CoolFace
Modelpublic

deepvk/plato-9b

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
9likes17downloads
Model Card

plato-9b

plato-9b is a fine-tuned version of the `google/gemma-2-9b-it` model for generating responses in the Russian language. This 9-billion parameter model excels at conversational tasks, offering rich contextual understanding and fine-grained results.

Usage

To use plato-9b with the transformers library:

python
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepvk/plato-9b")
model = AutoModelForCausalLM.from_pretrained("deepvk/plato-9b")

input_text = "Что стоит посетить в России?"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids

output = model.generate(input_ids, max_length=150, do_sample=True, temperature=0.7)
response = tokenizer.decode(output[0], skip_special_tokens=True)
print(response)
# Что стоит посетить в России?
# 1. Красная площадь и Кремль в Москве
# 2. Эрмитаж в Санкт-Петербурге
# 3. Байкал
# 4. Соловецкие острова
# 5. Камчатка и её вулканы
# 6. Золотое Кольцо
# 7. Казанский Кремль
# 8. Алтай
# 9. Астраханская область и Волго-Донской канал
# 10. Кавказские горы и Черноморское побережье
# 
# Каждое из этих мест предлагает уникальные культурные, исторические и природные достопримечательности,
# которые делают Россию столь удивительной и разнообразной страной.

Dataset

We applied both Supervised Fine-Tuning (SFT) and Preference Optimization (PO). For SFT, we used an 8B token instruction dataset, with 4B tokens consisting of dialogues and the rest covering math, biology, chemistry, code, and general knowledge. The PO dataset contains 200M tokens featuring common knowledge instructions. We trained on both datasets for several epochs.

Evaluation

To evaluate, we applied LLM-as-a-judge approach on academic tasks. Specifically, we used arena-general-ru and arena-hard-ru with gpt4o judge and gpt4o-mini baseline.

arena-general-ru

ModelScoreScore w/ SC
gpt-4o-2024-11-2081.87 (-2.04, +1.81)78.42 (-2.39, +2.33)
gpt-4o-mini-2024-07-1850.00 (-0.00, +0.00)50.00 (-0.00, +0.00)
deepvk/plato-9b41.27 (-2.18, +2.24)32.13 (-1.97, +2.05)
t-tech/T-lite-it-1.038.52 (-2.04, +2.98)30.38 (-1.90, +3.15)
google/gemma-2-9b-it27.46 (-2.06, +1.74)25.80 (-2.09, +1.98)
Qwen/Qwen2.5-7B-Instruct24.60 (-2.36, +2.38)23.67 (-2.36, +2.28)
IlyaGusev/saiga_gemma2_9b17.83 (-1.95, +1.66)18.46 (-2.22, +1.69)

arena-hard-ru

ModelScoreScore w/ SC
gpt-4o-2024-11-2085.70 (-1.45, +1.38)80.19 (-1.99, +2.04)
gpt-4o-mini-2024-07-1850.00 (-0.00, +0.00)50.00 (-0.00, +0.00)
t-tech/T-lite-it-1.034.80 (-1.98, +2.38)26.99 (-1.74, +2.67)
deepvk/plato-9b31.81 (-1.92, +1.90)24.25 (-1.71, +1.84)
Qwen/Qwen2.5-7B-Instruct20.84 (-1.99, +1.67)17.70 (-1.63, +1.68)
google/gemma-2-9b-it12.98 (-1.36, +1.57)12.97 (-1.46, +1.69)
IlyaGusev/saiga_gemma2_9b9.72 (-1.34, +1.50)10.64 (-1.40, +1.78)

Citation

Both authors contribute equally, order is alphabetical.

@misc{deepvk2024plato-9b,
    title={plato-9b},
    author={Eliseev, Anton and Semin, Kirill},
    url={https://huggingface.co/deepvk/plato-9b},
    publisher={Hugging Face}
    year={2025},
}