INDEX

Explanations

terms related to legal and regulatory language, particularly regarding statutes and amendments

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits



-0.45

AutoModerator

-0.45

ronpa

-0.44

 przekon

-0.43

DataBind

-0.42

KommentareTeilen

-0.42

newArrayList

-0.41

 RemoteException

-0.41



-0.41

 volet

-0.41

POSITIVE LOGITS

CloseOperation

0.59

 thereof

0.48

DockStyle

0.43

+#+

0.42

 skippy

0.40

 Rough

0.36

 transfieras

0.36

StoryboardSegue

0.35

 Slick

0.35

Uno

0.35

Activations Density 0.084%

terms related to legal and regulatory language, particularly regarding statutes and amendments

No Comments

No Known Activations