INDEX

Explanations

people's names related to different crime incidents

oai_token-act-pair · gpt-3.5-turbo

names and identifiers associated with people involved in criminal activities or incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 conventions

-0.78

 endorsements

-0.77

 understatement

-0.74

 announcements

-0.71

 resolutions

-0.70

ournal

-0.70

 insights

-0.69

 analytics

-0.69

 pointers

-0.69

 legislatures

-0.68

POSITIVE LOGITS

Jr

1.32

III

1.21

Sr

1.13

aka

0.98

 Tsarnaev

0.98

vich

0.94

endez

0.91

II

0.91

IV

0.90

gui

0.90

Activations Density 0.261%

people's names related to different crime incidents

names and identifiers associated with people involved in criminal activities or incidents

No Comments

No Known Activations