trainer_state.json

{
  "best_metric": 0.6978054046630859,
  "best_model_checkpoint": "add_BERT_no_pretrain_sst2/checkpoint-527",
  "epoch": 6.0,
  "global_step": 3162,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.0004900379506641366,
      "loss": 0.7276,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6978054046630859,
      "eval_runtime": 1.3727,
      "eval_samples_per_second": 635.255,
      "eval_steps_per_second": 5.1,
      "step": 527
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048003795066413666,
      "loss": 0.6872,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.7013314366340637,
      "eval_runtime": 1.3699,
      "eval_samples_per_second": 636.538,
      "eval_steps_per_second": 5.11,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047003795066413663,
      "loss": 0.6873,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6995900273323059,
      "eval_runtime": 1.3659,
      "eval_samples_per_second": 638.425,
      "eval_steps_per_second": 5.125,
      "step": 1581
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0004604554079696395,
      "loss": 0.6868,
      "step": 2108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4908256880733945,
      "eval_loss": NaN,
      "eval_runtime": 1.3643,
      "eval_samples_per_second": 639.154,
      "eval_steps_per_second": 5.131,
      "step": 2108
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.000453168880455408,
      "loss": 0.6864,
      "step": 2635
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4908256880733945,
      "eval_loss": NaN,
      "eval_runtime": 1.3451,
      "eval_samples_per_second": 648.288,
      "eval_steps_per_second": 5.204,
      "step": 2635
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.000443168880455408,
      "loss": 0.0,
      "step": 3162
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4908256880733945,
      "eval_loss": NaN,
      "eval_runtime": 1.3487,
      "eval_samples_per_second": 646.555,
      "eval_steps_per_second": 5.19,
      "step": 3162
    },
    {
      "epoch": 6.0,
      "step": 3162,
      "total_flos": 5.771431023476736e+16,
      "train_loss": 0.5792089338019104,
      "train_runtime": 1859.6741,
      "train_samples_per_second": 1810.774,
      "train_steps_per_second": 14.169
    }
  ],
  "max_steps": 26350,
  "num_train_epochs": 50,
  "total_flos": 5.771431023476736e+16,
  "trial_name": null,
  "trial_params": null
}