---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- FelixChao/WestSeverus-7B-DPO-v2
- CultriX/Wernicke-7B-v8
- vanillaOVO/supermario_v2
base_model:
- FelixChao/WestSeverus-7B-DPO-v2
- CultriX/Wernicke-7B-v8
- vanillaOVO/supermario_v2
model-index:
- name: Wernicke-7B-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: 71.84
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-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: 88.63
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-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: 65.22
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-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: 73.91
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-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: 84.61
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-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: 67.63
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CultriX/Wernicke-7B-dpo
      name: Open LLM Leaderboard
---

# Edit:
* DPO-finetune using truthy-dpo dataset and CultriX/Wernicke-7B-v9. 

# Wernicke-7B-v9

Wernicke-7B-v9 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [FelixChao/WestSeverus-7B-DPO-v2](https://huggingface.co/FelixChao/WestSeverus-7B-DPO-v2)
* [CultriX/Wernicke-7B-v8](https://huggingface.co/CultriX/Wernicke-7B-v8)
* [vanillaOVO/supermario_v2](https://huggingface.co/vanillaOVO/supermario_v2)

## 🧩 Configuration

```yaml
models:
  - model: FelixChao/WestSeverus-7B-DPO-v2
    parameters:
      density: 0.50
      weight: 0.35
  - model: CultriX/Wernicke-7B-v8
    parameters:
      density: 0.50
      weight: 0.35
  - model: vanillaOVO/supermario_v2
    parameters:
      density: 0.50
      weight: 0.30
merge_method: dare_ties
base_model: FelixChao/WestSeverus-7B-DPO-v2
parameters:
  int8_mask: true
dtype: float16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "CultriX/Wernicke-7B-v9"
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](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_CultriX__Wernicke-7B-dpo)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |75.31|
|AI2 Reasoning Challenge (25-Shot)|71.84|
|HellaSwag (10-Shot)              |88.63|
|MMLU (5-Shot)                    |65.22|
|TruthfulQA (0-shot)              |73.91|
|Winogrande (5-shot)              |84.61|
|GSM8k (5-shot)                   |67.63|