INDEX

Explanations

actions related to prevention and protection

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

actions to improve or change states

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

BagConstraints

-0.59

艋

-0.53

 Chwiliwch

-0.50

alb

-0.49

oredCriteria

-0.49

 trif

-0.47

цезда

-0.47

 Bayou

-0.47

 anth

-0.47

 actionPerformed

-0.47

POSITIVE LOGITS

 başına

0.41

 مشين

0.41

 kurtka

0.38

 kobieta

0.37

 olacak

0.36

 ilustracja

0.36

 lumineuse

0.36

 féminine

0.36

 auprès

0.36

 Inscrivez

0.36

Activations Density 0.074%

actions related to prevention and protection

actions to improve or change states

No Comments

No Known Activations