CultriX / MergeCeption-7B-v1

MergeCeption-7B-v1

MergeCeption-7B-v1 is a merge of the following models using LazyMergekit :

• CultriX/MonaTrix-v4
• mlabonne/OmniTruthyBeagle-7B-v0
• Kukedlc/NeuralMaxime-7B-slerp
• CultriX/NeuralTrixlaser-bf16

🧩 Configuration

base_model: CultriX/MonaTrix-v4
merge_method: dare_ties
models:
  - model: CultriX/MonaTrix-v4
    parameters:
      density: 0.65
      weight: 0.4
  - model: mlabonne/OmniTruthyBeagle-7B-v0
    parameters:
      density: 0.65
      weight: 0.4
  - model: Kukedlc/NeuralMaxime-7B-slerp
    parameters:
      density: 0.65
      weight: 0.4
  - model: CultriX/NeuralTrixlaser-bf16
    parameters:
      density: 0.6
      weight: 0.35
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "CultriX/MergeCeption-7B-v1"
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"])