INDEX

Explanations

phrases related to large, powerful, or dangerous entities

oai_token-act-pair · gpt-3.5-turbo

references to the term "beast" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

monds

-0.76

enza

-0.74

arters

-0.74

psons

-0.74

pai

-0.74

mit

-0.74

 polymer

-0.73

bered

-0.72

licted

-0.71

activated

-0.71

POSITIVE LOGITS

 beasts

1.37

 beast

1.24

 Beasts

0.99

 carc

0.92

 Beast

0.89

osaurs

0.81

hawk

0.80

 Incarn

0.79

monster

0.79

wagon

0.78

Activations Density 0.008%

phrases related to large, powerful, or dangerous entities

references to the term "beast" and its variations

No Comments

No Known Activations