INDEX

Explanations

references to privilege

oai_token-act-pair · gpt-3.5-turbo

mentions of privilege and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Estimated

-0.72

 Animated

-0.66

ource

-0.64

atom

-0.63

Plot

-0.61

 Reported

-0.60

english

-0.59

 estimates

-0.59

Fig

-0.59

atl

-0.58

POSITIVE LOGITS

 privilege

4.30

 privileges

2.40

 privileged

2.06

 Priv

2.05

ilege

1.97

priv

1.51

Priv

1.39

 entitlement

1.27

 pleasure

1.21

ileged

1.19

Activations Density 0.010%

references to privilege

mentions of privilege and related concepts

No Comments

No Known Activations