model card with HF metadata
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# mathcode-llm
|
| 2 |
|
| 3 |
A decoder-only transformer language model **built from scratch** -- no
|
|
@@ -57,5 +72,5 @@ python train.py --config configs/mathcode.yaml \
|
|
| 57 |
|
| 58 |
## Status
|
| 59 |
|
| 60 |
-
Training checkpoint: **step 100 /
|
| 61 |
(13,107,200 tokens seen, val loss inf)
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
pipeline_tag: text-generation
|
| 4 |
+
tags:
|
| 5 |
+
- llm
|
| 6 |
+
- from-scratch
|
| 7 |
+
- mathematics
|
| 8 |
+
- code
|
| 9 |
+
- nature-of-code
|
| 10 |
+
license: mit
|
| 11 |
+
language:
|
| 12 |
+
- en
|
| 13 |
+
- code
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
# mathcode-llm
|
| 17 |
|
| 18 |
A decoder-only transformer language model **built from scratch** -- no
|
|
|
|
| 72 |
|
| 73 |
## Status
|
| 74 |
|
| 75 |
+
Training checkpoint: **step 100 / 4577**
|
| 76 |
(13,107,200 tokens seen, val loss inf)
|