INDEX

Explanations

references to theft and criminal activities

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

thieves and related concepts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 poffible

-0.30

 pleaſure

-0.29

 sopp

-0.28

 station

-0.28

 token

-0.28

 purpoſe

-0.27

 Majefty

-0.27

 Huss

-0.26

CloseHandle

-0.26

PendingIntent

-0.26

POSITIVE LOGITS

 Thief

0.59

 EconPapers

0.59

 Cleaner

0.58

 Cleaners

0.58

 thief

0.57

AndEndTag

0.57

 Crimes

0.56

crimes

0.56

 Cleaning

0.55

 thieves

0.55

Activations Density 0.129%

references to theft and criminal activities

thieves and related concepts

No Comments

No Known Activations