INDEX

Explanations

words related to religious figures, especially priests

oai_token-act-pair · gpt-3.5-turbo

references to religious figures, specifically priests and their roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

BAT

-0.71

Ø§ÙĦ

-0.69

 Retro

-0.67

DonaldTrump

-0.67

ilater

-0.67

opia

-0.66

 Guarant

-0.66

 Wonderland

-0.65

issors

-0.65

oday

-0.63

POSITIVE LOGITS

esses

1.59

ess

1.25

 priests

1.00

 priest

0.97

 ordained

0.87

hood

0.87

esan

0.87

mares

0.82

ciples

0.81

essing

0.78

Activations Density 0.043%

words related to religious figures, especially priests

references to religious figures, specifically priests and their roles

No Comments

No Known Activations