Edit model card

FINGU-AI/Qwen2.5-32B-Lora-HQ-e-635

Overview

FINGU-AI/Qwen2.5-32B-Lora-HQ-e-635 is a powerful causal language model designed for a variety of natural language processing (NLP) tasks, including machine translation, text generation, and chat-based applications. This model is particularly useful for translating between Korean and Uzbek, as well as supporting other custom NLP tasks through flexible input.

Model Details

  • Model ID: FINGU-AI/Qwen2.5-32B-Lora-HQ-e-635
  • Architecture: Causal Language Model (LM)
  • Parameters: 32 billion
  • Precision: Torch BF16 for efficient GPU memory usage
  • Attention: SDPA (Scaled Dot-Product Attention)
  • Primary Use Case: Translation (e.g., Korean to Uzbek), text generation, and dialogue systems.

Example Usage

Installation

Make sure to install the required packages:

pip install torch transformers

Loading the Model

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Model and Tokenizer
model_id = 'FINGU-AI/Qwen2.5-32B-Lora-HQ-e-635'
model = AutoModelForCausalLM.from_pretrained(model_id, attn_implementation="sdpa", torch_dtype=torch.bfloat16)
tokenizer = AutoTokenizer.from_pretrained(model_id)
model.to('cuda')

# Input Messages for Translation
messages = [
    {"role": "system", "content": "translate korean to Uzbek"},
    {"role": "user", "content": """์ƒˆ๋กœ์šด ์€ํ–‰ ๊ณ„์ขŒ๋ฅผ ๊ฐœ์„คํ•˜๋Š” ์ ˆ์ฐจ๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค:

1. ๊ณ„์ขŒ ๊ฐœ์„ค ๋ชฉ์ ๊ณผ ์‹ ๋ถ„ ํ™•์ธ์„ ์œ„ํ•œ ์„œ๋ฅ˜ ์ œ์ถœ
2. ์„œ๋ฅ˜ ๊ฒ€ํ†  ๊ณผ์ •์„ ๊ฑฐ์น˜๋Š” ๊ฒƒ
3. ๊ณ ๊ฐ๋‹˜์˜ ์‹ ์› ํ™•์ธ ์ ˆ์ฐจ๋ฅผ ์ง„ํ–‰ํ•˜๋Š” ๊ฒƒ
4. ๋ชจ๋“  ์ ˆ์ฐจ๊ฐ€ ์™„๋ฃŒ๋˜๋ฉด ๊ณ„์ขŒ ๊ฐœ์„ค์ด ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.

๊ณ„์ขŒ ๊ฐœ์„ค์„ ์›ํ•˜์‹œ๋Š” ๊ฒฝ์šฐ, ์‹ ๋ถ„์ฆ๊ณผ ํ•จ๊ป˜ ๋ฐฉ๋ฌธํ•ด ์ฃผ์‹œ๋ฉด ๋ฉ๋‹ˆ๋‹ค.
    """},
]

# Tokenize and Generate Response
input_ids = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to('cuda')

outputs = model.generate(
    input_ids,
    max_new_tokens=500,
    do_sample=True,
)

# Decode and Print the Translation
response = outputs[0][input_ids.shape[-1]:]
print(tokenizer.decode(response, skip_special_tokens=True))
Downloads last month
11
Inference API
Unable to determine this modelโ€™s pipeline type. Check the docs .

Model tree for FINGU-AI/Qwen2.5-32B-Lora-HQ-e-635

Base model

Qwen/Qwen2.5-32B
Adapter
(8)
this model