INDEX

Explanations

instances of the word "vote"

oai_token-act-pair · gpt-3.5-turbo

references to voting

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

ptoms

-0.78

DragonMagazine

-0.78

emer

-0.75

peria

-0.74

cel

-0.72

 millenn

-0.71

 Flavoring

-0.69

etheless

-0.68

 Palestin

-0.68

Whit

-0.68

POSITIVE LOGITS

 Leave

0.82

room

0.81

 Vote

0.80

rooms

0.79

 votes

0.79

vote

0.77

 irregularities

0.74

 ballots

0.74

arate

0.73

 vote

0.72

Activations Density 0.026%

instances of the word "vote"

references to voting

No Comments

No Known Activations