INDEX

Explanations

phrases related to alert systems and safety warnings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

detect an

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bbene

-0.69

₁,

-0.57

 bekasi

-0.56

 Thesaurus

-0.56

 myſelf

-0.55

zköz

-0.55

 larmes

-0.55

oarece

-0.54

 Jefus

-0.54

 Theſe

-0.54

POSITIVE LOGITS

rungsseite

0.63

AlterField

0.59

styleable

0.57

новништво

0.56

 begrü

0.54

AutoScaleMode

0.53

 welcome

0.53

GEBURTSDATUM

0.53

 welcomed

0.52

argo

0.52

Activations Density 0.155%

phrases related to alert systems and safety warnings

detect an

No Comments

No Known Activations