hfl
/

llama-3-chinese-8b-gguf

Model card Files Files and versions

hfl-rc commited on Apr 29

Commit

00cabd5

•

1 Parent(s): 2110c05

Update README.md

Files changed (1) hide show

README.md +13 -11

README.md CHANGED Viewed

@@ -21,17 +21,19 @@ Further details (performance, usage, etc.) should refer to GitHub project page:
 Metric: PPL, **lower is better**
-| Quant |     Size |  PPL |
-| :---: | -------: | ---: |
-| Q2_K  |  2.96 GB | 17.7212 +/- 0.59814    |
-| Q3_K  |  3.74 GB | 8.6303 +/- 0.28481     |
-| Q4_0  |  4.34 GB | 8.2513 +/- 0.27102     |
-| Q4_K  |  4.58 GB | 7.8897 +/- 0.25830     |
-| Q5_0  |  5.21 GB | 7.7975 +/- 0.25639     |
-| Q5_K  |  5.34 GB | 7.7062 +/- 0.25218     |
-| Q6_K  |  6.14 GB | 7.6600 +/- 0.25043     |
-| Q8_0  |  7.95 GB | 7.6512 +/- 0.25064     |
-|  F16  | 14.97 GB | 7.6389 +/- 0.25001     |
 ## Others

 Metric: PPL, **lower is better**
+The model name with `-im` suffix is generated with important matrix, which has generally better performance.
+| Quant |     Size |                 PPL |             PPL (`-im`) |
+| :---: | -------: | ------------------: | ----------------------: |
+| Q2_K  |  2.96 GB | 17.7212 +/- 0.59814 | **14.9583 +/- 0.50455** |
+| Q3_K  |  3.74 GB |  8.6303 +/- 0.28481 |  **8.4423 +/- 0.28087** |
+| Q4_0  |  4.34 GB |  8.2513 +/- 0.27102 |  **7.9077 +/- 0.25525** |
+| Q4_K  |  4.58 GB |  7.8897 +/- 0.25830 |  **7.8279 +/- 0.25542** |
+| Q5_0  |  5.21 GB |  7.7975 +/- 0.25639 |  **7.7724 +/- 0.25625** |
+| Q5_K  |  5.34 GB |  7.7062 +/- 0.25218 |  **7.6902 +/- 0.25170** |
+| Q6_K  |  6.14 GB |  7.6600 +/- 0.25043 |  **7.6412 +/- 0.24949** |
+| Q8_0  |  7.95 GB |  7.6512 +/- 0.25064 |      7.6512 +/- 0.25064 |
+|  F16  | 14.97 GB |  7.6389 +/- 0.25001 |                     N/A |
 ## Others