aipib_-_phillama-prune3-awq

Quantization made by Richard Erkhov.

Github

Discord

Request more models

phillama-prune3 - AWQ

• Model creator: https://huggingface.co/aipib/
• Original model: https://huggingface.co/aipib/phillama-prune3/

Original model description:

tags:

• merge
• mergekit
• lazymergekit
• raincandy-u/phillama-3.8b-v1 base_model:
• raincandy-u/phillama-3.8b-v1
• raincandy-u/phillama-3.8b-v1

---

phillama-prune3

phillama-prune3 is a merge of the following models using LazyMergekit:

• raincandy-u/phillama-3.8b-v1
• raincandy-u/phillama-3.8b-v1

🧩 Configuration

slices:
  - sources:
      - model: raincandy-u/phillama-3.8b-v1
        layer_range: [0, 22]
  - sources:
      - model: raincandy-u/phillama-3.8b-v1
        layer_range: [26, 32]
merge_method: passthrough
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "aipib/phillama-prune3"
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"])