hfl
/

hfl-rc commited on
Commit
026be4b
1 Parent(s): afea6a3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -0
README.md CHANGED
@@ -9,6 +9,24 @@ language:
9
 
10
  This repository contains the GGUF-v3 models (llama.cpp compatible) for **Chinese-Alpaca-2-1.3B**.
11
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  For Hugging Face version, please see: https://huggingface.co/hfl/chinese-alpaca-2-1.3b
13
 
14
  Please refer to [https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/) for more details.
 
9
 
10
  This repository contains the GGUF-v3 models (llama.cpp compatible) for **Chinese-Alpaca-2-1.3B**.
11
 
12
+
13
+ ## Performance
14
+
15
+ Metric: PPL, lower is better
16
+
17
+ | Quant | original | imatrix (`-im`) |
18
+ |-----|------|------|
19
+ | Q2_K | 19.9339 +/- 0.29752 | 18.8935 +/- 0.28558 |
20
+ | Q3_K | 17.2487 +/- 0.27668 | 17.2950 +/- 0.27994 |
21
+ | Q4_K | 16.4583 +/- 0.26453 | 16.2688 +/- 0.26216 |
22
+ | Q5_K | 15.7547 +/- 0.25207 | 16.0190 +/- 0.25782 |
23
+ | Q6_K | 15.8166 +/- 0.25359 | 15.7357 +/- 0.25210 |
24
+
25
+ *The model with `-im` suffix is generated with important matrix, which has generally better performance (not always though).*
26
+
27
+
28
+ ## Others
29
+
30
  For Hugging Face version, please see: https://huggingface.co/hfl/chinese-alpaca-2-1.3b
31
 
32
  Please refer to [https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/) for more details.