INDEX

Explanations

punctuation marks and specific formatting related to lists or categorizations

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

Text between parentheses

np_acts-logits-general · gemini-2.0-flash

parentheses with slashes

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ویکی‌پدی

-0.57

rrggbb

-0.54

setVerticalGroup

-0.53

httphttps

-0.50

تقاوى

-0.49

 ModelRenderer

-0.49

 OMITBAD

-0.48

findpost

-0.46

oa̍t

-0.45

MigrationBuilder

-0.45

POSITIVE LOGITS

 ebenfalls

0.54

 transgender

0.46

 Glen

0.46

 Billy

0.45

同じく

0.45

 whose

0.44

 szint

0.44

 Pharmacy

0.44

 South

0.44

 West

0.44

Activations Density 0.768%

punctuation marks and specific formatting related to lists or categorizations

Text between parentheses

parentheses with slashes

No Comments

No Known Activations