INDEX

Explanations

prominent punctuation marks and formatting elements

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

legal citations and formal language

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

featureID

-0.69

contentLoaded

-0.64

 propOrder

-0.62

GEBURTSDATUM

-0.62

ParallelGroup

-0.59

 unknownFields

-0.59

webElementXpaths

-0.59

ंदीखरीदारी

-0.57

Personensuche

-0.56

InjectAttribute

-0.56

POSITIVE LOGITS

Kaieteur

0.36

 snippetHide

0.36

Justice

0.35

※

0.34

 зале

0.33

LikeLike

0.33

 vrais

0.33

providedIn

0.31

Replication

0.31

 plufieurs

0.31

Activations Density 0.001%

prominent punctuation marks and formatting elements

legal citations and formal language

No Comments

No Known Activations