{
  "best_global_step": 2844,
  "best_metric": 0.3926401436328888,
  "best_model_checkpoint": "bert_base_km_100_v1_qqp/checkpoint-2844",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 9954,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.8982822895050049,
      "learning_rate": 4.9e-05,
      "loss": 0.483,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7904031659658669,
      "eval_combined_score": 0.7393263714458869,
      "eval_f1": 0.6882495769259068,
      "eval_loss": 0.44323667883872986,
      "eval_runtime": 25.4507,
      "eval_samples_per_second": 1588.561,
      "eval_steps_per_second": 6.208,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.774909734725952,
      "learning_rate": 4.8e-05,
      "loss": 0.3749,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8175117487014593,
      "eval_combined_score": 0.7824289104441479,
      "eval_f1": 0.7473460721868365,
      "eval_loss": 0.3926401436328888,
      "eval_runtime": 25.2361,
      "eval_samples_per_second": 1602.072,
      "eval_steps_per_second": 6.261,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.4883244037628174,
      "learning_rate": 4.7e-05,
      "loss": 0.2912,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8258224091021519,
      "eval_combined_score": 0.7952857619106002,
      "eval_f1": 0.7647491147190486,
      "eval_loss": 0.3976278007030487,
      "eval_runtime": 25.289,
      "eval_samples_per_second": 1598.716,
      "eval_steps_per_second": 6.248,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.621770143508911,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2189,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8304229532525352,
      "eval_combined_score": 0.7960264408151279,
      "eval_f1": 0.7616299283777206,
      "eval_loss": 0.4470672607421875,
      "eval_runtime": 25.3207,
      "eval_samples_per_second": 1596.715,
      "eval_steps_per_second": 6.24,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6303458213806152,
      "learning_rate": 4.5e-05,
      "loss": 0.1641,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8310165718525847,
      "eval_combined_score": 0.7998136439158609,
      "eval_f1": 0.768610715979137,
      "eval_loss": 0.4701743721961975,
      "eval_runtime": 25.3192,
      "eval_samples_per_second": 1596.81,
      "eval_steps_per_second": 6.24,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.171130180358887,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.126,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8332426416027702,
      "eval_combined_score": 0.8016329185636992,
      "eval_f1": 0.7700231955246282,
      "eval_loss": 0.5648208856582642,
      "eval_runtime": 25.1891,
      "eval_samples_per_second": 1605.057,
      "eval_steps_per_second": 6.273,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.9093337059021,
      "learning_rate": 4.3e-05,
      "loss": 0.1003,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8328468958694039,
      "eval_combined_score": 0.7983261461033024,
      "eval_f1": 0.7638053963372011,
      "eval_loss": 0.6059678196907043,
      "eval_runtime": 25.1988,
      "eval_samples_per_second": 1604.439,
      "eval_steps_per_second": 6.27,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "step": 9954,
      "total_flos": 3.3506166766980096e+17,
      "train_loss": 0.2511948379905966,
      "train_runtime": 4263.6329,
      "train_samples_per_second": 4266.854,
      "train_steps_per_second": 16.676
    }
  ],
  "logging_steps": 1,
  "max_steps": 71100,
  "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": 3.3506166766980096e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}