Hugging Face's logo

fblgit
/
cybertron-v4-qw7B-UNAMGS

Text Generation
Transformers
Safetensors
English
qwen2
Generated from Trainer
conversational
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
1
cybertron-v4-qw7B-UNAMGS / README.md
fblgit's picture
fblgit
Update README.md
b58fe7d verified
preview code
|
raw
history blame contribute delete
9.06 kB
metadata
language:
  - en
license: other
library_name: transformers
tags:
  - generated_from_trainer
base_model:
  - Qwen/Qwen2.5-7B-Instruct
datasets:
  - Magpie-Align/Magpie-Qwen2.5-Pro-1M-v0.1
license_name: qwen
license_link: https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE
model-index:
  - name: cybertron-v4-qw7B-UNAMGS
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 60.84
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 37.71
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 29.91
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 10.85
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 12.69
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 38.89
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=fblgit/cybertron-v4-qw7B-UNAMGS
          name: Open LLM Leaderboard

cybertron-v4-qw7B-UNAMGS

UNA IS BACK Cybertron v4 UNA-MGS, Based on the amazing Qwen2.5 7B

SCORING #1 7-8B LLM WITH NO CONTAMINATION 21.11.2024 with avg. 31.82

cybertron-v4-MGS

This special edition went thru UNA at MLP layers just like miniclaus-1.5B

Here we use our novel approach called MGS. Its up to you to figure out what it means. On top of that we used UNA: Uniform Neural Alignment

Cybertron V4 went thru SFT with MGS & UNA over Magpie-Align/Magpie-Qwen2.5-Pro-1M-v0.1 dataset.

Contamination Benchmark

https://gair-nlp.github.io/benbench/

  • MATH:
5gram-Qwen2.5-7B-Instruct-orgn-MATH-test.jsonl: 37.52666666666667
5gram-Qwen2.5-7B-Instruct-orgn-MATH-train.jsonl: 46.36666666666667

vs

5gram-UNA-cybertron-v4-qw7B-MGS-orgn-MATH-test.jsonl: 37.42666666666667
5gram-UNA-cybertron-v4-qw7B-MGS-orgn-MATH-train.jsonl: 46.053333333333335

vs

5gram-Homer-v0.5-orgn-MATH-test.jsonl: 38.77333333333333
5gram-Homer-v0.5-orgn-MATH-train.jsonl: 47.16666666666667

Quantz

Available at bartowski repo

https://huggingface.co/bartowski/cybertron-v4-qw7B-UNAMGS-GGUF

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 31.82
IFEval (0-Shot) 60.84
BBH (3-Shot) 37.71
MATH Lvl 5 (4-Shot) 29.91
GPQA (0-shot) 10.85
MuSR (0-shot) 12.69
MMLU-PRO (5-shot) 38.89

MGS & UNA & Details

  • MGS, 1+1 = 2 and not 3
  • UNA, 1+1 = 2 obviously

We also followed https://arxiv.org/pdf/2410.21228 insights.

Training procedure

1 Epoch as usual.

Built with Axolotl

datasets:
  - path: Magpie-Align/Magpie-Qwen2.5-Pro-1M-v0.1
    split: train
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    roles:
      user: ["human", "user"]
      assistant: ["gpt", "assistant", "ai"]
      system: ["system"]

Training hyperparameters

The following hyperparameters were used during training:

  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 64
  • total_eval_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss
0.7824 0.0003 1 0.5555
0.5489 0.0503 144 0.4848
0.5348 0.1006 288 0.4732
0.5256 0.1509 432 0.4670
0.5172 0.2012 576 0.4621
0.4882 0.2515 720 0.4578
0.4848 0.3018 864 0.4550
0.4678 0.3520 1008 0.4522
0.4686 0.4023 1152 0.4502
0.4775 0.4526 1296 0.4474
0.4464 0.5029 1440 0.4454
0.4772 0.5532 1584 0.4438
0.4546 0.6035 1728 0.4425
0.4661 0.6538 1872 0.4411
0.4569 0.7041 2016 0.4399
0.4529 0.7544 2160 0.4390
0.4409 0.8047 2304 0.4380
0.4405 0.8550 2448 0.4370
0.4642 0.9053 2592 0.4363
0.4566 0.9556 2736 0.4359

Framework versions

  • PEFT 0.13.2
  • Transformers 4.45.2 (UNA & MGS patch)
  • Pytorch 2.3.0+cu121
  • Datasets 3.0.1
  • Tokenizers 0.20.1

Citations 

@misc{thebeagle-v2,
  title={TheBeagle v2: MGS}, 
  author={Xavier Murias},
  year={2024},
  publisher = {HuggingFace},
  journal = {HuggingFace repository},
  howpublished = {\url{https://huggingface.co/fblgit/TheBeagle-v2beta-32B-MGS}},
}

@misc{Magpie,
    title={Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing}, 
    author={Zhangchen Xu and Fengqing Jiang and Luyao Niu and Yuntian Deng and Radha Poovendran and Yejin Choi and Bill Yuchen Lin},
    year={2024},
    eprint={2406.08464},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

@misc{qwen2.5,
    title = {Qwen2.5: A Party of Foundation Models},
    url = {https://qwenlm.github.io/blog/qwen2.5/},
    author = {Qwen Team},
    month = {September},
    year = {2024}
}

@article{qwen2,
      title={Qwen2 Technical Report}, 
      author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
      journal={arXiv preprint arXiv:2407.10671},
      year={2024}
}

@article{xu2024benchmarking,
      title={Benchmarking Benchmark Leakage in Large Language Models},
      author={Xu, Ruijie and Wang, Zengzhi and Fan, Run-Ze and Liu, Pengfei},
      year={2024},
      journal={arXiv preprint arXiv:2404.18824},
      url={https://arxiv.org/abs/2404.18824}
}