INDEX

Explanations

expressions indicating completion or status updates of actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.02

2:0.20

3:0.09

4:0.14

5:0.04

6:0.14

7:0.09

8:0.04

9:0.03

10:0.06

11:0.05

Negative Logits

 Cosponsors

-1.61

pring

-1.43

inline

-1.36

 Winged

-1.27

eed

-1.24

catentry

-1.22

campus

-1.20

estic

-1.18

 earliest

-1.17

istar

-1.17

POSITIVE LOGITS

})

1.43

 Virtue

1.26

]}

1.25

ECK

1.24

 Mahjong

1.22

�

1.18

 Vegas

1.18

>)

1.16

])

1.16

 Herz

1.16

Activations Density 0.001%

expressions indicating completion or status updates of actions

No Comments

No Known Activations