checkpoint-150/trainer_state.json

{
  "best_metric": 0.6410807371139526,
  "best_model_checkpoint": "output/checkpoint-50",
  "epoch": 0.12025253031365868,
  "eval_steps": 50,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.5558,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2e-05,
      "loss": 0.5679,
      "step": 50
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.6776119402985075,
      "eval_loss": 0.6410807371139526,
      "eval_runtime": 61.731,
      "eval_samples_per_second": 5.427,
      "eval_steps_per_second": 1.361,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.6108,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.6925373134328359,
      "eval_loss": 0.6733226776123047,
      "eval_runtime": 56.5238,
      "eval_samples_per_second": 5.927,
      "eval_steps_per_second": 1.486,
      "step": 100
    },
    {
      "epoch": 0.12,
      "learning_rate": 6.444444444444446e-05,
      "loss": 0.5865,
      "step": 150
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.6865671641791045,
      "eval_loss": 0.7122625708580017,
      "eval_runtime": 61.5948,
      "eval_samples_per_second": 5.439,
      "eval_steps_per_second": 1.364,
      "step": 150
    }
  ],
  "logging_steps": 50,
  "max_steps": 1247,
  "num_train_epochs": 1,
  "save_steps": 50,
  "total_flos": 5.55301623576576e+16,
  "trial_name": null,
  "trial_params": null
}