INDEX

Explanations

phrases related to manipulation and deception

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

tricked into believing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

the

-0.39

WebVitals

-0.38

usercontent

-0.33

dry

-0.31

 Rising

-0.31

 elkaar

-0.31

zeichnen

-0.31

 onderhoud

-0.30

 հղումներ

-0.30

 native

-0.30

POSITIVE LOGITS

 betweenstory

0.65

ValueStyle

0.61

 surla

0.59

AndEndTag

0.59

 terpaksa

0.58

 Forced

0.55

 tricked

0.55

Forced

0.55

 coer

0.54

forced

0.52

Activations Density 0.019%

phrases related to manipulation and deception

tricked into believing

No Comments

No Known Activations