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

0:0.08

1:0.07

2:0.07

3:0.08

4:0.09

5:0.08

6:0.07

7:0.09

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

aughs

-1.52

utherford

-1.51

olson

-1.47

oult

-1.45

pez

-1.40

 Rutherford

-1.39

uliffe

-1.39

UGH

-1.37

 millenn

-1.37

 tradem

-1.35

POSITIVE LOGITS

esa

1.37

 Chimera

1.36

vre

1.35

 Liang

1.34

 wreckage

1.31

Bai

1.29

 confines

1.28

��極

1.28

 Petra

1.24

 Mesa

1.24

Activations Density 0.000%

No Known Activations

This feature has no known activations.