INDEX

Explanations

phrases indicating action or change in context

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

0:0.02

1:0.01

2:0.08

3:0.08

4:0.21

5:0.03

6:0.03

7:0.26

8:0.03

9:0.05

10:0.09

11:0.05

Negative Logits

BuyableInstoreAndOnline

-1.45

quartered

-1.44

currently

-1.44

disabled

-1.43

aligned

-1.43

eligible

-1.40

neutral

-1.40

placed

-1.39

irable

-1.39

icted

-1.39

POSITIVE LOGITS

 herd

1.42

 reins

1.33

 wedge

1.31

 youthful

1.27

 defenses

1.25

 cynicism

1.21

 sarc

1.19

 partisans

1.19

 expectations

1.18

 defences

1.18

Activations Density 0.163%

phrases indicating action or change in context

No Comments

No Known Activations