trainer_state.json

{
  "best_metric": 0.6171163320541382,
  "best_model_checkpoint": "tiny_bert_rand_50_v1_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.7138214111328125,
      "learning_rate": 4.9e-05,
      "loss": 0.6134,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6171163320541382,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4916,
      "eval_samples_per_second": 2121.452,
      "eval_steps_per_second": 10.17,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9771915078163147,
      "learning_rate": 4.8e-05,
      "loss": 0.601,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6195900440216064,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3924,
      "eval_samples_per_second": 2657.785,
      "eval_steps_per_second": 12.741,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1553252935409546,
      "learning_rate": 4.7e-05,
      "loss": 0.5781,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6259943246841431,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3832,
      "eval_samples_per_second": 2721.81,
      "eval_steps_per_second": 13.048,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9693224430084229,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5374,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6836050152778625,
      "eval_loss": 0.6578595042228699,
      "eval_matthews_correlation": 0.042582902474918326,
      "eval_runtime": 0.3804,
      "eval_samples_per_second": 2741.559,
      "eval_steps_per_second": 13.143,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.186906337738037,
      "learning_rate": 4.5e-05,
      "loss": 0.4953,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6538830399513245,
      "eval_loss": 0.6752046942710876,
      "eval_matthews_correlation": 0.08801919690309869,
      "eval_runtime": 0.3951,
      "eval_samples_per_second": 2640.126,
      "eval_steps_per_second": 12.656,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.2683961391448975,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4567,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6222435235977173,
      "eval_loss": 0.7260271310806274,
      "eval_matthews_correlation": 0.07575741604681904,
      "eval_runtime": 0.4037,
      "eval_samples_per_second": 2583.494,
      "eval_steps_per_second": 12.385,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 1345426898614272.0,
      "train_loss": 0.5469665293600044,
      "train_runtime": 42.0475,
      "train_samples_per_second": 10168.271,
      "train_steps_per_second": 40.43
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 1345426898614272.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}