This is an extension of ELECTRA small model, trained on the OpenWebText corpus. The training task (discriminative LM / replaced-token-detection) can be generalized to any transformer type. Here, we train an ALBERT model under the same scheme.

(figure from Clark et al. 2020)

ELECTRA uses discriminative LM / replaced-token-detection for pretraining. This involves a generator (a Masked LM model) creating examples for a discriminator to classify as original or replaced for each token.

The generator generalizes to any *ForMaskedLM model and the discriminator could be any *ForTokenClassification model. Therefore, we can extend the task to ALBERT models, not just BERT as in the original paper.

NOTE: this ALBERT model uses a BERT WordPiece tokenizer.

The pytorch module that implements this task is available here.

Further implementation information here, and here is the script that created this model.

This specific model was trained with the following params: