INDEX

Explanations

emotional expressions of satisfaction or gratification

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

1:0.02

2:0.05

3:0.06

4:0.12

5:0.02

6:0.35

7:0.12

8:0.03

9:0.03

10:0.06

11:0.07

Negative Logits

 Isle

-1.49

nesday

-1.38

 Carbuncle

-1.32

 WORLD

-1.28

enium

-1.25

 Families

-1.25

 Gazette

-1.23

Ll

-1.22

 yacht

-1.21

[|

-1.16

POSITIVE LOGITS

pel

1.66

achy

1.61

angu

1.57

conscious

1.53

ustomed

1.47

iru

1.47

usional

1.43

oslav

1.41

assy

1.38

aspers

1.37

Activations Density 0.032%

emotional expressions of satisfaction or gratification

No Comments

No Known Activations