Percival 01 7b Slerp

🚀 Percival_01-7b-slerp

The 2nd best 7B model on the OPENLLM LeaderBoard!

Percival_01-7b-slerp is a merge of the following models using LazyMergekit:

• liminerity/M7-7b
• Gille/StrangeMerges_32-7B-slerp

🚀 Quick Start

Percival_01-7b-slerp is a powerful merged model. You can start using it following the steps below.

✨ Features

• High Ranking: It ranks 2nd among 7B models on the OPENLLM LeaderBoard.
• Model Merging: Merged from liminerity/M7-7b and Gille/StrangeMerges_32-7B-slerp using LazyMergekit.

📦 Installation

You need to install the necessary libraries to use this model. Run the following command:

!pip install -qU transformers accelerate

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer
import transformers
import torch

model = "AurelPx/Percival_01-7b-slerp"
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"])

📚 Documentation

🧩 Configuration

The following is the configuration for merging the models:

slices:
  - sources:
      - model: liminerity/M7-7b
        layer_range: [0, 32]
      - model: Gille/StrangeMerges_32-7B-slerp
        layer_range: [0, 32]
merge_method: slerp
base_model: liminerity/M7-7b
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

📄 License

This project is licensed under the apache-2.0 license.