INDEX

Explanations

mentions of document credits or acknowledgments

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

copyright, credits, license

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

 climá

-0.44

 Forsyth

-0.36

mäß

-0.36

tyfik

-0.35

 contact

-0.35

derung

-0.35

ioctl

-0.35

 positive

-0.35

 nonnegative

-0.35

tirol

-0.34

POSITIVE LOGITS

RTEX

0.56

 defaultstate

0.56

ConstraintMaker

0.52

 initComponents

0.51

 CommonModule

0.51

DockStyle

0.51

 للاسماء

0.50

 FetchType

0.49

 Chwiliwch

0.49

Rohy

0.48

Activations Density 0.000%

mentions of document credits or acknowledgments

copyright, credits, license

No Comments

No Known Activations