INDEX

Explanations

words and phrases related to political violence and oppression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

لاق

-0.46

 meren

-0.45

 concernés

-0.44

 bertahan

-0.43

 cuadrados

-0.43

 timeType

-0.43

 beraber

-0.42

 efektif

-0.42

plaatst

-0.42

 conocen

-0.42

POSITIVE LOGITS

 innocent

1.24

innocent

1.06

 defen

1.05

 innoc

1.01

 unarmed

0.92

 inocente

0.91

 helpless

0.89

 civilians

0.88

 vulnerable

0.87

 harmless

0.86

Activations Density 0.700%

words and phrases related to political violence and oppression

No Comments

No Known Activations