INDEX

Explanations

details related to laws or regulations

oai_token-act-pair · gpt-3.5-turbo

phrases concerning social policies and discrimination issues

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

itiveness

-0.60

Drag

-0.60

polit

-0.56

âĦ¢:

-0.55

\":

-0.54

 nutshell

-0.50

omics

-0.50

ISS

-0.49

atch

-0.49

gem

-0.48

POSITIVE LOGITS

 beware

0.65

instead

0.64

 instead

0.56

anwhile

0.55

CHO

0.53

 thereby

0.53

entimes

0.52

 presumably

0.52

 draining

0.51

 soaking

0.50

Activations Density 1.970%

details related to laws or regulations

phrases concerning social policies and discrimination issues

No Comments

No Known Activations