Hugging Face Overview
L3.3-70B-Magnum-v4-SE: Claude-esque Prose for Creative Applications
This 70 billion parameter model, developed by Doctor-Shotgun, is a "Special Edition" fine-tune of the meta-llama/Llama-3.3-70B-Instruct base model, implemented as an rsLoRA adapter. The primary objective of the Magnum series, including this v4-SE, is to emulate the prose style and quality of the Claude 3 Sonnet/Opus models at a local scale, often exhibiting "Claude-isms" in its output.
Key Capabilities
- Claude-like Prose: Specifically fine-tuned to mimic the sophisticated writing style of Claude 3 Sonnet/Opus models.
- Creative Writing & Roleplay: Optimized for generating engaging narratives, character interactions, and descriptive content.
- Llama 3.3 Base: Benefits from the strong foundation of the Llama 3.3 architecture.
- Extended Context: Supports a context length of 32768 tokens, allowing for more extensive and coherent interactions.
- Flexible Usage: Follows the Llama 3 prompt format and includes recommended sampler settings (temperature = 1.1, min_p = 0.1) for optimal performance.
Good For
- Creative Writers: Ideal for generating story plots, character dialogues, and descriptive passages.
- Roleplay Enthusiasts: Excels in creating immersive and dynamic roleplaying scenarios.
- Developers seeking Claude-style output: Provides a local alternative for generating text with a similar stylistic flair to Claude 3 models.
Limitations
Intended for creative writing and roleplay, this model may exhibit biases similar to contemporary LLM-based roleplay models and the Claude 3 series. Outputs should be considered fictional and not relied upon for factual information or advice.