README.md

metadata

language:
  - en
license: mit
tags:
  - ORPO
datasets:
  - mlabonne/orpo-dpo-mix-40k
model-index:
  - name: Barcenas-14b-Phi-3-medium-ORPO
    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: 73.2
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 65.79
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 51.03
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 92.6
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 71.45
            name: pearson
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 69.06
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 84.6
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 73.55
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          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: 67.01
            name: f1-macro
        source:
          url: >-
            https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Danielbrdz/Barcenas-14b-Phi-3-medium-ORPO
          name: Open Portuguese LLM Leaderboard

Barcenas-14b-Phi-3-medium-ORPO

Model trained with the innovative ORPO method, based on the robust VAGOsolutions/SauerkrautLM-Phi-3-medium.

The model was trained with the dataset: mlabonne/orpo-dpo-mix-40k, which combines diverse data sources to enhance conversational capabilities and contextual understanding.

Made with ❤️ in Guadalupe, Nuevo Leon, Mexico 🇲🇽

Open Portuguese LLM Leaderboard Evaluation Results

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

Metric	Value
Average	72.03
ENEM Challenge (No Images)	73.20
BLUEX (No Images)	65.79
OAB Exams	51.03
Assin2 RTE	92.60
Assin2 STS	71.45
FaQuAD NLI	69.06
HateBR Binary	84.60
PT Hate Speech Binary	73.55
tweetSentBR	67.01