INDEX

Explanations

quotes or references related to religious texts

oai_token-act-pair · gpt-3.5-turbo

phrases related to spiritual or religious teachings

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

assic

-0.87

ormal

-0.85

agame

-0.76

 Collider

-0.75

Ops

-0.74

CTV

-0.74

 interns

-0.73

 stunts

-0.73

 Curve

-0.73

 Deadline

-0.71

POSITIVE LOGITS

 divine

1.07

 pious

1.06

 thou

1.05

 prayer

1.04

 heavenly

1.03

 consec

1.02

 holy

1.02

 everlasting

1.01

 unbel

1.00

 Blessed

1.00

Activations Density 0.586%

quotes or references related to religious texts

phrases related to spiritual or religious teachings

No Comments

No Known Activations