Daredevil 8B

🚀 Daredevil-8B

Daredevil-8B is a mega-merge model aiming to maximize MMLU performance. As of May 27, 2024, it holds the highest MMLU score among Llama 3 8B models. In practice, a high MMLU score is a significant advantage for Llama 3 models.

🚀 Quick Start

Daredevil-8B is a merge of multiple models using LazyMergekit. You can use it as an enhanced alternative to meta-llama/Meta-Llama-3-8B-Instruct.

✨ Features

• High MMLU Score: On May 27, 2024, it achieved the highest MMLU score among Llama 3 8B models on the Open LLM Leaderboard.
• Versatile Applications: Can be used as an improved version of meta-llama/Meta-Llama-3-8B-Instruct.
• Censored and Uncensored Versions: A censored version is available here, and an uncensored version can be found at mlabonne/Daredevil-8B-abliterated.

📦 Installation

The model can be used directly through the Hugging Face platform. You can also use the following command to install the necessary libraries for local use:

!pip install -qU transformers accelerate

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Daredevil-8B"
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.bfloat16,
    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

Applications

You can use it as an improved version of meta-llama/Meta-Llama-3-8B-Instruct. This is a censored model. For an uncensored version, see mlabonne/Daredevil-8B-abliterated. It has been tested on LM Studio using the "Llama 3" preset.

Quantization

• GGUF: https://huggingface.co/mlabonne/Daredevil-8B-GGUF

Evaluation

Open LLM Leaderboard

Daredevil-8B is the best-performing 8B model on the Open LLM Leaderboard in terms of MMLU score (May 27, 2024).

Nous

Daredevil-8B is the best-performing 8B model on Nous' benchmark suite (evaluation performed using LLM AutoEval, May 27, 2024). See the entire leaderboard here.

Model Family Tree

Configuration

models:
  - model: NousResearch/Meta-Llama-3-8B
    # No parameters necessary for base model
  - model: nbeerbower/llama-3-stella-8B
    parameters:
      density: 0.6
      weight: 0.16
  - model: Hastagaras/llama-3-8b-okay
    parameters:
      density: 0.56
      weight: 0.1
  - model: nbeerbower/llama-3-gutenberg-8B
    parameters:
      density: 0.6
      weight: 0.18
  - model: openchat/openchat-3.6-8b-20240522
    parameters:
      density: 0.56
      weight: 0.12
  - model: Kukedlc/NeuralLLaMa-3-8b-DT-v0.1
    parameters:
      density: 0.58
      weight: 0.18
  - model: cstr/llama3-8b-spaetzle-v20
    parameters:
      density: 0.56
      weight: 0.08
  - model: mlabonne/ChimeraLlama-3-8B-v3
    parameters:
      density: 0.56
      weight: 0.08
  - model: flammenai/Mahou-1.1-llama3-8B
    parameters:
      density: 0.55
      weight: 0.05
  - model: KingNish/KingNish-Llama3-8b
    parameters:
      density: 0.55
      weight: 0.05
merge_method: dare_ties
base_model: NousResearch/Meta-Llama-3-8B
dtype: bfloat16

📄 License

The license for this model is "other".