DataVortexS-10.7B-dpo-v1.6
Our Team
| Research & Engineering | Product Management |
|---|---|
| Kwangseok Yang | Seunghyun Choi |
| Jeongwon Choi | Hyoseok Choi |
Model Details
Base Model
Trained On
- OS: Ubuntu 22.04
- GPU: H100 80GB 4ea
- transformers: v4.36.2
Instruction format
It follows ChatML format.
E.g.
text = """\
<|im_start|>system
당신은 사람들이 정보를 찾을 수 있도록 도와주는 인공지능 비서입니다.<|im_end|>
<|im_start|>user
대한민국의 수도는 어디야?<|im_end|>
<|im_start|>assistant
대한민국의 수도는 서울입니다.<|im_end|>
<|im_start|>user
서울 인구는 총 몇 명이야?<|im_end|>
<|im_start|>assistant
"""
Model Benchmark
Ko LM Eval Harness
| Task | 0-shot | 5-shot | 10-shot | 50-shot |
|---|---|---|---|---|
| kobest_boolq | 0.920118 | 0.92442 | 0.929443 | 0.927317 |
| kobest_copa | 0.727263 | 0.778936 | 0.804812 | 0.815761 |
| kobest_hellaswag | 0.433039 | 0.465922 | 0.459741 | 0.471022 |
| kobest_sentineg | 0.764909 | 0.93946 | 0.937002 | 0.931962 |
| Average | 0.711332 | 0.777185 | 0.78275 | 0.786516 |
Ko-LLM-Leaderboard
| Average | Ko-ARC | Ko-HellaSwag | Ko-MMLU | Ko-TruthfulQA | Ko-CommonGen V2 |
|---|---|---|---|---|---|
| 59.22 | 53.84 | 67.9 | 52.37 | 64.6 | 57.38 |
Implementation Code
This model contains the chat_template instruction format.
You can use the code below.
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto
model = AutoModelForCausalLM.from_pretrained("Edentns/DataVortexS-10.7B-dpo-v1.6")
tokenizer = AutoTokenizer.from_pretrained("Edentns/DataVortexS-10.7B-dpo-v1.6")
messages = [
{"role": "system", "content": "당신은 사람들이 정보를 찾을 수 있도록 도와주는 인공지능 비서입니다."},
{"role": "user", "content": "대한민국의 수도는 어디야?"},
{"role": "assistant", "content": "대한민국의 수도는 서울입니다."},
{"role": "user", "content": "서울 인구는 총 몇 명이야?"}
]
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])
License
The model is licensed under the cc-by-nc-sa-4.0 license, which allows others to copy, modify, and share the work non-commercially, as long as they give appropriate credit and distribute any derivative works under the same license.