R136a1 commited on
Commit
7c64888
1 Parent(s): 0d33874

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -12,12 +12,14 @@ language:
12
 
13
  ## Model details
14
 
15
- Quantized at 3.18bpw with hb 6. Can run full 4K context on 16GB VRAM
16
 
17
  Perplexity:
18
 
19
  Base = 6.5820
20
 
 
 
21
  3.18 h6 = 6.6928
22
 
23
  Dataset = [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)
 
12
 
13
  ## Model details
14
 
15
+ Quantized at 3.18bpw with hb 6. Can run full 4K context on 16GB VRAM. 8.13bpw also available.
16
 
17
  Perplexity:
18
 
19
  Base = 6.5820
20
 
21
+ 8.13 = 6.5535
22
+
23
  3.18 h6 = 6.6928
24
 
25
  Dataset = [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)