facebookresearch / memoryLinks
Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense feed-forward layers, providing dedicated capacity to store and retrieve information cheaply.
☆337Updated 6 months ago
Alternatives and similar repositories for memory
Users that are interested in memory are comparing it to the libraries listed below
Sorting:
- prime-rl is a codebase for decentralized async RL training at scale☆341Updated this week
- 🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.☆379Updated last week
- ☆178Updated 6 months ago
- [ICML 2024] CLLMs: Consistency Large Language Models☆394Updated 7 months ago
- PyTorch building blocks for the OLMo ecosystem☆234Updated this week
- LongRoPE is a novel method that can extends the context window of pre-trained LLMs to an impressive 2048k tokens.☆231Updated 10 months ago
- Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache☆108Updated last month
- Scalable toolkit for efficient model reinforcement☆438Updated this week
- Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024☆311Updated last month
- Tina: Tiny Reasoning Models via LoRA☆258Updated 3 weeks ago
- ☆193Updated 4 months ago
- Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch☆174Updated 5 months ago
- Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling☆395Updated last month
- A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).☆255Updated this week
- Block Transformer: Global-to-Local Language Modeling for Fast Inference (NeurIPS 2024)☆158Updated 2 months ago
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆239Updated 4 months ago
- An extension of the nanoGPT repository for training small MOE models.☆148Updated 3 months ago
- Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"☆315Updated 7 months ago
- SkyRL-v0: Train Real-World Long-Horizon Agents via Reinforcement Learning☆422Updated this week
- Efficient LLM Inference over Long Sequences☆377Updated 2 weeks ago
- Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs☆176Updated last week
- Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.☆198Updated 11 months ago
- Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models☆169Updated last month
- Build your own visual reasoning model☆381Updated this week
- [NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models☆221Updated last month
- Normalized Transformer (nGPT)☆183Updated 7 months ago
- Code for the paper: "Learning to Reason without External Rewards"☆295Updated this week
- A project to improve skills of large language models☆423Updated this week
- EvaByte: Efficient Byte-level Language Models at Scale☆101Updated 2 months ago
- PyTorch implementation of models from the Zamba2 series.☆182Updated 4 months ago