trainer_state.json

{
  "best_metric": 1.0985816717147827,
  "best_model_checkpoint": "hBERTv1_no_pretrain_mnli/checkpoint-16364",
  "epoch": 9.0,
  "global_step": 36819,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 1.1037,
      "step": 4091
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0994353294372559,
      "eval_runtime": 45.3863,
      "eval_samples_per_second": 216.255,
      "eval_steps_per_second": 2.269,
      "step": 4091
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 1.0988,
      "step": 8182
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986318588256836,
      "eval_runtime": 45.3182,
      "eval_samples_per_second": 216.58,
      "eval_steps_per_second": 2.273,
      "step": 8182
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 1.0987,
      "step": 12273
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0989253520965576,
      "eval_runtime": 45.2425,
      "eval_samples_per_second": 216.942,
      "eval_steps_per_second": 2.277,
      "step": 12273
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 1.0987,
      "step": 16364
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985816717147827,
      "eval_runtime": 45.2148,
      "eval_samples_per_second": 217.075,
      "eval_steps_per_second": 2.278,
      "step": 16364
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 1.0987,
      "step": 20455
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985941886901855,
      "eval_runtime": 45.411,
      "eval_samples_per_second": 216.137,
      "eval_steps_per_second": 2.268,
      "step": 20455
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 1.0987,
      "step": 24546
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.098624587059021,
      "eval_runtime": 46.2085,
      "eval_samples_per_second": 212.407,
      "eval_steps_per_second": 2.229,
      "step": 24546
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 1.0986,
      "step": 28637
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986192226409912,
      "eval_runtime": 46.0885,
      "eval_samples_per_second": 212.96,
      "eval_steps_per_second": 2.235,
      "step": 28637
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 1.0986,
      "step": 32728
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0986138582229614,
      "eval_runtime": 46.1524,
      "eval_samples_per_second": 212.665,
      "eval_steps_per_second": 2.232,
      "step": 32728
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 1.0986,
      "step": 36819
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.098609447479248,
      "eval_runtime": 46.0831,
      "eval_samples_per_second": 212.985,
      "eval_steps_per_second": 2.235,
      "step": 36819
    },
    {
      "epoch": 9.0,
      "step": 36819,
      "total_flos": 5.2342526347680154e+17,
      "train_loss": 1.099236006078723,
      "train_runtime": 48831.8135,
      "train_samples_per_second": 402.096,
      "train_steps_per_second": 4.189
    }
  ],
  "max_steps": 204550,
  "num_train_epochs": 50,
  "total_flos": 5.2342526347680154e+17,
  "trial_name": null,
  "trial_params": null
}