INDEX

Explanations

proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

names of notable individuals

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

uers

-0.83

gradation

-0.83

anium

-0.81

runner

-0.80

runners

-0.78

Ã§a

-0.78

lete

-0.76

arist

-0.75

stasy

-0.75

aeper

-0.75

POSITIVE LOGITS

BuyableInstoreAndOnline

0.89

 Norris

0.81

 Jarvis

0.75

 Gord

0.72

Downloadha

0.67

 Malfoy

0.65

 Connor

0.63

dayName

0.63

 Hatch

0.63

 Skinner

0.63

Activations Density 0.043%

proper nouns and names

names of notable individuals

No Comments

No Known Activations