INDEX

Explanations

references to the name "Alexey"

oai_token-act-pair · gpt-3.5-turbo

instances of the substring "ey" in names or titles

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

mint

-0.87

istani

-0.78

atoon

-0.77

atchewan

-0.72

ULAR

-0.69

EED

-0.63

adium

-0.62

IFA

-0.61

 Archdemon

-0.60

ï

-0.60

POSITIVE LOGITS

ewitness

1.08

kj

0.88

er

0.77

ey

0.76

ield

0.76

estinal

0.75

outube

0.75

esy

0.74

giene

0.74

velt

0.70

Activations Density 0.015%

references to the name "Alexey"

instances of the substring "ey" in names or titles

No Comments

No Known Activations