EXAONE 3.5 7.8B Korean-English Translation
This model is LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct fine-tuned with LoRA and merged for Korean-English translation tasks.
Model Details
- Base Model: LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct
- Fine-tuning Method: LoRA (Low-Rank Adaptation)
- LoRA Config:
- Rank (r): 64
- Alpha: 128
- Dropout: 0.05
- Target Modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained(
"ray5273/EXAONE-3.5-7.8B-KorEng-Translation",
torch_dtype="auto",
device_map="auto",
trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(
"ray5273/EXAONE-3.5-7.8B-KorEng-Translation",
trust_remote_code=True
)
messages = [
{"role": "system", "content": "You are a helpful translator."},
{"role": "user", "content": "Translate the following to English: 안녕하세요, 만나서 반갑습니다."}
]
input_ids = tokenizer.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
return_tensors="pt"
).to(model.device)
output = model.generate(input_ids, max_new_tokens=256)
print(tokenizer.decode(output[0], skip_special_tokens=True))
License
This model follows the EXAONE AI Model License Agreement 1.1 - NC. See the EXAONE License for details.