Naphula

AI & ML interests

Building new model tools, merges (7B-70B), ablations, finetunes, and datasets.

Recent Activity

published a model about 9 hours ago

Naphula-Archives/C79-24B-v1-Q4_0-GGUF

new activity about 10 hours ago

24B-Suite/Mergedonia-Suite-24B-v1-GGUF:expect

updated a model 1 day ago

Naphula-Archives/ms-2509-ultra-uncensored-heretic-v1-textonly

View all activity

Organizations

upvoted 5 papers 1 day ago

No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces

Paper • 2502.04959 • Published Feb 7, 2025 • 12

Accurate and Efficient Low-Rank Model Merging in Core Space

Paper • 2509.17786 • Published Sep 22, 2025 • 3

upvoted a collection 1 day ago

Favorite Merge Methods

Collection

12 items • Updated 1 day ago • 1

upvoted 3 papers 2 days ago

Reconstructions of Einstein-Aether Gravity from Barrow Agegraphic and New Barrow Agegraphic Dark Energy models: Examinations and Observational Limits

Paper • 2410.19897 • Published Jun 11, 2025 • 1

Inflation in light of ACT/SPT: a new perspective from Weyl gravity

Paper • 2512.10862 • Published Jan 8 • 1

CABS: Conflict-Aware and Balanced Sparsification for Enhancing Model Merging

Paper • 2503.01874 • Published Feb 26, 2025 • 1

upvoted a collection 3 days ago

Riot Gremlins 👹🙀

Collection

7B Models For merging. for RP on my TINY rig at Q6. Without a bloody POD. Merge ideas/sketches. Results will go in 'Babsies Models.' • 47 items • Updated 11 days ago • 2

upvoted a paper 3 days ago

Self-Distillation Enables Continual Learning

Paper • 2601.19897 • Published Jan 27 • 29

upvoted a paper 5 days ago

Functionality-Oriented LLM Merging on the Fisher--Rao Manifold

Paper • 2603.04972 • Published Mar 5 • 3

upvoted a paper 10 days ago

Magistral

Paper • 2506.10910 • Published Jun 12, 2025 • 68

upvoted 4 papers 11 days ago

FuseChat: Knowledge Fusion of Chat Models

Paper • 2408.07990 • Published Aug 15, 2024 • 15

On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

Paper • 2508.05629 • Published Aug 7, 2025 • 191

QLoRA: Efficient Finetuning of Quantized LLMs

Paper • 2305.14314 • Published May 23, 2023 • 61

Resolving Interference When Merging Models

Paper • 2306.01708 • Published Jun 2, 2023 • 18

upvoted an article 16 days ago

Article

Merge Large Language Models with mergekit

Jan 9, 2024

•

154

upvoted an article 18 days ago

Article

Welcome Gemma 4: Frontier multimodal intelligence on device

19 days ago

•

867

upvoted a paper 19 days ago

Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch

Paper • 2311.03099 • Published Nov 6, 2023 • 32