INDEX

Explanations

elements related to rankings or top positions

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

images and managers

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

########.

-0.56

-0.50

 censiti

-0.50

 EconPapers

-0.44

ciatore

-0.43

bital

-0.43

 Ranks

-0.43

timewa

-0.40

 Cartier

-0.40

 Answering

-0.40

POSITIVE LOGITS

ngdoc

0.49

Rujuakan

0.47

 muchacha

0.41

참고

0.40

 Wikiseite

0.40

labelledby

0.39

 chi̍t

0.37

Détails

0.35

Rptr

0.35

 TextAppearance

0.35

Activations Density 0.010%

elements related to rankings or top positions

images and managers

No Comments

No Known Activations