INDEX

Explanations

mentions of voters and voting-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to voting and voters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

SourceFile

-0.90

ity

-0.81

Else

-0.72

 Chef

-0.71

ness

-0.70

cer

-0.69

kefeller

-0.68

uran

-0.68

sa

-0.67

Ski

-0.65

POSITIVE LOGITS

 turnout

0.93

ndum

0.93

 polled

0.88

 electing

0.79

 disenfranch

0.79

 ballots

0.77

hip

0.77

 disillusion

0.75

 voting

0.74

 canv

0.74

Activations Density 0.032%

mentions of voters and voting-related terms

terms related to voting and voters

No Comments

No Known Activations