INDEX

Explanations

language related to governance and the governing process

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

governed by

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

ا

-0.59

 ingrat

-0.56

Pic

-0.52

Anita

-0.51

 Anita

-0.51

 Rembrandt

-0.51

 Nicky

-0.50

テンツ

-0.50

 Farina

-0.50

 malic

-0.50

POSITIVE LOGITS

 governed

1.03

 governing

1.02

 governs

0.94

 Governing

0.91

 govern

0.91

 gouvern

0.88

Governing

0.84

 governance

0.84

govern

0.82

 GOVERN

0.82

Activations Density 0.007%

language related to governance and the governing process

governed by

No Comments

No Known Activations