INDEX

Explanations

phrases related to showcasing talents, investments, and unique abilities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

0:0.06

1:0.04

2:0.11

3:0.07

4:0.12

5:0.08

6:0.04

7:0.03

8:0.21

9:0.11

10:0.06

11:0.03

Negative Logits

 Maintenance

-1.40

 Debor

-1.32

 Reverend

-1.19

 maintenance

-1.17

 aside

-1.16

appointed

-1.15

 Merrill

-1.14

 Nottingham

-1.14

 reckoning

-1.14

 ascert

-1.14

POSITIVE LOGITS

wcsstore

1.44

Girls

1.43

 feats

1.33

 performances

1.31

emale

1.30

vec

1.25

Pokemon

1.24

APE

1.23

Pokémon

1.22

Kids

1.20

Activations Density 0.007%

phrases related to showcasing talents, investments, and unique abilities

No Comments

No Known Activations