INDEX

Explanations

time-related information

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Times or dates

np_acts-logits-general · gemini-2.0-flash

email metadata and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ⓧ

-0.60

 NgModule

-0.59

__(/*!

-0.58

 continuant

-0.58

########.

-0.56

oneof

-0.54

KURZBESCHREIBUNG

-0.50

EndContext

-0.50

AntiForgeryToken

-0.50

 szóci

-0.49

POSITIVE LOGITS

 myſelf

0.71

 itſelf

0.64

 themſelves

0.63

 himſelf

0.62

 Welles

0.62

 ſaid

0.61

Efq

0.60

 Monfieur

0.59

τια

0.59

 виправивши

0.59

Activations Density 0.144%

time-related information

Times or dates

email metadata and dates

No Known Activations