INDEX

Explanations

phrases indicating an increase in quantity or activity

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

0:0.10

1:0.03

2:0.08

3:0.07

4:0.06

5:0.08

6:0.04

7:0.26

8:0.04

9:0.04

10:0.11

11:0.04

Negative Logits

utral

-2.51

 deter

-2.47

 disappro

-2.25

 disbel

-2.14

)</

-2.13

 damp

-2.07

 dissu

-2.05

 Disp

-2.04

 phased

-2.04

 neutrality

-1.99

POSITIVE LOGITS

apan

3.22

umbn

3.14

ila

2.77

ebus

2.47

pas

2.41

hire

2.30

iverpool

2.23

airo

2.22

ilian

2.20

的

2.20

Activations Density 0.000%

phrases indicating an increase in quantity or activity

No Comments

No Known Activations