INDEX

Explanations

sentences that contain punctuation

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

New Auto-Interp

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.02

1:0.01

2:0.06

3:0.06

4:0.15

5:0.02

6:0.30

7:0.05

8:0.11

9:0.05

10:0.06

11:0.06

Negative Logits

 Corm

-1.42

 Roses

-1.37

 Newman

-1.29

Uz

-1.29

Cla

-1.27

��

-1.26

 retirees

-1.26

 Sheen

-1.26

 Pesh

-1.26

Uh

-1.24

POSITIVE LOGITS

acceptable

1.62

legraph

1.58

ifix

1.55

exit

1.46

etheless

1.46

shoot

1.46

hesion

1.46

Print

1.46

 cannabin

1.43

ved

1.41

Activations Density 0.005%

sentences that contain punctuation

No Comments

No Known Activations