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

0:0.07

1:0.08

2:0.07

3:0.08

4:0.10

5:0.08

6:0.08

7:0.07

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

phia

-1.47

��

-1.43

itage

-1.42

alties

-1.34

urances

-1.32

lishing

-1.31

leaf

-1.31

................

-1.28

teness

-1.27

ateurs

-1.25

POSITIVE LOGITS

 Sexy

1.29

ById

1.24

 avatar

1.20

 pseudonym

1.19

 shorthand

1.16

 alias

1.14

 chimpan

1.11

ogging

1.11

imes

1.11

 Kardashian

1.10

Activations Density 0.000%

No Known Activations

This feature has no known activations.