README.md

metadata

language:
  - en
license: cc-by-nc-sa-4.0
datasets:
  - Intel/orca_dpo_pairs
  - argilla/distilabel-math-preference-dpo
  - kyujinpy/orca_math_dpo
pipeline_tag: text-generation
model-index:
  - name: Sakura-SOLRCA-Math-Instruct-DPO-v1
    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: 71.25
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          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: 88.48
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          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: 66.21
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          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: 72.12
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          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: 82.87
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          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: 63.84
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
          name: Open LLM Leaderboard

Sakura-SOLRCA-Math-Instruct-DPO-v1

Model Details

Model Developers Kyujin Han (kyujinpy)

Method
Using DPO method.
With Intel/orca_dpo_pairs and argilla/distilabel-math-preference-dpo.

I shared the merge version kyujinpy/orca_math_dpo.

I will share the information about my model. (training and code)
Please see: ⭐Sakura-SOLAR.

Model Benchmark

Open leaderboard

• Follow up as link.

Model	Average	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K
Sakura-SOLRCA-Math-Instruct-DPO-v2	74.17	71.25	88.52	66.13	72.16	83.03	63.91
Sakura-SOLRCA-Math-Instruct-DPO-v1	74.13	71.25	88.48	66.21	72.12	82.87	63.84
Sakura-SOLRCA-Instruct-DPO	74.05	71.16	88.49	66.17	72.10	82.95	63.46
Sakura-SOLAR-Instruct-DPO-v2	74.14	70.90	88.41	66.48	71.86	83.43	63.76
kyujinpy/Sakura-SOLAR-Instruct	74.40	70.99	88.42	66.33	71.79	83.66	65.20

Implementation Code

### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1"
OpenOrca = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)

---

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	74.13
AI2 Reasoning Challenge (25-Shot)	71.25
HellaSwag (10-Shot)	88.48
MMLU (5-Shot)	66.21
TruthfulQA (0-shot)	72.12
Winogrande (5-shot)	82.87
GSM8k (5-shot)	63.84