• Yih-Dar's avatar
    Fix mixed precision issue in TF DistilBert (#6915) · a75e3198
    Yih-Dar authored
    * Remove hard-coded uses of float32 to fix mixed precision use in TF Distilbert
    
    * fix style
    
    * fix gelu dtype issue in TF Distilbert
    
    * fix numeric overflow while using half precision
    a75e3198
modeling_tf_distilbert.py 46.2 KB