INDEX

Explanations

references to death or "dead" in various contexts

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

"dead"

np_acts-logits-general · gemini-2.0-flash

dead, deadly, dead end

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

 pinulongan

-0.79

 estekak

-0.75

tocin

-0.73

 autorytatywna

-0.67

 متعلقه

-0.66

ummings

-0.65

zzleHttp

-0.64

@[

-0.64

baomidou

-0.64

complexContent

-0.63

POSITIVE LOGITS

 dead

1.16

Dead

1.04

 Dead

1.04

 DEAD

1.02

dead

0.98

DEAD

0.92

 muerta

0.69

 мерт

0.65

isDead

0.64

$_"

0.60

Activations Density 0.044%

references to death or "dead" in various contexts

"dead"

dead, deadly, dead end

No Comments

No Known Activations