+16,
  "trial_name": null,
  "trial_params": null
}