INDEX

Explanations

the presence of the term "ass" in various contexts

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

ending in "ass" or "ess"

np_acts-logits-general · gemini-2.0-flash

assumptions, assassins

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

ization

-0.61

 Buren

-0.58

]_{\

-0.57

../../

-0.55

🏽

-0.54

../../../

-0.54

hares

-0.53

 Secondly

-0.53

endphp

-0.53

POSITIVE LOGITS

 Ress

0.72

 Vass

0.71

Ress

0.71

mybatisplus

0.69

ASS

0.68

iss

0.65

 Coss

0.65

SSS

0.65

 Bess

0.63

 Sass

0.62

Activations Density 0.269%

the presence of the term "ass" in various contexts

ending in "ass" or "ess"

assumptions, assassins

No Comments

No Known Activations