run-3/checkpoint-40/trainer_state.json

{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-3/checkpoint-8",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.62,
      "grad_norm": 0.94398033618927,
      "learning_rate": 7.138113611405621e-06,
      "loss": 0.6981,
      "step": 5
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6864166259765625,
      "eval_runtime": 1.3249,
      "eval_samples_per_second": 48.305,
      "eval_steps_per_second": 6.038,
      "step": 8
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.5316386222839355,
      "learning_rate": 1.1103732284408743e-05,
      "loss": 0.683,
      "step": 10
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.5028849840164185,
      "learning_rate": 1.031060854980812e-05,
      "loss": 0.6564,
      "step": 15
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.680694580078125,
      "eval_runtime": 1.3565,
      "eval_samples_per_second": 47.179,
      "eval_steps_per_second": 5.897,
      "step": 16
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5775301456451416,
      "learning_rate": 9.517484815207495e-06,
      "loss": 0.6436,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6852684020996094,
      "eval_runtime": 1.3445,
      "eval_samples_per_second": 47.601,
      "eval_steps_per_second": 5.95,
      "step": 24
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.4549338221549988,
      "learning_rate": 8.72436108060687e-06,
      "loss": 0.6098,
      "step": 25
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.3741567134857178,
      "learning_rate": 7.931237346006246e-06,
      "loss": 0.609,
      "step": 30
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6952018737792969,
      "eval_runtime": 1.3401,
      "eval_samples_per_second": 47.757,
      "eval_steps_per_second": 5.97,
      "step": 32
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.38969138264656067,
      "learning_rate": 7.138113611405621e-06,
      "loss": 0.6091,
      "step": 35
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2588692009449005,
      "learning_rate": 6.344989876804997e-06,
      "loss": 0.6043,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7042198181152344,
      "eval_runtime": 1.3815,
      "eval_samples_per_second": 46.326,
      "eval_steps_per_second": 5.791,
      "step": 40
    }
  ],
  "logging_steps": 5,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 5002699790009952.0,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.1420981778248994e-05,
    "per_device_train_batch_size": 24
  }
}