trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6170077323913574,
  "best_model_checkpoint": "bert_base_rand_100_v2_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": 1.0025776624679565,
      "learning_rate": 4.9e-05,
      "loss": 0.6165,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6170077323913574,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6754,
      "eval_samples_per_second": 1544.361,
      "eval_steps_per_second": 7.403,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.582245111465454,
      "learning_rate": 4.8e-05,
      "loss": 0.5893,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6883988380432129,
      "eval_loss": 0.6173972487449646,
      "eval_matthews_correlation": 0.01845565733408863,
      "eval_runtime": 0.6896,
      "eval_samples_per_second": 1512.377,
      "eval_steps_per_second": 7.25,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8974895477294922,
      "learning_rate": 4.7e-05,
      "loss": 0.5377,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6730585098266602,
      "eval_loss": 0.6450505256652832,
      "eval_matthews_correlation": 0.10603855839346674,
      "eval_runtime": 0.6911,
      "eval_samples_per_second": 1509.187,
      "eval_steps_per_second": 7.235,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.176020622253418,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4833,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.668264627456665,
      "eval_loss": 0.7213912010192871,
      "eval_matthews_correlation": 0.10879024574677724,
      "eval_runtime": 0.6856,
      "eval_samples_per_second": 1521.385,
      "eval_steps_per_second": 7.293,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.716684579849243,
      "learning_rate": 4.5e-05,
      "loss": 0.4189,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6442952752113342,
      "eval_loss": 0.6887880563735962,
      "eval_matthews_correlation": 0.0983853715855654,
      "eval_runtime": 0.6891,
      "eval_samples_per_second": 1513.517,
      "eval_steps_per_second": 7.256,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.1573657989501953,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3763,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.641418993473053,
      "eval_loss": 0.7857490181922913,
      "eval_matthews_correlation": 0.12453997012075438,
      "eval_runtime": 0.7316,
      "eval_samples_per_second": 1425.606,
      "eval_steps_per_second": 6.834,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5036639653000177,
      "train_runtime": 100.6474,
      "train_samples_per_second": 4248.0,
      "train_steps_per_second": 16.891
    }
  ],
  "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
}