mbahrsnc_-_mini-mcqueen-exl2

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mini-mcqueen - EXL2

• Model creator: https://huggingface.co/mbahrsnc/
• Original model: https://huggingface.co/mbahrsnc/mini-mcqueen/

Available sizes

| Branch | Bits | Description | | ----- | ---- | ------- | ------ | ------ | ------ | ------ | ------------ | | 8_0 | 8.0 | Maximum quality that ExLlamaV2 can produce, near unquantized performance. | | 6_5 | 6.5 | Very similar to 8.0, good tradeoff of size vs performance, recommended. | | 5_0 | 5.0 | Slightly lower quality vs 6.5, but usable on 8GB cards. | | 4_25 | 4.25 | GPTQ equivalent bits per weight, slightly higher quality. | | 3_5 | 3.5 | Lower quality, only use if you have to. |

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/mbahrsnc_-_mini-mcqueen-exl2 mini-mcqueen-6_5

With huggingface hub:

pip3 install huggingface-hub

To download a specific branch, use the --revision parameter. For example, to download the 6.5 bpw branch: Linux:

huggingface-cli download mbahrsnc_-_mini-mcqueen-exl2 --revision 6_5 --local-dir mini-mcqueen-6_5 --local-dir-use-symlinks False

Windows (which apparently doesn't like _ in folders sometimes?):

huggingface-cli download mbahrsnc_-_mini-mcqueen-exl2 --revision 6_5 --local-dir mini-mcqueen-6.5 --local-dir-use-symlinks False

Original model description:

base_model:

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

---

mini-mcqueen

mini-mcqueen is a merge of the following models using LazyMergekit:

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

🧩 Configuration

slices:
  - sources:
      - model: cognitivecomputations/TinyDolphin-2.8-1.1b
        layer_range: [0, 21]
        output_weight: 0.4
      - model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
        layer_range: [0, 21]
        output_weight: 0.6
merge_method: slerp
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
parameters:
  t:
    - filter: self_attn
      value: [0, 0.75, 0.5, 0.85, 1]
    - filter: mlp
      value: [1, 0.75, 0.85, 0.5, 0.25]
    - value: 0.75
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mbahrsnc/mini-mcqueen"
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"])