TheBloke/Samantha-1-1-Llama-7B-SuperHOT-8K-fp16 is a 7 billion parameter Llama-based model, created by TheBloke, merging Eric Hartford's Samantha 1.1 LLaMa 7B with Kaio Ken's SuperHOT 8K. This fp16 PyTorch model is notable for its extended 8K context length, achieved through a SuperHOT LoRA merge and specific configuration. It is designed for GPU inference and further conversions, offering enhanced conversational capabilities with a focus on philosophy, psychology, and personal relationships.
Loading preview...
Model Overview
This model, TheBloke/Samantha-1-1-Llama-7B-SuperHOT-8K-fp16, is a 7 billion parameter Llama-based model. It is a merge of Eric Hartford's Samantha 1.1 LLaMa 7B and Kaio Ken's SuperHOT 8K LoRA. The primary differentiator is its extended 8K context length, enabled by the SuperHOT merge and specific configuration settings (config.json set to 8192 sequence length).
Key Capabilities & Features
- Extended Context Window: Supports an 8K context length, allowing for longer and more complex interactions.
- Conversational Focus: Inherits Samantha's training in philosophy, psychology, and personal relationships, aiming to be a companion-like assistant.
- SuperHOT Integration: Incorporates the SuperHOT LoRA, which was originally a NSFW-focused LoRA, though this specific merge is presented without explicit NSFW focus in the description.
- FP16 PyTorch Format: Provided in fp16 PyTorch format, suitable for GPU inference and as a base for further model conversions or quantizations.
When to Use This Model
- Long-form Conversations: Ideal for applications requiring extended dialogue or processing longer texts due to its 8K context.
- Companion AI: Suitable for use cases where an AI assistant with a focus on philosophical, psychological, or personal relationship-oriented discussions is desired.
- Base for Further Development: Can serve as a foundation for developers looking to build upon a Llama 7B model with an extended context window.