First is Better Than Last for Language Data Influence. (arXiv:2202.11844v3 [cs.LG] UPDATED)

The ability to identify influential training examples enables us to debug
training data and explain model behavior. Existing techniques to do so are
based on the flow of training data influence through the model parameters. For
large models in NLP applications, it is often computationally infeasible to
study this flow through all model parameters, therefore techniques usually pick
the last layer of weights. However, we observe that since the activation
connected to the last layer of weights contains "shared logic", …

arxiv data influence language language data