INDEX

Explanations

No Explanations Found

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

0:0.10

1:0.13

2:0.04

3:0.07

4:0.06

5:0.13

6:0.02

7:0.02

8:0.09

9:0.10

10:0.09

11:0.09

Negative Logits

 Strongh

-1.62

phia

-1.55

iddles

-1.46

 Examination

-1.44

 Colon

-1.37

trap

-1.36

fac

-1.36

 Crim

-1.35

lia

-1.34

nesota

-1.34

POSITIVE LOGITS

UTE

1.75

UNCH

1.66

Reviewer

1.62

ICAN

1.56

tumblr

1.51

VK

1.51

DragonMagazine

1.41

 personalized

1.34

Maximum

1.33

DonaldTrump

1.32

Activations Density 0.000%

No Known Activations

This feature has no known activations.