INDEX

Explanations

phrases that indicate a quantity or measurement

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

about preceding quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 présence

-0.49

 isolado

-0.43

 זאת

-0.43

 presence

-0.42

eden

-0.41

rasında

-0.40

 זו

-0.40

Acceso

-0.40

 bezpiecze

-0.39

XmlIgnore

-0.39

POSITIVE LOGITS

 ungefähr

0.76

 about

0.76

about

0.72

 ungefär

0.68

 ongeveer

0.67

 ABOUT

0.65

 roughly

0.63

 propOrder

0.63

 About

0.63

ftagPool

0.63

Activations Density 0.024%

phrases that indicate a quantity or measurement

about preceding quantities

No Comments

No Known Activations