Ksgk-fy/M7Percival_010.19-0.06-0.41-0.8-0.94-0.69-7B

🧩 Configuration

slices:
  - sources:
      - model: liminerity/M7-7b
        layer_range: [0, 32]
      - model: AurelPx/Percival_01-7b-slerp
        layer_range: [0, 32]
merge_method: slerp
base_model: liminerity/M7-7b
parameters:
  t:
    - filter: self_attn
      value: [0.19008531186131294, 0.06343621644332342, 0.4083317855158569, 0.7960607977104741, 0.9409166775477182]
    - filter: mlp
      value: [0.8099146881386871, 0.9365637835566766, 0.2039392022895259, 0.2039392022895259, 0.05908332245228176]
    - value: 0.6944218577360742
dtype: bfloat16
random_seed: 0
    ```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Ksgk-fy/M7Percival_010.19-0.06-0.41-0.8-0.94-0.69-7B"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)
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"])