INDEX

Explanations

terms related to "fraud" or "fraudulent" activities

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

Tokens starting with "Fra" or "Fri"

np_acts-logits-general · gemini-2.0-flash

words starting with fr or fra

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

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

 aigu

-0.51

verwijspagina

-0.48

 Secret

-0.48

 syscall

-0.47

+</

-0.47

 secret

-0.47

 menyes

-0.47

 Deum

-0.46

WithMany

-0.46

 feit

-0.45

POSITIVE LOGITS

Fra

0.67

fra

0.65

 disambiguazione

0.64

/*

0.62

 GenerationType

0.62

Fra

0.61

TagMode

0.58

Frat

0.57

Santo

0.56

GraphicsUnit

0.56

Activations Density 0.121%

terms related to "fraud" or "fraudulent" activities

Tokens starting with "Fra" or "Fri"

words starting with fr or fra

No Comments

No Known Activations