INDEX

Explanations

location-related terms and phrases

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

location and web addresses

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

uttgart

-0.54

 scott

-0.51

canestro

-0.50

 inversion

-0.47

RefNanny

-0.46

titleMargin

-0.46

XmlEnum

-0.45

ModelState

-0.44

 schal

-0.44

 caption

-0.44

POSITIVE LOGITS

 Wohn

0.47

bootstrapcdn

0.47

__':

0.44

__':

0.43

 resides

0.43

__":

0.42

 Address

0.40

__":

0.40

 woning

0.40

 weiße

0.39

Activations Density 0.000%

location-related terms and phrases

location and web addresses

No Comments

No Known Activations