AI & LLM

huggingface-tokenizers @zechenzhangAGI

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds....

sentencepiece @zechenzhangAGI

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k...

grpo-rl-training @zechenzhangAGI

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

miles-rl-training @zechenzhangAGI

Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training...

openrlhf-training @zechenzhangAGI

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models...

simpo-training @zechenzhangAGI

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4...

slime-rl-training @zechenzhangAGI

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM...

torchforge-rl-training @zechenzhangAGI

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms....

fine-tuning-with-trl @zechenzhangAGI

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment,...

verl-rl-training @zechenzhangAGI

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when...

axolotl @zechenzhangAGI

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO,...

llama-factory @zechenzhangAGI

Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA,...

peft-fine-tuning @zechenzhangAGI

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models...

unsloth @zechenzhangAGI

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

nnsight-remote-interpretability @zechenzhangAGI

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote...

pyvene-interventions @zechenzhangAGI

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention...

sparse-autoencoder-training @zechenzhangAGI

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network...

transformer-lens-interpretability @zechenzhangAGI

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate...

nemo-curator @zechenzhangAGI

GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16×...

ray-data @zechenzhangAGI

Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images....

constitutional-ai @zechenzhangAGI

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with...

llamaguard @zechenzhangAGI

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual...

nemo-guardrails @zechenzhangAGI

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation,...

huggingface-accelerate @zechenzhangAGI

Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for...