INDEX

Explanations

phrases indicating a social norm or common practice

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

2:0.10

3:0.07

4:0.28

5:0.02

6:0.02

7:0.22

8:0.04

9:0.04

10:0.07

11:0.07

Negative Logits

urat

-1.46

uctions

-1.41

 prematurely

-1.39

eness

-1.36

antic

-1.35

raped

-1.33

 sucked

-1.32

 extracted

-1.31

 healed

-1.31

ournal

-1.30

POSITIVE LOGITS

 throughout

1.45

��

1.43

Nation

1.41

OPLE

1.37

Elsewhere

1.35

 Category

1.30

Lum

1.30

 Amnesty

1.27

Ic

1.27

Geh

1.26

Activations Density 0.000%

phrases indicating a social norm or common practice

No Comments

No Known Activations