README.md

metadata

language:
  - ru
license: apache-2.0
datasets:
  - 0x7194633/fialka-v3-data
pipeline_tag: text-generation
model-index:
  - name: fialka-13B-v3
    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: 30.97
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          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: 48.83
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          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: 26.36
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          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: 40.58
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          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: 59.43
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          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: 1.29
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=0x7194633/fialka-13B-v3
          name: Open LLM Leaderboard

Fialka v3.0 13B

Description

Fialka language models are trained to follow instructions and maintain communication in Russian.

Usage

The model has a query format as in zephyr.

<|user|>
Что такое мем?</s>
<|assistant|>
Мем (англ. meme) — это единица культурной информации, которая распространяется в социальных сетях и других онлайн-платформах с помощью цифровых технологий или через физический контакт. Мемы могут быть связаны между собой тематически или иметь общие черты, такие как использование определенных слов или фраз для создания определенного настроения или выражения эмоций. Они также могут содержать информацию о культуре, истории или науке, которую можно использовать для обучения новым вещам или расширения кругозора.

Check out the space to use the model in UI without downloading.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	34.58
AI2 Reasoning Challenge (25-Shot)	30.97
HellaSwag (10-Shot)	48.83
MMLU (5-Shot)	26.36
TruthfulQA (0-shot)	40.58
Winogrande (5-shot)	59.43
GSM8k (5-shot)	1.29