trainer_state.json

{
  "best_metric": 0.4978097081184387,
  "best_model_checkpoint": "output_dir/checkpoint-168",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 168,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.7504690431519699,
      "eval_loss": 0.6844255328178406,
      "eval_runtime": 31.6654,
      "eval_samples_per_second": 10.516,
      "eval_steps_per_second": 0.189,
      "step": 21
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7504690431519699,
      "eval_loss": 0.6841882467269897,
      "eval_runtime": 29.3795,
      "eval_samples_per_second": 11.334,
      "eval_steps_per_second": 0.204,
      "step": 42
    },
    {
      "epoch": 2.38,
      "grad_norm": 1.0868926048278809,
      "learning_rate": 7.61904761904762e-06,
      "loss": 0.6614,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9393939393939394,
      "eval_loss": 0.6861188411712646,
      "eval_runtime": 29.5784,
      "eval_samples_per_second": 11.258,
      "eval_steps_per_second": 0.203,
      "step": 63
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.15668202764976957,
      "eval_loss": 0.6874817609786987,
      "eval_runtime": 30.604,
      "eval_samples_per_second": 10.881,
      "eval_steps_per_second": 0.196,
      "step": 84
    },
    {
      "epoch": 4.76,
      "grad_norm": 1.4174736738204956,
      "learning_rate": 5.2380952380952384e-06,
      "loss": 0.5543,
      "step": 100
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.2608695652173913,
      "eval_loss": 0.682039737701416,
      "eval_runtime": 31.2133,
      "eval_samples_per_second": 10.669,
      "eval_steps_per_second": 0.192,
      "step": 105
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.8636363636363636,
      "eval_loss": 0.5800318717956543,
      "eval_runtime": 29.3466,
      "eval_samples_per_second": 11.347,
      "eval_steps_per_second": 0.204,
      "step": 126
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.8235294117647058,
      "eval_loss": 0.5767043232917786,
      "eval_runtime": 29.766,
      "eval_samples_per_second": 11.187,
      "eval_steps_per_second": 0.202,
      "step": 147
    },
    {
      "epoch": 7.14,
      "grad_norm": 2.605710506439209,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.4746,
      "step": 150
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.9159891598915989,
      "eval_loss": 0.4978097081184387,
      "eval_runtime": 29.603,
      "eval_samples_per_second": 11.249,
      "eval_steps_per_second": 0.203,
      "step": 168
    }
  ],
  "logging_steps": 50,
  "max_steps": 210,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 156512489226240.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}