INDEX

Explanations

information related to historical events or biographies

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

sections or headings within a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

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

hovah

-0.67

jee

-0.66

querque

-0.59

',

-0.59

ullivan

-0.59

alled

-0.56

isconsin

-0.54

 eleph

-0.54

kef

-0.54

oÄŁ

-0.53

POSITIVE LOGITS

Appearances

0.79

tro

0.75

Gameplay

0.73

Background

0.69

Spoiler

0.67

src

0.66

 Edit

0.65

isode

0.64

pmwiki

0.63

 Animation

0.62

Activations Density 0.038%

information related to historical events or biographies

sections or headings within a document

No Comments

No Known Activations