INDEX

Explanations

phrases related to disguising or camouflaging

oai_token-act-pair · gpt-3.5-turbo

instances of words related to masquerade or disguise

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 RELEASE

-0.65

CSV

-0.63

NEC

-0.62

 fertility

-0.60

 conversions

-0.59

RSS

-0.59

 Pace

-0.58

 Stern

-0.58

 Schumer

-0.58

Bei

-0.58

POSITIVE LOGITS

quer

1.10

querade

1.03

ading

1.00

ÃŃa

0.90

aders

0.86

ior

0.84

ader

0.79

idon

0.77

ware

0.76

mop

0.76

Activations Density 0.024%

phrases related to disguising or camouflaging

instances of words related to masquerade or disguise

No Comments

No Known Activations