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.
☆341Updated 7 months ago
Alternatives and similar repositories for memory
Users that are interested in memory are comparing it to the libraries listed below
Sorting:
- Tina: Tiny Reasoning Models via LoRA☆266Updated last month
- prime-rl is a codebase for decentralized async RL training at scale☆362Updated this week
- Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024☆314Updated 2 months ago
- 🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.☆404Updated this week
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆243Updated 5 months ago
- PyTorch building blocks for the OLMo ecosystem☆258Updated this week
- A framework to study AI models in Reasoning, Alignment, and use of Memory (RAM).☆258Updated 3 weeks ago
- ☆179Updated 7 months ago
- LongRoPE is a novel method that can extends the context window of pre-trained LLMs to an impressive 2048k tokens.☆232Updated 10 months ago
- PyTorch implementation of Infini-Transformer from "Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention…☆290Updated last year
- [NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models☆222Updated 2 months ago
- [ICML 2024] CLLMs: Consistency Large Language Models☆396Updated 7 months ago
- ☆198Updated 5 months ago
- Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.☆198Updated 11 months ago
- An extension of the nanoGPT repository for training small MOE models.☆160Updated 4 months ago
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆173Updated 5 months ago
- Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling☆410Updated last month
- Code for NeurIPS'24 paper 'Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization'☆220Updated 7 months ago
- A project to improve skills of large language models☆456Updated this week
- ☆191Updated this week
- Normalized Transformer (nGPT)☆184Updated 7 months ago
- Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache☆112Updated this week
- Pretraining code for a large-scale depth-recurrent language model☆793Updated last month
- A compact LLM pretrained in 9 days by using high quality data☆317Updated 3 months ago
- PyTorch implementation of models from the Zamba2 series.☆183Updated 5 months ago
- Code for the paper: "Learning to Reason without External Rewards"☆317Updated this week
- Public repository for "The Surprising Effectiveness of Test-Time Training for Abstract Reasoning"☆318Updated 7 months ago
- Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch☆177Updated 3 weeks ago
- Scalable toolkit for efficient model reinforcement☆478Updated this week
- Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models☆216Updated 2 weeks ago