INDEX

Explanations

phrases associated with legal processes and actions

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

focusing on specific objectives

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

性和

-0.37

bigsqcup

-0.36

 antig

-0.34

fil

-0.33

完

-0.33

 circ

-0.32

and

-0.31

dur

-0.30

sab

-0.30

 extension

-0.30

POSITIVE LOGITS

verwijspagina

0.74

MLLoader

0.73

makeConstraints

0.72

RenderAtEndOf

0.71

 تضيفلها

0.70

expandindo

0.69

AnchorTagHelper

0.69

webElementXpaths

0.68

EDEFAULT

0.68

InputTagHelper

0.67

Activations Density 0.091%

phrases associated with legal processes and actions

focusing on specific objectives

No Comments

No Known Activations