models:

  - model: unsloth/mistral-7b

    parameters:

      weight: 1

      density: 1

  - model: unsloth/mistral-7b-v0.2

    parameters:

      weight: 1

      density: 1

merge_method: ties

base_model: unsloth/mistral-7b

parameters:

  weight: 1

  density: 1

  normalize: true

  int8_mask: false

dtype: float16

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "JoPmt/Trismal-7B-Base-v1v2-Ties"
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"])