69 lines
1.7 KiB
Markdown
69 lines
1.7 KiB
Markdown
---
|
|
license: apache-2.0
|
|
tags:
|
|
- moe
|
|
- frankenmoe
|
|
- merge
|
|
- mergekit
|
|
- maywell/PiVoT-0.1-Starling-LM-RP
|
|
- WizardLM/WizardMath-7B-V1.1
|
|
base_model:
|
|
- maywell/PiVoT-0.1-Starling-LM-RP
|
|
- WizardLM/WizardMath-7B-V1.1
|
|
---
|
|
|
|
# Rose-2x7B
|
|
|
|
Rose-2x7B is a Mixure of Experts (MoE) made with the following models using [Mergekit](https://github.com/cg123/mergekit):
|
|
* [maywell/PiVoT-0.1-Starling-LM-RP](https://huggingface.co/maywell/PiVoT-0.1-Starling-LM-RP)
|
|
* [WizardLM/WizardMath-7B-V1.1](https://huggingface.co/WizardLM/WizardMath-7B-V1.1)
|
|
|
|
```bash
|
|
mergekit-moe mergekit_moe.yaml merge --copy-tokenizer --device cuda --low-cpu-memory
|
|
```
|
|
|
|
## 🧩 Configuration
|
|
|
|
```yaml
|
|
base_model: uproai/ros-7b-v1
|
|
experts:
|
|
- source_model: maywell/PiVoT-0.1-Starling-LM-RP
|
|
positive_prompts:
|
|
- "storywriting"
|
|
- "write"
|
|
- "scene"
|
|
- "story"
|
|
- "character"
|
|
- source_model: WizardLM/WizardMath-7B-V1.1
|
|
positive_prompts:
|
|
- "reason"
|
|
- "math"
|
|
- "mathematics"
|
|
- "solve"
|
|
- "count"
|
|
tokenizer_source: union
|
|
```
|
|
|
|
## 💻 Usage
|
|
|
|
```python
|
|
!pip install -qU transformers bitsandbytes accelerate
|
|
|
|
from transformers import AutoTokenizer
|
|
import transformers
|
|
import torch
|
|
|
|
model = "uproai/Rose-2x7B"
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained(model)
|
|
pipeline = transformers.pipeline(
|
|
"text-generation",
|
|
model=model,
|
|
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
|
|
)
|
|
|
|
messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
|
|
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
|
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
|
print(outputs[0]["generated_text"])
|
|
``` |