Hugging Face's logo

hfl
/
chinese-alpaca-2-7b-gguf

GGUF
Chinese
English
Model card Files Community
1
hfl-rc commited on
Commit
2d5ea85
•
1 Parent(s): e02a1e4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -18,9 +18,12 @@ Metric: PPL, lower is better
18
  |-----|------|------|
19
  | Q2_K | 10.3705 +/- 0.14109 | 11.7442 +/- 0.16034 |
20
  | Q3_K | 8.8478 +/- 0.12085 | 8.7104 +/- 0.11925 |
 
21
  | Q4_K | 8.3294 +/- 0.11396 | 8.3034 +/- 0.11391 |
 
22
  | Q5_K | 8.2361 +/- 0.11298 | 8.2136 +/- 0.11281 |
23
  | Q6_K | 8.1956 +/- 0.11259 | 8.1852 +/- 0.11246 |
 
24
 
25
  *The model with `-im` suffix is generated with important matrix, which has generally better performance (not always though).*
26
 
 
18
  |-----|------|------|
19
  | Q2_K | 10.3705 +/- 0.14109 | 11.7442 +/- 0.16034 |
20
  | Q3_K | 8.8478 +/- 0.12085 | 8.7104 +/- 0.11925 |
21
+ | Q4_0 | 8.6418 +/- 0.11876 | - |
22
  | Q4_K | 8.3294 +/- 0.11396 | 8.3034 +/- 0.11391 |
23
+ | Q5_0 | 8.3320 +/- 0.11411 | - |
24
  | Q5_K | 8.2361 +/- 0.11298 | 8.2136 +/- 0.11281 |
25
  | Q6_K | 8.1956 +/- 0.11259 | 8.1852 +/- 0.11246 |
26
+ | Q8_0 | 8.1784 +/- 0.11232 | - |
27
 
28
  *The model with `-im` suffix is generated with important matrix, which has generally better performance (not always though).*
29