INDEX

Explanations

mentions of specific locations and organizational titles

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

organizations or locations

np_acts-logits-general · gemini-2.0-flash

branches and offices

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 systematic

-0.42

 théorie

-0.36

AndEndTag

-0.35

 चीज़ों

-0.34

writeFieldEnd

-0.34

ViewFeatures

-0.34

 tæn

-0.32

 Exacts

-0.32

 lecciones

-0.32

 fucking

-0.31

POSITIVE LOGITS

setVerticalGroup

0.59

Branch

0.55

 Branch

0.52

 BRANCH

0.50

Territory

0.50

 branch

0.49

 couvercle

0.49

 Branches

0.48

Station

0.47

 الرياضيه

0.47

Activations Density 0.189%

mentions of specific locations and organizational titles

organizations or locations

branches and offices

No Comments

No Known Activations