kyujinpy's picture
Adding Evaluation Results (#1)
49133ed verified
metadata
language:
  - en
license: cc-by-nc-sa-4.0
library_name: transformers
datasets:
  - garage-bAInd/Open-Platypus
pipeline_tag: text-generation
model-index:
  - name: PlatYi-34B-200k-Q-FastChat
    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: 64.93
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          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: 84.46
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          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: 77.13
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          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: 48.38
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          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: 80.74
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          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: 51.48
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/PlatYi-34B-200k-Q-FastChat
          name: Open LLM Leaderboard

PlatYi-34B-200k-Q-FastChat

Model Details

Model Developers Kyujin Han (kyujinpy)

Input Models input text only.

Output Models generate text only.

Model Architecture
PlatYi-34B-200k-Q-FastChat is an auto-regressive language model based on the Yi-34B transformer architecture.

Blog Link
Blog: [Coming soon...]
Github: [Coming soon...]

Base Model
01-ai/Yi-34B-200K

Training Dataset
garage-bAInd/Open-Platypus.

Notice
While training, I used QLoRA.
lora_r values is 64.

Apply prompting
References by FastChat.

Model Benchmark

Open leaderboard

  • Follow up as link.
Model Average ARC HellaSwag MMLU TruthfulQA Winogrande GSM8K
PlatYi-34B-200k-Q-FastChat 67.85 64.93 84.46 77.13 48.38 80.74 51.48
PlatYi-34B-Llama-Q-FastChat 68.31 66.31 85.25 78.37 53.62 82.16 44.35
Yi-34B 69.42 64.59 85.69 76.35 56.23 83.03 50.64

Implementation Code

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

repo = "kyujinpy/PlatYi-34B-200k-Q-FastChat"
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. 67.85
AI2 Reasoning Challenge (25-Shot) 64.93
HellaSwag (10-Shot) 84.46
MMLU (5-Shot) 77.13
TruthfulQA (0-shot) 48.38
Winogrande (5-shot) 80.74
GSM8k (5-shot) 51.48