Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
|
| 3 |
+
## Llama-Engineer-Evol-7B
|
| 4 |
+
|
| 5 |
+
This is a version of Meta's chat instruction-tuned Llama 2 further fine-tuned on over 80,000 coding samples.
|
| 6 |
+
|
| 7 |
+
The dataset is a combination of [Evol-Instruct-Code-80k-v1](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) from [nikrosh](https://huggingface.co/nickrosh), a replication of the Evol-Instruct-Code as described in the WizardCoder Paper, and [Teknium](https://huggingface.co/teknium)'s [GPTeacher](https://github.com/teknium1/GPTeacher/blob/main/Codegen/codegen-instruct.json). Special thanks to these folks for putting these datasets together.
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
### Prompt Format
|
| 11 |
+
The reccomended model usage is a variant of the standard Llama 2 format:
|
| 12 |
+
```
|
| 13 |
+
[INST] <<SYS>>
|
| 14 |
+
You are a helpful programming assistant. Always answer as helpfully as possible. Be direct in your response and get to the answer right away. Responses should be short.
|
| 15 |
+
<</SYS>>
|
| 16 |
+
{your prompt}[/INST]
|
| 17 |
+
```
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
### Next Steps
|
| 21 |
+
- Prune the dataset and possibly fine-tune for longer.
|
| 22 |
+
- Run benchmarks.
|
| 23 |
+
- Provide GGML and GPTQ.
|