INDEX

Explanations

instances of the word "here"

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

here are

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 незавершена

-0.64

 CURIAM

-0.56

 Мексичка

-0.56

 CallOverrides

-0.56

TagMode

-0.54

 ویکی‌پدی

-0.52

verwijspagina

-0.52

 Administrativna

-0.50

Geplaatst

-0.48

municipi

-0.48

POSITIVE LOGITS

are

0.82

is

0.72

 comes

0.62

 follows

0.48

we

0.48

 lies

0.43

 outlines

0.42

 هستیم

0.41

inafter

0.41

êm

0.41

Activations Density 0.014%

instances of the word "here"

here are

No Comments

No Known Activations