INDEX

Explanations

references to coaches

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_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.9.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

0:0.09

1:0.07

2:0.08

3:0.08

4:0.08

5:0.07

6:0.09

7:0.08

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

scl

-2.85

 Caldwell

-2.68

)=(

-2.59

cock

-2.56

Fla

-2.52

 shenan

-2.41

 Shuttle

-2.40

CB

-2.36

penter

-2.34

 Michaels

-2.30

POSITIVE LOGITS

 Niger

2.80

Europe

2.40

ashtra

2.38

hunt

2.29

 infancy

2.28

NAT

2.23

 doping

2.22

 FISA

2.19

UK

2.18

 Disease

2.18

Activations Density 0.000%

references to coaches

No Comments

No Known Activations