INDEX

Explanations

mentions of religious or sacred structures such as shrines

oai_token-act-pair · gpt-3.5-turbo

references to shrines and related concepts

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

eering

-0.83

 laundering

-0.77

 Winchester

-0.76

 PowerPoint

-0.73

eers

-0.72

 conditioning

-0.66

 Dragonbound

-0.64

 Hebdo

-0.64

eer

-0.63

 Trojan

-0.62

POSITIVE LOGITS

anch

1.01

ines

0.98

abies

0.94

acy

0.93

ubs

0.92

iffs

0.91

ower

0.90

ith

0.90

otted

0.90

anches

0.89

Activations Density 0.013%

mentions of religious or sacred structures such as shrines

references to shrines and related concepts

No Comments

No Known Activations