INDEX

Explanations

comprehensive and detailed descriptions or indexes of various topics

oai_token-act-pair · gpt-3.5-turbo

phrases indicating comprehensive summaries or reviews of various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

lease

-0.62

 dreaded

-0.61

soever

-0.61

ammers

-0.61

 detectors

-0.60

 emit

-0.60

ween

-0.59

 rescind

-0.59

bees

-0.58

alties

-0.58

POSITIVE LOGITS

 chron

0.84

 exposition

0.81

DragonMagazine

0.75

endium

0.75

LECT

0.74

 detailing

0.73

 COMPLE

0.72

 overview

0.69

Detailed

0.69

helm

0.68

Activations Density 0.351%

comprehensive and detailed descriptions or indexes of various topics

phrases indicating comprehensive summaries or reviews of various subjects

No Comments

No Known Activations