INDEX

Explanations

HTML attributes and parameters within a code structure

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

1:0.03

2:0.26

3:0.10

4:0.06

5:0.04

6:0.07

7:0.05

8:0.05

9:0.06

10:0.08

11:0.13

Negative Logits

 RCMP

-1.27

 comprom

-1.27

 expulsion

-1.25

 mural

-1.24

IPM

-1.23

cus

-1.22

 juven

-1.21

 Offensive

-1.18

 Lans

-1.18

 turf

-1.17

POSITIVE LOGITS

iji

1.52

Mouse

1.33

ongyang

1.30

atson

1.30

votes

1.29

mouse

1.29

ouse

1.28

poke

1.27

UTF

1.27

ongh

1.27

Activations Density 0.007%

HTML attributes and parameters within a code structure

No Comments

No Known Activations