INDEX

Explanations

references to physical locations and actions relating to movement or transportation

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

Head Attr Weights

0:0.03

1:0.03

2:0.13

3:0.07

4:0.29

5:0.04

6:0.04

7:0.12

8:0.03

9:0.03

10:0.09

11:0.05

Negative Logits

 Flavoring

-1.78

fter

-1.27

Factor

-1.25

 mater

-1.20

adding

-1.18

я

-1.17

ankind

-1.17

lance

-1.17

Reviewer

-1.15

 proportion

-1.15

POSITIVE LOGITS

yards

1.53

armac

1.37

jug

1.37

 benches

1.36

nih

1.36

 aisle

1.34

 scenic

1.33

 stairs

1.32

 extremes

1.30

 ditch

1.29

Activations Density 0.004%

references to physical locations and actions relating to movement or transportation

No Known Activations