INDEX

Explanations

references to public policy topics

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

"policy" followed by another word

np_acts-logits-general · gemini-2.0-flash

policy schedule, changes, efforts

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

 ThemeData

-0.60

BarStyle

-0.60

 BoxFit

-0.59

 GenerationType

-0.58

__':

-0.56

żd

-0.55

 cherchés

-0.55

دانشنامهٔ

-0.55

存于互联网档案馆

-0.55

gister

-0.55

POSITIVE LOGITS

maker

0.82

making

0.77

 makers

0.74

 policies

0.73

Policies

0.69

holders

0.68

holder

0.67

makers

0.67

 Policies

0.62

matig

0.60

Activations Density 0.076%

references to public policy topics

"policy" followed by another word

policy schedule, changes, efforts

No Comments

No Known Activations