AnonymousCS/bert-large-uncased-Twitter-toxicity
Full fine-tune vs google-bert/bert-large-uncased. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 3/3, Attention 192/192, other 98/98
Full fine-tune vs google-bert/bert-large-uncased. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Embeddings 3/3, Attention 192/192, other 98/98