INDEX

Explanations

the word 'the' followed by a noun

oai_token-act-pair · gpt-3.5-turbo

instances of the end-of-text token

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

 alongside

-0.59

 besides

-0.55

—"

-0.55

—-

-0.54

 according

-0.54

Ãĥ

-0.54

 during

-0.53

 elsewhere

-0.53

 beforehand

-0.53

 theirs

-0.52

POSITIVE LOGITS

odore

0.75

resa

0.73

orem

0.69

urgy

0.64

atre

0.63

DragonMagazine

0.62

 holiest

0.61

oret

0.59

 coolest

0.59

ory

0.58

Activations Density 0.155%

the word 'the' followed by a noun

instances of the end-of-text token

No Comments

No Known Activations