INDEX

Explanations

Code

np_max-act · gemini-2.0-flash

This neuron detects module or component names appearing in package/import or namespace/use declarations (i.e. tokens naming code modules or bundles).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 FAMILY

-0.08

 gastr

-0.07

(phone

-0.07

س

-0.07

 Devon

-0.07

SYSTEM

-0.06

 واست

-0.06

Bd

-0.06

 explicit

-0.06

 more

-0.06

POSITIVE LOGITS

-effects

0.07

حداث

0.07

 يمكن

0.06

launch

0.06

وسط

0.06

UILayout

0.06

_MAT

0.06

hardt

0.06

/owl

0.06

산업

0.06

Activations Density 0.004%

Code

This neuron detects module or component names appearing in package/import or namespace/use declarations (i.e. tokens naming code modules or bundles).

No Comments

No Known Activations