Kiv
Follow
Stars
A Native-PyTorch Library for LLM Fine-tuning
Blazingly 🔥 fast 🚀 memory vulnerabilities, written in 100% safe Rust. 🦀
NVIDIA Linux open GPU with P2P support
lightweight, standalone C++ inference engine for Google's Gemma models.
Finetune Llama 3, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
tiktoken is a fast BPE tokeniser for use with OpenAI's models.
Extremely simple implementation of path patching (aka causal scrubbing) in PyTorch.
Stanford NLP Python Library for Understanding and Improving PyTorch Models via Interventions
Calculate token/s & GPU memory requirement for any LLM. Supports llama.cpp/ggml/bnb/QLoRA quantization
Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
Named tensors with first-class dimensions for PyTorch
Must-read Papers on Textual Adversarial Attack and Defense
🦜🔗 Build context-aware reasoning applications
sanjeevanahilan / nanoChatGPT
Forked from karpathy/nanoGPTA crude RLHF layer on top of nanoGPT with Gumbel-Softmax trick
Hackable and optimized Transformers building blocks, supporting a composable construction.
A library for mechanistic interpretability of GPT-style language models
The CUDA version of the RWKV language model ( https://github.com/BlinkDL/RWKV-LM )
RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference,…
Implementation of Invariant Point Attention, used for coordinate refinement in the structure module of Alphafold2, as a standalone Pytorch module
Duelyst is a digital collectible card game and turn-based strategy hybrid, developed by Counterplay Games.
A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
Cinder is Meta's internal performance-oriented production version of CPython.