INDEX

Explanations

phrases and descriptions related to infidelity or illicit romantic relationships

oai_token-act-pair · gpt-3.5-turbo

references to romantic or extramarital relationships

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

solid

-0.87

srf

-0.74

ACTED

-0.71

printed

-0.71

ovi

-0.71

externalActionCode

-0.70

owicz

-0.70

imm

-0.69

gments

-0.68

umbing

-0.68

POSITIVE LOGITS

 affair

0.93

ional

0.79

ttes

0.77

 involving

0.73

uality

0.73

 revolving

0.72

 contrace

0.71

ilial

0.68

cov

0.66

ually

0.66

Activations Density 0.022%

phrases and descriptions related to infidelity or illicit romantic relationships

references to romantic or extramarital relationships

No Known Activations