slices:
  - sources:
    - model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
      layer_range: [0, 8]
  - sources:
    - model: Nondzu/Mistral-7B-Instruct-v0.2-code-ft
      layer_range: [4, 14]
  - sources:
    - model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
      layer_range: [10, 20]
  - sources:
    - model: Nondzu/Mistral-7B-Instruct-v0.2-code-ft
      layer_range: [16, 26]
  - sources:
    - model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
      layer_range: [22, 32]
merge_method: passthrough
base_model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
dtype: float16

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "frost19k/dolphin-2.8-mistral-11b-v02-code-ft"
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"])