R136a1
/

MXLewd-L2-20B-exl2

Text Generation

Not-For-All-Audiences

nsfw

Inference Endpoints

Model card Files Files and versions Community

R136a1 commited on Oct 16, 2023

Commit

f162ad7

•

1 Parent(s): fd51c67

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -15,8 +15,8 @@ language:
 First attempt to quantize a 20B model so it can run on 16GB VRAM with the highest quality possible.
 Quantized at 3.23bpw with hb 6
 Perplexity:
-# Base = 6.4744
-# 3.23 h6 = 6.5369
 Dataset = [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)
 ## Prompt Format

 First attempt to quantize a 20B model so it can run on 16GB VRAM with the highest quality possible.
 Quantized at 3.23bpw with hb 6
 Perplexity:
+Base = 6.4744
+3.23 h6 = 6.5369
 Dataset = [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)
 ## Prompt Format