Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
{}
|
| 3 |
+
---
|
| 4 |
+
# Model Details
|
| 5 |
+
- **Architecture**: Basic/default GPT-2, decoder only
|
| 6 |
+
- **Num params**: ~204M
|
| 7 |
+
- **Num tokens seen**: ~1.3 B
|
| 8 |
+
- **Dataset**: USPTO subset of The Pile interleaved with PubMed Abstracts subset of The Pile.
|
| 9 |
+
- Interleaved with probabilities [0.5, 0.5], respectively (first argument is for USPTO, second is for PubMedAbs)
|