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": 0.9709584712982178,
      "learning_rate": 0.0009264705882352942,
      "loss": 6.89,
      "step": 500
    },
    {
      "epoch": 29.41176470588235,
      "grad_norm": 0.5043423771858215,
      "learning_rate": 0.0008529411764705882,
      "loss": 1.6253,
      "step": 1000
    },
    {
      "epoch": 44.11764705882353,
      "grad_norm": 0.24679997563362122,
      "learning_rate": 0.0007794117647058824,
      "loss": 0.0784,
      "step": 1500
    },
    {
      "epoch": 58.8235294117647,
      "grad_norm": 0.13042904436588287,
      "learning_rate": 0.0007058823529411765,
      "loss": 0.011,
      "step": 2000
    },
    {
      "epoch": 73.52941176470588,
      "grad_norm": 0.1388075053691864,
      "learning_rate": 0.0006323529411764706,
      "loss": 0.0133,
      "step": 2500
    },
    {
      "epoch": 88.23529411764706,
      "grad_norm": 0.21229323744773865,
      "learning_rate": 0.0005588235294117647,
      "loss": 0.0072,
      "step": 3000
    },
    {
      "epoch": 102.94117647058823,
      "grad_norm": 0.055483490228652954,
      "learning_rate": 0.0004852941176470588,
      "loss": 0.0052,
      "step": 3500
    },
    {
      "epoch": 117.6470588235294,
      "grad_norm": 0.06371592730283737,
      "learning_rate": 0.0004117647058823529,
      "loss": 0.0032,
      "step": 4000
    },
    {
      "epoch": 132.35294117647058,
      "grad_norm": 0.019501326605677605,
      "learning_rate": 0.0003382352941176471,
      "loss": 0.0028,
      "step": 4500
    },
    {
      "epoch": 147.05882352941177,
      "grad_norm": 0.18367668986320496,
      "learning_rate": 0.0002647058823529412,
      "loss": 0.0034,
      "step": 5000
    },
    {
      "epoch": 161.76470588235293,
      "grad_norm": 0.026915088295936584,
      "learning_rate": 0.00019117647058823528,
      "loss": 0.0029,
      "step": 5500
    },
    {
      "epoch": 176.47058823529412,
      "grad_norm": 0.014995796605944633,
      "learning_rate": 0.00011764705882352942,
      "loss": 0.0018,
      "step": 6000
    },
    {
      "epoch": 191.1764705882353,
      "grad_norm": 0.04031387344002724,
      "learning_rate": 4.411764705882353e-05,
      "loss": 0.0015,
      "step": 6500
    },
    {
      "epoch": 200.0,
      "step": 6800,
      "total_flos": 1.1838393679872e+17,
      "train_loss": 0.6357850695752045,
      "train_runtime": 11166.7529,
      "train_samples_per_second": 2.436,
      "train_steps_per_second": 0.609
    }
  ],
  "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": 1.1838393679872e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}