trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6191930770874023,
  "best_model_checkpoint": "bert_base_rand_5_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": 0.798346996307373,
      "learning_rate": 4.9e-05,
      "loss": 0.6104,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6191930770874023,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6716,
      "eval_samples_per_second": 1553.047,
      "eval_steps_per_second": 7.445,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9842283725738525,
      "learning_rate": 4.8e-05,
      "loss": 0.5898,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6864812970161438,
      "eval_loss": 0.6220880746841431,
      "eval_matthews_correlation": 0.08320014042985302,
      "eval_runtime": 0.6722,
      "eval_samples_per_second": 1551.583,
      "eval_steps_per_second": 7.438,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.379241943359375,
      "learning_rate": 4.7e-05,
      "loss": 0.5378,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6586769223213196,
      "eval_loss": 0.6418293118476868,
      "eval_matthews_correlation": 0.04521871317410409,
      "eval_runtime": 0.69,
      "eval_samples_per_second": 1511.669,
      "eval_steps_per_second": 7.247,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.134252071380615,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4934,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6558005809783936,
      "eval_loss": 0.6848466396331787,
      "eval_matthews_correlation": 0.12073105148250744,
      "eval_runtime": 0.701,
      "eval_samples_per_second": 1487.937,
      "eval_steps_per_second": 7.133,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2500391006469727,
      "learning_rate": 4.5e-05,
      "loss": 0.434,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.665388286113739,
      "eval_loss": 0.6772574186325073,
      "eval_matthews_correlation": 0.11415043333891922,
      "eval_runtime": 0.668,
      "eval_samples_per_second": 1561.328,
      "eval_steps_per_second": 7.485,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.0308282375335693,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3853,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6299136877059937,
      "eval_loss": 0.7984164953231812,
      "eval_matthews_correlation": 0.05305184855723889,
      "eval_runtime": 0.7002,
      "eval_samples_per_second": 1489.655,
      "eval_steps_per_second": 7.141,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.5084464176028383,
      "train_runtime": 99.0348,
      "train_samples_per_second": 4317.168,
      "train_steps_per_second": 17.166
    }
  ],
  "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
}