trainer_state.json

{
  "best_metric": 1.0984379053115845,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_mnli/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": 1.0988,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0988268852233887,
      "eval_runtime": 21.0346,
      "eval_samples_per_second": 466.613,
      "eval_steps_per_second": 3.661,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 1.0987,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3183902190524707,
      "eval_loss": 1.0985690355300903,
      "eval_runtime": 21.0362,
      "eval_samples_per_second": 466.577,
      "eval_steps_per_second": 3.66,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 1.0987,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.098904013633728,
      "eval_runtime": 21.0601,
      "eval_samples_per_second": 466.047,
      "eval_steps_per_second": 3.656,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.0987,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.098731517791748,
      "eval_runtime": 21.0671,
      "eval_samples_per_second": 465.892,
      "eval_steps_per_second": 3.655,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 1.0987,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0984379053115845,
      "eval_runtime": 21.029,
      "eval_samples_per_second": 466.737,
      "eval_steps_per_second": 3.662,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0986,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0986697673797607,
      "eval_runtime": 21.0827,
      "eval_samples_per_second": 465.548,
      "eval_steps_per_second": 3.652,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 1.0986,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0993210077285767,
      "eval_runtime": 21.0085,
      "eval_samples_per_second": 467.192,
      "eval_steps_per_second": 3.665,
      "step": 21476
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 1.0986,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985081195831299,
      "eval_runtime": 21.0338,
      "eval_samples_per_second": 466.631,
      "eval_steps_per_second": 3.661,
      "step": 24544
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.1e-05,
      "loss": 1.0986,
      "step": 27612
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985286235809326,
      "eval_runtime": 21.1222,
      "eval_samples_per_second": 464.677,
      "eval_steps_per_second": 3.645,
      "step": 27612
    },
    {
      "epoch": 10.0,
      "learning_rate": 4e-05,
      "loss": 1.0986,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986618995666504,
      "eval_runtime": 21.0328,
      "eval_samples_per_second": 466.652,
      "eval_steps_per_second": 3.661,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "step": 30680,
      "total_flos": 1.1616577644396544e+17,
      "train_loss": 1.0986627252969157,
      "train_runtime": 26385.2507,
      "train_samples_per_second": 744.17,
      "train_steps_per_second": 5.814
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 1.1616577644396544e+17,
  "trial_name": null,
  "trial_params": null
}