You can deploy this model and start earning money today!

RichardErkhov/
OT20230122_-_karasu-base-slerp-exl2

Name: OT20230122_-_karasu-base-slerp-exl2
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

karasu-base-slerp - EXL2

Model creator: https://huggingface.co/OT20230122/
Original model: https://huggingface.co/OT20230122/karasu-base-slerp/

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
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/OT20230122_-_karasu-base-slerp-exl2 karasu-base-slerp-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 OT20230122_-_karasu-base-slerp-exl2 --revision 6_5 --local-dir karasu-base-slerp-6_5 --local-dir-use-symlinks False

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

huggingface-cli download OT20230122_-_karasu-base-slerp-exl2 --revision 6_5 --local-dir karasu-base-slerp-6.5 --local-dir-use-symlinks False

Original model description:

tags:

merge
mergekit
lazymergekit
lightblue/karasu-1.1B
TinyLlama/TinyLlama-1.1B-Chat-v1.0 base_model:
lightblue/karasu-1.1B
TinyLlama/TinyLlama-1.1B-Chat-v1.0

karasu-base-slerp

karasu-base-slerp is a merge of the following models using LazyMergekit:

🧩 Configuration

yaml

slices:
  - sources:
      - model: lightblue/karasu-1.1B
        layer_range: [0, 16]
      - model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
        layer_range: [6, 22]
merge_method: slerp
base_model: lightblue/karasu-1.1B
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "OT20230122/karasu-base-slerp"
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"])

RichardErkhov/OT20230122_-_karasu-base-slerp-exl2