BERT_vocab: Check if mask words are in the model vocabulary.

In FMAT: The Fill-Mask Association Test

Check if mask words are in the model vocabulary.

Description

Check if mask words are in the model vocabulary.

Usage

BERT_vocab(
  models,
  mask.words,
  add.tokens = FALSE,
  add.verbose = FALSE,
  weight.decay = 1
)

Arguments

models	A character vector of model names at HuggingFace.
mask.words	Option words filling in the mask.
add.tokens	Add new tokens (for out-of-vocabulary words or phrases) to model vocabulary? Defaults to FALSE. Default method of producing the new token embeddings is computing the (equally weighted) average subword token embeddings. To change the weights of different subwords, specify weight.decay. It just adds tokens temporarily without changing the raw model file.
add.verbose	Print subwords of each new token? Defaults to FALSE.
weight.decay	Decay factor of relative importance of multiple subwords. Defaults to 1 (see weight_decay() for computational details). A smaller decay value would give greater weight to the former subwords than to the latter subwords. The i-th subword has raw weight = decay ^ i. decay = 1: all subwords are equally important (default) 0 < decay < 1: first subwords are more important decay > 1: last subwords are more important For example, decay = 0.5 would give 0.5 and 0.25 (with normalized weights 0.667 and 0.333) to two subwords (e.g., "individualism" = 0.667 "individual" + 0.333 "##ism").

Value

A data.table of model name, mask word, real token (replaced if out of vocabulary), and token id (0~N).

See Also

BERT_download()

BERT_info()

FMAT_run()

Examples

## Not run: 
models = c("bert-base-uncased", "bert-base-cased")
BERT_info(models)

BERT_vocab(models, c("bruce", "Bruce"))

BERT_vocab(models, 2020:2025)  # some are out-of-vocabulary
BERT_vocab(models, 2020:2025, add.tokens=TRUE)  # add vocab

BERT_vocab(models,
           c("individualism", "artificial intelligence"),
           add.tokens=TRUE)

## End(Not run)

FMAT documentation built on Jan. 12, 2026, 9:08 a.m.