INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.07

1:0.06

2:0.07

3:0.07

4:0.09

5:0.08

6:0.08

7:0.08

8:0.08

9:0.09

10:0.09

11:0.07

Negative Logits

 reflex

-1.57

 regulators

-1.50

 ensures

-1.37

 inhib

-1.36

*/

-1.34

 tech

-1.29

 intern

-1.27

 defund

-1.27

 detects

-1.26

 mounting

-1.26

POSITIVE LOGITS

 largeDownload

1.94

buquerque

1.58

soDeliveryDate

1.56

verning

1.53

Vote

1.52

ighed

1.44

arial

1.42

 Newsp

1.40

liest

1.33

ワン

1.31

Activations Density 0.000%

No Known Activations

This feature has no known activations.