INDEX

Explanations

politically-related terms and statements

oai_token-act-pair · gpt-3.5-turbo

references to political figures and government actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

MpServer

-0.74

occ

-0.67

Pv

-0.65

exist

-0.64

Tor

-0.63

Atl

-0.63

 classmate

-0.62

Tek

-0.62

acters

-0.62

ahi

-0.62

POSITIVE LOGITS

 appoint

1.05

 legisl

0.93

erity

0.92

 appointing

0.91

 aides

0.86

 policymakers

0.84

 approving

0.84

 cabinet

0.83

 appointed

0.83

 administrations

0.83

Activations Density 0.908%

politically-related terms and statements

references to political figures and government actions

No Comments

No Known Activations