INDEX

Explanations

descriptions of things as extremely interesting, captivating, or engrossing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "fascinating" and related expressions

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

IFrame

Link

Negative Logits

abad

-0.71

xia

-0.66

 approvals

-0.65

vr

-0.65

gs

-0.65

uts

-0.64

uters

-0.64

hers

-0.63

pex

-0.63

aper

-0.63

POSITIVE LOGITS

ly

1.05

 fascinating

0.86

ĸļ

0.82

 mysteries

0.80

 Flavoring

0.74

ously

0.74

ingly

0.74

 fascination

0.73

 curiosity

0.71

 tales

0.70

Activations Density 0.034%

descriptions of things as extremely interesting, captivating, or engrossing

instances of the word "fascinating" and related expressions

No Known Activations