|
--- |
|
license: apache-2.0 |
|
tags: |
|
- moe |
|
- mixtral |
|
- merge |
|
--- |
|
|
|
# MetaModel_moe |
|
|
|
This model is a Mixure of Experts (MoE) made with [mergekit](https://github.com/cg123/mergekit) (mixtral branch). It uses the following base models: |
|
* [gagan3012/MetaModel](https://huggingface.co/gagan3012/MetaModel) |
|
* [jeonsworld/CarbonVillain-en-10.7B-v2](https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v2) |
|
* [jeonsworld/CarbonVillain-en-10.7B-v4](https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v4) |
|
* [TomGrc/FusionNet_linear](https://huggingface.co/TomGrc/FusionNet_linear) |
|
|
|
## 🧩 Configuration |
|
|
|
```yaml |
|
base_model: gagan3012/MetaModel |
|
gate_mode: hidden |
|
dtype: bfloat16 |
|
experts: |
|
- source_model: gagan3012/MetaModel |
|
- source_model: jeonsworld/CarbonVillain-en-10.7B-v2 |
|
- source_model: jeonsworld/CarbonVillain-en-10.7B-v4 |
|
- source_model: TomGrc/FusionNet_linear |
|
``` |
|
|
|
## 💻 Usage |
|
|
|
```python |
|
!pip install -qU transformers bitsandbytes accelerate |
|
|
|
from transformers import AutoTokenizer |
|
import transformers |
|
import torch |
|
|
|
model = "gagan3012/MetaModel_moe" |
|
|
|
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"]) |
|
``` |
|
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) |
|
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_gagan3012__MetaModel_moe) |
|
|
|
| Metric | Value | |
|
|-----------------------|---------------------------| |
|
| Avg. | 74.42 | |
|
| ARC (25-shot) | 71.25 | |
|
| HellaSwag (10-shot) | 88.4 | |
|
| MMLU (5-shot) | 66.26 | |
|
| TruthfulQA (0-shot) | 71.86 | |
|
| Winogrande (5-shot) | 83.35 | |
|
| GSM8K (5-shot) | 65.43 | |
|
|
