INDEX

Explanations

instances of the word "vote" with varying relevance

oai_token-act-pair · gpt-3.5-turbo

references to voting and electoral processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 unden

-0.69

 millenn

-0.69

 Palestin

-0.66

Kut

-0.65

 Flavoring

-0.65

algia

-0.64

 thous

-0.61

 colon

-0.61

 pilgr

-0.60

idays

-0.60

POSITIVE LOGITS

vote

0.94

 totals

0.90

board

0.88

 tally

0.85

 votes

0.81

 cast

0.81

 count

0.81

 bloc

0.80

 counted

0.78

boarding

0.75

Activations Density 0.044%

instances of the word "vote" with varying relevance

references to voting and electoral processes

No Comments

No Known Activations