INDEX

Explanations

references to liberalism or liberal viewpoints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

Personendaten

-0.72

 Roskov

-0.60

دانشنامهٔ

-0.49

 BoxFit

-0.49

SequentialGroup

-0.48

windowFixed

-0.47

 Nutri

-0.46

ump

-0.45

uey

-0.45

awtextra

-0.45

POSITIVE LOGITS

liberal

0.55

 liberalization

0.55

eseorang

0.54

 galinha

0.53

 liberals

0.51

 électroniques

0.50

 unlock

0.50

 desblo

0.49

 unlocks

0.49

 desbloquear

0.49

Activations Density 0.336%

references to liberalism or liberal viewpoints

No Comments

No Known Activations