INDEX

Explanations

security vulnerabilities

np_max-act · gemini-2.0-flash

statements about users or anyone easily accessing, reading, or extracting files/code without authorization, highlighting client-side security weaknesses and decompilation risks.

oai_token-act-pair · gpt-5 Triggered by @vetterc0

New Auto-Interp

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Inline

-0.08

 music

-0.07

��

-0.07

nder

-0.07

 Adapt

-0.07

��

-0.07

 CONVERT

-0.07

 nurse

-0.06

pf

-0.06

 desarrollo

-0.06

POSITIVE LOGITS

trg

0.08

桕

0.08

Ctrls

0.08

阄

0.07

fsp

0.07

年龄段

0.07

检察

0.07

VIP

0.07

.fromRGBO

0.07

(QWidget

0.07

Activations Density 0.052%

security vulnerabilities

statements about users or anyone easily accessing, reading, or extracting files/code without authorization, highlighting client-side security weaknesses and decompilation risks.

No Comments

No Known Activations