README.md

metadata

language:
  - en
license: other
library_name: transformers
tags:
  - chat
  - qwen
  - qwen2
  - finetune
  - chatml
base_model: Qwen/Qwen2-72B-Instruct
model_name: MaziyarPanahi/calme-2.1-qwen2-72b
license_name: tongyi-qianwen
license_link: https://huggingface.co/Qwen/Qwen2-72B-Instruct/blob/main/LICENSE
pipeline_tag: text-generation
inference: false
model_creator: MaziyarPanahi
quantized_by: MaziyarPanahi
model-index:
  - name: calme-2.1-qwen2-72b
    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: 81.63
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          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: 57.33
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          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: 36.03
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          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: 17.45
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          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: 20.15
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          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: 49.05
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=MaziyarPanahi/calme-2.1-qwen2-72b
          name: Open LLM Leaderboard

MaziyarPanahi/calme-2.1-qwen2-72b

This is a fine-tuned version of the Qwen/Qwen2-72B-Instruct model. It aims to improve the base model across all benchmarks.

⚡ Quantized GGUF

All GGUF models are available here: MaziyarPanahi/calme-2.1-qwen2-72b-GGUF

🏆 Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	43.61
IFEval (0-Shot)	81.63
BBH (3-Shot)	57.33
MATH Lvl 5 (4-Shot)	36.03
GPQA (0-shot)	17.45
MuSR (0-shot)	20.15
MMLU-PRO (5-shot)	49.05

Tasks	Version	Filter	n-shot	Metric	Value		Stderr
truthfulqa_mc2	2	none	0	acc	0.6761	±	0.0148

Tasks	Version	Filter	n-shot	Metric	Value		Stderr
winogrande	1	none	5	acc	0.8248	±	0.0107

Tasks	Version	Filter	n-shot	Metric	Value		Stderr
arc_challenge	1	none	25	acc	0.6852	±	0.0136
		none	25	acc_norm	0.7184	±	0.0131

Tasks	Version	Filter	n-shot	Metric	Value		Stderr
gsm8k	3	strict-match	5	exact_match	0.8582	±	0.0096
		flexible-extract	5	exact_match	0.8893	±	0.0086

Prompt Template

This model uses ChatML prompt template:

<|im_start|>system
{System}
<|im_end|>
<|im_start|>user
{User}
<|im_end|>
<|im_start|>assistant
{Assistant}

How to use

# Use a pipeline as a high-level helper

from transformers import pipeline

messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe = pipeline("text-generation", model="MaziyarPanahi/calme-2.1-qwen2-72b")
pipe(messages)


# Load model directly

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MaziyarPanahi/calme-2.1-qwen2-72b")
model = AutoModelForCausalLM.from_pretrained("MaziyarPanahi/calme-2.1-qwen2-72b")