metadata
license: cc-by-sa-4.0
Synatra-7B-v0.3-Translation๐ง
Support Me
์๋ํธ๋ผ๋ ๊ฐ์ธ ํ๋ก์ ํธ๋ก, 1์ธ์ ์์์ผ๋ก ๊ฐ๋ฐ๋๊ณ ์์ต๋๋ค. ๋ชจ๋ธ์ด ๋ง์์ ๋์ จ๋ค๋ฉด ์ฝ๊ฐ์ ์ฐ๊ตฌ๋น ์ง์์ ์ด๋จ๊น์?
Wanna be a sponser? (Please) Contact me on Telegram AlzarTakkarsen
Model Details
Base Model
mistralai/Mistral-7B-Instruct-v0.1
Datasets sharegpt_deepl_ko_translation
Filtered version of above dataset included.
Trained On
A100 80GB * 1
Instruction format
It follows ChatML format and Alpaca(No-Input) format.
<|im_start|>system
์ฃผ์ด์ง ๋ฌธ์ฅ์ ํ๊ตญ์ด๋ก ๋ฒ์ญํด๋ผ.<|im_end|>
<|im_start|>user
{instruction}<|im_end|>
<|im_start|>assistant
<|im_start|>system
์ฃผ์ด์ง ๋ฌธ์ฅ์ ์์ด๋ก ๋ฒ์ญํด๋ผ.<|im_end|>
<|im_start|>user
{instruction}<|im_end|>
<|im_start|>assistant
Ko-LLM-Leaderboard
On Benchmarking...
Implementation Code
Since, chat_template already contains insturction format above. You can use the code below.
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto
model = AutoModelForCausalLM.from_pretrained("maywell/Synatra-7B-v0.3-Translation")
tokenizer = AutoTokenizer.from_pretrained("maywell/Synatra-7B-v0.3-Translation")
messages = [
{"role": "user", "content": "๋ฐ๋๋๋ ์๋ ํ์์์ด์ผ?"},
]
encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
model_inputs = encodeds.to(device)
model.to(device)
generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])