last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1259181532004197,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 9.99476390346275e-06,
      "loss": 1.6498,
      "step": 500
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.430292611666101,
      "eval_f1": 0.31503797170432934,
      "eval_loss": 1.56297767162323,
      "eval_runtime": 713.0904,
      "eval_samples_per_second": 190.071,
      "eval_steps_per_second": 0.743,
      "step": 600
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.989517313746065e-06,
      "loss": 1.5882,
      "step": 1000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.4599964585577476,
      "eval_f1": 0.38184708517142385,
      "eval_loss": 1.539364218711853,
      "eval_runtime": 713.8848,
      "eval_samples_per_second": 189.86,
      "eval_steps_per_second": 0.742,
      "step": 1200
    }
  ],
  "max_steps": 953000,
  "num_train_epochs": 100,
  "total_flos": 5.046794543497728e+16,
  "trial_name": null,
  "trial_params": null
}