license: cc-by-nc-4.0 tags:

• merge
• mergekit
• lazymergekit
• AIDC-ai-business/Marcoroni-7B-v3
• EmbeddedLLM/Mistral-7B-Merge-14-v0.1 model-index:
• name: Marcoro14-7B-slerp results:
  • task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics:
    • type: acc_norm value: 69.8 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics:
    • type: acc_norm value: 87.13 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics:
    • type: acc value: 65.11 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics:
    • type: mc2 value: 63.54 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics:
    • type: acc value: 81.61 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics:
    • type: acc value: 70.89 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mlabonne/Marcoro14-7B-slerp name: Open LLM Leaderboard

Marcoro14-7B-slerp

This model is a merge of the following models made with mergekit:

• AIDC-ai-business/Marcoroni-7B-v3
• EmbeddedLLM/Mistral-7B-Merge-14-v0.1

🏆 Evaluation

Marcoro14-7B-slerp is the best-performing 7B LLM on the Open LLM Leaderboard (rank 1 below is 9B):

I also evaluated it using Nous' benchmark suite and obtained the following results:

AGIEval

Average: 44.66%

GPT4ALL

Average: 76.24%

TruthfulQA

Average: 64.15%

Bigbench

Average: 45.64%

Average score: 57.67%

🧩 Configuration

slices:
  - sources:
      - model: AIDC-ai-business/Marcoroni-7B-v3
        layer_range: [0, 32]
      - model: EmbeddedLLM/Mistral-7B-Merge-14-v0.1
        layer_range: [0, 32]
merge_method: slerp
base_model: AIDC-ai-business/Marcoroni-7B-v3
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

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Marcoro14-7B-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"])

Output:

A large language model is a type of artificial intelligence (AI) system that has been trained on vast amounts of text data. It's designed to understand and generate human-like language, making predictions on what words or phrases might come next in a sentence or document. These models use complex algorithms and neural network architectures to learn from the data and improve their performance over time. Some well-known large language models include GPT-3 from OpenAI and BERT from Google.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here