INDEX

Explanations

phrases related to opting out or unsubscribing

oai_token-act-pair · gpt-3.5-turbo

phrases related to opting out or cancelling subscriptions

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

pmwiki

-0.59

 Lent

-0.52

 Zimmer

-0.50

RPG

-0.50

prototype

-0.49

LOD

-0.48

escription

-0.48

pher

-0.47

Ort

-0.47

itialized

-0.47

POSITIVE LOGITS

 anytime

0.63

ourt

0.58

pload

0.57

any

0.57

atever

0.57

ritical

0.55

¶æ

0.53

mop

0.53

izons

0.53

£ı

0.52

Activations Density 0.022%

phrases related to opting out or unsubscribing

phrases related to opting out or cancelling subscriptions

No Comments

No Known Activations