compressa-ai
/

Llama-3-70B-Instruct-OmniQuant

Text Generation

text-generation-inference

Inference Endpoints

4-bit precision

Model card Files Files and versions Community

Vasily Alexeev commited on Apr 28, 2024

Commit

a630053

·

1 Parent(s): 0c79aa9

fix naming style in tables

Files changed (1) hide show

README.md +8 -8

README.md CHANGED Viewed

@@ -25,18 +25,18 @@ Quantized with [OmniQuant](https://github.com/OpenGVLab/OmniQuant).
 ### PPL (↓)
-|           | wiki |
-| --------- | ---- |
-| FP        | 5,33 |
-| Quantized | 5,90 |
 ### Accuracy on English Benchmarks, % (↑)
-|           | piqa | arc_easy | arc_challenge | boolq | hellaswag | winogrande |
-| --------- | ---- | -------- | ------------- | ----- | --------- | ---------- |
-| FP        | 81,5 | 86,2     | 61,9          | 87,4  | 63,7      | 75,8       |
-| Quantized | 80,7 | 85,8     | 61,4          | 87,0  | 62,7      | 73,0       |
 ### Summary

 ### PPL (↓)
+|               | wiki |
+| ------------- | ---- |
+| FP            | 5,33 |
+| **Quantized** | 5,90 |
 ### Accuracy on English Benchmarks, % (↑)
+|               | piqa | arc_easy | arc_challenge | boolq | hellaswag | winogrande |
+| ------------- | ---- | -------- | ------------- | ----- | --------- | ---------- |
+| FP            | 81,5 | 86,2     | 61,9          | 87,4  | 63,7      | 75,8       |
+| **Quantized** | 80,7 | 85,8     | 61,4          | 87,0  | 62,7      | 73,0       |
 ### Summary