allknowingroger's picture
Adding Evaluation Results (#1)
1962d70 verified
metadata
license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - Kukedlc/Neural4gsm8k
  - PetroGPT/WestSeverus-7B-DPO
base_model:
  - Kukedlc/Neural4gsm8k
  - PetroGPT/WestSeverus-7B-DPO
model-index:
  - name: NeuralWestSeverus-7B-slerp
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 41.36
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 33.41
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 6.87
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 2.8
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 15.41
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 23.75
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/NeuralWestSeverus-7B-slerp
          name: Open LLM Leaderboard

NeuralWestSeverus-7B-slerp

NeuralWestSeverus-7B-slerp is a merge of the following models using LazyMergekit:

🧩 Configuration

slices:
  - sources:
      - model: Kukedlc/Neural4gsm8k
        layer_range: [0, 32]
      - model: PetroGPT/WestSeverus-7B-DPO
        layer_range: [0, 32]
merge_method: slerp
base_model: Kukedlc/Neural4gsm8k
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "allknowingroger/NeuralWestSeverus-7B-slerp"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 20.60
IFEval (0-Shot) 41.36
BBH (3-Shot) 33.41
MATH Lvl 5 (4-Shot) 6.87
GPQA (0-shot) 2.80
MuSR (0-shot) 15.41
MMLU-PRO (5-shot) 23.75