INDEX

Explanations

phrases that indicate a violation or conflict with certain established standards or regulations

oai_token-act-pair · gpt-3.5-turbo

the repetition of the word "of" and its association with concepts or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 partName

-0.49

 istg

-0.47

forward

-0.47

 Cosponsors

-0.45

helle

-0.45

resy

-0.45

 showc

-0.44

 Canaver

-0.44

webkit

-0.44

Ô

-0.44

POSITIVE LOGITS

 theirs

0.57

 their

0.52

 ours

0.52

 course

0.51

 sorts

0.51

the

0.49

our

0.49

 hers

0.48

 those

0.46

 tomorrow

0.46

Activations Density 0.712%

phrases that indicate a violation or conflict with certain established standards or regulations

the repetition of the word "of" and its association with concepts or processes

No Comments

No Known Activations