INDEX

Explanations

links or buttons used for interactions on websites

oai_token-act-pair · gpt-3.5-turbo

engagement-related actions and commands, particularly on social media platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Â¶

-0.78

clus

-0.75

gew

-0.70

cephal

-0.63

paces

-0.61

bush

-0.61

authent

-0.59

 pupp

-0.59

rica

-0.59

anthrop

-0.59

POSITIVE LOGITS

 Cancel

0.72

duction

0.70

 Email

0.70

azo

0.67

 haste

0.66

Reason

0.64

});

0.63

ãĥĥãĥī

0.63

TextColor

0.59

uterte

0.58

Activations Density 0.177%

links or buttons used for interactions on websites

engagement-related actions and commands, particularly on social media platforms

No Comments

No Known Activations