INDEX

Explanations

questions starting with the word "What"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "What" and its variations, particularly in questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

interstitial

-0.73

-0.62

Ń·

-0.62

Bey

-0.62

Discover

-0.58

atis

-0.56

recy

-0.55

ffe

-0.55

ean

-0.55

uchs

-0.54

POSITIVE LOGITS

soever

1.43

 happened

1.15

 happens

1.11

?!

1.05

 else

1.01

 happ

1.00

?!"

0.97

!?

0.95

!?"

0.88

 bothers

0.83

Activations Density 0.077%

questions starting with the word "What"

instances of the word "What" and its variations, particularly in questions

No Comments

No Known Activations