INDEX

Explanations

references to legal cases and court decisions

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

previous or prior cases

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

Personendaten

-0.64

DockStyle

-0.63

 Taktlose

-0.53

 estekak

-0.53

Tikang

-0.51

 صوتيه

-0.49

Derbyniad

-0.49

 kaarangay

-0.47

 propOrder

-0.45

 utafitiHapana

-0.44

POSITIVE LOGITS

ViewImports

0.52

 vorige

0.48

 previous

0.47

Previous

0.44

 poprzed

0.40

setViewportView

0.40

previous

0.39

 preview

0.38

 Previous

0.38

 eerder

0.38

Activations Density 0.060%

references to legal cases and court decisions

previous or prior cases

No Comments

No Known Activations