INDEX

Explanations

instances of the word "love."

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

1:0.01

2:0.11

3:0.08

4:0.12

5:0.02

6:0.10

7:0.05

8:0.02

9:0.04

10:0.19

11:0.17

Negative Logits

omorph

-1.33

 anim

-1.32

 Values

-1.26

 antiqu

-1.25

Relations

-1.25

values

-1.25

acons

-1.24

 Parenthood

-1.22

ocrates

-1.21

avers

-1.21

POSITIVE LOGITS

 hurry

1.55

�

1.47

 Cheong

1.42

heard

1.41

 succeeded

1.40

ornia

1.35

 roundup

1.32

soDeliveryDate

1.31

 waited

1.28

isconsin

1.27

Activations Density 0.001%

instances of the word "love."

No Comments

No Known Activations