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.
☆343Updated 9 months ago
Alternatives and similar repositories for memory
Users that are interested in memory are comparing it to the libraries listed below
Sorting:
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆245Updated 7 months ago
- Tina: Tiny Reasoning Models via LoRA☆281Updated last month
- Normalized Transformer (nGPT)☆188Updated 9 months ago
- PyTorch building blocks for the OLMo ecosystem☆286Updated last week
- Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024☆334Updated 4 months ago
- Decentralized RL Training at Scale☆592Updated this week
- 🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.☆459Updated 3 weeks ago
- A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).☆284Updated last week
- [ICML 2024] CLLMs: Consistency Large Language Models☆403Updated 10 months ago
- Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling☆443Updated 3 months ago
- ☆196Updated 9 months ago
- A project to improve skills of large language models☆553Updated last week
- Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.☆200Updated last year
- Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch☆179Updated 2 months ago
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆172Updated 8 months ago
- PyTorch implementation of Infini-Transformer from "Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention…☆291Updated last year
- Code for NeurIPS'24 paper 'Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization'☆229Updated last month
- An extension of the nanoGPT repository for training small MOE models.☆185Updated 6 months ago
- PyTorch implementation of models from the Zamba2 series.☆185Updated 7 months ago
- Pretraining and inference code for a large-scale depth-recurrent language model☆826Updated last week
- Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models☆220Updated last week
- LongRoPE is a novel method that can extends the context window of pre-trained LLMs to an impressive 2048k tokens.☆248Updated last year
- ☆217Updated 7 months ago
- Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"☆328Updated 9 months ago
- Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache☆123Updated last month
- Exploring Applications of GRPO☆247Updated 3 weeks ago
- Code for the paper: "Learning to Reason without External Rewards"☆353Updated 2 months ago
- A simplified implementation for experimenting with RLVR on GSM8K, This repository provides a starting point for exploring reasoning.☆122Updated 7 months ago
- Block Transformer: Global-to-Local Language Modeling for Fast Inference (NeurIPS 2024)☆161Updated 5 months ago
- [NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models☆229Updated 4 months ago