INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

0:0.08

1:0.08

2:0.08

3:0.08

4:0.08

5:0.09

6:0.08

7:0.08

8:0.08

9:0.07

10:0.09

11:0.07

Negative Logits

 Aires

-2.71

ulner

-2.46

bright

-2.43

 Soph

-2.33

irens

-2.30

scl

-2.27

hea

-2.25

tmp

-2.23

 Bohem

-2.22

nel

-2.19

POSITIVE LOGITS

 Publication

2.63

ipedia

2.51

 textual

2.50

 encyclopedia

2.46

 preferential

2.39

iety

2.37

 aggregate

2.32

 favorably

2.30

 Publications

2.26

 Article

2.26

Activations Density 0.000%

No Known Activations

This feature has no known activations.