INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

place names, particularly cities and countries

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

rodu

-0.63

 motto

-0.58

 constitu

-0.58

xual

-0.58

etheless

-0.57

espie

-0.54

minist

-0.54

antry

-0.53

 anecd

-0.52

 Hamp

-0.52

POSITIVE LOGITS

IRC

0.75

isi

0.62

ADVERTISEMENT

0.59

kies

0.59

 Airways

0.58

/?

0.58

/#

0.58

Flight

0.58

 suburbs

0.58

 Airlines

0.56

Activations Density 0.540%

locations or places

place names, particularly cities and countries

No Comments

No Known Activations