---
language:
- ko
- en
license:
- apache-2.0
pipeline_tag: text-generation
model-index:
- name: ko-wand-136M
  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: 21.33
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      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: 25.0
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      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: 23.58
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      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: 50.68
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      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: 49.17
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      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: 0.0
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=instructkr/ko-wand-136M
      name: Open LLM Leaderboard
---
# **ko-wand-136M**

**ko-wand-136M**는 [**insturctkr**](https://instruct.kr)에서 사전학습한 SLM입니다.

# Model Description
[maywell/korean_textbooks](https://huggingface.co/datasets/maywell/korean_textbooks)와 한국어 말뭉치를 이용하여 사전학습 되었습니다.

## Model Info

미스트랄 아키텍쳐를 기반으로 완전히 랜덤 가중치를 시작으로 사전학습 된 모델입니다. Instruction 튜닝되지 않았습니다.

## Training Details
| Batch Size | Token Seen | lr |
|---|---|---|
| 1024 | 2.5B | 2e-3 (cosine)|

## License
apache-2.0
# [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_instructkr__ko-wand-136M)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |28.29|
|AI2 Reasoning Challenge (25-Shot)|21.33|
|HellaSwag (10-Shot)              |25.00|
|MMLU (5-Shot)                    |23.58|
|TruthfulQA (0-shot)              |50.68|
|Winogrande (5-shot)              |49.17|
|GSM8k (5-shot)                   | 0.00|