INDEX

Explanations

references to Catholic institutions and beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.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

0:0.03

1:0.02

2:0.06

3:0.14

4:0.04

5:0.04

6:0.21

7:0.16

8:0.04

9:0.03

10:0.09

11:0.07

Negative Logits

��

-1.52

��

-1.20

 teamwork

-1.15

eki

-1.05

azor

-0.98

 Ogre

-0.97

 pandemonium

-0.96

omorph

-0.96

Downloadha

-0.94

 seams

-0.94

POSITIVE LOGITS

ACP

1.22

 Sonia

1.15

 Teresa

1.11

 Liberties

1.06

 Association

1.06

 Socialist

1.03

hemy

1.01

Universal

0.99

 Catherine

0.99

phia

0.99

Activations Density 0.014%

references to Catholic institutions and beliefs

No Comments

No Known Activations