models/distilbert-base-uncased-offensive/checkpoint-1500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0134228187919465,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.67,
      "grad_norm": 5.409279823303223,
      "learning_rate": 4.161073825503356e-05,
      "loss": 0.4761,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8593023255813953,
      "eval_f1": 0.7055961070559611,
      "eval_loss": 0.3657611906528473,
      "eval_precision": 0.847953216374269,
      "eval_recall": 0.6041666666666666,
      "eval_runtime": 0.6409,
      "eval_samples_per_second": 1341.936,
      "eval_steps_per_second": 84.261,
      "step": 745
    },
    {
      "epoch": 1.34,
      "grad_norm": 8.624994277954102,
      "learning_rate": 3.3221476510067115e-05,
      "loss": 0.3911,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8430232558139535,
      "eval_f1": 0.7133757961783439,
      "eval_loss": 0.40293434262275696,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.7,
      "eval_runtime": 0.7138,
      "eval_samples_per_second": 1204.827,
      "eval_steps_per_second": 75.652,
      "step": 1490
    },
    {
      "epoch": 2.01,
      "grad_norm": 2.4068655967712402,
      "learning_rate": 2.4832214765100674e-05,
      "loss": 0.329,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2980,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 412170241115040.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}