Update README.md
Browse files
README.md
CHANGED
|
@@ -4,18 +4,54 @@ datasets:
|
|
| 4 |
- bigcode/the-stack-dedup
|
| 5 |
tags:
|
| 6 |
- code
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
---
|
| 8 |
|
| 9 |
|
| 10 |
-
# replit-code-v1-3b
|
| 11 |
|
| 12 |
-
`replit-code-v1-3b` is a 2.7B Causal Language Model focused on Code Completion
|
| 13 |
|
| 14 |
-
The training mixture includes 20 different languages
|
| 15 |
<br/>
|
| 16 |
`Markdown`, `Java`, `JavaScript`, `Python`, `TypeScript`, `PHP`, `SQL`, `JSX`, `reStructuredText`, `Rust`, `C`, `CSS`, `Go`, `C++`, `HTML`, `Vue`, `Ruby`, `Jupyter Notebook`, `R`, `Shell`
|
| 17 |
|
| 18 |
-
In total, the training dataset contains 175B tokens, which were repeated over 3 epochs -- in total, `replit-code-v1-3b` has been trained on 525B tokens (~195 tokens per parameter).
|
| 19 |
|
| 20 |
|
| 21 |
## How to use the model
|
|
|
|
| 4 |
- bigcode/the-stack-dedup
|
| 5 |
tags:
|
| 6 |
- code
|
| 7 |
+
language:
|
| 8 |
+
- code
|
| 9 |
+
programming_language:
|
| 10 |
+
- Markdown
|
| 11 |
+
- Java
|
| 12 |
+
- JavaScript
|
| 13 |
+
- Python
|
| 14 |
+
- TypeScript
|
| 15 |
+
- PHP
|
| 16 |
+
- SQL
|
| 17 |
+
- JSX
|
| 18 |
+
- reStructuredText
|
| 19 |
+
- Rust
|
| 20 |
+
- C
|
| 21 |
+
- CSS
|
| 22 |
+
- Go
|
| 23 |
+
- C++
|
| 24 |
+
- HTML
|
| 25 |
+
- Vue
|
| 26 |
+
- Ruby
|
| 27 |
+
- Jupyter Notebook
|
| 28 |
+
- R
|
| 29 |
+
- Shell
|
| 30 |
+
model-index:
|
| 31 |
+
- name: replit-code-v1-3b
|
| 32 |
+
results:
|
| 33 |
+
- task:
|
| 34 |
+
type: text-generation
|
| 35 |
+
dataset:
|
| 36 |
+
type: openai_humaneval
|
| 37 |
+
name: HumanEval (Python)
|
| 38 |
+
metrics:
|
| 39 |
+
- name: pass@1
|
| 40 |
+
type: pass@1
|
| 41 |
+
value: 0.219
|
| 42 |
+
verified: false
|
| 43 |
---
|
| 44 |
|
| 45 |
|
| 46 |
+
# replit-code-v1-3b [Test it ]
|
| 47 |
|
| 48 |
+
`replit-code-v1-3b` is a 2.7B Causal Language Model focused on **Code Completion**. The model has been trained on a subset of the Stack Dedup v1.2 dataset.
|
| 49 |
|
| 50 |
+
The training mixture includes **20 different languages**, listed here in descending order of number of tokens:
|
| 51 |
<br/>
|
| 52 |
`Markdown`, `Java`, `JavaScript`, `Python`, `TypeScript`, `PHP`, `SQL`, `JSX`, `reStructuredText`, `Rust`, `C`, `CSS`, `Go`, `C++`, `HTML`, `Vue`, `Ruby`, `Jupyter Notebook`, `R`, `Shell`
|
| 53 |
|
| 54 |
+
In total, the training dataset contains 175B tokens, which were repeated over 3 epochs -- in total, `replit-code-v1-3b` has been trained on **525B** tokens (~195 tokens per parameter).
|
| 55 |
|
| 56 |
|
| 57 |
## How to use the model
|