INDEX

Explanations

issues related to politics, governance, and economic control by powerful entities

oai_token-act-pair · gpt-3.5-turbo

phrases related to corporate influence and power in society

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

yssey

-0.71

mary

-0.67

canon

-0.65

 Dresden

-0.63

 condol

-0.62

 Ramadan

-0.62

ortality

-0.62

ppo

-0.61

estyles

-0.61

 Islands

-0.61

POSITIVE LOGITS

 corporations

1.49

 bureaucrats

1.35

 billionaires

1.28

 elites

1.27

 vested

1.22

 greedy

1.22

 dictators

1.19

 lobbyists

1.18

 wealthy

1.18

 unaccount

1.16

Activations Density 0.375%

issues related to politics, governance, and economic control by powerful entities

phrases related to corporate influence and power in society

No Comments

No Known Activations