Update README.md
Browse files
README.md
CHANGED
@@ -2,16 +2,17 @@
|
|
2 |
license: mit
|
3 |
---
|
4 |
|
5 |
-
|
6 |
|
|
|
7 |
Uses the olmo default architecture.
|
8 |
|
|
|
9 |
- Heads: 8
|
10 |
- Layers: 8
|
11 |
- Dimension model: 512
|
12 |
- Dimension mlp: 4096
|
13 |
|
14 |
-
Training Data
|
15 |
-
|
16 |
Pretraining:
|
17 |
- 5B Tokens C4 (preprocessed, from olmo-data.org)
|
|
|
2 |
license: mit
|
3 |
---
|
4 |
|
5 |
+
# Bingus-v0.1-60M-Base
|
6 |
|
7 |
+
A not-so-state-of-the-art 60M parameter transformer model.
|
8 |
Uses the olmo default architecture.
|
9 |
|
10 |
+
### Specs
|
11 |
- Heads: 8
|
12 |
- Layers: 8
|
13 |
- Dimension model: 512
|
14 |
- Dimension mlp: 4096
|
15 |
|
16 |
+
### Training Data
|
|
|
17 |
Pretraining:
|
18 |
- 5B Tokens C4 (preprocessed, from olmo-data.org)
|