INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron activates on code patterns inspecting the user-agent string—e.g. calls to navigator.userAgent (or server-side user-agent headers) and indexOf/regex checks for browser or OS identifiers.

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

 stringWithFormat

-0.07

्वप

-0.07

ullah

-0.07

 sched

-0.07

spd

-0.06

model

-0.06

Hostname

-0.06

 autop

-0.06

esinde

-0.06

POSITIVE LOGITS

 Jeans

0.07

 shore

0.06

�

0.06

claration

0.06

_NET

0.06

(Y

0.06

GIS

0.06

<S

0.06

 صنعتی

0.06

Activations Density 0.005%

code

The neuron activates on code patterns inspecting the user-agent string—e.g. calls to navigator.userAgent (or server-side user-agent headers) and indexOf/regex checks for browser or OS identifiers.

No Comments

No Known Activations