INDEX

Explanations

indulge

np_max-act · gemini-2.0-flash

The neuron specifically detects occurrences of the word “indulge” (and its close morphological variants like “indulgence,” “indulged,” etc.).

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

应

-0.07

 chiến

-0.07

 dương

-0.07

 Sector

-0.06

RenderTarget

-0.06

 обнаруж

-0.06

general

-0.06

 viable

-0.06

 black

-0.06

 counts

-0.06

POSITIVE LOGITS

 indul

0.10

 indulge

0.10

 treat

0.08

-eslint

0.07

 platí

0.07

od

0.07

”↵

0.07

datable

0.07

 Treat

0.07

φο

0.07

Activations Density 0.007%

indulge

The neuron specifically detects occurrences of the word “indulge” (and its close morphological variants like “indulgence,” “indulged,” etc.).

No Comments

No Known Activations