INDEX

Explanations

email addresses

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

references to information sources and communication methods

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

Head Attr Weights

0:0.06

1:0.03

2:0.18

3:0.06

4:0.09

5:0.06

6:0.04

7:0.03

8:0.10

9:0.21

10:0.05

11:0.03

Negative Logits

 principled

-1.37

wcs

-1.25

 ILCS

-1.20

 toler

-1.19

��

-1.18

faced

-1.17

milo

-1.16

 succeeded

-1.13

SPONSORED

-1.11

 ethic

-1.11

POSITIVE LOGITS

 Stard

1.32

 whereabouts

1.20

ascript

1.16

apr

1.12

 Wear

1.12

 wearer

1.12

glers

1.11

inarily

1.11

 vendor

1.10

 Presents

1.10

Activations Density 0.006%

email addresses

references to information sources and communication methods

No Known Activations