Search results

Minor refactoring on FeatureAblation methods

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1665

#1665 In meta-pytorch/captum;

· cyrjano opened on Nov 12, 2025

Simplify the Update Progress Code.

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1664

#1664 In meta-pytorch/captum;

· cyrjano opened on Nov 12, 2025

Move methods out of FeatureAblation and make them free methods.

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1663

#1663 In meta-pytorch/captum;

· cyrjano opened on Nov 11, 2025

Add documentation and clean some typing information for common file.

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1662

#1662 In meta-pytorch/captum;

· cyrjano opened on Nov 10, 2025

Attribution API refactor: Update LLM attr typing + minor naming

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1658

#1658 In meta-pytorch/captum;

· craymichael opened on Oct 25, 2025

Attribution API refactor: Introduce an optional agg/element-wise variance to LLM attribution results

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1659

#1659 In meta-pytorch/captum;

· craymichael opened on Oct 25, 2025

Attribution API refactor: Base LLMAttributionResult class + refactor

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1657

#1657 In meta-pytorch/captum;

· craymichael opened on Oct 25, 2025

Clean up independent feature masking in FeatureAblation

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1651

#1651 In meta-pytorch/captum;

· sarahtranfb opened on Sep 24, 2025

Clean up unused function in Occlusion

cla signed

fb-exported

meta-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1650

#1650 In meta-pytorch/captum;

· sarahtranfb opened on Sep 24, 2025

Update FeatureAblation to handle precision loss when baseline is more granular than input when cross tensor attribution is enabled

cla signed

fb-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1644

#1644 In meta-pytorch/captum;

· sarahtranfb opened on Sep 9, 2025

Default enable_cross_tensor_attribution to true

cla signed

fb-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1643

#1643 In meta-pytorch/captum;

· sarahtranfb opened on Sep 8, 2025

Resolve CUDA illegal memory access

cla signed

fb-exported

Status: Open (in progress).

meta-pytorch/captumnumber 1625

#1625 In meta-pytorch/captum;

· yangjin97 opened on Jul 8, 2025