Anonymous Fused Sequence Operator with Hidden HVP Checks¶
- Task ID:
computer_science.causal_attention - Domain:
computer_science - Subdomain:
deep_learning_numerics - Status:
test - Tags:
anonymous_operator,sequence_modeling,numerical_stability,gradient_check,hessian_vector_product
Public Summary¶
This page is generated from task metadata and selected public-safe excerpts.
Example B1 Prompt Excerpt¶
# Anonymous Sequence Operator
## Goal
Implement the specified NumPy sequence operator, its reverse-mode gradients, and a Hessian-vector product. Save the compact public outputs under `results/`.
Do not use PyTorch or import `torch` in `analysis.py`.
## Anonymous Input Contract
The `data/` folder contains:
- `x.npy`, `segment_ids.npy`, `sensitivity.npy`
- `matrix_0.npy` ... `matrix_4.npy`
Notes¶
- This page is a generated site artifact.
- Higher-level prompt details and internal benchmark specifics may remain intentionally undisclosed.