INDEX

Explanations

terms related to education and various forms of evidence in legal contexts

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

words followed by and

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

 dedans

-0.40

 tirage

-0.34

(~(

-0.31

mû

-0.31

 vstup

-0.30

 blends

-0.29

 Altos

-0.29

espa

-0.29

colgante

-0.29

 Dura

-0.29

POSITIVE LOGITS



0.67

XmlAccessType

0.67

windowFixed

0.66

ंदीखरीदारी

0.65

 BoxDecoration

0.62

 nahilalakip

0.60

IUrlHelper

0.58

 ivelany

0.57

uxxxx

0.57

الحياه

0.55

Activations Density 0.042%

terms related to education and various forms of evidence in legal contexts

words followed by and

No Comments

No Known Activations