INDEX

Explanations

prepositions and phrases indicating relationships or roles

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

code or error

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

 تضيفلها

-0.46

 chargés

-0.45

Tembelea

-0.45

 chargé

-0.44



-0.44

 ſou

-0.43

 Magyar

-0.43

 ſever

-0.43

 ſeveral

-0.42

 faſt

-0.42

POSITIVE LOGITS

RectangleBorder

0.48

 dAtA

0.43

 AttributeSet

0.43

 tekem

0.41

 stesse

0.41

 kife

0.41

 farlo

0.41

Tikang

0.40

 resourceCulture

0.39

♥♥

0.39

Activations Density 0.032%

prepositions and phrases indicating relationships or roles

code or error

No Comments

No Known Activations