Update README.md
Browse files
README.md
CHANGED
|
@@ -4,7 +4,9 @@ license: apache-2.0
|
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
datasets:
|
| 7 |
-
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
## Using MDLM
|
|
@@ -18,6 +20,7 @@ model_name = 'kuleshov-group/mdlm-no_flashattn-fp32-owt'
|
|
| 18 |
model = AutoModelForMaskedLM.from_pretrained(model_name)
|
| 19 |
```
|
| 20 |
|
|
|
|
| 21 |
|
| 22 |
## Model Details
|
| 23 |
The model, which has a context length of `1024` and is similar in size to GPT2-medium with approximately `130 million` non-embedding parameters,
|
|
|
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
datasets:
|
| 7 |
+
- Skylion007/openwebtext
|
| 8 |
+
metrics:
|
| 9 |
+
- perplexity
|
| 10 |
---
|
| 11 |
|
| 12 |
## Using MDLM
|
|
|
|
| 20 |
model = AutoModelForMaskedLM.from_pretrained(model_name)
|
| 21 |
```
|
| 22 |
|
| 23 |
+
For more details, please see our github repository: [MDLM](https://github.com/kuleshov-group/mdlm)
|
| 24 |
|
| 25 |
## Model Details
|
| 26 |
The model, which has a context length of `1024` and is similar in size to GPT2-medium with approximately `130 million` non-embedding parameters,
|