Model Overview

Blackroot/Nous-Hermes-Llama2-13b-Storywriter is a 13 billion parameter language model created by merging two distinct models: NousResearch's Nous-Hermes-Llama2-13b and Blackroot's Llama-2-13B-Storywriter-LORA. This combination aims to leverage the general capabilities of Nous-Hermes with the specialized creative writing focus of the Storywriter LORA.

Key Characteristics

• Architecture: Based on the Llama 2 family, specifically a 13 billion parameter variant.
• Merged Origin: Combines a general-purpose instruction-tuned model (Nous-Hermes) with a LoRA fine-tuned for storytelling.
• Unaligned Output: The model has not undergone any alignment, censorship, or output manipulation. Users should be aware that it may produce unexpected or unfiltered content.

Prompt Format

This model utilizes the Alpaca instruct format for prompts, inherited from the Nous-Hermes base model. This format supports both simple instructions and instructions with additional context:

• Simple Instruction:

  ### Instruction:
  <prompt>
  
  ### Response:
  <leave a newline blank for model to respond>

• Instruction with Input:

  ### Instruction:
  <prompt>
  
  ### Input:
  <additional context>
  
  ### Response:
  <leave a newline blank for model to respond>

Use Cases

This model is particularly well-suited for:

• Creative Writing: Generating stories, narratives, and other forms of creative text.
• Unfiltered Content Generation: For applications where raw, unaligned text outputs are desired or required.
• Exploratory AI Research: Investigating the capabilities of merged models and unaligned language generation.