README.md

metadata

tags:
  - merge
  - mergekit
  - lazymergekit
  - datatab/Serbian-Mistral-Orca-Slim-v1
  - mlabonne/AlphaMonarch-7B
  - datatab/YugoGPT-Alpaca-v1-epoch1-good
base_model:
  - datatab/Serbian-Mistral-Orca-Slim-v1
  - mlabonne/AlphaMonarch-7B
  - datatab/YugoGPT-Alpaca-v1-epoch1-good

datatab/Serbian-Mistral-Orca-Slim-v1

datatab/Serbian-Mistral-Orca-Slim-v1 is a merge of the following models using LazyMergekit:

• datatab/Serbian-Mistral-Orca-Slim-v1
• mlabonne/AlphaMonarch-7B
• datatab/YugoGPT-Alpaca-v1-epoch1-good

🧩 Configuration

models:
  - model: datatab/Serbian-Mistral-Orca-Slim-v1
    parameters:
      weight: 1.0
  - model: mlabonne/AlphaMonarch-7B
    parameters:
      weight: 1.0
  - model: datatab/YugoGPT-Alpaca-v1-epoch1-good
    parameters:
      weight: 1.0
merge_method: linear
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "datatab/datatab/Serbian-Mistral-Orca-Slim-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"])