INDEX

Explanations

instances of questioning or interrogation

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

1:0.03

2:0.08

3:0.12

4:0.02

5:0.05

6:0.06

7:0.20

8:0.07

9:0.13

10:0.08

11:0.11

Negative Logits

lins

-1.13

 Printed

-1.10

Wra

-1.02

sterdam

-1.00

ahime

-0.99

 Babel

-0.99

 Printing

-0.98

EEE

-0.98

ーク

-0.97

 Shant

-0.96

POSITIVE LOGITS

 motives

1.25

 elicit

1.24

 incrim

1.18

answered

1.13

 patriotism

1.05

 ambush

1.04

 loyalty

1.04

 credentials

1.03

essional

1.01

 motivations

1.00

Activations Density 0.008%

instances of questioning or interrogation

No Comments

No Known Activations