README.md

metadata

language:
  - pt
  - en
license: cc
tags:
  - text-generation-inference
  - transformers
  - mistral
  - mixtral
  - gguf
  - brazil
  - brasil
  - portuguese
model-index:
  - name: CabraMixtral-8x7b
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: ENEM Challenge (No Images)
          type: eduagarcia/enem_challenge
          split: train
          args:
            num_few_shot: 3
        metrics:
          - type: acc
            value: 78.17
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BLUEX (No Images)
          type: eduagarcia-temp/BLUEX_without_images
          split: train
          args:
            num_few_shot: 3
        metrics:
          - type: acc
            value: 64.12
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: OAB Exams
          type: eduagarcia/oab_exams
          split: train
          args:
            num_few_shot: 3
        metrics:
          - type: acc
            value: 55.49
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Assin2 RTE
          type: assin2
          split: test
          args:
            num_few_shot: 15
        metrics:
          - type: f1_macro
            value: 90.95
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Assin2 STS
          type: eduagarcia/portuguese_benchmark
          split: test
          args:
            num_few_shot: 15
        metrics:
          - type: pearson
            value: 77.63
            name: pearson
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: FaQuAD NLI
          type: ruanchaves/faquad-nli
          split: test
          args:
            num_few_shot: 15
        metrics:
          - type: f1_macro
            value: 78.93
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HateBR Binary
          type: ruanchaves/hatebr
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: f1_macro
            value: 78
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: PT Hate Speech Binary
          type: hate_speech_portuguese
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: f1_macro
            value: 69.54
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: tweetSentBR
          type: eduagarcia/tweetsentbr_fewshot
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: f1_macro
            value: 72.83
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=botbot-ai/CabraMixtral-8x7b
          name: Open Portuguese LLM Leaderboard

BotBot Cabra Mixtral 8x7b

Esse modelo é um finetune do Mixtral 8x7b com o dataset Cabra 30k. Esse modelo é optimizado para português. Ele apresenta melhoria em varios benchmarks brasileiros em comparação com o modelo base.

Conheça os nossos outros modelos: Cabra.

dataset: Cabra 30k

Dataset interno para finetuning. Vamos lançar em breve.

Quantização / GGUF

Colocamos diversas versões (GGUF) quantanizadas no branch "quantanization".

Exemplo

<s> [INST] who is Elon Musk? [/INST]Elon Musk é um empreendedor, inventor e capitalista americano. Ele é o fundador, CEO e CTO da SpaceX, CEO da Neuralink e fundador do The Boring Company. Musk também é o proprietário do Twitter.</s>

Uso

O modelo é destinado, por agora, a fins de pesquisa. As áreas e tarefas de pesquisa possíveis incluem:

• Pesquisa sobre modelos gerativos.
• Investigação e compreensão das limitações e viéses de modelos gerativos.

Proibido para uso comercial. Somente pesquisa.

Evals

Open Portuguese LLM Leaderboard Evaluation Results

Detailed results can be found here and on the 🚀 Open Portuguese LLM Leaderboard

Metric	Value
Average	73.96
ENEM Challenge (No Images)	78.17
BLUEX (No Images)	64.12
OAB Exams	55.49
Assin2 RTE	90.95
Assin2 STS	77.63
FaQuAD NLI	78.93
HateBR Binary	78
PT Hate Speech Binary	69.54
tweetSentBR	72.83