README.md

metadata

tags:
  - merge
  - mergekit
  - lazymergekit
  - allknowingroger/Multimerge-12B-MoE
base_model:
  - allknowingroger/Multimerge-12B-MoE
  - allknowingroger/Multimerge-12B-MoE
license: apache-2.0

Multimerge-19B-pass

Multimerge-19B-pass is a merge of the following models using LazyMergekit:

• allknowingroger/Multimerge-12B-MoE
• allknowingroger/Multimerge-12B-MoE

🧩 Configuration

slices:
  - sources:
    - model: allknowingroger/Multimerge-12B-MoE
      layer_range: [8, 32]
  - sources:
    - model: allknowingroger/Multimerge-12B-MoE
      layer_range: [8, 32]
merge_method: passthrough
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "allknowingroger/Multimerge-19B-pass"
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"])