INDEX

Explanations

phrases related to government and politics

oai_token-act-pair · gpt-3.5-turbo

references to government and military entities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

iphate

-0.89

Õ

-0.80

Ó

-0.75

phe

-0.74

wn

-0.71

ocene

-0.71

ailable

-0.71

ocalyptic

-0.71

Downloadha

-0.71

utic

-0.69

POSITIVE LOGITS

 officials

1.77

 spokesman

1.51

 Deputy

1.46

 lawmakers

1.43

 spokeswoman

1.42

 aides

1.40

 investigators

1.39

 negotiators

1.38

 prosecutors

1.33

 authorities

1.31

Activations Density 0.270%

phrases related to government and politics

references to government and military entities

No Comments

No Known Activations