INDEX

Explanations

quotations in text

oai_token-act-pair · gpt-3.5-turbo

dialogue and expressions attributed to speakers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 vulner

-0.50

 brill

-0.47

NetMessage

-0.44

 sacrific

-0.42

omorphic

-0.42

 este

-0.41

 simulac

-0.41

ographical

-0.41

 corrections

-0.40

ework

-0.40

POSITIVE LOGITS

 Needless

0.56

ONSORED

0.56

elvet

0.52

 Lastly

0.49

swick

0.48

emphasis

0.47

iHUD

0.46

 Flavoring

0.46

 Whereas

0.45

 Alternatively

0.45

Activations Density 0.865%

quotations in text

dialogue and expressions attributed to speakers

No Comments

No Known Activations