README.md

metadata

language:
  - pt
license: apache-2.0
datasets:
  - nicholasKluge/Pt-Corpus
model-index:
  - name: Mistral-7B-v0.2-Base_ptbr
    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: 64.94
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 53.96
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 45.42
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 90.11
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 72.51
            name: pearson
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 69.04
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 79.62
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 58.52
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          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: 62.32
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
          name: Open Portuguese LLM Leaderboard

É um modelo base pré-treinado com cerca de 1b tokens em portugues iniciado com os pesos oficiais do modelo, deve ser utilizado para fine tuning.

Obs: Aguardando resultados oficiais

	Mistral Base PTBR	Mistral Base	Melhoria
assin2_rte	90,2	87,74	2,46
assin2_sts	72,45	67,05	5,4
bluex	53,27	53,27	0
enem	64,66	62,42	2,24
faquad_nli	68,11	47,63	20,48
hatebr_offensive_binary	79,65	77,63	2,02
oab_exams	45,42	45,24	0,18
portuguese_hate_speech_binary	59,18	55,72	3,46

Open Portuguese LLM Leaderboard Evaluation Results

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

Metric	Value
Average	66.27
ENEM Challenge (No Images)	64.94
BLUEX (No Images)	53.96
OAB Exams	45.42
Assin2 RTE	90.11
Assin2 STS	72.51
FaQuAD NLI	69.04
HateBR Binary	79.62
PT Hate Speech Binary	58.52
tweetSentBR	62.32