INDEX

Explanations

phrases related to things being hidden or buried

oai_token-act-pair · gpt-3.5-turbo

references to hidden or obscured elements and truths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

undai

-0.80

cial

-0.79

hyde

-0.78

trak

-0.77

mbuds

-0.73

FUL

-0.73

-0.72

PORT

-0.71

OY

-0.71

idas

-0.71

POSITIVE LOGITS

 rubble

1.49

 bushes

1.29

 clouds

1.28

 wreckage

1.26

 weeds

1.19

 fabric

1.18

 obscurity

1.17

 folds

1.16

 shadows

1.15

 woods

1.14

Activations Density 0.355%

phrases related to things being hidden or buried

references to hidden or obscured elements and truths

No Comments

No Known Activations