INDEX

Explanations

references to significant achievements and long-term efforts

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_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.05

1:0.04

2:0.10

3:0.08

4:0.16

5:0.08

6:0.04

7:0.03

8:0.12

9:0.15

10:0.06

11:0.03

Negative Logits

bors

-1.32

 beverages

-1.28

 proxies

-1.23

 bedrooms

-1.22

 Hamb

-1.21

horn

-1.21

heit

-1.16

ervatives

-1.14

sensitive

-1.13

 fries

-1.13

POSITIVE LOGITS

 teamwork

1.41

 Congratulations

1.34

 convergence

1.30

=~=~

1.30

ADRA

1.28

izational

1.25

operation

1.24

BuyableInstoreAndOnline

1.23

\\\\\\\\

1.23

Completed

1.22

Activations Density 0.001%

references to significant achievements and long-term efforts

No Comments

No Known Activations