trainer_state.json

{
  "best_metric": 0.6930913925170898,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_qnli/checkpoint-2457",
  "epoch": 8.0,
  "global_step": 6552,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.6934,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4938678381841479,
      "eval_loss": 0.6931551694869995,
      "eval_runtime": 11.6345,
      "eval_samples_per_second": 469.553,
      "eval_steps_per_second": 3.696,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.6933,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6932691335678101,
      "eval_runtime": 11.6609,
      "eval_samples_per_second": 468.488,
      "eval_steps_per_second": 3.688,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.6932,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930913925170898,
      "eval_runtime": 11.6537,
      "eval_samples_per_second": 468.777,
      "eval_steps_per_second": 3.69,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6932,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4946000366099213,
      "eval_loss": 0.6932663917541504,
      "eval_runtime": 11.6584,
      "eval_samples_per_second": 468.591,
      "eval_steps_per_second": 3.688,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.6932,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931136250495911,
      "eval_runtime": 11.7253,
      "eval_samples_per_second": 465.915,
      "eval_steps_per_second": 3.667,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6932,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931105256080627,
      "eval_runtime": 11.6627,
      "eval_samples_per_second": 468.416,
      "eval_steps_per_second": 3.687,
      "step": 4914
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.6932,
      "step": 5733
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6931049823760986,
      "eval_runtime": 11.6589,
      "eval_samples_per_second": 468.567,
      "eval_steps_per_second": 3.688,
      "step": 5733
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 0.6932,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5053999633900788,
      "eval_loss": 0.6930971145629883,
      "eval_runtime": 11.6722,
      "eval_samples_per_second": 468.037,
      "eval_steps_per_second": 3.684,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "step": 6552,
      "total_flos": 2.4786690056912896e+16,
      "train_loss": 0.6932271507924584,
      "train_runtime": 5675.5068,
      "train_samples_per_second": 922.763,
      "train_steps_per_second": 7.215
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 2.4786690056912896e+16,
  "trial_name": null,
  "trial_params": null
}