INDEX

Explanations

terms related to privilege and social identity

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

Informal, negative sentiment

np_acts-logits-general · gemini-2.0-flash

profanity and negative sentiment

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 muß

-0.90

・・・・・

-0.80

 mußte

-0.77

 Usted

-0.77

 sculptured

-0.76

 läßt

-0.76

 skall

-0.75

 Moslem

-0.73

omiast

-0.73

 egli

-0.73

POSITIVE LOGITS

 shitty

1.43

 fucking

1.41

 weirdly

1.36

idk

1.32

 fucked

1.31

 FUCKING

1.28

 goddamn

1.24

tbh

1.23

 fuck

1.20

 shit

1.19

Activations Density 3.595%

terms related to privilege and social identity

Informal, negative sentiment

profanity and negative sentiment

No Comments

No Known Activations