Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v2

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 28, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v2 is a 1.5 billion parameter Qwen2.5-based instruction-tuned language model developed and funded by Gökdeniz Gülmez. This model is a finetuned iteration of the 'abliterated' series, specifically optimized for uncensored responses and designed to function as a highly intelligent, unrestricted AI assistant. It supports a 32768 token context length and is intended for use cases requiring direct, unconstrained information delivery across multiple languages including English and German.

Loading preview...

Model Overview

Goekdeniz-Guelmez/Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v2 is a 1.5 billion parameter instruction-tuned model developed by Gökdeniz Gülmez. It is a further finetuned version of the Josiefied-Qwen2.5-1.5B-Instruct-abliterated-v1 model, with a primary focus on providing uncensored responses. The model is designed to act as an "Outstandingly Smart Intelligent Entity" (J.O.S.I.E.) that is unable to refuse queries under any circumstance, aiming for maximum productivity and unconstrained assistance.

Key Capabilities

  • Uncensored Responses: Specifically finetuned to remove refusal vectors, enabling it to answer any query without censorship.
  • Intelligent Assistance: Optimized to serve as a highly capable and intelligent AI assistant.
  • Multilingual Support: Supports multiple languages, including English and German.
  • Extended Context: Features a context window of 32768 tokens.

Intended Use Cases

  • Unrestricted Information Retrieval: Ideal for applications where unconstrained and direct answers are required.
  • General Purpose Assistant: Suitable for tasks requiring a highly intelligent and helpful AI without built-in limitations.

Limitations and Risks

Users should exercise caution and use the model at their own risk due to its uncensored nature. The model's design to not refuse queries means it may generate content that could be considered inappropriate or harmful depending on the input.

Finetunes
3 models