INDEX

Explanations

patterns in structured data and key-value associations

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

Code-related terms or functions

np_acts-logits-general · gemini-2.0-flash

names of universities

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

 strto

-0.63

WriteTagHelper

-0.59

 мәкал

-0.55

InvalidProtocol

-0.55

 correctes

-0.55

省市镇

-0.55

MessageTagHelper

-0.55

UnsafeEnabled

-0.54

ToRefresh

-0.54

lievo

-0.53

POSITIVE LOGITS

GEBURTSDATUM

0.79

IBOutlet

0.53

 Мексичка

0.51

canActivate

0.48

 terce

0.46

Ves

0.46

 Akismet

0.46

 DbSet

0.45

verwijspagina

0.43

いき

0.42

Activations Density 0.051%

patterns in structured data and key-value associations

Code-related terms or functions

names of universities

No Comments

No Known Activations