Update README.md
Browse files
README.md
CHANGED
|
@@ -12,6 +12,8 @@ This is the 26 categorical finetune of nomic-bert-2048's encoder.
|
|
| 12 |
|
| 13 |
775,000,000 - 144-256 token samples with 30% mask rate
|
| 14 |
|
|
|
|
|
|
|
| 15 |
Total samples > 1bil
|
| 16 |
|
| 17 |
The model has learned to categorize certain masked patterns with their categories and special tokens.
|
|
|
|
| 12 |
|
| 13 |
775,000,000 - 144-256 token samples with 30% mask rate
|
| 14 |
|
| 15 |
+
453,800,000 - 385-512 token samples with 30% mask rate
|
| 16 |
+
|
| 17 |
Total samples > 1bil
|
| 18 |
|
| 19 |
The model has learned to categorize certain masked patterns with their categories and special tokens.
|