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

3:0.07

4:0.08

5:0.09

6:0.08

7:0.07

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

ゼウス

-3.12

soDeliveryDate

-2.68

 Debt

-2.52

 contag

-2.51

 Protest

-2.50

 epid

-2.48

pmwiki

-2.39

 Narc

-2.35

 trillions

-2.34

 Rosenstein

-2.33

POSITIVE LOGITS

++

2.91

+.

2.78

inged

2.47

irlf

2.45

ilk

2.44

+,

2.43

(_

2.42

 skiing

2.37

++

2.34

KDE

2.31

Activations Density 0.000%

No Known Activations

This feature has no known activations.