INDEX

Explanations

words related to ballot initiatives and political campaigns

oai_token-act-pair · gpt-3.5-turbo

references to ballot initiatives

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

Kut

-0.78

rir

-0.75

hered

-0.73

PLIED

-0.73

ews

-0.68

auga

-0.66

ĺħ

-0.66

elsen

-0.64

sei

-0.64

 tremend

-0.64

POSITIVE LOGITS

 ballots

0.88

zeb

0.87

ination

0.86

 canv

0.81

room

0.78

asso

0.78

boxes

0.77

otle

0.76

 Compass

0.75

eering

0.75

Activations Density 0.016%

words related to ballot initiatives and political campaigns

references to ballot initiatives

No Comments

No Known Activations