INDEX

Explanations

references to humans being dehumanized or treated as inhuman in some way

oai_token-act-pair · gpt-3.5-turbo

terms related to humanity and its dehumanization

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

rypt

-0.85

CHAT

-0.74

 LINE

-0.71

PM

-0.70

REL

-0.70

URN

-0.69

APS

-0.67

 Channel

-0.66

EG

-0.65

ieri

-0.65

POSITIVE LOGITS

itarian

1.23

ity

1.05

 beings

1.03

ciating

0.95

 captives

0.90

izations

0.90

human

0.89

itary

0.88

ization

0.86

itar

0.86

Activations Density 0.014%

references to humans being dehumanized or treated as inhuman in some way

terms related to humanity and its dehumanization

No Known Activations