INDEX

Explanations

references to individuals and their associated actions or statuses

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

names, titles, and pronouns

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

 ComVisible

-0.38

 financeiro

-0.36

 originais

-0.34

 civilización

-0.33

 diário

-0.33

 confirmación

-0.32

 eixo

-0.31

 abstrato

-0.31

 Übere

-0.31

 auténtico

-0.30

POSITIVE LOGITS

contentLoaded

0.53

ceğine

0.51

icitis

0.50

TypedDataSet

0.49

外部リンク

0.48

UnsafeEnabled

0.47

surate

0.47

Mitglieder

0.47

ocumented

0.46

partic

0.46

Activations Density 0.179%

references to individuals and their associated actions or statuses

names, titles, and pronouns

No Comments

No Known Activations