INDEX

Explanations

phrases indicating subscription or support for journals or publications

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

2:0.09

3:0.18

4:0.10

5:0.04

6:0.10

7:0.10

8:0.05

9:0.07

10:0.08

11:0.09

Negative Logits

tsky

-1.42

gery

-1.33

geries

-1.14

itiz

-1.12

 respectively

-1.09

$.

-1.09

_.

-1.08

 Bord

-1.06

»

-1.03

trace

-1.02

POSITIVE LOGITS

オ

1.41

�

1.25

闘

1.23

DragonMagazine

1.22

ギ

1.22

�

1.18

rocal

1.17

�

1.15

ジ

1.11

udo

1.09

Activations Density 0.001%

phrases indicating subscription or support for journals or publications

No Comments

No Known Activations