dmf-archive / PILFLinks
PILF: A IPWT-inspired bionic continual learning experiment focus on mitigate catastrophic forgetting with Surprise-gated Mixture of Experts (MoE) model.
☆36Updated 2 months ago
Alternatives and similar repositories for PILF
Users that are interested in PILF are comparing it to the libraries listed below
Sorting:
- A playground to make it easy to try crazy things☆33Updated 3 months ago
- Pivotal Token Search☆125Updated 2 months ago
- Live-bending a foundation model’s output at neural network level.☆264Updated 5 months ago
- ☆196Updated 4 months ago
- Hierarchical topic segmentation of meeting transcripts using embeddings and divisive clustering.☆53Updated last year
- Testing platform for covert data exfiltration techniques where sensitive documents are embedded into vector representations and tunneled …☆69Updated last month
- Heirarchical Navigable Small Worlds☆101Updated last month
- Dead Simple LLM Abliteration☆231Updated 7 months ago
- Pytorch script hot swap: Change code without unloading your LLM from VRAM☆125Updated 5 months ago
- A star for organising blocks and playing with transformers.☆23Updated last year
- A library for incremental loading of large PyTorch checkpoints☆56Updated 2 years ago
- ☆230Updated 6 months ago
- 33B Chinese LLM, DPO QLORA, 100K context, AirLLM 70B inference with single 4GB GPU☆13Updated last year
- Your AI research assistant☆79Updated 5 months ago
- Implement recursion using English as the programming language and an LLM as the runtime.☆238Updated 2 years ago
- Visual inference exploration & experimentation playground☆95Updated 9 months ago
- Applying the ideas of Deepseek R1 to computer use☆216Updated 7 months ago
- Editor with LLM generation tree exploration☆77Updated 7 months ago
- ☆67Updated 3 months ago
- Neurox control helm chart details☆30Updated 4 months ago
- Build data processing and data analysis pipelines that leverage the power of LLMs 🧠☆220Updated last week
- Parallel thinking for LLMs. Confidence‑gated, strategy‑driven, offline‑friendly☆250Updated last week
- A tiny autograd engine with a Jax-like API☆74Updated 2 months ago
- a curated list of data for reasoning ai☆137Updated last year
- Mistral7B playing DOOM☆137Updated last year
- SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?☆107Updated this week
- Benchmark that evaluates LLMs using 759 NYT Connections puzzles extended with extra trick words☆146Updated last week
- Build Secure and Compliant AI agents and MCP Servers. YC W23☆151Updated 3 months ago
- Craziness.☆29Updated 7 months ago
- Advanced Python Function Debugging with MCP Integration.☆57Updated 3 months ago