INDEX

Explanations

No Explanations Found

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

0:0.05

1:0.04

2:0.10

3:0.09

4:0.11

5:0.05

6:0.06

7:0.10

8:0.07

9:0.06

10:0.11

11:0.11

Negative Logits

alien

-1.47

uchin

-1.45

adium

-1.36

inton

-1.32

yles

-1.30

cellent

-1.29

tube

-1.29

 defective

-1.28

oner

-1.27

 Guantanamo

-1.26

POSITIVE LOGITS

UTC

1.54

notations

1.52

Tro

1.52

fml

1.40

BIP

1.36

 thereafter

1.36

Malley

1.36

sequently

1.33

DragonMagazine

1.31

 Cheong

1.30

Activations Density 0.000%

No Known Activations

This feature has no known activations.