INDEX

Explanations

interactions between people and their surroundings

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

appearances and visibility

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

 surla

-0.48

="@+

-0.43

 locu

-0.40

 εμπ

-0.40

 يتيمه

-0.38

 "../../../../

-0.38

WithIOException

-0.38

要做

-0.38

luia

-0.37

NIF

-0.36

POSITIVE LOGITS

 visible

0.49

 visibles

0.44

twimg

0.44

orianCalendar

0.44

messageInfo

0.43

 Meksiku

0.43

 näky

0.41

everywhere

0.41

 للمعارف

0.41

Appear

0.41

Activations Density 0.104%

interactions between people and their surroundings

appearances and visibility

No Comments

No Known Activations