INDEX

Explanations

phrases that involve measurement or control of physical entities

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

0:0.01

1:0.02

2:0.08

3:0.04

4:0.14

5:0.02

6:0.07

7:0.34

8:0.02

9:0.03

10:0.08

11:0.08

Negative Logits

 Pact

-1.47

ocre

-1.35

���

-1.34

nces

-1.34

��極

-1.33

gery

-1.32

Fav

-1.30

Zak

-1.29

 pays

-1.28

 prejud

-1.26

POSITIVE LOGITS

 joystick

1.67

 stairs

1.60

animate

1.60

 mouse

1.57

 cursor

1.56

 arrows

1.56

�

1.55

 buttons

1.54

 gaze

1.51

 hither

1.51

Activations Density 0.009%

phrases that involve measurement or control of physical entities

No Known Activations