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

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.07

2:0.08

3:0.08

4:0.06

5:0.08

6:0.10

7:0.09

8:0.09

9:0.08

10:0.07

11:0.07

Negative Logits

heading

-1.44

 unspecified

-1.39

tesy

-1.38

isite

-1.37

otten

-1.37

lehem

-1.35

romeda

-1.35

lined

-1.34

anced

-1.27

saf

-1.26

POSITIVE LOGITS

 Wiki

1.50

 Architects

1.42

ustomed

1.33

RPG

1.33

 Occupations

1.33

Dun

1.32

mini

1.31

Nation

1.31

Wiki

1.30

ebook

1.30

Activations Density 0.000%

No Known Activations

This feature has no known activations.