INDEX

Explanations

mentions or descriptions of things that are physically lightweight or not heavy

oai_token-act-pair · gpt-3.5-turbo

instances of the word "light" or its associations

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

halla

-0.98

ettings

-0.84

isner

-0.74

utor

-0.73

OPLE

-0.73

apons

-0.72

ĳå£«

-0.72

OUP

-0.70

berus

-0.69

 Forbidden

-0.68

POSITIVE LOGITS

 bulb

1.21

hearted

1.17

weights

1.15

bul

1.15

ening

1.15

enment

1.10

 bulbs

1.09

lights

1.04

heartedly

1.02

ener

0.97

Activations Density 0.027%

mentions or descriptions of things that are physically lightweight or not heavy

instances of the word "light" or its associations

No Comments

No Known Activations