INDEX

Explanations

commands related to game updates and maintenance actions

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_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

Not in Any Lists

No Comments

Head Attr Weights

0:0.29

1:0.04

2:0.06

3:0.08

4:0.06

5:0.01

6:0.08

7:0.15

8:0.01

9:0.03

10:0.09

11:0.03

Negative Logits

enegger

-3.71

 Hendricks

-3.49

 Friedman

-3.41

 Suzuki

-3.37

Hamilton

-3.37

 Twain

-3.32

 motorcycles

-3.18

Ford

-3.17

BMW

-3.17

Kaw

-3.17

POSITIVE LOGITS

 Bungie

8.03

 Destiny

7.72

Dest

4.93

 Grimoire

4.32

 Guardians

4.24

 Activision

4.00

 Andromeda

3.89

romeda

3.82

 Osiris

3.75

 Halo

3.73

Activations Density 0.007%

commands related to game updates and maintenance actions

No Comments

No Known Activations