Edit model card

This model was derived from the bert-base-uncased checkpoint by replacing the GELU with ReLU activation function and further pre-training through several iterations to adapt it to the change of the activation function.

Downloads last month
120
Safetensors
Model size
110M params
Tensor type
I64
·
F32
·

Datasets used to train mpiorczynski/relu-bert-base-uncased