INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.08

1:0.07

2:0.07

3:0.08

4:0.08

5:0.08

6:0.07

7:0.09

8:0.07

9:0.09

10:0.08

11:0.08

Negative Logits

Zup

-2.77

Nib

-2.49

iewicz

-2.31

stan

-2.28

Py

-2.27

aper

-2.22

 financed

-2.21

lez

-2.21

Ski

-2.21

Sno

-2.17

POSITIVE LOGITS

 violet

2.65

 Celtics

2.37

FTWARE

2.35

aito

2.30

 affirm

2.29

mons

2.27

 Torah

2.26

orate

2.19

til

2.19

 Dare

2.19

Activations Density 0.000%

No Known Activations

This feature has no known activations.