Hugging Face's logo

v000000
/
L3.1-Niitorm-8B-LATCOSx2-Version-GGUFs-IMATRIX

GGUF
llama
Merge
llama-cpp
Inference Endpoints
conversational
Model card Files Community
L3.1-Niitorm-8B-LATCOSx2-Version-GGUFs-IMATRIX
File size: 832 Bytes 
62b350b
 
 
 
 
 
 
 
5c66ff8
 
35080b6
 
6b0287e
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
 
---
base_model: v000000/L3.1-Niitorm-8B-LATCOSx2
tags:
- llama
- merge
- llama-cpp
---

# Llama-3.1-Niitorm-8B-LATCOSx2 

![d48ca23f-9063-4a66-a6b8-0abcbfe26dc5.jpg](https://cdn-uploads.huggingface.co/production/uploads/64f74b6e6389380c77562762/96he_28zzbVZBoh29QlIm.jpeg)

This is a test, <b>"v000000/L3.1-Niitorm-8B-t0.0001"</b> but merged one extra time with <b>"akjindal53244/Llama-3.1-Storm-8B"</b>. Using a new merging algorithm I wrote "LATCOS", which is non linear interpolation and cosine vector similarity between tensors in both magnitude and direction.
This attempts to find the smoothest possible interpolation and make them work more seamlessly together by taking into account the vector direction where both models agree. Seems a lot smarter, but also more compliant which could be a negative since it's less dynamic.