INDEX

Explanations

nouns associated with advantages or special rights

oai_token-act-pair · gpt-3.5-turbo

mentions of privilege in various contexts

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

ood

-0.81

GH

-0.75

Interstitial

-0.68

ready

-0.68

tra

-0.65

leaf

-0.65

urgy

-0.65

ailed

-0.65

psc

-0.64

 Adin

-0.64

POSITIVE LOGITS

ilege

1.54

 privilege

1.24

ileged

1.22

 Priv

0.89

 afforded

0.86

 privileges

0.81

itism

0.79

 holders

0.77

othal

0.74

 privileged

0.73

Activations Density 0.015%

nouns associated with advantages or special rights

mentions of privilege in various contexts

No Comments

No Known Activations