INDEX

Explanations

references to people in a position of power

oai_token-act-pair · gpt-3.5-turbo

references to political figures and their associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

wcs

-0.84

wcsstore

-0.83

rooms

-0.80

hops

-0.79

catentry

-0.78

taboola

-0.78

 thereof

-0.75

ouls

-0.74

isphere

-0.71

Forge

-0.70

POSITIVE LOGITS

 former

1.21

 disgr

1.21

 comedian

1.19

 novelist

1.16

 longtime

1.13

 actress

1.13

 legendary

1.12

 billionaire

1.12

 actor

1.09

 famed

1.09

Activations Density 0.349%

references to people in a position of power

references to political figures and their associations

No Comments

No Known Activations