Downloads · 30 days
4
7% of all-time downloads
JayhC/C-Based-2x7B-8bpw-h8-exl2
C-Based-2x7B-8bpw-h8-exl2 is a text generation model from JayhC. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as cc-by-nc-4.0.
<br/<br/ 8bpw/h8 exl2 quantization of Undi95/C-Based-2x7B using default exllamav2 calibration dataset.
Downloads · 30 days
4
7% of all-time downloads
All-time downloads
56
Public
Repo size
13 GB
Likes
1
Public
Click a slice to open those files.
.safetensors13 GB · 100%
From the Hugging Face model README
<br/><br/> 8bpw/h8 exl2 quantization of Undi95/C-Based-2x7B using default exllamav2 calibration dataset.
ORIGINAL CARD:
<!-- description start -->This repo contains fp16 files of C-Based-2x7B.
Created by taking the best benchmark scoring model with the smallest size on HF leaderboard as of today (29/03/2024) which is zhengr/MixTAO-7Bx2-MoE-v8.1 and merging to it some MoE done by myself using the human feedback data from Chaiverse specifically to have high level of RP, intelligence and usage.
Since this is a frankenmoe, I really don't know what the result will be leaderboard side, but what interest me is the human interaction realism anyway (for RP/ERP).
<!-- description end --> <!-- prompt-template start -->Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{system prompt}
### Input:
{prompt}
### Response:
{output}
If you want to support me, you can here.