INDEX

Explanations

proper nouns related to locations or people

oai_token-act-pair · gpt-3.5-turbo

proper names and significant identifiers related to individuals and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ĳå£«

-0.82

Ĥª

-0.78

ãĥ¼ãĥ³

-0.78

ãĥ¼ãĥĨ

-0.76

anamo

-0.75

ãĤ©

-0.73

é¾įåĸļå£«

-0.72

Zen

-0.72

PDATE

-0.68

ãĤ®

-0.68

POSITIVE LOGITS

adders

0.89

 Luthor

0.87

oyd

0.86

opez

0.85

utenant

0.85

ibrary

0.84

yrics

0.82

uggage

0.82

ibrarian

0.81

orem

0.77

Activations Density 0.049%

proper nouns related to locations or people

proper names and significant identifiers related to individuals and places

No Comments

No Known Activations