INDEX

Explanations

proper nouns related to geographical locations, specifically focusing on the city of Perth

oai_token-act-pair · gpt-3.5-turbo

references to the location of Perth and its significance

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

displayText

-0.70

xual

-0.69

BOOK

-0.66

ãĥĻ

-0.64

OHN

-0.64

 SERVICE

-0.62

ariat

-0.62

 chore

-0.61

termination

-0.61

Magikarp

-0.60

POSITIVE LOGITS

shire

1.54

 Lumpur

0.98

burgh

0.91

ington

0.85

 Perth

0.81

sburg

0.81

uckland

0.80

awar

0.79

lins

0.78

patrick

0.78

Activations Density 0.008%

proper nouns related to geographical locations, specifically focusing on the city of Perth

references to the location of Perth and its significance

No Comments

No Known Activations