INDEX

Explanations

references to concepts related to light or to objects with the word "light" in them

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "Light."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

PsyNetMessage

-0.86

CVE

-0.74

ĳå£«

-0.73

Fas

-0.72

apons

-0.72

ä½ľ

-0.71

ettings

-0.70

ersion

-0.66

ODUCT

-0.65

AAP

-0.65

POSITIVE LOGITS

bringer

1.07

lights

1.00

light

1.00

ening

0.97

 bulbs

0.96

 bulb

0.95

beam

0.95

hearted

0.95

weight

0.95

bul

0.94

Activations Density 0.013%

references to concepts related to light or to objects with the word "light" in them

references to the concept of "Light."

No Comments

No Known Activations