INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.04

2:0.09

3:0.07

4:0.08

5:0.09

6:0.09

7:0.10

8:0.07

9:0.09

10:0.08

11:0.08

Negative Logits

 compan

-1.52

 reconc

-1.52

terness

-1.47

kj

-1.44

isms

-1.42

 cruc

-1.42

 simultane

-1.40

omach

-1.39

cens

-1.38

Ranked

-1.36

POSITIVE LOGITS

 Conservation

1.38

1.29

 Crate

1.26

 Kick

1.23

 Finish

1.22

 Avalanche

1.19

ample

1.18

 Advantage

1.18

 Trick

1.17

 Corps

1.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.