INDEX

Explanations

parenthetical statements or references within the text

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

0:0.08

1:0.09

2:0.08

3:0.08

4:0.07

5:0.08

6:0.08

7:0.07

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

 Stanton

-2.51

 Speedway

-2.33

baugh

-2.31

 Ralph

-2.31

 strut

-2.28

 Harriet

-2.27

HRC

-2.25

RFC

-2.24

 patented

-2.24

NRS

-2.23

POSITIVE LOGITS

 Nightmares

2.60

hatt

2.41

oks

2.41

sheets

2.39

allah

2.27

oux

2.27

inged

2.26

icated

2.26

hab

2.25

NULL

2.20

Activations Density 0.000%

parenthetical statements or references within the text

No Comments

No Known Activations