INDEX

Explanations

references to the effects of various phenomena or substances

oai_token-act-pair · gpt-3.5-turbo

phrases discussing the effects of various factors or conditions

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

-0.81

zan

-0.70

swick

-0.69

 Patriarch

-0.67

imb

-0.66

apest

-0.65

MQ

-0.64

gdala

-0.63

 Malt

-0.63

mbuds

-0.62

POSITIVE LOGITS

effects

1.27

iveness

1.02

 effects

0.99

Effects

0.97

Effect

0.90

 havoc

0.88

 impact

0.86

 reverber

0.85

 consequences

0.83

 impacts

0.83

Activations Density 0.033%

references to the effects of various phenomena or substances

phrases discussing the effects of various factors or conditions

No Comments

No Known Activations