INDEX

Explanations

instances of addresses mentioned in text

oai_token-act-pair · gpt-3.5-turbo

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

fps

-0.78

lime

-0.76

umbers

-0.74

drm

-0.73

uj

-0.72

icer

-0.69

nature

-0.66

kus

-0.65

iatrics

-0.65

cia

-0.65

POSITIVE LOGITS

 addr

0.93

chool

0.80

 Address

0.79

holder

0.76

 address

0.75

holders

0.73

able

0.71

onica

0.68

 addresses

0.68

entric

0.68

Activations Density 0.033%

instances of addresses mentioned in text

No Comments

No Known Activations