NeuralFusion-7b-Dare-Ties/README.md

license, tags, base_model, model-index

license	tags	base_model	model-index
apache-2.0	merge mergekit lazymergekit Kukedlc/NeuralMaxime-7B-slerp Kukedlc/Fasciculus-Arcuatus-7B-slerp Kukedlc/NeoCortex-7B-slerp	Kukedlc/NeuralMaxime-7B-slerp Kukedlc/Fasciculus-Arcuatus-7B-slerp Kukedlc/NeoCortex-7B-slerp	name results NeuralFusion-7b-Dare-Ties task dataset metrics source type name text-generation Text Generation name type config split args AI2 Reasoning Challenge (25-Shot) ai2_arc ARC-Challenge test num_few_shot 25 type value name acc_norm 73.21 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type split args HellaSwag (10-Shot) hellaswag validation num_few_shot 10 type value name acc_norm 88.96 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args MMLU (5-Shot) cais/mmlu all test num_few_shot 5 type value name acc 64.77 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args TruthfulQA (0-shot) truthful_qa multiple_choice validation num_few_shot 0 type value mc2 73.32 url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args Winogrande (5-shot) winogrande winogrande_xl validation num_few_shot 5 type value name acc 85.56 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args GSM8k (5-shot) gsm8k main test num_few_shot 5 type value name acc 69.83 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/NeuralFusion-7b-Dare-Ties Open LLM Leaderboard

NeuralFusion-7b-Dare-Ties

NeuralFusion-7b-Dare-Ties is a merge of the following models using LazyMergekit:

• Kukedlc/NeuralMaxime-7B-slerp
• Kukedlc/Fasciculus-Arcuatus-7B-slerp
• Kukedlc/NeoCortex-7B-slerp

🧩 Configuration

models:
  - model: mlabonne/Monarch-7B
    # no parameters necessary for base model
  - model: Kukedlc/NeuralMaxime-7B-slerp
    parameters:
      density: 0.65
      weight: 0.36
  - model: Kukedlc/Fasciculus-Arcuatus-7B-slerp
    parameters:
      density: 0.6
      weight: 0.34
  - model: Kukedlc/NeoCortex-7B-slerp
    parameters:
      density: 0.6
      weight: 0.3
merge_method: dare_ties
base_model: mlabonne/Monarch-7B
parameters:
  int8_mask: true
dtype: bfloat16
random_seed: 0

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Kukedlc/NeuralFusion-7b-Dare-Ties"
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"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	75.94
AI2 Reasoning Challenge (25-Shot)	73.21
HellaSwag (10-Shot)	88.96
MMLU (5-Shot)	64.77
TruthfulQA (0-shot)	73.32
Winogrande (5-shot)	85.56
GSM8k (5-shot)	69.83