README.md

metadata

tags:
  - text-generation
license: cc-by-nc-sa-4.0
language:
  - ko
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
pipeline_tag: text-generation
datasets:
  - beomi/KoAlpaca-v1.1a
  - jojo0217/korean_rlhf_dataset
  - kyujinpy/OpenOrca-KO
  - nlpai-lab/kullm-v2
widget:
  - text: |
      <|system|>
      You are a chatbot who answers User's questions.
      <|user|>
      대한민국의 수도는 어디야?
      <|assistant|>

DataVortexTL-1.1B-v0.1

Model Details

Base Model

TinyLlama/TinyLlama-1.1B-Chat-v1.0

Trained On

• OS: Ubuntu 20.04
• GPU: H100 80GB 1ea
• transformers: v4.36.2

Dataset

• beomi/KoAlpaca-v1.1a
• jojo0217/korean_rlhf_dataset
• kyujinpy/OpenOrca-KO
• nlpai-lab/kullm-v2

Instruction format

It follows TinyLlama format.

E.g.

text = """\
<|system|>
당신은 사람들이 정보를 찾을 수 있도록 도와주는 인공지능 비서입니다.</s>
<|user|>
대한민국의 수도는 어디야?</s>
<|assistant|>
대한민국의 수도는 서울입니다.</s>
<|user|>
서울 인구는 총 몇 명이야?</s>
"""

Model Benchmark

Ko LM Eval Harness

Task	0-shot	5-shot	10-shot	50-shot
kobest_boolq	0.334282	0.516446	0.500478	0.498941
kobest_copa	0.515061	0.504321	0.492927	0.50809
kobest_hellaswag	0.36253	0.357733	0.355873	0.376502
kobest_sentineg	0.481146	0.657411	0.687417	0.635703
Average	0.42325475	0.50897775	0.50917375	0.504809

Ko-LLM-Leaderboard

Average	Ko-ARC	Ko-HellaSwag	Ko-MMLU	Ko-TruthfulQA	Ko-CommonGen V2
31.5	25.26	33.53	24.56	43.34	30.81

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/DataVortexTL-1.1B-v0.1")
tokenizer = AutoTokenizer.from_pretrained("Edentns/DataVortexTL-1.1B-v0.1")

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.