INDEX

Explanations

the word 'demon' or variations of it

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Mellon

-0.67

artney

-0.64

 Bacon

-0.64

abouts

-0.63

 Howe

-0.62

arkin

-0.62

 Norn

-0.61

arella

-0.61

ippi

-0.61

 Morales

-0.60

POSITIVE LOGITS

stration

1.74

strate

1.47

str

1.25

ization

1.11

iac

1.11

izing

1.04

ica

1.00

izations

1.00

isation

0.99

ising

0.97

Activations Density 0.017%

the word 'demon' or variations of it

No Comments

No Known Activations