INDEX

Explanations

mentions of voter-related topics or activities

oai_token-act-pair · gpt-3.5-turbo

mentions of voters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atern

-0.74

uran

-0.72

 Fury

-0.69

ãĤ¨ãĥ«

-0.68

aul

-0.65

artment

-0.64

occ

-0.63

ijk

-0.62

Wilson

-0.61

OUS

-0.61

POSITIVE LOGITS

 voters

1.32

 Voters

1.17

 ballots

0.98

ndum

0.97

oters

0.91

 votes

0.91

 electorate

0.88

 voter

0.87

 ballot

0.84

 voting

0.83

Activations Density 0.011%

mentions of voter-related topics or activities

mentions of voters

No Comments

No Known Activations