INDEX

Explanations

people using specific words or actions as a characteristic or qualifier

oai_token-act-pair · gpt-3.5-turbo

the word "who" in various contexts relating to people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ãĥĥãĥī

-0.68

 Darling

-0.66

AN

-0.66

 Urug

-0.60

Pai

-0.60

Oz

-0.59

........

-0.59

Els

-0.59

QUI

-0.58

PokÃ©

-0.58

POSITIVE LOGITS

 benefited

0.86

 accompanies

0.79

oped

0.76

atoon

0.76

soever

0.75

ortium

0.74

 violates

0.73

arnaev

0.72

ikk

0.72

 participated

0.72

Activations Density 0.168%

people using specific words or actions as a characteristic or qualifier

the word "who" in various contexts relating to people

No Comments

No Known Activations