INDEX

Explanations

phrases that indicate permission or allowance

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

allowing actions

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

 fizer

-0.36

 gloire

-0.34

 Région

-0.33

 rapides

-0.33

 cadeia

-0.31

 utama

-0.31

 dificuldades

-0.31

 estés

-0.31

 measures

-0.30

ostante

-0.30

POSITIVE LOGITS

 мәкал

0.69

parsedMessage

0.65

IntoConstraints

0.65

Билгалдахарш

0.65

 Allowed

0.59

 permission

0.58

exitRule

0.57

 GenerationType

0.57

ſſung

0.57

Activations Density 0.050%

phrases that indicate permission or allowance

allowing actions

No Comments

No Known Activations