INDEX

Explanations

negative statements about public figures and accusations of wrongdoing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 rät

-0.58

 parad

-0.55

inetti

-0.50

 mobilité

-0.49

 Negoti

-0.49

 optimis

-0.49

IntoConstraints

-0.48

Intu

-0.48

coltà

-0.48

unzel

-0.47

POSITIVE LOGITS

 attacks

1.30

 insults

1.19

 slander

1.18

 attack

1.14

 attacking

1.11

 insulting

1.08

 Attacks

1.05

Attacks

1.04

 accusations

1.03

 insult

1.02

Activations Density 0.578%

negative statements about public figures and accusations of wrongdoing

No Comments

No Known Activations