INDEX

Explanations

phrases indicating ownership or personal perspectives

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

2:0.17

3:0.15

4:0.14

5:0.05

6:0.08

7:0.06

8:0.05

9:0.09

10:0.08

11:0.04

Negative Logits

 beck

-1.36

 Pist

-1.31

bub

-1.28

OPA

-1.27

gged

-1.25

omed

-1.24

 Samar

-1.22

 magically

-1.21

 Kenn

-1.20

lite

-1.18

POSITIVE LOGITS

 CLSID

1.62

ibliography

1.58

 裏�

1.57

フ

1.54

enhagen

1.51

ンジ

1.48

エル

1.45

ividual

1.42

1.39

 david

1.37

Activations Density 0.002%

phrases indicating ownership or personal perspectives

No Comments

No Known Activations