Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations directly in Jupyter notebooks explaining the behavior of Transformer-based language models (like GPT2, BERT, RoBERTA, T5, and T0).
Here I noticed that the tokenizer adds a partial token suffix instead of partial token prefix. Having a suffix instead of prefix is not configurable in the config.
Following your guide for identifying model configuration
returns:
Then
returns:
Here I noticed that the tokenizer adds a partial token suffix instead of partial token prefix. Having a suffix instead of prefix is not configurable in the config.