File size: 2,119 Bytes
242184b 2b7961b 242184b 2b7961b 242184b 2b7961b 242184b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
---
license: apache-2.0
tags:
- moe
- merge
- mergekit
- NotAiLOL/Med-Yi-1.5-9B
- NotAiLOL/Yi-1.5-dolphin-9B
- NotAiLOL/Yi-1.5-9B-coder
- NotAiLOL/Math-Yi-1.5-9B
base_model:
- NotAiLOL/Med-Yi-1.5-9B
- NotAiLOL/Yi-1.5-dolphin-9B
- NotAiLOL/Yi-1.5-9B-coder
- NotAiLOL/Math-Yi-1.5-9B
---
# Eclipse-Yi-1.5-4x9b
Eclipse-Yi-1.5-4x9b is a Mixture of Experts (MoE) made with the following models:
* [NotAiLOL/Med-Yi-1.5-9B](https://huggingface.co/NotAiLOL/Med-Yi-1.5-9B)
* [NotAiLOL/Yi-1.5-dolphin-9B](https://huggingface.co/NotAiLOL/Yi-1.5-dolphin-9B)
* [NotAiLOL/Yi-1.5-9B-coder](https://huggingface.co/NotAiLOL/Yi-1.5-9B-coder)
* [NotAiLOL/Math-Yi-1.5-9B](https://huggingface.co/NotAiLOL/Math-Yi-1.5-9B)
## 🧩 Configuration
```yaml
base_model: 01-ai/Yi-1.5-9B
gate_mode: hidden
dtype: bfloat16
experts:
- source_model: NotAiLOL/Med-Yi-1.5-9B
positive_prompts:
- "explain"
- "write"
- source_model: NotAiLOL/Yi-1.5-dolphin-9B
positive_prompts:
- "chat"
- "assistant"
- "tell me"
- "explain"
- "I want"
- source_model: NotAiLOL/Yi-1.5-9B-coder
positive_prompts:
- "chat"
- "assistant"
- "tell me"
- "explain"
- "I want"
- source_model: NotAiLOL/Math-Yi-1.5-9B
positive_prompts:
- "reason"
- "math"
- "mathematics"
- "solve"
- "count"
```
## 💻 Usage
```python
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "NotAiLOL/Eclipse-Yi-1.5-4x9b"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.bfloat16, "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"])
``` |