INDEX

Explanations

numbers and various coding structures or annotations

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

numerical data

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

posedge

-0.88

aarrggbb

-0.86

enumii

-0.81

InjectAttribute

-0.81

 المعيارى

-0.77

EndContext

-0.76

RotationOrder

-0.75

ésultats

-0.75

Personensuche

-0.73

fjspx

-0.73

POSITIVE LOGITS

Until

0.31

true

0.30

 reicht

0.29

 Ahnung

0.27

 Kontexten

0.26

อล

0.25

Icon

0.25

0.24

Only

0.24

 útiles

0.24

Activations Density 0.280%

numbers and various coding structures or annotations

numerical data

No Comments

No Known Activations

numbers and various coding structures or annotations

numerical data

No Comments

No Known Activations