INDEX

Explanations

titles or references to nobility and military ranks

oai_token-act-pair · gpt-3.5-turbo

mentions of nobility and military affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

okÃ©

-0.83

ovie

-0.83

abwe

-0.79

arcity

-0.78

alyst

-0.76

 scanner

-0.74

 tracker

-0.73

orer

-0.72

 projector

-0.70

antz

-0.70

POSITIVE LOGITS

 Edinburgh

0.86

 Duchess

0.84

burgh

0.79

Prince

0.79

Royal

0.79

Prin

0.76

 Caribbean

0.76

 Brunswick

0.74

Pont

0.73

 Mercy

0.71

Activations Density 0.173%

titles or references to nobility and military ranks

mentions of nobility and military affiliations

No Comments

No Known Activations