INDEX

Explanations

phrases indicating a potential threat to safety or security

oai_token-act-pair · gpt-3.5-turbo

phrases indicating threats to safety or security

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DragonMagazine

-0.86

wcsstore

-0.75

 Cosponsors

-0.72

zek

-0.69

aram

-0.68

zig

-0.67

ket

-0.66

VPN

-0.64

ãĤ´ãĥ³

-0.64

sort

-0.64

POSITIVE LOGITS

 livelihood

1.13

 viability

1.07

 humankind

1.01

 morale

0.98

 health

0.97

 wellbeing

0.97

 biodiversity

0.96

 humanity

0.93

 productivity

0.92

 mankind

0.91

Activations Density 0.171%

phrases indicating a potential threat to safety or security

phrases indicating threats to safety or security

No Comments

No Known Activations