INDEX

Explanations

elements of data or statistics related to structured lists or categories

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

0:0.10

1:0.04

2:0.08

3:0.05

4:0.04

5:0.05

6:0.17

7:0.03

8:0.07

9:0.25

10:0.03

11:0.04

Negative Logits

Lew

-4.19

Gy

-3.76

HUD

-3.72

zb

-3.49

Bie

-3.40

 Gund

-3.28

 Yugoslavia

-3.16

 Croatian

-3.16

Lew

-3.10

 Hamm

-3.04

POSITIVE LOGITS

 Express

8.06

Express

7.81

express

7.66

 express

7.31

expression

5.53

 Expression

5.32

 expressing

4.64

 expressions

4.48

 expression

4.36

 expresses

4.20

Activations Density 0.005%

elements of data or statistics related to structured lists or categories

No Comments

No Known Activations