INDEX

Explanations

instances of the word "attack" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GEBURTSDATUM

-0.72

{~

-0.57

СТВА

-0.56

activités

-0.55

 فريبيس

-0.54

icali

-0.54

frac

-0.54

("$.

-0.53

figsize

-0.53

ddots

-0.53

POSITIVE LOGITS

 Monfieur

0.84

 Attacks

0.74

Attacks

0.71

WireFormatLite

0.65

 Taktlose

0.65

 MessageBoxIcon

0.65

 ATTACK

0.64

attacks

0.63

ATTACK

0.63

 attacks

0.63

Activations Density 0.162%

instances of the word "attack" in various contexts

No Comments

No Known Activations