Adding `safetensors` variant of this model (#1)

upload flax model

allow flax

add missing char tokens to vocab (with embeddings close to [UNK])

Migrate model card from transformers-repo

Update tokenizer configuration

Add tokenizer configuration

Update tokenizer.json

Update config.json

Update config.json

Update config.json

Update tokenizer_config.json

Update tf_model.h5

Update model.ckpt.data-00000-of-00001

Update pretraining_output_eval_results.txt

Update model.ckpt.index

Update pytorch_model.bin

Update vocab.txt

Update config.json

Update checkpoint

Update model.ckpt.meta

initial commit