INDEX

Explanations

references to places or periods where people lived

oai_token-act-pair · gpt-3.5-turbo

instances of the word "lived" to indicate life experiences or places someone has lived

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

sonian

-0.84

uyomi

-0.80

essee

-0.79

elight

-0.79

achu

-0.74

aneers

-0.73

enei

-0.73

ion

-0.73

vernment

-0.72

reme

-0.71

POSITIVE LOGITS

lihood

1.02

 Wage

0.87

vic

0.81

 Reincarn

0.78

liest

0.77

hart

0.77

 upstairs

0.74

lier

0.73

ths

0.72

 Rent

0.70

Activations Density 0.025%

references to places or periods where people lived

instances of the word "lived" to indicate life experiences or places someone has lived

No Comments

No Known Activations