Sumandora / remove-refusals-with-transformers
Implements harmful/harmless refusal removal using pure HF Transformers
☆19Updated 5 months ago
Related projects ⓘ
Alternatives and complementary repositories for remove-refusals-with-transformers
- An unsupervised model merging algorithm for Transformers-based language models.☆99Updated 6 months ago
- Model REVOLVER, a human in the loop model mixing system.☆33Updated last year
- Low-Rank adapter extraction for fine-tuned transformers model☆162Updated 6 months ago
- ☆27Updated last year
- 5X faster 60% less memory QLoRA finetuning☆21Updated 5 months ago
- Easily view and modify JSON datasets for large language models☆62Updated last month
- entropix style sampling + GUI☆25Updated 2 weeks ago
- Train Llama Loras Easily☆29Updated last year
- After my server ui improvements were successfully merged, consider this repo a playground for experimenting, tinkering and hacking around…☆56Updated 2 months ago
- All the world is a play, we are but actors in it.☆47Updated 4 months ago
- An easy-to-understand framework for LLM samplers that rewind and revise generated tokens☆105Updated 2 weeks ago
- RAG implementation for Ooba characters. dynamically spins up new qdrant vector DB and manages retrieval and commits for conversations ba…☆44Updated last year
- Experimental sampler to make LLMs more creative☆30Updated last year
- Dynamic parameter modulation for oobabooga's text-generation-webui that adjusts generation parameters to better mirror user affect.☆34Updated last year
- Accepts a Hugging Face model URL, automatically downloads and quantizes it using Bits and Bytes.☆38Updated 8 months ago
- Science-driven chatbot development☆55Updated 6 months ago
- idea: https://github.com/nyxkrage/ebook-groupchat/☆81Updated 2 months ago
- ☆64Updated 5 months ago
- A KoboldAI-like memory extension for oobabooga's text-generation-webui☆107Updated 2 weeks ago
- Interact with a AI Game-engine that keep building its rules and world as you play, adapted to your gameplay.☆43Updated 4 months ago
- ☆104Updated 7 months ago
- BUD-E (Buddy) is an open-source voice assistant framework that facilitates seamless interaction with AI models and APIs, enabling the cre…☆28Updated 3 months ago
- ☆72Updated last year
- Cohere Toolkit is a collection of prebuilt components enabling users to quickly build and deploy RAG applications.☆25Updated last week
- A pipeline parallel training script for LLMs.☆83Updated this week
- Automated prompting and scoring framework to evaluate LLMs using updated human knowledge prompts☆111Updated last year
- Steer LLM outputs towards a certain topic/subject and enhance response capabilities using activation engineering by adding steering vecto…☆42Updated 7 months ago
- Traing PRO extension for oobabooga WebUI - recent dev version☆46Updated 2 months ago
- An Extension for oobabooga/text-generation-webui☆36Updated last year