INDEX

Explanations

references to arcade games and locations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to video game titles and related gaming terminology

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.05

1:0.09

2:0.18

3:0.05

4:0.09

5:0.08

6:0.03

7:0.06

8:0.05

9:0.16

10:0.06

11:0.03

Negative Logits

jri

-1.63

assad

-1.58

 Dalai

-1.48

 destabil

-1.44

arent

-1.41

ilateral

-1.40

moderate

-1.34

 infl

-1.33

answered

-1.33

 Tillerson

-1.32

POSITIVE LOGITS

Driver

1.48

 editions

1.46

 Prix

1.44

 EDITION

1.43

Switch

1.42

Reader

1.41

 Pinball

1.41

 Chicken

1.40

 Classics

1.40

ドラゴン

1.40

Activations Density 0.008%

references to arcade games and locations

references to video game titles and related gaming terminology

No Comments

No Known Activations