checkpoint-33/trainer_state.json

{
  "best_metric": NaN,
  "best_model_checkpoint": "/tmp/model/checkpoint-33",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 26
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 28
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 29
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 31
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 32
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 33
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.07692307692307693,
      "eval_f1_macro": 0.020408163265306124,
      "eval_f1_micro": 0.07692307692307693,
      "eval_f1_weighted": 0.010989010989010992,
      "eval_loss": NaN,
      "eval_precision_macro": 0.01098901098901099,
      "eval_precision_micro": 0.07692307692307693,
      "eval_precision_weighted": 0.00591715976331361,
      "eval_recall_macro": 0.14285714285714285,
      "eval_recall_micro": 0.07692307692307693,
      "eval_recall_weighted": 0.07692307692307693,
      "eval_runtime": 1.2533,
      "eval_samples_per_second": 51.861,
      "eval_steps_per_second": 3.989,
      "step": 33
    }
  ],
  "logging_steps": 1,
  "max_steps": 99,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 5.895722366867866e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}