Murtadha Ahmed

Results 3 issues of Murtadha Ahmed

``` causal_mask = self.bias[:, :, key_length - query_length : key_length, :key_length] ``` but in Structured Prompting the key_length exceeds the max_positions. How to address this issue. Thank you.

First of all, we highly appreciate your efforts implementing Numbskull library, which is very helpful and practical. However, we encountered some difficulties to deeply understand some points in the code...