INDEX

Explanations

words related to self-interest and conflict

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

concepts related to self-interest and exploitation in societal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.05

2:0.10

3:0.14

4:0.05

5:0.09

6:0.03

7:0.04

8:0.06

9:0.20

10:0.07

11:0.04

Negative Logits

anwhile

-1.34

ramid

-1.31

 sidx

-1.28

idium

-1.25

ugu

-1.24

iago

-1.22

untled

-1.21

yip

-1.21

BuyableInstoreAndOnline

-1.15

asury

-1.15

POSITIVE LOGITS

 Buddhism

1.17

dom

1.16

 Conscious

1.16

 alike

1.15

 Catholicism

1.15

 perv

1.14

 metic

1.13

 Mormonism

1.13

 Manifest

1.10

 Nietzsche

1.09

Activations Density 0.033%

words related to self-interest and conflict

concepts related to self-interest and exploitation in societal contexts

No Comments

No Known Activations