INDEX

Explanations

terms related to specific geographical locations and organizations

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names

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

merce

-0.65

 Reincarn

-0.64

 SERV

-0.64

 WARN

-0.64

ãģ®éŃĶ

-0.63

 PROC

-0.62

ÙĴ

-0.62

llah

-0.62

 thous

-0.62

ÙĲ

-0.61

POSITIVE LOGITS

heed

1.03

emon

0.84

ttle

0.81

ounge

0.81

achev

0.78

igans

0.78

vel

0.77

henko

0.77

anova

0.76

imer

0.75

Activations Density 0.246%

terms related to specific geographical locations and organizations

proper nouns and names

No Comments

No Known Activations