INDEX

Explanations

Keywords related to negative characteristics, controversies, and political figures

oai_token-act-pair · gpt-3.5-turbo

references to dramatic or significant events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

essor

-0.64

 sidx

-0.63

afety

-0.62

ect

-0.59

idge

-0.59

peg

-0.57

lder

-0.56

 referral

-0.55

enhagen

-0.55

 affiliate

-0.54

POSITIVE LOGITS

\\\\\\\\

0.64

 humankind

0.62

 uttered

0.59

 culmin

0.59

ulhu

0.58

Mur

0.58

 akin

0.57

selage

0.57

 culminated

0.56

 Starfleet

0.56

Activations Density 1.382%

Keywords related to negative characteristics, controversies, and political figures

references to dramatic or significant events

No Comments

No Known Activations