INDEX

Explanations

proper nouns, particularly names

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

names, people, or affiliations

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

 våre

-0.34

 myndig

-0.33

 hendes

-0.33

 skrift

-0.32

 ferien

-0.32

 كورة

-0.32

 piele

-0.31

 dumne

-0.31

 nicio

-0.31

 betekenis

-0.31

POSITIVE LOGITS

IUrlHelper

0.71

expandindo

0.71

awtextra

0.68

NewUrlParser

0.65

ſelben

0.65

 CanadaChoose

0.65

LayoutStyle

0.64

 otomatig

0.63

Портали

0.62

səhifə

0.62

Activations Density 0.076%

proper nouns, particularly names

names, people, or affiliations

No Comments

No Known Activations