INDEX

Explanations

elements in a document that prompt the user to click on something

oai_token-act-pair · gpt-3.5-turbo

references to clicking actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Chancellor

-0.68

 Scotia

-0.68

 Yard

-0.65

 Cumber

-0.61

 Warsaw

-0.60

 Croat

-0.60

 Berk

-0.60

 Lutheran

-0.59

oglu

-0.59

egal

-0.59

POSITIVE LOGITS

lish

1.04

antry

0.87

prints

0.81

dress

0.81

ety

0.81

through

0.78

roots

0.74

pad

0.74

wheel

0.73

oola

0.73

Activations Density 0.028%

elements in a document that prompt the user to click on something

references to clicking actions

No Comments

No Known Activations