INDEX

Explanations

cities and locations, particularly police departments and specific enforcement agencies

oai_token-act-pair · gpt-3.5-turbo

references to specific cities and locations

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

morrow

-0.76

alin

-0.75

thia

-0.74

ãĥĦ

-0.73

renheit

-0.73

DonaldTrump

-0.71

cedes

-0.71

thood

-0.71

riel

-0.71

uria

-0.69

POSITIVE LOGITS

 metropolitan

1.04

 skyline

1.01

 Metropolitan

1.00

 branch

0.99

 suburb

0.97

 area

0.97

 Police

0.95

 suburbs

0.95

 chapter

0.95

 Convention

0.93

Activations Density 0.130%

cities and locations, particularly police departments and specific enforcement agencies

references to specific cities and locations

No Comments

No Known Activations