INDEX

Explanations

specific references to different nationalities or citizenship statuses

oai_token-act-pair · gpt-3.5-turbo

references to various nationalities and citizenship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Nut

-0.79

Stretch

-0.77

 Lans

-0.77

Ward

-0.75

 Tooth

-0.74

Leader

-0.74

 Hath

-0.73

Har

-0.73

 Hind

-0.71

ww

-0.70

POSITIVE LOGITS

 nationals

1.03

ignty

0.96

 extrad

0.94

 residing

0.94

 abroad

0.94

 smugglers

0.93

 deported

0.93

doms

0.92

 visas

0.91

 fleeing

0.90

Activations Density 0.043%

specific references to different nationalities or citizenship statuses

references to various nationalities and citizenship

No Comments

No Known Activations