Commit
•
b4f0fad
1
Parent(s):
41e8485
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
license_name: tongyi-qianwen
|
4 |
+
license_link: https://huggingface.co/Qwen/Qwen2-7B/blob/main/LICENSE
|
5 |
+
language:
|
6 |
+
- en
|
7 |
+
pipeline_tag: text-generation
|
8 |
+
tags:
|
9 |
+
- chat
|
10 |
+
- qwen
|
11 |
+
- qwen2
|
12 |
+
- finetune
|
13 |
+
- chatml
|
14 |
+
- OpenHermes-2.5
|
15 |
+
- HelpSteer2
|
16 |
+
- Orca
|
17 |
+
- SlimOrca
|
18 |
+
library_name: transformers
|
19 |
+
inference: false
|
20 |
+
model_creator: MaziyarPanahi
|
21 |
+
quantized_by: MaziyarPanahi
|
22 |
+
base_model: Qwen/Qwen2-7B
|
23 |
+
model_name: Qwen2-7B-Instruct-v0.8
|
24 |
+
datasets:
|
25 |
+
- nvidia/HelpSteer2
|
26 |
+
- teknium/OpenHermes-2.5
|
27 |
+
- microsoft/orca-math-word-problems-200k
|
28 |
+
- Open-Orca/SlimOrca
|
29 |
+
---
|
30 |
+
|
31 |
+
<img src="./qwen2-fine-tunes-maziyar-panahi.webp" alt="Qwen2 fine-tune" width="500" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
32 |
+
|
33 |
+
# MaziyarPanahi/Qwen2-7B-Instruct-v0.8
|
34 |
+
|
35 |
+
This is a fine-tuned version of the `Qwen/Qwen2-7B` model. It aims to improve the base model across all benchmarks.
|
36 |
+
|
37 |
+
# ⚡ Quantized GGUF
|
38 |
+
|
39 |
+
All GGUF models are available here: [MaziyarPanahi/Qwen2-7B-Instruct-v0.8](https://huggingface.co/MaziyarPanahi/Qwen2-7B-Instruct-v0.8)
|
40 |
+
|
41 |
+
# 🏆 [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
42 |
+
|
43 |
+
coming soon!
|
44 |
+
|
45 |
+
|
46 |
+
# Prompt Template
|
47 |
+
|
48 |
+
This model uses `ChatML` prompt template:
|
49 |
+
|
50 |
+
```
|
51 |
+
<|im_start|>system
|
52 |
+
{System}
|
53 |
+
<|im_end|>
|
54 |
+
<|im_start|>user
|
55 |
+
{User}
|
56 |
+
<|im_end|>
|
57 |
+
<|im_start|>assistant
|
58 |
+
{Assistant}
|
59 |
+
````
|
60 |
+
|
61 |
+
# How to use
|
62 |
+
|
63 |
+
|
64 |
+
```python
|
65 |
+
|
66 |
+
# Use a pipeline as a high-level helper
|
67 |
+
|
68 |
+
from transformers import pipeline
|
69 |
+
|
70 |
+
messages = [
|
71 |
+
{"role": "user", "content": "Who are you?"},
|
72 |
+
]
|
73 |
+
pipe = pipeline("text-generation", model="MaziyarPanahi/Qwen2-7B-Instruct-v0.8")
|
74 |
+
pipe(messages)
|
75 |
+
|
76 |
+
|
77 |
+
# Load model directly
|
78 |
+
|
79 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
80 |
+
|
81 |
+
tokenizer = AutoTokenizer.from_pretrained("MaziyarPanahi/Qwen2-7B-Instruct-v0.8")
|
82 |
+
model = AutoModelForCausalLM.from_pretrained("MaziyarPanahi/Qwen2-7B-Instruct-v0.8")
|
83 |
+
```
|