INDEX

Explanations

proper nouns and terms related to legal and political issues

oai_token-act-pair · gpt-3.5-turbo

references to controversial or significant issues and topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ĸļ

-0.63

 Pyro

-0.60

 Soph

-0.56

¥µ

-0.56

ahime

-0.55

OTAL

-0.55

ptroller

-0.53

umar

-0.52

FI

-0.52

 Romeo

-0.52

POSITIVE LOGITS

 Haram

0.54

Ħ¢

0.52

 poses

0.50

ervatives

0.50

 flour

0.49

MIN

0.48

 Getty

0.48

 supporters

0.48

ervative

0.48

 violates

0.47

Activations Density 0.671%

proper nouns and terms related to legal and political issues

references to controversial or significant issues and topics

No Comments

No Known Activations