trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 200.0,
  "eval_steps": 500,
  "global_step": 6800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 14.705882352941176,
      "grad_norm": 0.8522269129753113,
      "learning_rate": 0.0009264705882352942,
      "loss": 1.7538,
      "step": 500
    },
    {
      "epoch": 29.41176470588235,
      "grad_norm": 0.24410583078861237,
      "learning_rate": 0.0008529411764705882,
      "loss": 0.0574,
      "step": 1000
    },
    {
      "epoch": 44.11764705882353,
      "grad_norm": 0.32126477360725403,
      "learning_rate": 0.0007794117647058824,
      "loss": 0.0246,
      "step": 1500
    },
    {
      "epoch": 58.8235294117647,
      "grad_norm": 0.1633734405040741,
      "learning_rate": 0.0007058823529411765,
      "loss": 0.0177,
      "step": 2000
    },
    {
      "epoch": 73.52941176470588,
      "grad_norm": 1.6365108489990234,
      "learning_rate": 0.0006323529411764706,
      "loss": 0.0407,
      "step": 2500
    },
    {
      "epoch": 88.23529411764706,
      "grad_norm": 0.11134446412324905,
      "learning_rate": 0.0005588235294117647,
      "loss": 0.0197,
      "step": 3000
    },
    {
      "epoch": 102.94117647058823,
      "grad_norm": 1.3109087944030762,
      "learning_rate": 0.0004852941176470588,
      "loss": 0.0146,
      "step": 3500
    },
    {
      "epoch": 117.6470588235294,
      "grad_norm": 0.1169649288058281,
      "learning_rate": 0.0004117647058823529,
      "loss": 0.0043,
      "step": 4000
    },
    {
      "epoch": 132.35294117647058,
      "grad_norm": 0.061263132840394974,
      "learning_rate": 0.0003382352941176471,
      "loss": 0.0044,
      "step": 4500
    },
    {
      "epoch": 147.05882352941177,
      "grad_norm": 0.019590700045228004,
      "learning_rate": 0.0002647058823529412,
      "loss": 0.0025,
      "step": 5000
    },
    {
      "epoch": 161.76470588235293,
      "grad_norm": 0.011288847774267197,
      "learning_rate": 0.00019117647058823528,
      "loss": 0.0024,
      "step": 5500
    },
    {
      "epoch": 176.47058823529412,
      "grad_norm": 0.011768895201385021,
      "learning_rate": 0.00011764705882352942,
      "loss": 0.0017,
      "step": 6000
    },
    {
      "epoch": 191.1764705882353,
      "grad_norm": 0.010613554157316685,
      "learning_rate": 4.411764705882353e-05,
      "loss": 0.0015,
      "step": 6500
    },
    {
      "epoch": 200.0,
      "step": 6800,
      "total_flos": 1.175134666752e+17,
      "train_loss": 0.14310694140546462,
      "train_runtime": 11074.5138,
      "train_samples_per_second": 2.438,
      "train_steps_per_second": 0.614
    }
  ],
  "logging_steps": 500,
  "max_steps": 6800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 200,
  "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": 1.175134666752e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}