trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 798,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38,
      "eval_f1": 0.5937051387034437,
      "eval_loss": 0.5818731188774109,
      "eval_runtime": 2.6676,
      "eval_samples_per_second": 12.746,
      "eval_steps_per_second": 12.746,
      "step": 100
    },
    {
      "epoch": 0.75,
      "eval_f1": 0.6147189338467695,
      "eval_loss": 0.5344187617301941,
      "eval_runtime": 2.5934,
      "eval_samples_per_second": 13.11,
      "eval_steps_per_second": 13.11,
      "step": 200
    },
    {
      "epoch": 1.13,
      "eval_f1": 0.616141567791633,
      "eval_loss": 0.7152214050292969,
      "eval_runtime": 2.6291,
      "eval_samples_per_second": 12.932,
      "eval_steps_per_second": 12.932,
      "step": 300
    },
    {
      "epoch": 1.5,
      "eval_f1": 0.6262624949298642,
      "eval_loss": 0.5542092323303223,
      "eval_runtime": 2.8526,
      "eval_samples_per_second": 11.919,
      "eval_steps_per_second": 11.919,
      "step": 400
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.8671679197994987e-05,
      "loss": 0.6563,
      "step": 500
    },
    {
      "epoch": 1.88,
      "eval_f1": 0.6518478762537062,
      "eval_loss": 0.5657368898391724,
      "eval_runtime": 2.8114,
      "eval_samples_per_second": 12.093,
      "eval_steps_per_second": 12.093,
      "step": 500
    },
    {
      "epoch": 2.26,
      "eval_f1": 0.6421228870345159,
      "eval_loss": 0.602083683013916,
      "eval_runtime": 2.8782,
      "eval_samples_per_second": 11.813,
      "eval_steps_per_second": 11.813,
      "step": 600
    },
    {
      "epoch": 2.63,
      "eval_f1": 0.6347334214391783,
      "eval_loss": 0.6153014302253723,
      "eval_runtime": 2.7323,
      "eval_samples_per_second": 12.444,
      "eval_steps_per_second": 12.444,
      "step": 700
    },
    {
      "epoch": 3.0,
      "step": 798,
      "total_flos": 1287131566841280.0,
      "train_loss": 0.5280582647873345,
      "train_runtime": 227.8039,
      "train_samples_per_second": 3.503,
      "train_steps_per_second": 3.503
    }
  ],
  "logging_steps": 500,
  "max_steps": 798,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1287131566841280.0,
  "trial_name": null,
  "trial_params": null
}