Update README.md
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ Small multimodal-ready transformer checkpoint trained on the project's education
|
|
| 12 |
|
| 13 |
- Parameters: approximately 229M
|
| 14 |
- Text decoder: 12 layers, 768 hidden size, 12 attention heads
|
| 15 |
-
- Vocabulary: 50,257
|
| 16 |
- Training samples: 1,000
|
| 17 |
- Checkpoint: global step 350
|
| 18 |
|
|
|
|
| 12 |
|
| 13 |
- Parameters: approximately 229M
|
| 14 |
- Text decoder: 12 layers, 768 hidden size, 12 attention heads
|
| 15 |
+
- Vocabulary: 50,257 BPE tokens
|
| 16 |
- Training samples: 1,000
|
| 17 |
- Checkpoint: global step 350
|
| 18 |
|