INDEX

Explanations

instances of guilty pleas and related legal terminology

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

guilty plea agreement

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

complexContent

-0.71

 estekak

-0.62

 الحره

-0.57

AnimationsModule

-0.53

tableFuture

-0.52

ExtendWith

-0.51

RegistryLite

-0.51

تقاوى

-0.51

VersionUID

-0.50

makeConstraints

-0.49

POSITIVE LOGITS

 confessed

0.48

 admitted

0.48

 voluntarily

0.44

 confes

0.43

tangentMode

0.42

 confession

0.41

 confess

0.41

 confessions

0.41

 surrender

0.40

 accepts

0.39

Activations Density 0.018%

instances of guilty pleas and related legal terminology

guilty plea agreement

No Comments

No Known Activations