INDEX

Explanations

words related to classified information

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

annis

-0.83

soDeliveryDate

-0.76

Kay

-0.70

htar

-0.70

hner

-0.68

alach

-0.66

Virgin

-0.66

LI

-0.66

kj

-0.64

 Beard

-0.64

POSITIVE LOGITS

 briefings

1.26

 documents

1.07

 material

1.02

 information

1.01

 markings

0.98

 intel

0.98

 materials

0.94

 papers

0.92

oslov

0.85

 briefing

0.85

Activations Density 0.038%

words related to classified information

No Comments

No Known Activations