INDEX

Explanations

words related to opposing or being against something

oai_token-act-pair · gpt-3.5-turbo

references to anti-establishment sentiments or movements

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

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

 laure

-0.79

ĸļ

-0.75

 precincts

-0.73

 pads

-0.69

 Brus

-0.68

 dots

-0.67

 leaps

-0.67

 addresses

-0.67

 somew

-0.66

 Brist

-0.66

POSITIVE LOGITS

government

1.40

social

1.38

establishment

1.35

Semitic

1.34

Semitism

1.32

immigrant

1.29

corruption

1.29

smoking

1.29

discrimination

1.28

capitalist

1.21

Activations Density 0.030%

words related to opposing or being against something

references to anti-establishment sentiments or movements

No Comments

No Known Activations