checkpoint-529/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 529,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 1.6355,
      "step": 100
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 1.4037,
      "step": 200
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 1.4183,
      "step": 300
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.3881,
      "step": 400
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.3891,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6137948717948718,
      "eval_loss": 1.3015451431274414,
      "eval_runtime": 4.1821,
      "eval_samples_per_second": 119.557,
      "eval_steps_per_second": 15.064,
      "step": 529
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 23.2,
      "eval_f1": 35.17014759120022,
      "step": 529
    }
  ],
  "logging_steps": 100,
  "max_steps": 26450,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 8.922877128186266e+16,
  "trial_name": null,
  "trial_params": null
}