INDEX

Explanations

specific identifiers and numerical values associated with biological or experimental data

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

units and measurements

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

 ujednoznacz

-0.62

ConstraintMaker

-0.59

delwed

-0.57

verwijspagina

-0.55

ChildScrollView

-0.54

htbp

-0.53

 الحره

-0.50

 AssemblyVersion

-0.50

HtmlAttribute

-0.48

numerusform

-0.47

POSITIVE LOGITS

libft

0.38

 ravi

0.37

ArrowToggle

0.36

 gyhoeddwyd

0.34

 associate

0.33

 robar

0.33

 rara

0.33

 partially

0.32

 приятно

0.31

 catch

0.30

Activations Density 0.064%

specific identifiers and numerical values associated with biological or experimental data

units and measurements

No Comments

No Known Activations