INDEX

Explanations

mentions of the country "Afghanistan."

oai_token-act-pair · gpt-3.5-turbo

mentions and references to Afghanistan

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

yss

-0.83

vous

-0.83

Hunt

-0.80

creen

-0.77

icles

-0.74

*/(

-0.70

ateur

-0.68

ection

-0.68

amins

-0.68

tainment

-0.68

POSITIVE LOGITS

istan

1.09

 Afghanistan

1.08

ghan

1.05

 Kabul

1.04

 Albania

1.01

 Afghan

0.99

 Afgh

0.98

 Taliban

0.96

 Afghans

0.94

 Pakistan

0.84

Activations Density 0.013%

mentions of the country "Afghanistan."

mentions and references to Afghanistan

No Comments

No Known Activations