INDEX

Explanations

content related to alerts and security issues

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

phrases related to video content and multimedia viewing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

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

0:0.02

1:0.09

2:0.07

3:0.09

4:0.03

5:0.16

6:0.07

7:0.09

8:0.07

9:0.11

10:0.08

11:0.07

Negative Logits

cember

-1.41

 Gork

-1.38

zhen

-1.35

ellation

-1.34

ォ

-1.32

 Tend

-1.32

Sov

-1.29

Sap

-1.29

 Turing

-1.27

 hindsight

-1.26

POSITIVE LOGITS

 disabling

1.49

avascript

1.44

 disable

1.41

 loading

1.35

 traffic

1.32

 BELOW

1.28

 seizure

1.27

Submit

1.25

 captcha

1.22

 CLICK

1.21

Activations Density 0.002%

content related to alerts and security issues

phrases related to video content and multimedia viewing

No Comments

No Known Activations