Commit
•
0c44282
1
Parent(s):
4cefaa9
Update README.md
Browse files
README.md
CHANGED
|
@@ -5,4 +5,16 @@ language:
|
|
| 5 |
metrics:
|
| 6 |
- perplexity
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
metrics:
|
| 6 |
- perplexity
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
This model was created by [ilnikolaev](https://huggingface.co/ilnikolaev)
|
| 11 |
+
|
| 12 |
+
Trained from scratch using Tensorflow Keras
|
| 13 |
+
(200mb Russian Comments from 2ch)[https://www.kaggle.com/datasets/fizzzgen/65mb-of-dvach-conversations] dataset used
|
| 14 |
+
|
| 15 |
+
*Type: decoder-only
|
| 16 |
+
*Vocabulary size: 32000
|
| 17 |
+
*Hidden size: 768
|
| 18 |
+
*FFN size: 3072
|
| 19 |
+
*Attention heads: 24
|
| 20 |
+
*Decoder layers: 4
|