INDEX

Explanations

places or scenarios where people are physically standing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "in."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gem

-0.79

catentry

-0.77

eni

-0.75

ľ

-0.69

cia

-0.69

ebin

-0.66

elf

-0.65

etheless

-0.64

gans

-0.64

Ĥİ

-0.64

POSITIVE LOGITS

 front

1.16

 seiz

1.09

awe

1.01

 queues

1.01

 line

0.99

 silence

0.97

 ambush

0.96

 circles

0.95

 obscurity

0.95

 limbo

0.94

Activations Density 0.105%

places or scenarios where people are physically standing

instances of the word "in."

No Comments

No Known Activations