INDEX

Explanations

specific lines of code or textual references within a document

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

line or sentence

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

 Europe

-0.30

 terület

-0.30

setPadding

-0.29

paddingTop

-0.28

tage

-0.28

 güç

-0.27

 Islands

-0.27

ing

-0.26

app

-0.26

losen

-0.26

POSITIVE LOGITS

AndEndTag

1.10

RenderAtEndOf

0.91

 المعيارى

0.90

GEBURTSDATUM

0.86

richTextPanel

0.84

IntoConstraints

0.79

posedge

0.78

GOTREF

0.78

 betweenstory

0.78

invokeLater

0.76

Activations Density 0.049%

specific lines of code or textual references within a document

line or sentence

No Comments

No Known Activations