INDEX

Explanations

phrases related to states or governmental entities

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "state" and its context within governance or legislation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

icago

-0.89

aurus

-0.83

ogle

-0.79

eger

-0.78

 Aires

-0.75

byn

-0.69

poons

-0.69

nob

-0.68

lyak

-0.68

aq

-0.66

POSITIVE LOGITS

 legislature

1.30

 Legislature

1.25

 legislatures

1.23

 troopers

1.18

 trooper

1.16

wide

1.10

 legislator

1.10

 senate

1.05

 senator

1.00

 legislators

0.99

Activations Density 0.058%

phrases related to states or governmental entities

mentions of the word "state" and its context within governance or legislation

No Comments

No Known Activations