INDEX

Explanations

religious mentions or exclamations

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "God" or variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Luxem

-0.74

 contin

-0.62

 detachment

-0.62

Altern

-0.61

phrine

-0.61

URRENT

-0.61

akeru

-0.61

 adul

-0.60

DERR

-0.59

 auth

-0.58

POSITIVE LOGITS

father

1.27

frey

1.25

mother

1.20

 forbid

1.08

parents

1.04

win

1.01

zilla

0.99

god

0.97

 bless

0.96

 Almighty

0.95

Activations Density 0.032%

religious mentions or exclamations

mentions of the word "God" or variations of it

No Comments

No Known Activations