INDEX

Explanations

mentions of densely populated areas

oai_token-act-pair · gpt-3.5-turbo

references to population density or populated areas

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aina

-0.93

ift

-0.90

href

-0.80

same

-0.77

hum

-0.76

kick

-0.75

yer

-0.75

utils

-0.73

smoking

-0.73

abol

-0.73

POSITIVE LOGITS

 populated

1.38

 populate

1.10

 inhabited

0.95

ãĤ¤ãĥĪ

0.89

 populous

0.88

 democracies

0.84

 unpop

0.83

 dwellings

0.82

 sectors

0.76

 Pengu

0.76

Activations Density 0.014%

mentions of densely populated areas

references to population density or populated areas

No Comments

No Known Activations