Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,6 @@ license: mit
|
|
| 3 |
datasets:
|
| 4 |
- HuggingFaceCode/stack-v3-train
|
| 5 |
---
|
| 6 |
-
My new
|
| 7 |
|
| 8 |
This model was trained on ~5.5B tokens of stack-v3-train, and only understands python. Attempted FIM but didn't work because I'm a dummy. Bananamind Base Bench 1.1 code_completion Elo is 1418.6, which is a meaningful jump over 42M-CodeCompletion.
|
|
|
|
| 3 |
datasets:
|
| 4 |
- HuggingFaceCode/stack-v3-train
|
| 5 |
---
|
| 6 |
+
My new 128M code completion model, named Vega 1.5 both because it has a new tokenizer... and I don't like how "VegaLM1" looks anyway.
|
| 7 |
|
| 8 |
This model was trained on ~5.5B tokens of stack-v3-train, and only understands python. Attempted FIM but didn't work because I'm a dummy. Bananamind Base Bench 1.1 code_completion Elo is 1418.6, which is a meaningful jump over 42M-CodeCompletion.
|