INDEX

Explanations

references to various government bureaus and agencies

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

Bureau of X or manager

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 wish

-0.33

TAR

-0.33

Wish

-0.32

 Angio

-0.30

absch

-0.30

 think

-0.30

Position

-0.30

 conqu

-0.30

COLOG

-0.30

 lieber

-0.29

POSITIVE LOGITS

 Bureau

0.85

 BUREAU

0.82

Bureau

0.79

bureau

0.74

 bureaus

0.70

 bureau

0.69

 autorytatywna

0.67

 kaarangay

0.65

EndGlobalSection

0.63

buro

0.62

Activations Density 0.004%

references to various government bureaus and agencies

Bureau of X or manager

No Comments

No Known Activations