INDEX

Explanations

words related to criminal activities and villainous behavior

oai_token-act-pair · gpt-3.5-turbo

terms related to crime or criminal behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rien

-0.91

eele

-0.83

tis

-0.80

icrobial

-0.80

rians

-0.80

mberg

-0.79

vation

-0.76

ffen

-0.76

ovember

-0.73

foreseen

-0.73

POSITIVE LOGITS

lords

0.84

ãĥ³ãĤ¸

0.82

 killers

0.74

 spree

0.73

 mastermind

0.72

 thugs

0.72

 gangs

0.71

 Boko

0.71

 blackmail

0.71

doms

0.70

Activations Density 0.316%

words related to criminal activities and villainous behavior

terms related to crime or criminal behavior

No Comments

No Known Activations