en license: apache-2.0 tags:
merge
mergekit
lazymergekit
argilla/CapybaraHermes-2.5-Mistral-7B
argilla/distilabeled-OpenHermes-2.5-Mistral-7B base_model:
argilla/CapybaraHermes-2.5-Mistral-7B
argilla/distilabeled-OpenHermes-2.5-Mistral-7B model-index:
name: KangalKhan-Ruby-7B-Fixed 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: 67.24 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed 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: 85.22 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed 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: 63.21 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed 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: 56.49 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed 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: 77.98 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed 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: 61.94 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-Ruby-7B-Fixed name: Open LLM Leaderboard

KangalKhan-Ruby-7B

I suggest using ChatML (Use whatever system prompt you like, this is just an example!):

<|im_start|>system
You are a friendly assistant.<|im_end|>
<|im_start|>user
Hello, what are you?<|im_end|>
<|im_start|>assistant
I am an AI language model designed to assist users with information and answer their questions. How can I help you today?<|im_end|>

Q4_K_S GGUF:
https://huggingface.co/Yuma42/KangalKhan-Ruby-7B-Fixed-GGUF

More GGUF variants by mradermacher:
WARNING: I have observed that these versions output typos in rare cases. If you have the same problem, use my Q4_K_S GGUF above. https://huggingface.co/mradermacher/KangalKhan-Ruby-7B-Fixed-GGUF

KangalKhan-Ruby-7B is a merge of the following models using LazyMergekit:

🧩 Configuration

yaml

slices:
  - sources:
      - model: argilla/CapybaraHermes-2.5-Mistral-7B
        layer_range: [0, 32]
      - model: argilla/distilabeled-OpenHermes-2.5-Mistral-7B
        layer_range: [0, 32]
merge_method: slerp
base_model:  argilla/CapybaraHermes-2.5-Mistral-7B
parameters:
  t:
    - filter: self_attn
      value: [1, 0.5, 0.7, 0.3, 0]
    - filter: mlp
      value: [0, 0.5, 0.3, 0.7, 1]
    - value: 0.5
dtype: bfloat16

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Yuma42/KangalKhan-Ruby-7B"
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.	68.68
AI2 Reasoning Challenge (25-Shot)	67.24
HellaSwag (10-Shot)	85.22
MMLU (5-Shot)	63.21
TruthfulQA (0-shot)	56.49
Winogrande (5-shot)	77.98
GSM8k (5-shot)	61.94

RichardErkhov/Yuma42_-_KangalKhan-Ruby-7B-Fixed-8bits

Original model description:

KangalKhan-Ruby-7B

🧩 Configuration

💻 Usage

Open LLM Leaderboard Evaluation Results

RichardErkhov/
Yuma42_-_KangalKhan-Ruby-7B-Fixed-8bits