INDEX

Explanations

phrases related to holding someone or something accountable

oai_token-act-pair · gpt-3.5-turbo

phrases related to accountability and responsibility

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endix

-0.71

ibel

-0.69

 deceive

-0.67

icycle

-0.67

naire

-0.65

ologist

-0.63

shire

-0.63

mph

-0.63

 Daylight

-0.61

inx

-0.61

POSITIVE LOGITS

hold

1.09

holders

1.03

holder

0.99

holding

0.98

 sway

0.95

erness

0.91

 hold

0.89

held

0.89

 accountable

0.82

fast

0.78

Activations Density 0.038%

phrases related to holding someone or something accountable

phrases related to accountability and responsibility

No Comments

No Known Activations