INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.07

1:0.07

2:0.09

3:0.09

4:0.07

5:0.08

6:0.07

7:0.08

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

wat

-1.48

endo

-1.44

oday

-1.37

etta

-1.36

pak

-1.34

 Symphony

-1.32

kun

-1.32

 Austral

-1.29

Austral

-1.29

Sword

-1.28

POSITIVE LOGITS

 recess

1.42

entary

1.41

 pandemonium

1.38

 strangers

1.37

 predicament

1.36

═

1.35

 aisle

1.34

 childish

1.33

 locker

1.32

 deprivation

1.31

Activations Density 0.000%

No Known Activations

This feature has no known activations.