INDEX

Explanations

variations of the word "bot"

oai_token-act-pair · gpt-3.5-turbo

references to "bot" or "botanical" concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

dress

-0.68

mund

-0.64

PLIED

-0.64

 Trib

-0.61

eways

-0.61

mble

-0.61

journal

-0.61

 Truth

-0.60

baugh

-0.58

 Institution

-0.56

POSITIVE LOGITS

anical

1.31

wana

1.01

anie

0.97

cham

0.95

zeb

0.94

ania

0.94

anic

0.92

Bot

0.91

anooga

0.91

nets

0.90

Activations Density 0.007%

variations of the word "bot"

references to "bot" or "botanical" concepts

No Comments

No Known Activations