INDEX

Explanations

country names and related adjectives

oai_token-act-pair · gemini-2.0-flash

countries and nationalities

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

 initComponents

-0.80

رشف

-0.78

\{\\

-0.77

 nahilalakip

-0.76

enderror

-0.75

batore

-0.75

 للمعارف

-0.75

 يتيمه

-0.74

providedIn

-0.72

 الحره

-0.71

POSITIVE LOGITS

 light

0.42

 Light

0.40

瓜

0.40

vij

0.40

 Fibre

0.38

 bilin

0.38

 Commute

0.37

na

0.37

angi

0.36

Light

0.36

Activations Density 0.923%

country names and related adjectives

countries and nationalities

No Comments

No Known Activations