INDEX

Explanations

keywords and phrases indicating technical specifications or capabilities in a context related to technology and data processing

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

approaches him; softer than

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

 שוליים

-1.02

 فريبيس

-0.99

:✨

-0.99

Rohy

-0.93

 ujednoznacz

-0.92

 Италијани

-0.91

 مرئيه

-0.91



-0.90

GEBURTSDATUM

-0.89

+#+#

-0.89

POSITIVE LOGITS

is

0.30

land

0.26

hij

0.25

汤

0.25

 thấm

0.24

co

0.23

pan

0.22

dos

0.22

š

0.22

Activations Density 0.005%

keywords and phrases indicating technical specifications or capabilities in a context related to technology and data processing

approaches him; softer than

No Comments

No Known Activations