INDEX

Explanations

homophobic and homosexual-related words or phrases

oai_token-act-pair · gpt-3.5-turbo

mentions of homosexuality and related topics

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

 sender

-0.72

utions

-0.70

çĦ

-0.66

 valley

-0.63

AAF

-0.63

raltar

-0.63

 Canaver

-0.61

 Weir

-0.61

 Mehran

-0.60

 Camb

-0.60

POSITIVE LOGITS

emade

1.35

osexual

1.17

estead

1.13

ework

1.07

eless

0.93

eline

0.91

lyn

0.90

ily

0.89

eland

0.85

ilia

0.85

Activations Density 0.033%

homophobic and homosexual-related words or phrases

mentions of homosexuality and related topics

No Comments

No Known Activations