INDEX

Explanations

references to winning or receiving prestigious awards or honors, particularly trophies

oai_token-act-pair · gpt-3.5-turbo

mentions of trophies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

utenberg

-0.81

weeney

-0.79

trak

-0.78

duino

-0.73

erman

-0.68

 experimenting

-0.65

OVA

-0.65

²¾

-0.64

SOURCE

-0.64

avin

-0.63

POSITIVE LOGITS

 Trophy

1.09

 trophy

0.97

 trophies

0.95

phies

0.88

 plaque

0.84

brates

0.79

aceous

0.73

brate

0.73

holder

0.72

ilage

0.70

Activations Density 0.033%

references to winning or receiving prestigious awards or honors, particularly trophies

mentions of trophies

No Comments

No Known Activations