INDEX

Explanations

names of people and places

oai_token-act-pair · gpt-3.5-turbo

references to food and dining experiences

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 anecd

-0.57

 encount

-0.57

 Ambro

-0.56

 hemor

-0.54

phasis

-0.54

eatures

-0.53

 pestic

-0.53

 Strait

-0.50

 Hilbert

-0.50

 Azerb

-0.49

POSITIVE LOGITS

 anymore

0.54

?!

0.52

 tattoo

0.52

 selfies

0.50

 raping

0.49

 Joker

0.49

quit

0.49

 tattoos

0.49

 porn

0.49

issors

0.48

Activations Density 1.466%

names of people and places

references to food and dining experiences

No Comments

No Known Activations