README.md

metadata

base_model:
  - TinyLlama/TinyLlama-1.1B-Chat-v1.0
  - cognitivecomputations/TinyDolphin-2.8-1.1b
tags:
  - merge
  - mergekit
  - lazymergekit
  - TinyLlama/TinyLlama-1.1B-Chat-v1.0
  - cognitivecomputations/TinyDolphin-2.8-1.1b

tiny-frankenstein

tiny-frankenstein is a merge of the following models using LazyMergekit:

• TinyLlama/TinyLlama-1.1B-Chat-v1.0
• cognitivecomputations/TinyDolphin-2.8-1.1b

🧩 Configuration

slices:
  - sources:
    - model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
      layer_range: [0, 14]
  - sources:
    - model: cognitivecomputations/TinyDolphin-2.8-1.1b
      layer_range: [14, 21]
merge_method: passthrough
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mbahrsnc/tiny-frankenstein"
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"])