INDEX

Explanations

keywords and phrases related to questioning and challenging established ideas or norms

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

challenge or question existence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

MLLoader

-0.44

 חיצוניים

-0.44

ThroughAttribute

-0.42

∮

-0.42

PostInfinity

-0.42

 ویکی‌پدی

-0.41

`}`,

-0.41

wra

-0.40

izoen

-0.40

夜景

-0.39

POSITIVE LOGITS

canza

0.41

 passions

0.40

CrossRef

0.40

OGND

0.39

 exploration

0.39

 explorations

0.39

 tradisi

0.36

asaran

0.35

PerformLayout

0.35

 pytania

0.35

Activations Density 0.044%

keywords and phrases related to questioning and challenging established ideas or norms

challenge or question existence

No Comments

No Known Activations