INDEX

Explanations

words related to military or authoritative structures

oai_token-act-pair · gpt-3.5-turbo

specific names, places, or numerical values mentioned in the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ties

-0.52

tons

-0.50

otin

-0.48

dollar

-0.47

Gap

-0.47

vironment

-0.46

addons

-0.46

Benz

-0.45

++++

-0.45

-------------

-0.45

POSITIVE LOGITS

rul

0.59

 awoken

0.55

 condem

0.53

 cryst

0.53

 exting

0.51

idious

0.51

 Voting

0.49

 shenan

0.49

espie

0.49

 Brilliant

0.48

Activations Density 3.507%

words related to military or authoritative structures

specific names, places, or numerical values mentioned in the document

No Comments

No Known Activations