INDEX

Explanations

institutional and geographic terms, particularly related to businesses or locations

oai_token-act-pair · gpt-3.5-turbo

references to specific geographic locations and fictional entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

mble

-0.93

yip

-0.91

xual

-0.88

vich

-0.85

cius

-0.84

ongyang

-0.81

ced

-0.81

berman

-0.76

htaking

-0.75

cript

-0.74

POSITIVE LOGITS

 Arkham

0.99

 Nights

0.97

 Asylum

0.96

 City

0.93

 Alley

0.90

 Strip

0.84

 Gotham

0.83

 Gazette

0.82

 Horror

0.81

 Evening

0.81

Activations Density 0.010%

institutional and geographic terms, particularly related to businesses or locations

references to specific geographic locations and fictional entities

No Comments

No Known Activations