INDEX

Explanations

words related to groups of people, communities, and countries

oai_token-act-pair · gpt-3.5-turbo

references to specific groups or communities within society

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

eatures

-0.65

 Stain

-0.62

 Dover

-0.62

 Heist

-0.61

 Anch

-0.58

 Beaut

-0.57

optim

-0.57

Buff

-0.56

IPO

-0.55

 Valkyrie

-0.55

POSITIVE LOGITS

who

0.95

 alike

0.88

 whom

0.86

 swear

0.84

who

0.82

 rejoice

0.71

 perished

0.68

whose

0.65

 behaved

0.65

gdala

0.64

Activations Density 0.371%

words related to groups of people, communities, and countries

references to specific groups or communities within society

No Comments

No Known Activations