INDEX

Explanations

references to individuals with leadership roles or titles

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

after personal pronouns

np_acts-logits-general · gemini-2.0-flash

legal, programming, personal statements

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.43

die

-0.41

${

-0.41

guan

-0.41

 दू

-0.41

най

-0.40

type

-0.40

 every

-0.39

am

-0.39

("

-0.39

POSITIVE LOGITS

SourceChecksum

1.16

DockStyle

0.91

SharedDtor

0.90

 '\\;'

0.89

parsedMessage

0.88

tagHelperRunner

0.88

 AttributeSet

0.88

InstrumentedTest

0.87

 समीक्षक

0.86

AsUp

0.85

Activations Density 0.034%

references to individuals with leadership roles or titles

after personal pronouns

legal, programming, personal statements

No Comments

No Known Activations