INDEX

Explanations

mentions of government officials or members of parliament

oai_token-act-pair · gpt-3.5-turbo

words related to governmental officials or cabinet members

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Samurai

-0.79

 Nation

-0.75

Mine

-0.69

ASED

-0.68

 Participant

-0.67

DonaldTrump

-0.65

Listener

-0.65

 Campus

-0.65

gran

-0.64

Fu

-0.64

POSITIVE LOGITS

 ministers

1.14

arians

1.07

arian

0.93

 Ministers

0.88

 ministerial

0.88

 minister

0.86

hips

0.85

utical

0.80

isters

0.79

 secretaries

0.78

Activations Density 0.027%

mentions of government officials or members of parliament

words related to governmental officials or cabinet members

No Comments

No Known Activations