aipib_-_karasu-aya-dareties2-awq

Quantization made by Richard Erkhov.

Github

Discord

Request more models

karasu-aya-dareties2 - AWQ

• Model creator: https://huggingface.co/aipib/
• Original model: https://huggingface.co/aipib/karasu-aya-dareties2/

Original model description:

base_model:

• yuiseki/karasu-aya-ja-v0.1
• aipib/karasu-slerp1 tags:
• merge
• mergekit
• lazymergekit
• yuiseki/karasu-aya-ja-v0.1
• aipib/karasu-slerp1

---

karasu-aya-dareties2

karasu-aya-dareties2 is a merge of the following models using LazyMergekit:

• yuiseki/karasu-aya-ja-v0.1
• aipib/karasu-slerp1

🧩 Configuration

slices:
- sources:
  - layer_range: [0, 22]
    model: yuiseki/karasu-aya-ja-v0.1
    parameters:
      density: [1, 0.7, 0.1]
      weight: 1.0
  - layer_range: [0, 22]
    model: aipib/karasu-slerp1
    parameters:
      density: 0.33
      weight:
        - filter: mlp
          value: 0.5
        - value: 0
merge_method: dare_ties
base_model: aipib/karasu-slerp1
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "aipib/karasu-aya-dareties2"
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"])