Simran Arora

upload new tasks

4

Adding three new datasets from the Based paper: FDA, SWDE, and SQUAD-Completion Paper reference: https://arxiv.org/abs/2402.18668

understanding context length behaviors

1

Hi, I have a quick question. What is the behavior of the harness if the input examples exceed the model's sequence length in long-document tasks (and how do few-shot examples...

[BUG]: AMD GEMM

3

### Bug description My issue was closed even though it does not seem resolved to me: https://github.com/modular/modular/issues/5503 ### Steps to reproduce I followed @raiseirql 's suggestion: ```bash modular/max/kernels/benchmarks/gpu# mojo bench_matmul.mojo...