checkpoint-2000/trainer_state.json

{
  "best_metric": 0.6859038472175598,
  "best_model_checkpoint": "deberta_v3_finetuned_predicting_effective_arguments/checkpoint-2000",
  "epoch": 0.9667673716012085,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "learning_rate": 9.518375241779499e-06,
      "loss": 0.8315,
      "step": 500
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.7444673180580139,
      "eval_runtime": 29.7252,
      "eval_samples_per_second": 123.33,
      "eval_steps_per_second": 30.849,
      "step": 500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.034816247582205e-06,
      "loss": 0.7232,
      "step": 1000
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.7441162467002869,
      "eval_runtime": 29.4226,
      "eval_samples_per_second": 124.598,
      "eval_steps_per_second": 31.167,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.552224371373308e-06,
      "loss": 0.7269,
      "step": 1500
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.7453668117523193,
      "eval_runtime": 29.5754,
      "eval_samples_per_second": 123.955,
      "eval_steps_per_second": 31.006,
      "step": 1500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.068665377176016e-06,
      "loss": 0.7185,
      "step": 2000
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.6859038472175598,
      "eval_runtime": 29.569,
      "eval_samples_per_second": 123.981,
      "eval_steps_per_second": 31.012,
      "step": 2000
    }
  ],
  "max_steps": 10340,
  "num_train_epochs": 5,
  "total_flos": 1822810502059560.0,
  "trial_name": null,
  "trial_params": null
}