trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.963369963369964,
  "global_step": 170,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.59,
      "learning_rate": 0.00018823529411764707,
      "loss": 3.2221,
      "step": 10
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00017647058823529413,
      "loss": 2.3963,
      "step": 20
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001647058823529412,
      "loss": 2.1803,
      "step": 30
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00015294117647058822,
      "loss": 2.0863,
      "step": 40
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0001411764705882353,
      "loss": 2.0213,
      "step": 50
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.00012941176470588237,
      "loss": 1.9832,
      "step": 60
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.00011764705882352942,
      "loss": 1.9548,
      "step": 70
    },
    {
      "epoch": 4.69,
      "learning_rate": 0.00010588235294117647,
      "loss": 1.9278,
      "step": 80
    },
    {
      "epoch": 5.27,
      "learning_rate": 9.411764705882353e-05,
      "loss": 1.908,
      "step": 90
    },
    {
      "epoch": 5.86,
      "learning_rate": 8.23529411764706e-05,
      "loss": 1.8947,
      "step": 100
    },
    {
      "epoch": 6.45,
      "learning_rate": 7.058823529411765e-05,
      "loss": 1.8833,
      "step": 110
    },
    {
      "epoch": 7.03,
      "learning_rate": 5.882352941176471e-05,
      "loss": 1.8635,
      "step": 120
    },
    {
      "epoch": 7.62,
      "learning_rate": 4.705882352941177e-05,
      "loss": 1.8532,
      "step": 130
    },
    {
      "epoch": 8.21,
      "learning_rate": 3.529411764705883e-05,
      "loss": 1.8499,
      "step": 140
    },
    {
      "epoch": 8.79,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 1.8394,
      "step": 150
    },
    {
      "epoch": 9.38,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.8459,
      "step": 160
    },
    {
      "epoch": 9.96,
      "learning_rate": 0.0,
      "loss": 1.8359,
      "step": 170
    },
    {
      "epoch": 9.96,
      "step": 170,
      "total_flos": 1.059141821617152e+17,
      "train_loss": 2.0321115493774413,
      "train_runtime": 22233.3267,
      "train_samples_per_second": 7.852,
      "train_steps_per_second": 0.008
    }
  ],
  "max_steps": 170,
  "num_train_epochs": 10,
  "total_flos": 1.059141821617152e+17,
  "trial_name": null,
  "trial_params": null
}