You can deploy this model and start earning money today!

RichardErkhov/
Yuma42_-_KangalKhan-PrimordialSapphire-7B-gguf

Name: Yuma42_-_KangalKhan-PrimordialSapphire-7B-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

KangalKhan-PrimordialSapphire-7B - GGUF

Model creator: https://huggingface.co/Yuma42/
Original model: https://huggingface.co/Yuma42/KangalKhan-PrimordialSapphire-7B/

Name	Quant method	Size
KangalKhan-PrimordialSapphire-7B.Q2_K.gguf	Q2_K	2.53GB
KangalKhan-PrimordialSapphire-7B.IQ3_XS.gguf	IQ3_XS	2.81GB
KangalKhan-PrimordialSapphire-7B.IQ3_S.gguf	IQ3_S	2.96GB
KangalKhan-PrimordialSapphire-7B.Q3_K_S.gguf	Q3_K_S	2.95GB
KangalKhan-PrimordialSapphire-7B.IQ3_M.gguf	IQ3_M	3.06GB
KangalKhan-PrimordialSapphire-7B.Q3_K.gguf	Q3_K	3.28GB
KangalKhan-PrimordialSapphire-7B.Q3_K_M.gguf	Q3_K_M	3.28GB
KangalKhan-PrimordialSapphire-7B.Q3_K_L.gguf	Q3_K_L	3.56GB
KangalKhan-PrimordialSapphire-7B.IQ4_XS.gguf	IQ4_XS	3.67GB
KangalKhan-PrimordialSapphire-7B.Q4_0.gguf	Q4_0	3.83GB
KangalKhan-PrimordialSapphire-7B.IQ4_NL.gguf	IQ4_NL	3.87GB
KangalKhan-PrimordialSapphire-7B.Q4_K_S.gguf	Q4_K_S	3.86GB
KangalKhan-PrimordialSapphire-7B.Q4_K.gguf	Q4_K	4.07GB
KangalKhan-PrimordialSapphire-7B.Q4_K_M.gguf	Q4_K_M	4.07GB
KangalKhan-PrimordialSapphire-7B.Q4_1.gguf	Q4_1	4.24GB
KangalKhan-PrimordialSapphire-7B.Q5_0.gguf	Q5_0	4.65GB
KangalKhan-PrimordialSapphire-7B.Q5_K_S.gguf	Q5_K_S	4.65GB
KangalKhan-PrimordialSapphire-7B.Q5_K.gguf	Q5_K	4.78GB
KangalKhan-PrimordialSapphire-7B.Q5_K_M.gguf	Q5_K_M	4.78GB
KangalKhan-PrimordialSapphire-7B.Q5_1.gguf	Q5_1	5.07GB
KangalKhan-PrimordialSapphire-7B.Q6_K.gguf	Q6_K	5.53GB
KangalKhan-PrimordialSapphire-7B.Q8_0.gguf	Q8_0	7.17GB

Original model description:

language:

en license: apache-2.0 tags:
merge
mergekit
lazymergekit
argilla/CapybaraHermes-2.5-Mistral-7B
Yuma42/KangalKhan-RawEmerald-7B base_model:
argilla/CapybaraHermes-2.5-Mistral-7B
Yuma42/KangalKhan-RawEmerald-7B model-index:
name: KangalKhan-PrimordialSapphire-7B 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: 65.87 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B 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: 85.51 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B 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: 63.11 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B 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: 57.25 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B 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: 78.22 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B 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: 61.18 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Yuma42/KangalKhan-PrimordialSapphire-7B name: Open LLM Leaderboard

KangalKhan-PrimordialSapphire-7B

KangalKhan-PrimordialSapphire-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: Yuma42/KangalKhan-RawEmerald-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-PrimordialSapphire-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.	68.52
AI2 Reasoning Challenge (25-Shot)	65.87
HellaSwag (10-Shot)	85.51
MMLU (5-Shot)	63.11
TruthfulQA (0-shot)	57.25
Winogrande (5-shot)	78.22
GSM8k (5-shot)	61.18

RichardErkhov/Yuma42_-_KangalKhan-PrimordialSapphire-7B-gguf