INDEX

Explanations

names and surnames

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 miscarriage

-0.61

 Velvet

-0.59

 dawn

-0.59

 Serial

-0.58

bound

-0.58

wcs

-0.56

 Polaris

-0.55

Bound

-0.54

 Werewolf

-0.54

 Costco

-0.54

POSITIVE LOGITS

enegger

1.21

zyk

0.86

baum

0.85

sson

0.80

orst

0.80

jen

0.80

udeau

0.79

ouf

0.79

reau

0.78

kson

0.75

Activations Density 0.255%

names and surnames

proper nouns and names

No Comments

No Known Activations