INDEX

Explanations

references to car models and companies

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly brands, companies, and notable figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 lightly

-0.55

 ðŁĳ

-0.49

 VERS

-0.48

 seriously

-0.47

 accredited

-0.46

 concise

-0.45

âĢİ

-0.45

 indeed

-0.45

AND

-0.45

 today

-0.45

POSITIVE LOGITS

axter

0.64

 thereof

0.63

 respectively

0.59

 thereto

0.56

ccoli

0.55

itto

0.54

aughed

0.54

undai

0.53

cellaneous

0.52

oultry

0.52

Activations Density 0.963%

references to car models and companies

proper nouns, particularly brands, companies, and notable figures

No Comments

No Known Activations