INDEX

Explanations

political figures and related terms, particularly related to the Democratic Party

oai_token-act-pair · gpt-3.5-turbo

names of political figures and references to political organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Jericho

-0.84

 Newport

-0.75

obyl

-0.74

omsday

-0.73

 Viking

-0.72

esan

-0.70

ãĤ¤ãĥĪ

-0.68

 Rapt

-0.67

 Rutherford

-0.67

icester

-0.66

POSITIVE LOGITS

 Schultz

0.93

urations

0.80

 Wasserman

0.78

umenthal

0.78

ilon

0.72

insky

0.67

Leaks

0.65

 assassinated

0.65

ultz

0.64

 fing

0.62

Activations Density 0.069%

political figures and related terms, particularly related to the Democratic Party

names of political figures and references to political organizations

No Comments

No Known Activations