INDEX

Explanations

words related to mental states such as agitation, confusion, and paranoia

oai_token-act-pair · gpt-3.5-turbo

words or phrases associated with various forms of "ness," indicating quality or state

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ODE

-0.71

 Auschwitz

-0.65

bern

-0.64

verbs

-0.64

ORN

-0.62

ask

-0.62

WAR

-0.62

ellar

-0.61

amen

-0.60

veh

-0.60

POSITIVE LOGITS

iness

1.16

terness

0.97

ness

0.93

yy

0.87

nesses

0.85

ionage

0.82

ĪĴ

0.81

hip

0.79

yk

0.79

edly

0.73

Activations Density 0.027%

words related to mental states such as agitation, confusion, and paranoia

words or phrases associated with various forms of "ness," indicating quality or state

No Comments

No Known Activations