INDEX

Explanations

phrases related to opting in or out of something

oai_token-act-pair · gpt-3.5-turbo

phrases related to opting out or consent

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

issance

-0.75

 ILCS

-0.72

ModLoader

-0.71

Lau

-0.69

 Helic

-0.66

enegger

-0.61

 Yard

-0.60

Mek

-0.59

 Arabia

-0.59

 crowded

-0.59

POSITIVE LOGITS

exit

0.99

out

0.99

posted

0.89

level

0.84

cert

0.84

up

0.83

mode

0.83

outs

0.83

present

0.82

aye

0.81

Activations Density 0.026%

phrases related to opting in or out of something

phrases related to opting out or consent

No Comments

No Known Activations