INDEX

Explanations

mentions of and references to "presidency."

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "presidency."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ku

-0.78

Bra

-0.72

FOX

-0.67

LESS

-0.66

ibr

-0.66

ASC

-0.65

Fort

-0.63

gra

-0.62

Var

-0.62

Yar

-0.62

POSITIVE LOGITS

 presidency

1.28

 Presidency

1.22

manship

0.95

OTUS

0.91

orship

0.81

 office

0.81

velt

0.80

olicy

0.79

 Jinping

0.79

uary

0.74

Activations Density 0.006%

mentions of and references to "presidency."

references to the concept of "presidency."

No Comments

No Known Activations