INDEX

Explanations

the presence of the word "to" indicating various actions or intentions

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

0:0.01

1:0.01

2:0.06

3:0.07

4:0.05

5:0.03

6:0.35

7:0.18

8:0.04

9:0.03

10:0.06

11:0.05

Negative Logits

 Canal

-1.43

ipel

-1.41

wikipedia

-1.32

forth

-1.25

 Tycoon

-1.24

─

-1.23

�

-1.23

Loop

-1.22

 Falk

-1.21

dm

-1.19

POSITIVE LOGITS

llah

1.49

rows

1.38

itto

1.30

 passers

1.28

earances

1.27

ample

1.26

 continu

1.25

 recip

1.25

gat

1.25

 confir

1.25

Activations Density 0.013%

the presence of the word "to" indicating various actions or intentions

No Comments

No Known Activations