INDEX

Explanations

words related to censorship

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.07

1:0.08

2:0.07

3:0.09

4:0.08

5:0.08

6:0.09

7:0.08

8:0.07

9:0.07

10:0.08

11:0.06

Negative Logits

 livest

-2.63

 colle

-2.45

 defe

-2.36

ewater

-2.36

 Dunn

-2.30

 Rove

-2.28

Closure

-2.28

utorial

-2.26

oult

-2.25

Capt

-2.24

POSITIVE LOGITS

NEO

3.17

NK

2.91

IST

2.85

Soy

2.81

 Airbus

2.76

 Nissan

2.76

 Mazda

2.73

 Musk

2.72

Yar

2.71

ismo

2.70

Activations Density 0.000%

words related to censorship

No Comments

No Known Activations