INDEX

Explanations

names of individuals in various contexts

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

names followed by verbs

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

IContainer

-0.46

:✨

-0.46

featureID

-0.43

Probably

-0.42

XmlAccessType

-0.41

ursa

-0.41

Tracce

-0.41

ise

-0.40

Dor

-0.40

الع

-0.40

POSITIVE LOGITS

Manbalar

0.52

Javadoc

0.45

 Houſe

0.44

konomi

0.43

NewUrlParser

0.43

PDOException

0.43

 calendriers

0.43

RegistryLite

0.42

Bibliograf

0.42

ніципа

0.42

Activations Density 0.031%

names of individuals in various contexts

names followed by verbs

No Comments

No Known Activations