INDEX

Explanations

words that indicate significant contributions or attributes of individuals in artistic or performance contexts

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

names and parts of names

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

 Normdatei

-0.75

yarnpkg

-0.73

featureID

-0.71

WithIOException

-0.68

 autorytatywna

-0.68

 propOrder

-0.66

 Roskov

-0.66

MessageOf

-0.66

 Wicidata

-0.65

ftagPool

-0.64

POSITIVE LOGITS

➞

0.48

IconModule

0.41

ViewImports

0.41

 Adamson

0.40

!*\

0.38

 Crespo

0.38

 artysty

0.36

an

0.36

tgz

0.35

ViewInit

0.35

Activations Density 0.320%

words that indicate significant contributions or attributes of individuals in artistic or performance contexts

names and parts of names

No Comments

No Known Activations