run-4/checkpoint-4276/trainer_state.json

{
  "best_metric": 0.4986084626480731,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-2138",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 4276,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4677268475210477,
      "grad_norm": 13.482518196105957,
      "learning_rate": 7.034296344775407e-06,
      "loss": 0.5589,
      "step": 500
    },
    {
      "epoch": 0.9354536950420954,
      "grad_norm": 10.600051879882812,
      "learning_rate": 6.308362769463187e-06,
      "loss": 0.5047,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4949962794780731,
      "eval_matthews_correlation": 0.4219587641967578,
      "eval_runtime": 0.7754,
      "eval_samples_per_second": 1345.165,
      "eval_steps_per_second": 85.121,
      "step": 1069
    },
    {
      "epoch": 1.4031805425631432,
      "grad_norm": 26.401622772216797,
      "learning_rate": 5.5824291941509675e-06,
      "loss": 0.3949,
      "step": 1500
    },
    {
      "epoch": 1.8709073900841908,
      "grad_norm": 9.348161697387695,
      "learning_rate": 4.856495618838748e-06,
      "loss": 0.3944,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4984108507633209,
      "eval_matthews_correlation": 0.4986084626480731,
      "eval_runtime": 0.8094,
      "eval_samples_per_second": 1288.531,
      "eval_steps_per_second": 81.537,
      "step": 2138
    },
    {
      "epoch": 2.3386342376052385,
      "grad_norm": 2.740610122680664,
      "learning_rate": 4.130562043526529e-06,
      "loss": 0.3296,
      "step": 2500
    },
    {
      "epoch": 2.8063610851262863,
      "grad_norm": 64.62920379638672,
      "learning_rate": 3.4046284682143094e-06,
      "loss": 0.3363,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6585824489593506,
      "eval_matthews_correlation": 0.49323508482582806,
      "eval_runtime": 0.8558,
      "eval_samples_per_second": 1218.682,
      "eval_steps_per_second": 77.117,
      "step": 3207
    },
    {
      "epoch": 3.2740879326473338,
      "grad_norm": 12.445216178894043,
      "learning_rate": 2.6786948929020896e-06,
      "loss": 0.2843,
      "step": 3500
    },
    {
      "epoch": 3.7418147801683816,
      "grad_norm": 59.14677047729492,
      "learning_rate": 1.9527613175898707e-06,
      "loss": 0.2771,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7595688700675964,
      "eval_matthews_correlation": 0.4963133749068313,
      "eval_runtime": 0.7761,
      "eval_samples_per_second": 1343.845,
      "eval_steps_per_second": 85.037,
      "step": 4276
    }
  ],
  "logging_steps": 500,
  "max_steps": 5345,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 150431219795616.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.760229920087626e-06,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 8,
    "seed": 19
  }
}