INDEX

Explanations

conditional phrases and statements that express hypotheticals or possibilities

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_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

0:0.01

1:0.01

2:0.06

3:0.11

4:0.27

5:0.06

6:0.05

7:0.07

8:0.10

9:0.06

10:0.06

11:0.08

Negative Logits

=>

-1.69

NetMessage

-1.68

�

-1.64

OSP

-1.39

Downloadha

-1.38

verages

-1.36

merce

-1.34

Compar

-1.32

��

-1.32

DragonMagazine

-1.32

POSITIVE LOGITS

 stayed

1.41

 stay

1.34

 stays

1.29

hit

1.25

sho

1.20

 recons

1.19

 younger

1.18

 frontman

1.17

 bumped

1.17

 cheated

1.16

Activations Density 0.093%

conditional phrases and statements that express hypotheticals or possibilities

No Comments

No Known Activations