INDEX

Explanations

names, especially emphasizing "Vic"

oai_token-act-pair · gpt-3.5-turbo

names and references to a specific individual or entity

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

*/(

-0.83

xual

-0.80

tenance

-0.78

manship

-0.76

Ô

-0.75

landish

-0.72

SpaceEngineers

-0.72

 EntityItem

-0.70

spread

-0.69

dress

-0.68

POSITIVE LOGITS

arious

1.13

inity

1.04

isions

1.02

entric

0.99

abulary

0.93

ente

0.91

ient

0.90

ision

0.89

iple

0.89

imated

0.89

Activations Density 0.014%

names, especially emphasizing "Vic"

names and references to a specific individual or entity

No Comments

No Known Activations