README.md

metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - mlabonne/NeuralBeagle14-7B
  - FelixChao/WestSeverus-7B-DPO-v2
  - CultriX/MergeTrix-7B-v2
base_model:
  - mlabonne/NeuralBeagle14-7B
  - FelixChao/WestSeverus-7B-DPO-v2
  - CultriX/MergeTrix-7B-v2
model-index:
  - name: OmniTrixAI
    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: 72.95
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          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: 88.52
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          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: 65.12
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          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: 70.12
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          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: 83.58
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          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: 70.51
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/OmniTrixAI
          name: Open LLM Leaderboard

EDIT:

Always check my space for the latest benchmark results for my models!

• https://huggingface.co/spaces/CultriX/Yet_Another_LLM_Leaderboard

OmniTrixAI

OmniTrixAI is a merge of the following models using LazyMergekit:

• mlabonne/NeuralBeagle14-7B
• FelixChao/WestSeverus-7B-DPO-v2
• CultriX/MergeTrix-7B-v2

🧩 Configuration

models:
  - model: senseable/WestLake-7B-v2
    # No parameters necessary for base model
  - model: mlabonne/NeuralBeagle14-7B
    parameters:
      density: 0.65
      weight: 0.40
  - model: FelixChao/WestSeverus-7B-DPO-v2
    parameters:
      density: 0.45
      weight: 0.26
  - model: CultriX/MergeTrix-7B-v2
    parameters:
      density: 0.55
      weight: 0.34
merge_method: dare_ties
base_model: senseable/WestLake-7B-v2
parameters:
  int8_mask: true
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "CultriX/OmniTrixAI"
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.	75.13
AI2 Reasoning Challenge (25-Shot)	72.95
HellaSwag (10-Shot)	88.52
MMLU (5-Shot)	65.12
TruthfulQA (0-shot)	70.12
Winogrande (5-shot)	83.58
GSM8k (5-shot)	70.51