INDEX

Explanations

references to authorship or creation

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

prepositions followed by 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

featureID

-0.40

CED

-0.39

[*]

-0.38

vst

-0.38

eteria

-0.38

 fand

-0.38

 Bist

-0.38

hiers

-0.37

 churn

-0.36

 bist

-0.35

POSITIVE LOGITS

 BoxFit

0.61

\{\\

0.54

ArgsConstructor

0.54

migrationBuilder

0.53

providedIn

0.51

ftagPool

0.51

AntiForgeryToken

0.50

ReusableCell

0.48

AutoScale

0.48

LabelTagHelper

0.45

Activations Density 0.019%

references to authorship or creation

prepositions followed by names

No Comments

No Known Activations