trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.0,
  "global_step": 555,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.7,
      "eval_loss": 3.552431106567383,
      "eval_runtime": 16.546,
      "eval_samples_per_second": 28.164,
      "eval_steps_per_second": 3.566,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 5.41,
      "eval_loss": 3.0274674892425537,
      "eval_runtime": 16.4613,
      "eval_samples_per_second": 28.309,
      "eval_steps_per_second": 3.584,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 8.11,
      "eval_loss": 1.879622459411621,
      "eval_runtime": 17.0563,
      "eval_samples_per_second": 27.321,
      "eval_steps_per_second": 3.459,
      "eval_wer": 1.0003040437823048,
      "step": 300
    },
    {
      "epoch": 10.81,
      "eval_loss": 0.679625391960144,
      "eval_runtime": 16.4506,
      "eval_samples_per_second": 28.327,
      "eval_steps_per_second": 3.587,
      "eval_wer": 0.7686226816661599,
      "step": 400
    },
    {
      "epoch": 13.51,
      "learning_rate": 0.0002988,
      "loss": 3.3102,
      "step": 500
    },
    {
      "epoch": 13.51,
      "eval_loss": 0.6372817158699036,
      "eval_runtime": 16.5643,
      "eval_samples_per_second": 28.133,
      "eval_steps_per_second": 3.562,
      "eval_wer": 0.7297050775311645,
      "step": 500
    },
    {
      "epoch": 15.0,
      "step": 555,
      "total_flos": 2.6702078188467456e+18,
      "train_loss": 3.0021808194684554,
      "train_runtime": 909.5121,
      "train_samples_per_second": 19.379,
      "train_steps_per_second": 0.61
    }
  ],
  "max_steps": 555,
  "num_train_epochs": 15,
  "total_flos": 2.6702078188467456e+18,
  "trial_name": null,
  "trial_params": null
}