INDEX

Explanations

phrases indicating confrontation or conflict avoidance

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

before or implying confrontation

np_acts-logits-general · gemini-2.0-flash

provocation or dared

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

databind

-0.57

DIPSETTING

-0.55

ngdoc

-0.51

addContainerGap

-0.51

 canlynol

-0.50

XmlAccessType

-0.49

íncia

-0.49

 setTotal

-0.48

errHandler

-0.47

 geprägt

-0.47

POSITIVE LOGITS

 provocation

1.50

 provoc

1.34

 provoke

1.33

 provoking

1.25

 provoked

1.20

 provocative

1.17

 provo

1.03

 dared

0.92

provoking

0.92

 provoca

0.91

Activations Density 0.227%

phrases indicating confrontation or conflict avoidance

before or implying confrontation

provocation or dared

No Comments

No Known Activations