INDEX

Explanations

phrases related to protection and shielding

oai_token-act-pair · gpt-3.5-turbo

phrases indicating protection or shielding from negative influences or threats

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

arse

-0.78

ios

-0.69

page

-0.69

Hello

-0.68

 isEnabled

-0.67

KC

-0.66

aucas

-0.65

laughter

-0.64

hello

-0.64

Tracker

-0.64

POSITIVE LOGITS

 afar

0.89

 Carbuncle

0.81

 demolition

0.64

 future

0.62

 Communism

0.60

 Arctic

0.60

adies

0.60

 Dominion

0.60

ption

0.59

 FEMA

0.59

Activations Density 0.064%

phrases related to protection and shielding

phrases indicating protection or shielding from negative influences or threats

No Comments

No Known Activations