You can deploy this model and start earning money today!

RichardErkhov/
jsfs11_-_MixtureofMerges-MoE-2x7b-SLERPv0.9-gguf

Name: jsfs11_-_MixtureofMerges-MoE-2x7b-SLERPv0.9-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

MixtureofMerges-MoE-2x7b-SLERPv0.9 - GGUF

Model creator: https://huggingface.co/jsfs11/
Original model: https://huggingface.co/jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9/

Name	Quant method	Size
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q2_K.gguf	Q2_K	4.43GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.IQ3_XS.gguf	IQ3_XS	4.95GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.IQ3_S.gguf	IQ3_S	5.22GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q3_K_S.gguf	Q3_K_S	5.2GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.IQ3_M.gguf	IQ3_M	5.35GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q3_K.gguf	Q3_K	5.78GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q3_K_M.gguf	Q3_K_M	5.78GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q3_K_L.gguf	Q3_K_L	6.27GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.IQ4_XS.gguf	IQ4_XS	6.5GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q4_0.gguf	Q4_0	6.78GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.IQ4_NL.gguf	IQ4_NL	6.85GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q4_K_S.gguf	Q4_K_S	6.84GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q4_K.gguf	Q4_K	7.25GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q4_K_M.gguf	Q4_K_M	7.25GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q4_1.gguf	Q4_1	7.52GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q5_0.gguf	Q5_0	8.26GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q5_K_S.gguf	Q5_K_S	8.26GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q5_K.gguf	Q5_K	8.51GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q5_K_M.gguf	Q5_K_M	8.51GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q5_1.gguf	Q5_1	9.01GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q6_K.gguf	Q6_K	9.84GB
MixtureofMerges-MoE-2x7b-SLERPv0.9.Q8_0.gguf	Q8_0	12.75GB

Original model description:

license: apache-2.0 tags:

merge
mergekit
lazymergekit
jsfs11/MixtureofMerges-MoE-2x7b-v7
jsfs11/MixtureofMerges-MoE-2x7bRP-v8 base_model:
jsfs11/MixtureofMerges-MoE-2x7b-v7
jsfs11/MixtureofMerges-MoE-2x7bRP-v8 model-index:
name: MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 73.12 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 88.76 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 65.0 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 74.83 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 83.58 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 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: 69.22 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9 name: Open LLM Leaderboard

MixtureofMerges-MoE-2x7b-SLERPv0.9

MixtureofMerges-MoE-2x7b-SLERPv0.9 is a merge of the following models using LazyMergekit:

🧩 Configuration

yaml

slices:
  - sources:
      - model: jsfs11/MixtureofMerges-MoE-2x7b-v7 
        layer_range: [0, 32]
      - model: jsfs11/MixtureofMerges-MoE-2x7bRP-v8
        layer_range: [0, 32]
merge_method: slerp
base_model: jsfs11/MixtureofMerges-MoE-2x7b-v7
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

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "jsfs11/MixtureofMerges-MoE-2x7b-SLERPv0.9"
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.	75.75
AI2 Reasoning Challenge (25-Shot)	73.12
HellaSwag (10-Shot)	88.76
MMLU (5-Shot)	65.00
TruthfulQA (0-shot)	74.83
Winogrande (5-shot)	83.58
GSM8k (5-shot)	69.22

RichardErkhov/jsfs11_-_MixtureofMerges-MoE-2x7b-SLERPv0.9-gguf