# CASE-Lab-UMD/LLM-Drop

<https://repo.masters-masters.com/en/repos/CASE-Lab-UMD/LLM-Drop>

[Home](https://repo.masters-masters.com/en) › [CASE-Lab-UMD](https://repo.masters-masters.com/en/owners/CASE-Lab-UMD) › [LLM-Drop](https://repo.masters-masters.com/en/repos/CASE-Lab-UMD/LLM-Drop)

CASE-Lab-UMD/LLM-Drop has 193 stars on GitHub. Its RepoMasters card: Prototype tier. The official implementation of the paper "Uncovering the Redundancy in…

| Field | Value |
| --- | --- |
| Owner | [CASE-Lab-UMD](https://repo.masters-masters.com/en/owners/CASE-Lab-UMD) |
| Stars | 193 |
| Forks | 23 |
| Language | [Python](https://repo.masters-masters.com/en/languages/python) |
| License | Apache-2.0 |
| Created | 2024 |
| Topics | attention-mechanism, deep-learning, large-language-models, machine-learning, model-compression, natural-language-processing |
| Tier | [Prototype (100 to 249 ★)](https://repo.masters-masters.com/en/ranks/rush) |
| View on GitHub | [CASE-Lab-UMD/LLM-Drop](https://github.com/CASE-Lab-UMD/LLM-Drop) |

## Repository

The official implementation of the paper "Uncovering the Redundancy in Transformers via a Unified Study of Layer Dropping (TMLR)".

The tier for repos with 100 to 249 ★. Published odds per card pulled: 70%. Rarity: Common.

## Nearby in Python

- [appletea233/OneSearch-VL · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/appletea233/OneSearch-VL)
- [sepehr071/iran-mcp · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/sepehr071/iran-mcp)
- [JoasASantos/php2shell-exploit · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/JoasASantos/php2shell-exploit)
- [charleso/git-cc · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/charleso/git-cc)
- [Becksteinlab/GromacsWrapper · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/Becksteinlab/GromacsWrapper)
- [coderanger/pychef · ★ 193 · Python · Prototype](https://repo.masters-masters.com/en/repos/coderanger/pychef)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.