trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6156706809997559,
  "best_model_checkpoint": "bert_base_rand_50_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": 1.2525804042816162,
      "learning_rate": 4.9e-05,
      "loss": 0.6124,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6156706809997559,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6736,
      "eval_samples_per_second": 1548.461,
      "eval_steps_per_second": 7.423,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5342663526535034,
      "learning_rate": 4.8e-05,
      "loss": 0.5879,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6845637559890747,
      "eval_loss": 0.6281534433364868,
      "eval_matthews_correlation": 0.08911780619538918,
      "eval_runtime": 0.6927,
      "eval_samples_per_second": 1505.607,
      "eval_steps_per_second": 7.218,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.4237747192382812,
      "learning_rate": 4.7e-05,
      "loss": 0.5343,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6759347915649414,
      "eval_loss": 0.640214204788208,
      "eval_matthews_correlation": 0.13475013027835261,
      "eval_runtime": 0.6746,
      "eval_samples_per_second": 1546.161,
      "eval_steps_per_second": 7.412,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.614173412322998,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4824,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6768935918807983,
      "eval_loss": 0.7317802906036377,
      "eval_matthews_correlation": 0.1245506910083033,
      "eval_runtime": 0.6812,
      "eval_samples_per_second": 1531.047,
      "eval_steps_per_second": 7.34,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.1389687061309814,
      "learning_rate": 4.5e-05,
      "loss": 0.4231,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6720997095108032,
      "eval_loss": 0.7201704978942871,
      "eval_matthews_correlation": 0.13265732178325915,
      "eval_runtime": 0.6971,
      "eval_samples_per_second": 1496.233,
      "eval_steps_per_second": 7.173,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.271090030670166,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3659,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6462128758430481,
      "eval_loss": 0.8767686486244202,
      "eval_matthews_correlation": 0.1385509173876989,
      "eval_runtime": 0.6919,
      "eval_samples_per_second": 1507.55,
      "eval_steps_per_second": 7.227,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5009848510517794,
      "train_runtime": 99.932,
      "train_samples_per_second": 4278.411,
      "train_steps_per_second": 17.012
    }
  ],
  "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
}