INDEX

Explanations

references to authority figures and their decisions

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.04

1:0.05

2:0.03

3:0.07

4:0.05

5:0.04

6:0.24

7:0.15

8:0.02

9:0.08

10:0.04

11:0.13

Negative Logits

 subscriptions

-3.20

 subscription

-2.96

 subscribed

-2.74

 subscribe

-2.60

 Patreon

-2.53

 subscrib

-2.46

acus

-2.44

◼

-2.40

amental

-2.40

 filibuster

-2.32

POSITIVE LOGITS

 officials

5.18

Officials

5.07

 Officials

4.79

 authorities

3.54

 official

3.50

Authorities

3.40

gov

3.31

 Superintendent

3.31

offic

3.30

 deputies

3.25

Activations Density 0.069%

references to authority figures and their decisions

No Comments

No Known Activations