Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
|
| 2 |
This quantized model:
|
| 3 |
```
|
|
@@ -19,4 +24,9 @@ vllm (pretrained=nvidia/Minitron-8B-Base), gen_kwargs: (None), limit: None, num_
|
|
| 19 |
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
| 20 |
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.5080|± |0.0138|
|
| 21 |
| | |strict-match | 5|exact_match|↑ |0.5064|± |0.0138|
|
| 22 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- fp8
|
| 4 |
+
- vllm
|
| 5 |
+
---
|
| 6 |
|
| 7 |
This quantized model:
|
| 8 |
```
|
|
|
|
| 24 |
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
| 25 |
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.5080|± |0.0138|
|
| 26 |
| | |strict-match | 5|exact_match|↑ |0.5064|± |0.0138|
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
The [original paper](https://arxiv.org/pdf/2407.14679) evals:
|
| 31 |
+
|
| 32 |
+
|