INDEX

Explanations

phrases relating to newsletters and subscriptions

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

0:0.01

1:0.01

2:0.03

3:0.25

4:0.16

5:0.05

6:0.14

7:0.04

8:0.16

9:0.04

10:0.03

11:0.02

Negative Logits

..."

-3.10

.""

-2.44

 toilets

-2.36

Reviewer

-2.27

 overc

-2.24

 Izan

-2.18

](

-2.16

rul

-2.15

ocks

-2.10

gling

-2.06

POSITIVE LOGITS

 newsletter

2.95

letters

2.76

 subscribe

2.71

 subscribers

2.71

 Monthly

2.70

 Subscribe

2.68

 Newsletter

2.64

 Sign

2.55

Month

2.54

Subscribe

2.51

Activations Density 0.003%

phrases relating to newsletters and subscriptions

No Comments

No Known Activations