SicariusSicariiStuff/Assistant_Pepe_70B
Assistant_Pepe_70B by SicariusSicariiStuff is a 70 billion parameter large language model based on the Llama 3.1 architecture, featuring a 32768-token context length. This model is uniquely optimized for lateral thinking, creative writing, and engaging, unfiltered banter, demonstrating an uncommon ability to solve trick questions not present in its training data. It is primarily intended for general tasks and "shitposting," offering a highly uncensored and distinct conversational style.
Loading preview...
Assistant_Pepe_70B: The Lateral Thinking LLM
Assistant_Pepe_70B, developed by SicariusSicariiStuff, is a 70 billion parameter model built on the Llama 3.1 architecture with a 32K context length. This model stands out for its unique "lateral thinking" capabilities, demonstrating an ability to solve complex trick questions that even frontier models struggle with, without having seen the answers in its training data. It is designed to maximize helpfulness and "shitposting" while reducing positivity, resulting in a highly engaging and unfiltered conversational experience.
Key Capabilities
- Exceptional Lateral Thinking: Uniquely solves trick questions, indicating emergent reasoning beyond typical LLM patterns.
- High Performance: Ranked #1 globally for 70B models on the UGI benchmark, outperforming Llama-3.3-70B-Instruct and Llama-3.1-70B-Instruct, and nearly matching Mistral-Large-Instruct-2411 in raw intelligence.
- Uncensored & Engaging: Features almost no censorship, excels at bantering, roasting, and provides a "quirky and unique" line of thinking.
- Creative Writing & Code Generation: Demonstrates excellent and extremely creative writing abilities, and can ship code.
- Long Context Coherency: Offers superb coherency at 32K tokens and very good coherency even at 64K, thanks to its Llama 3.1 70B base.
- No System Prompt Required: Simplifies usage by not needing a system prompt.
Good for
- Creative Content Generation: Ideal for generating highly creative and unconventional text.
- Roleplay & Interactive Storytelling: Its unique personality and lateral thinking make it suitable for dynamic and unpredictable roleplay scenarios.
- Unfiltered & Humorous Interactions: Perfect for applications requiring a model that can engage in witty banter, "shitposting," and even good-natured roasting.
- Exploring Emergent AI Capabilities: Researchers interested in emergent properties and lateral reasoning in LLMs will find this model particularly intriguing.
- General Assistant Tasks: Capable of handling a wide range of general tasks with its distinct approach.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.