Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
# CodeModernBERT-Crow-v1-Pre
|
2 |
|
3 |
## Model Description
|
@@ -48,7 +69,7 @@ A custom **BPE tokenizer** was trained for code and docstrings.
|
|
48 |
* **Number of layers**: 12
|
49 |
* **Attention heads**: 12
|
50 |
* **Intermediate size**: 3072
|
51 |
-
* **Max sequence length
|
52 |
* **RoPE positional encoding**: supported
|
53 |
|
54 |
---
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- bigcode/starcoderdata
|
5 |
+
- bigcode/starcoder2data-extras
|
6 |
+
language:
|
7 |
+
- en
|
8 |
+
tags:
|
9 |
+
- code
|
10 |
+
- python
|
11 |
+
- java
|
12 |
+
- javascript
|
13 |
+
- typescript
|
14 |
+
- go
|
15 |
+
- rust
|
16 |
+
- php
|
17 |
+
- ruby
|
18 |
+
- cpp
|
19 |
+
- c
|
20 |
+
- sql
|
21 |
+
---
|
22 |
# CodeModernBERT-Crow-v1-Pre
|
23 |
|
24 |
## Model Description
|
|
|
69 |
* **Number of layers**: 12
|
70 |
* **Attention heads**: 12
|
71 |
* **Intermediate size**: 3072
|
72 |
+
* **Max sequence length: 8192** (during training, inputs were limited to 1024 tokens)
|
73 |
* **RoPE positional encoding**: supported
|
74 |
|
75 |
---
|