INDEX

Explanations

prominent names and titles in various contexts

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

0:0.02

1:0.04

2:0.04

3:0.08

4:0.05

5:0.04

6:0.46

7:0.02

8:0.04

9:0.06

10:0.07

11:0.05

Negative Logits

ecause

-1.43

eg

-1.26

ographed

-1.26

 redund

-1.22

ittees

-1.22

perty

-1.21

 tuition

-1.20

ometimes

-1.20

 theirs

-1.18

 gown

-1.17

POSITIVE LOGITS

 Awakens

1.58

arth

1.53

LOCK

1.46

ゼ

1.38

 Shed

1.36

 Unleashed

1.35

 Execution

1.32

ケ

1.32

 Ahead

1.30

 Difference

1.29

Activations Density 0.094%

prominent names and titles in various contexts

No Comments

No Known Activations