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

1:0.07

2:0.08

3:0.09

4:0.08

5:0.07

6:0.06

7:0.07

8:0.08

9:0.08

10:0.07

11:0.10

Negative Logits

"}

-3.47

 Pengu

-2.96

asio

-2.54

ulet

-2.45

Sing

-2.45

Emer

-2.45

Remote

-2.43

 Emin

-2.38

thood

-2.37

Magikarp

-2.35

POSITIVE LOGITS

MIL

2.58

 Marines

2.47

KH

2.46

jury

2.37

 Chero

2.36

POW

2.35

Hil

2.27

 protective

2.26

 toughness

2.24

 Buchanan

2.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.