README.md

---
language:
- en
license: mit
model-index:
- name: MoMo-70B-lora-1.8.6-DPO
  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: 70.14
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      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: 86.03
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      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: 77.4
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      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: 69.0
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      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: 84.37
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      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: 76.8
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-lora-1.8.6-DPO
      name: Open LLM Leaderboard
---
# **Introduction**
MoMo-72B-lora-1.8.6-DPO is trained via Direct Preference Optimization([DPO](https://arxiv.org/abs/2305.18290)) from [MoMo-72B-LoRA-V1.4](https://huggingface.co/moreh/MoMo-72B-LoRA-V1.4) as its base model, with several optimizations in hyperparameters.  
[MoMo-72B-LoRA-V1.4](https://huggingface.co/moreh/MoMo-72B-LoRA-V1.4) is trained via Supervised Fine-Tuning (SFT) using [LoRA](https://arxiv.org/abs/2106.09685), with the QWEN-72B model as its base-model.  
Note that we did not exploit any form of weight merge.  
For leaderboard submission, the trained weight is realigned for compatibility with llama.  
MoMo-72B is trained using **[Moreh](https://moreh.io/)**'s [MoAI platform](https://moreh.io/product), which simplifies the training of large-scale models, and AMD's MI250 GPU.


## Details
### Used Librarys
- torch
- peft
### Used Datasets
- [slimorca](Open-Orca/SlimOrca)
- [truthy](https://huggingface.co/datasets/jondurbin/truthy-dpo-v0.1)
- [orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs)
- No other dataset was used
- No benchmark test set or the training set are used
  - [data contamination check](https://github.com/swj0419/detect-pretrain-code-contamination) result
    
| Model                        | ARC   | MMLU | TruthfulQA | GSM8K |
|------------------------------|-------|-------|-------|-------|
| **V1.8.6(result < 0.1, %)**| TBU |TBU | 0.73 | TBU |
### Used Environments
- AMD MI250 & MoAI platform
- Please visit https://moreh.io/product for more information about MoAI platform
- Or, contact us directly [[email protected]](mailto:[email protected])

## How to use

```python
# pip install transformers==4.35.2
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("moreh/MoMo-72B-lora-1.8.6-DPO")
model = AutoModelForCausalLM.from_pretrained(
    "moreh/MoMo-72B-lora-1.8.6-DPO"
)
```
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_moreh__MoMo-70B-lora-1.8.6-DPO)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |77.29|
|AI2 Reasoning Challenge (25-Shot)|70.14|
|HellaSwag (10-Shot)              |86.03|
|MMLU (5-Shot)                    |77.40|
|TruthfulQA (0-shot)              |69.00|
|Winogrande (5-shot)              |84.37|
|GSM8k (5-shot)                   |76.80|