Model Card for Model ID

BERT model trained on Bulgarian literature, Web, and other datasets - uncased.

Model Details

657M parameter BERT model trained on 19B (23B depending on tokenization) tokens for 5 epochs with Masked Language Modelling objective.

Tokenizer vocabulary size is 50176.
Model hidden dimension is 1024.
Feed-Forward dimension is 4096.
Hidden layer count is 48.
Developed by: Artificial Inteligence and Language Technologies Department at Institute of Information and Communication Technologies - Bulgarian Academy of Sciences.
Funded by: The model is pretrained within the CLaDA-BG: National Interdisciplinary Research E-Infrastructure for Bulgarian Language and Cultural heritage - member of the pan-European research consortia CLARIN-ERIC & DARIAH-ERIC, funded by the Ministry of Education and Science of Bulgaria (support for the Bulgarian National Roadmap for Research Infrastructure). The training was performed at the supercomputer HEMUS at IICT-BAS, part of the RIs of the CoE on Informatics and ICT, financed by the OP SESG (2014–2020), and co-financed by the European Union through the ESIF.
Model type: BERT
Language(s) (NLP): Bulgarian.
License: MIT

Uses

The model is intended to be used as a base model for fine-tuning tasks in NLP.

Direct Use

>>> from transformers import (
>>>     PreTrainedTokenizerFast,
>>>     BertForMaskedLM,
>>>     pipeline 
>>> )

>>> model = BertForMaskedLM.from_pretrained('AIaLT-IICT/bert_bg_lit_web_extra_large_uncased')
>>> tokenizer = PreTrainedTokenizerFast.from_pretrained('AIaLT-IICT/bert_bg_lit_web_extra_large_uncased')

>>> fill_mask = pipeline(
>>>     "fill-mask",
>>>     model=model,
>>>     tokenizer=tokenizer
>>> )


>>> fill_mask("Заради 3 завода няма да [MASK] нито есенниците неподхранени, нито зърното да поскъпне заради тях.")

[{'score': 0.3125033974647522,
  'token': 19273,
  'token_str': 'останат',
  'sequence': 'заради 3 завода няма да останат нито есенниците неподхранени, нито зърното да поскъпне заради тях.'},
 {'score': 0.24685998260974884,
  'token': 15953,
  'token_str': 'остави',
  'sequence': 'заради 3 завода няма да остави нито есенниците неподхранени, нито зърното да поскъпне заради тях.'},
 {'score': 0.2007855325937271,
  'token': 27509,
  'token_str': 'оставим',
  'sequence': 'заради 3 завода няма да оставим нито есенниците неподхранени, нито зърното да поскъпне заради тях.'},
 {'score': 0.10415761172771454,
  'token': 24533,
  'token_str': 'оставят',
  'sequence': 'заради 3 завода няма да оставят нито есенниците неподхранени, нито зърното да поскъпне заради тях.'},
 {'score': 0.016328580677509308,
  'token': 13989,
  'token_str': 'има',
  'sequence': 'заради 3 завода няма да има нито есенниците неподхранени, нито зърното да поскъпне заради тях.'}]

Out-of-Scope Use

The model is not trained on Next Sentence prediction so the [CLS] token embedding will not be useful out of the box. If you want to use the model for Sequence classification it is recommended to fine-tune it.

Recommendations

It is recommended to use the model for Token Classification and Sequence classification fine-tuning tasks. The model can be used within SentenceTransformers framework for producing embeddings.

Training Details

Training Data

Trained on 19B tokens mainly consisting of:

uonlp/CulturaX
MaCoCu-bg 2.0
Literature
others

Training Procedure

Trained with Masked Language Modelling with 20% masks for 5 epochs with tf32 mixed precision, 512 tokens context and batch size of 256*512 tokens.

Evaluation

The model is evaluated on the Masked Language Modelling objective on test split with 20% random masked tokens. It achieves test loss of 1.09 and test accuracy of 75.55%

Model Card Authors

Nikolay Paev, Kiril Simov

Model Card Contact

[email protected]

AIaLT-IICT
/

bert_bg_lit_web_extra_large_uncased