INDEX

Explanations

symbols and unique characters

New Auto-Interp

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

Head Attr Weights

0:0.06

1:0.02

2:0.13

3:0.07

4:0.04

5:0.03

6:0.35

7:0.03

8:0.03

9:0.08

10:0.05

11:0.05

Negative Logits

zona

-1.87

quartered

-1.41

 fertil

-1.35

 challeng

-1.28

 tuna

-1.28

 snipers

-1.26

wid

-1.25

xus

-1.23

 Loch

-1.21

 aggregation

-1.21

POSITIVE LOGITS

talk

1.65

essen

1.48

iren

1.47

uliffe

1.41

ament

1.32

imen

1.31

ê

1.30

agogue

1.29

enter

1.29

POSE

1.28

Activations Density 0.001%

symbols and unique characters

No Known Activations