INDEX

Explanations

mentions of morally tainted or unethical behavior

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "dirty" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

*/(

-0.97

HCR

-0.95

yip

-0.90

iculture

-0.83

icultural

-0.79

utra

-0.79

uther

-0.79

ajor

-0.78

ommel

-0.76

zai

-0.75

POSITIVE LOGITS

 laundry

1.06

 linen

0.98

 diaper

0.91

 tricks

0.87

 diapers

0.84

 dirty

0.84

 mole

0.79

ness

0.76

 cleaner

0.72

mouth

0.72

Activations Density 0.025%

mentions of morally tainted or unethical behavior

references to the concept of "dirty" in various contexts

No Comments

No Known Activations