INDEX

Explanations

mathematical symbols and equations in the text

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

numbers and equations

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

RegressionTest

-0.56

 <<<<<<<<<<<<<<

-0.52

WebElementEntity

-0.49

NameInMap

-0.47

IntoConstraints

-0.46

ThemeOverlay

-0.44

InstrumentedTest

-0.44

 nakalista

-0.42

 voici

-0.40

PreExecute

-0.40

POSITIVE LOGITS

0.41

rawQuery

0.39

мерть

0.37

 InputDecoration

0.36

 schließen

0.36

window

0.35

rabaya

0.35

 Städten

0.35

 złot

0.35

Див

0.35

Activations Density 0.080%

mathematical symbols and equations in the text

numbers and equations

No Comments

No Known Activations