trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.058806233460746836,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.6579,
      "step": 20
    },
    {
      "epoch": 0.01,
      "eval_accuracy": 0.574271023273468,
      "eval_loss": 0.6647623181343079,
      "eval_runtime": 404.381,
      "eval_samples_per_second": 14.417,
      "eval_steps_per_second": 3.606,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.764705882352942e-05,
      "loss": 0.6407,
      "step": 40
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.6174957156181335,
      "eval_loss": 0.9209257364273071,
      "eval_runtime": 392.6859,
      "eval_samples_per_second": 14.846,
      "eval_steps_per_second": 3.713,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.4622,
      "step": 60
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.9147512912750244,
      "eval_loss": 0.2529328763484955,
      "eval_runtime": 393.2855,
      "eval_samples_per_second": 14.824,
      "eval_steps_per_second": 3.707,
      "step": 60
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.529411764705883e-05,
      "loss": 0.271,
      "step": 80
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.9614065289497375,
      "eval_loss": 0.10574772953987122,
      "eval_runtime": 393.862,
      "eval_samples_per_second": 14.802,
      "eval_steps_per_second": 3.702,
      "step": 80
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.41764705882353e-05,
      "loss": 0.3011,
      "step": 100
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.9797598719596863,
      "eval_loss": 0.0672488585114479,
      "eval_runtime": 397.5693,
      "eval_samples_per_second": 14.664,
      "eval_steps_per_second": 3.667,
      "step": 100
    }
  ],
  "max_steps": 1700,
  "num_train_epochs": 1,
  "total_flos": 1.6617742871792326e+17,
  "trial_name": null,
  "trial_params": null
}