Tiger Gemma 9B v2 Overview

Tiger Gemma 9B v2 is a 9 billion parameter language model developed by TheDrummer, built upon the Gemma architecture. This iteration focuses on providing a "decensored" experience, aiming to minimize content refusals often encountered in base models while preserving the inherent qualities of Gemma. The model maintains a substantial context length of 16384 tokens, making it suitable for processing longer inputs and generating coherent, extended outputs.

Key Characteristics

• Decensored Output: Engineered to reduce content refusals, offering more permissive generation compared to its base model.
• Gemma Architecture: Retains the foundational characteristics and performance profile of the original Gemma 9B model.
• Context Length: Supports a 16384 token context window, beneficial for tasks requiring extensive contextual understanding.

Intended Use Cases

• Applications where less restrictive content generation is desired.
• Scenarios requiring a Gemma-based model with reduced refusal rates.
• Tasks benefiting from a 9 billion parameter model with a large context window.