p1746-lingua/ruRoberta-large-rucola-science

ruRoberta-large-rucola-science

This model is a fine-tuned version of p1746-lingua/ruRoberta-large-rucola on the dataset with errors in scientific texts. It predicts whether a given sentence contains errors.

Key Features

• Task: Binary classification (correct vs. contains errors)
• Training data: https://huggingface.co/datasets/p1746-lingua/correct_incorrect_sents (~2.4 labeled sentences)
• Max sequence length: 512 tokens
• Fine-tuning framework: PyTorch + Hugging Face transformers

Hyperparameters

Parameter	Value
Batch size	32
Learning rate	1e-5
Epochs	64
Warmup steps	100
Optimizer	adamw_bnb_8bit