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

2:0.09

3:0.09

4:0.06

5:0.09

6:0.07

7:0.07

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

 Puzzle

-2.87

 Puzz

-2.56

 Completed

-2.47

 Promise

-2.42

 Games

-2.41

 impe

-2.37

merce

-2.33

osuke

-2.31

 Poland

-2.28

........

-2.26

POSITIVE LOGITS

 Swanson

3.50

 Heller

2.97

ariat

2.80

 Gallagher

2.71

 Holden

2.65

 Hayes

2.63

 Hanson

2.60

vari

2.58

 observational

2.54

 Roth

2.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.