INDEX

Explanations

mentions of politicians and political activities

oai_token-act-pair · gpt-3.5-turbo

references to political figures and their positions or actions

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

 disturbance

-0.55

 fins

-0.55

 Frie

-0.51

sap

-0.50

 flux

-0.50

 curv

-0.49

 sharks

-0.49

 menstru

-0.49

arium

-0.49

 destro

-0.49

POSITIVE LOGITS

 Majority

0.63

bernatorial

0.63

bush

0.59

govtrack

0.56

hement

0.56

Leader

0.55

lander

0.54

racial

0.53

 unanimously

0.53

 Legislation

0.53

Activations Density 1.496%

mentions of politicians and political activities

references to political figures and their positions or actions

No Comments

No Known Activations