trainer_state.json

{
  "best_metric": 0.3913634121417999,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_rte/checkpoint-60",
  "epoch": 8.0,
  "global_step": 160,
  "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": 0.4089,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.3934378921985626,
      "eval_runtime": 0.6831,
      "eval_samples_per_second": 405.5,
      "eval_steps_per_second": 4.392,
      "step": 20
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.4084,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.39218229055404663,
      "eval_runtime": 0.6827,
      "eval_samples_per_second": 405.726,
      "eval_steps_per_second": 4.394,
      "step": 40
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.4078,
      "step": 60
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.3913634121417999,
      "eval_runtime": 0.6893,
      "eval_samples_per_second": 401.851,
      "eval_steps_per_second": 4.352,
      "step": 60
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4073,
      "step": 80
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.39413484930992126,
      "eval_runtime": 0.6787,
      "eval_samples_per_second": 408.129,
      "eval_steps_per_second": 4.42,
      "step": 80
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.4076,
      "step": 100
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.39272287487983704,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 405.661,
      "eval_steps_per_second": 4.393,
      "step": 100
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4065,
      "step": 120
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.39255064725875854,
      "eval_runtime": 0.6874,
      "eval_samples_per_second": 402.968,
      "eval_steps_per_second": 4.364,
      "step": 120
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.4013,
      "step": 140
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.47653429602888087,
      "eval_loss": 0.4076071083545685,
      "eval_runtime": 0.6808,
      "eval_samples_per_second": 406.858,
      "eval_steps_per_second": 4.406,
      "step": 140
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 0.3911,
      "step": 160
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.48375451263537905,
      "eval_loss": 0.4072694778442383,
      "eval_runtime": 0.6817,
      "eval_samples_per_second": 406.342,
      "eval_steps_per_second": 4.401,
      "step": 160
    },
    {
      "epoch": 8.0,
      "step": 160,
      "total_flos": 589240857526272.0,
      "train_loss": 0.4048383593559265,
      "train_runtime": 209.2775,
      "train_samples_per_second": 594.904,
      "train_steps_per_second": 4.778
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 50,
  "total_flos": 589240857526272.0,
  "trial_name": null,
  "trial_params": null
}