INDEX

Explanations

place names, especially smaller towns and regions

oai_token-act-pair · gemini-2.0-flash

dova

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emf

-0.62

RepeatedField

-0.58

ensio

-0.57

enz

-0.55

egy

-0.54

ents

-0.52

eload

-0.52

enski

-0.52

ens

-0.52

escent

-0.52

POSITIVE LOGITS

verwijspagina

0.52

IntoConstraints

0.50

utia

0.47

хьтан

0.46

 Normdatei

0.46

Unmarshaller

0.44

 AssemblyProduct

0.43

ⓘ

0.43

Uninitialized

0.42

ovací

0.42

Activations Density 0.854%

place names, especially smaller towns and regions

dova

No Comments

No Known Activations