trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6167530417442322,
  "best_model_checkpoint": "bert_base_rand_10_v1_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8418697118759155,
      "learning_rate": 4.9e-05,
      "loss": 0.6138,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6167530417442322,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6744,
      "eval_samples_per_second": 1546.589,
      "eval_steps_per_second": 7.414,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7829123735427856,
      "learning_rate": 4.8e-05,
      "loss": 0.5908,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6874400973320007,
      "eval_loss": 0.6230852603912354,
      "eval_matthews_correlation": 0.06872035940447745,
      "eval_runtime": 0.6832,
      "eval_samples_per_second": 1526.656,
      "eval_steps_per_second": 7.319,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.31274676322937,
      "learning_rate": 4.7e-05,
      "loss": 0.5407,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6778523325920105,
      "eval_loss": 0.6428341865539551,
      "eval_matthews_correlation": 0.13137634266049855,
      "eval_runtime": 0.6724,
      "eval_samples_per_second": 1551.215,
      "eval_steps_per_second": 7.436,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.515699625015259,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4904,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6366251111030579,
      "eval_loss": 0.7110387086868286,
      "eval_matthews_correlation": 0.1286880975824553,
      "eval_runtime": 0.6936,
      "eval_samples_per_second": 1503.807,
      "eval_steps_per_second": 7.209,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0114657878875732,
      "learning_rate": 4.5e-05,
      "loss": 0.4418,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6692233681678772,
      "eval_loss": 0.6772195100784302,
      "eval_matthews_correlation": 0.1075479635452414,
      "eval_runtime": 0.6716,
      "eval_samples_per_second": 1552.896,
      "eval_steps_per_second": 7.444,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.488552570343018,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3958,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6471716165542603,
      "eval_loss": 0.823727548122406,
      "eval_matthews_correlation": 0.07524548978135541,
      "eval_runtime": 0.7017,
      "eval_samples_per_second": 1486.482,
      "eval_steps_per_second": 7.126,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5122250669142779,
      "train_runtime": 99.4486,
      "train_samples_per_second": 4299.206,
      "train_steps_per_second": 17.094
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6749587903150080.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}