last-checkpoint/trainer_state.json

{
  "best_metric": 0.7603304380457284,
  "best_model_checkpoint": "/opt/ml/checkpoints/toxicity-target-type-identification-2023-02-13-13-22-05-747/model/checkpoint-355",
  "epoch": 3.0,
  "global_step": 1065,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7505285412262156,
      "eval_f1": 0.7603304380457284,
      "eval_loss": 0.700143575668335,
      "eval_precision": 0.7813181118720665,
      "eval_recall": 0.7505285412262156,
      "eval_runtime": 5.4125,
      "eval_samples_per_second": 174.779,
      "eval_steps_per_second": 21.986,
      "step": 355
    },
    {
      "epoch": 1.41,
      "learning_rate": 3.766830354166815e-05,
      "loss": 0.7919,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7505285412262156,
      "eval_f1": 0.745221599691193,
      "eval_loss": 1.095289707183838,
      "eval_precision": 0.758963982756652,
      "eval_recall": 0.7505285412262156,
      "eval_runtime": 5.6768,
      "eval_samples_per_second": 166.643,
      "eval_steps_per_second": 20.963,
      "step": 710
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.581272208641356e-05,
      "loss": 0.5218,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7484143763213531,
      "eval_f1": 0.7550927022068308,
      "eval_loss": 1.4217394590377808,
      "eval_precision": 0.7688482844579471,
      "eval_recall": 0.7484143763213531,
      "eval_runtime": 5.7254,
      "eval_samples_per_second": 165.23,
      "eval_steps_per_second": 20.785,
      "step": 1065
    }
  ],
  "max_steps": 10650,
  "num_train_epochs": 30,
  "total_flos": 415329154924194.0,
  "trial_name": null,
  "trial_params": null
}