trainer_state.json

{
  "best_metric": 0.9413175460369637,
  "best_model_checkpoint": "model_saves/deberta-large_lemon_5k_5_p3/checkpoint-536",
  "epoch": 5.0,
  "global_step": 1340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9411498339650488,
      "eval_loss": 0.414813369512558,
      "eval_runtime": 4.6665,
      "eval_samples_per_second": 939.038,
      "eval_steps_per_second": 7.5,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9413175460369637,
      "eval_loss": 0.4369977116584778,
      "eval_runtime": 4.6563,
      "eval_samples_per_second": 941.092,
      "eval_steps_per_second": 7.517,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9407529153948502,
      "eval_loss": 0.47772112488746643,
      "eval_runtime": 4.6554,
      "eval_samples_per_second": 941.269,
      "eval_steps_per_second": 7.518,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2552,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.940070886302396,
      "eval_loss": 0.5178443193435669,
      "eval_runtime": 4.6625,
      "eval_samples_per_second": 939.847,
      "eval_steps_per_second": 7.507,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9399143550352754,
      "eval_loss": 0.5831624865531921,
      "eval_runtime": 4.6653,
      "eval_samples_per_second": 939.273,
      "eval_steps_per_second": 7.502,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.009276016177971e+16,
      "train_loss": 0.22275368135366866,
      "train_runtime": 589.3982,
      "train_samples_per_second": 873.026,
      "train_steps_per_second": 6.821
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.009276016177971e+16,
  "trial_name": null,
  "trial_params": null
}