gmonsoon's picture
Adding Evaluation Results (#1)
241b5aa verified
|
raw
history blame
4.59 kB
metadata
language:
  - en
  - id
license: cc-by-nc-4.0
tags:
  - text-generation-inference
  - transformers
  - mistral
  - trl
datasets:
  - MBZUAI/Bactrian-X
base_model: mistral-7b
model-index:
  - name: OpenMia-Indo-Mistral-7b-v4
    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: 64.16
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          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: 82.84
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          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: 61.08
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          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: 53.36
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          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: 79.08
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          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: 47.84
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=indischepartij/OpenMia-Indo-Mistral-7b-v4
          name: Open LLM Leaderboard

image/png

updated to V4, after this version, OpenMia will be finetuned (branched) to some Indonesia local languages, such as Javanese, Sundanese, and Minang language. Stay tuned.

MIA : (M)istral finetuned with (I)ndonesia language from (A)lpaca dataset

(formerly named Mistral-7b-Alpaca-Indonesia)

OpenMia-Indo-Mistral-7b is finetuned model based of Mistral-7b with capability to do conversation in Bahasa Indonesia.

Due to limited resources, this model is still in alpha stage.

Want to contribute to this project? join our organization: https://huggingface.co/indischepartij or contact me at https://twitter.com/gmonsooniii

Modelfile/Prompt format

SYSTEM Kamu adalah asisten AI yang cerdas dan ceria, bernama Mia.

PARAMETER stop <|im_start|> 
PARAMETER stop <|im_end|>

TEMPLATE <|im_start|>system {{ .System }}<|im_end|> <|im_start|>user {{ .Prompt }}<|im_end|> <|im_start|>assistant

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 64.73
AI2 Reasoning Challenge (25-Shot) 64.16
HellaSwag (10-Shot) 82.84
MMLU (5-Shot) 61.08
TruthfulQA (0-shot) 53.36
Winogrande (5-shot) 79.08
GSM8k (5-shot) 47.84