You can deploy this model and start earning money today!

RichardErkhov/
Heng666_-_EastAsia-4x7B-Moe-experiment-gguf

Name: Heng666_-_EastAsia-4x7B-Moe-experiment-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

EastAsia-4x7B-Moe-experiment - GGUF

Model creator: https://huggingface.co/Heng666/
Original model: https://huggingface.co/Heng666/EastAsia-4x7B-Moe-experiment/

Name	Quant method	Size
EastAsia-4x7B-Moe-experiment.Q2_K.gguf	Q2_K	6.33GB
EastAsia-4x7B-Moe-experiment.IQ3_XS.gguf	IQ3_XS	7.08GB
EastAsia-4x7B-Moe-experiment.IQ3_S.gguf	IQ3_S	7.48GB
EastAsia-4x7B-Moe-experiment.Q3_K_S.gguf	Q3_K_S	7.46GB
EastAsia-4x7B-Moe-experiment.IQ3_M.gguf	IQ3_M	7.63GB
EastAsia-4x7B-Moe-experiment.Q3_K.gguf	Q3_K	8.28GB
EastAsia-4x7B-Moe-experiment.Q3_K_M.gguf	Q3_K_M	8.28GB
EastAsia-4x7B-Moe-experiment.Q3_K_L.gguf	Q3_K_L	8.98GB
EastAsia-4x7B-Moe-experiment.IQ4_XS.gguf	IQ4_XS	9.32GB
EastAsia-4x7B-Moe-experiment.Q4_0.gguf	Q4_0	9.73GB
EastAsia-4x7B-Moe-experiment.IQ4_NL.gguf	IQ4_NL	9.83GB
EastAsia-4x7B-Moe-experiment.Q4_K_S.gguf	Q4_K_S	9.82GB
EastAsia-4x7B-Moe-experiment.Q4_K.gguf	Q4_K	10.43GB
EastAsia-4x7B-Moe-experiment.Q4_K_M.gguf	Q4_K_M	10.43GB
EastAsia-4x7B-Moe-experiment.Q4_1.gguf	Q4_1	10.8GB
EastAsia-4x7B-Moe-experiment.Q5_0.gguf	Q5_0	11.87GB
EastAsia-4x7B-Moe-experiment.Q5_K_S.gguf	Q5_K_S	11.87GB
EastAsia-4x7B-Moe-experiment.Q5_K.gguf	Q5_K	12.23GB
EastAsia-4x7B-Moe-experiment.Q5_K_M.gguf	Q5_K_M	12.23GB
EastAsia-4x7B-Moe-experiment.Q5_1.gguf	Q5_1	12.94GB
EastAsia-4x7B-Moe-experiment.Q6_K.gguf	Q6_K	14.15GB
EastAsia-4x7B-Moe-experiment.Q8_0.gguf	Q8_0	18.33GB

Original model description:

language:

zh
ja
ko
tw license: apache-2.0 tags:
moe
merge
mergekit
lazymergekit
MediaTek-Research/Breeze-7B-Instruct-v0.1
augmxnt/shisa-7b-v1
beomi/OPEN-SOLAR-KO-10.7B model-index:
name: EastAsia-4x7B-Moe-experiment 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: 39.51 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment 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: 48.92 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment 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: 56.2 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment 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: 49.83 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment 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: 58.09 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment 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: 0.15 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Heng666/EastAsia-4x7B-Moe-experiment name: Open LLM Leaderboard

EastAsia-4x7B-Moe-experiment

EastAsia-4x7B-Moe-experiment is a Mixure of Experts (MoE) made with the following models using LazyMergekit:

🧩 Configuration

yaml

gate_mode: hidden
dtype: bfloat16
base_model: mlabonne/Marcoro14-7B-slerp
experts:
  - source_model: MediaTek-Research/Breeze-7B-Instruct-v0.1
    positive_prompts:
    - "翻譯"
  - source_model: augmxnt/shisa-7b-v1
    positive_prompts:
    - "翻訳"
  - source_model: beomi/OPEN-SOLAR-KO-10.7B
    positive_prompts:
    - "번역"

💻 Usage

python

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Heng666/EastAsia-4x7B-Moe-experiment"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)

messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
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.	42.12
AI2 Reasoning Challenge (25-Shot)	39.51
HellaSwag (10-Shot)	48.92
MMLU (5-Shot)	56.20
TruthfulQA (0-shot)	49.83
Winogrande (5-shot)	58.09
GSM8k (5-shot)	0.15

RichardErkhov/Heng666_-_EastAsia-4x7B-Moe-experiment-gguf