INDEX

Explanations

uses of deception or pretense

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

pretend, pretence, feign

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Rector

-0.51

ISR

-0.50

榄

-0.50

MDC

-0.48

MDC

-0.46

cemic

-0.46

 Fluss

-0.46

 gekomen

-0.45

lccn

-0.45

CHS

-0.45

POSITIVE LOGITS

 pretend

0.86

pret

0.78

 pretends

0.77

 pretended

0.74

Pret

0.71

 Pret

0.70

 pretending

0.65

Pseud

0.62

 PRET

0.61

 pret

0.59

Activations Density 0.006%

uses of deception or pretense

pretend, pretence, feign

No Comments

No Known Activations