INDEX

Explanations

results or findings that indicate relationships or effects in experimental data

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

preceding "leads to" or similar phrases

np_acts-logits-general · gemini-2.0-flash

leads to, shows, confirms

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

Aiheesta

-0.57

 okazji

-0.50

iana

-0.48

ftagPool

-0.47

Orléans

-0.47

aine

-0.45

soa

-0.45

 plegable

-0.45

 برانيه

-0.45

 المعيارى

-0.45

POSITIVE LOGITS

DeleteBehavior

0.70

 propOrder

0.70



0.69

RTLU

0.66

 تضيفلها

0.65

InjectAttribute

0.64

enerbah

0.62

<",

0.62

 ویکی‌پدیا

0.61

UnsafeEnabled

0.59

Activations Density 0.768%

results or findings that indicate relationships or effects in experimental data

preceding "leads to" or similar phrases

leads to, shows, confirms

No Comments

No Known Activations