RichardErkhov/Kukedlc_-_Fasciculus-Arcuatus-7B-slerp-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Fasciculus-Arcuatus-7B-slerp - bnb 8bits

• Model creator: https://huggingface.co/Kukedlc/
• Original model: https://huggingface.co/Kukedlc/Fasciculus-Arcuatus-7B-slerp/

Original model description:

license: apache-2.0 tags: - merge
- mergekit - lazymergekit - macadeliccc/MonarchLake-7B - Kukedlc/NeoCortex-7B-slerp base_model: - macadeliccc/MonarchLake-7B - Kukedlc/NeoCortex-7B-slerp model-index: - name: Fasciculus-Arcuatus-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: 73.55 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-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: 88.95 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-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: 64.65 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-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: 72.53 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-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: 85.71 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-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: 71.04 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Kukedlc/Fasciculus-Arcuatus-7B-slerp name: Open LLM Leaderboard

Fasciculus-Arcuatus-7B-slerp

Fasciculus-Arcuatus-7B-slerp is a merge of the following models using LazyMergekit:

• macadeliccc/MonarchLake-7B
• Kukedlc/NeoCortex-7B-slerp

🧩 Configuration

slices:
  - sources:
      - model: macadeliccc/MonarchLake-7B
        layer_range: [0, 32]
      - model: Kukedlc/NeoCortex-7B-slerp
        layer_range: [0, 32]
merge_method: slerp
base_model: Kukedlc/NeoCortex-7B-slerp
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 = "Kukedlc/Fasciculus-Arcuatus-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"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	76.07
AI2 Reasoning Challenge (25-Shot)	73.55
HellaSwag (10-Shot)	88.95
MMLU (5-Shot)	64.65
TruthfulQA (0-shot)	72.53
Winogrande (5-shot)	85.71
GSM8k (5-shot)	71.04

Model Family