INDEX

Explanations

references to individuals named "Jud"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Jud" and its variations, related to discussions about judges or judicial matters

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

 Afric

-0.69

 Mali

-0.66

fields

-0.65

Ø§ÙĦ

-0.65

xon

-0.64

forth

-0.64

xual

-0.64

oteric

-0.63

ULAR

-0.61

 Kerala

-0.61

POSITIVE LOGITS

gement

1.46

iciary

1.39

gment

1.27

gments

1.24

icial

1.24

jud

1.19

icator

1.01

Jud

1.00

gements

0.97

ging

0.97

Activations Density 0.014%

references to individuals named "Jud"

occurrences of the name "Jud" and its variations, related to discussions about judges or judicial matters

No Comments

No Known Activations