|
--- |
|
license: apache-2.0 |
|
tags: |
|
- moe |
|
- frankenmoe |
|
- merge |
|
- mergekit |
|
- lazymergekit |
|
- beomi/OPEN-SOLAR-KO-10.7B |
|
- hyeogi/SOLAR-10.7B-dpo-v1 |
|
- GAI-LLM/OPEN-SOLAR-KO-10.7B-mixed-v15 |
|
- megastudyedu/M-SOLAR-10.7B-v1.1-beta |
|
base_model: |
|
- beomi/OPEN-SOLAR-KO-10.7B |
|
- hyeogi/SOLAR-10.7B-dpo |
|
- GAI-LLM/OPEN-SOLAR-KO-10.7B-mixed-v15 |
|
- megastudyedu/M-SOLAR-10.7B-v1.1-beta |
|
|
|
--- |
|
|
|
# solar_merge_test_2 |
|
|
|
## ๐งฉ Configuration |
|
|
|
```yaml |
|
base_model: beomi/OPEN-SOLAR-KO-10.7B |
|
dtype: float16 |
|
experts: |
|
- source_model: beomi/OPEN-SOLAR-KO-10.7B |
|
positive_prompts: ["๋น์ ์ ์น์ ํ ๋ณดํธ์ ์ธ ์ด์์คํดํธ์ด๋ค."] |
|
- source_model: hyeogi/SOLAR-10.7B-dpo-v1 |
|
positive_prompts: ["๋น์ ์ ์น์ ํ ์ด์์คํดํธ์ด๋ค."] |
|
- source_model: GAI-LLM/OPEN-SOLAR-KO-10.7B-mixed-v15 |
|
positive_prompts: ["๋น์ ์ ์น์ ํ ์ด์์คํดํธ์ด๋ค."] |
|
- source_model: megastudyedu/M-SOLAR-10.7B-v1.1-beta |
|
positive_prompts: ["๋น์ ์ ์น์ ํ ์ด์์คํดํธ์ด๋ค."] |
|
gate_mode: cheap_embed |
|
tokenizer_source: base |
|
``` |
|
|
|
## ๐ป Usage |
|
|
|
```python |
|
!pip install -qU transformers bitsandbytes accelerate |
|
|
|
from transformers import AutoTokenizer |
|
import transformers |
|
import torch |
|
|
|
model = "jieunhan/solar_merge_test_2" |
|
|
|
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"]) |
|
``` |