trainer_state.json

{
  "best_metric": 1.0985050201416016,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_mnli_128/checkpoint-15340",
  "epoch": 10.0,
  "global_step": 30680,
  "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": 1.0987,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0987107753753662,
      "eval_runtime": 17.0184,
      "eval_samples_per_second": 576.728,
      "eval_steps_per_second": 4.525,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 1.0986,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986207723617554,
      "eval_runtime": 17.0318,
      "eval_samples_per_second": 576.276,
      "eval_steps_per_second": 4.521,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 1.0986,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0988447666168213,
      "eval_runtime": 17.0815,
      "eval_samples_per_second": 574.599,
      "eval_steps_per_second": 4.508,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.0986,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986438989639282,
      "eval_runtime": 17.0583,
      "eval_samples_per_second": 575.381,
      "eval_steps_per_second": 4.514,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 1.0986,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985050201416016,
      "eval_runtime": 17.2861,
      "eval_samples_per_second": 567.797,
      "eval_steps_per_second": 4.454,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0986,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0986541509628296,
      "eval_runtime": 17.023,
      "eval_samples_per_second": 576.574,
      "eval_steps_per_second": 4.523,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 1.0986,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0987762212753296,
      "eval_runtime": 17.3871,
      "eval_samples_per_second": 564.5,
      "eval_steps_per_second": 4.429,
      "step": 21476
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 1.0986,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985809564590454,
      "eval_runtime": 17.0016,
      "eval_samples_per_second": 577.297,
      "eval_steps_per_second": 4.529,
      "step": 24544
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.1e-05,
      "loss": 1.0986,
      "step": 27612
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985565185546875,
      "eval_runtime": 17.2613,
      "eval_samples_per_second": 568.613,
      "eval_steps_per_second": 4.461,
      "step": 27612
    },
    {
      "epoch": 10.0,
      "learning_rate": 4e-05,
      "loss": 1.0986,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986559391021729,
      "eval_runtime": 17.0342,
      "eval_samples_per_second": 576.193,
      "eval_steps_per_second": 4.52,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "step": 30680,
      "total_flos": 9.374910732304384e+16,
      "train_loss": 1.098636202457831,
      "train_runtime": 19983.4376,
      "train_samples_per_second": 982.569,
      "train_steps_per_second": 7.676
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 9.374910732304384e+16,
  "trial_name": null,
  "trial_params": null
}