FunAGI
/

LLaDA-8B-Base-gptqmodel-4bit

Text Generation

4-bit precision

Model card Files Files and versions Community

chentianqi commited on Mar 15

Commit

54d1f7a

·

verified ·

1 Parent(s): 2f7d2d1

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -43,8 +43,8 @@ This model has been 4-bit quantized Llada-8B-Base model with [GPTQModel](https:/
 | Dataset         | GPTQ-4bit | FP16 |
 |----------------|-------------|------|
-| mmlu          | TODO        | 65.9(5)   |
-| cmmlu         | TODO       | 69.9(5)    |
 | arc_challenge | 45.48       |  47.9(0)    |
 ## Example:

 | Dataset         | GPTQ-4bit | FP16 |
 |----------------|-------------|------|
+| mmlu          | 65.20       |    65.9 |
+| cmmlu         | 69.23     | 69.9  |
 | arc_challenge | 45.48       |  47.9(0)    |
 ## Example: