INDEX

Explanations

phrases related to geographical locations or descriptions of physical spaces

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

New Auto-Interp

Top Features by Cosine Similarity

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

1:0.03

2:0.20

3:0.06

4:0.18

5:0.06

6:0.03

7:0.03

8:0.10

9:0.13

10:0.06

11:0.03

Negative Logits

oday

-1.30

ulz

-1.27

 Drunk

-1.24

Bj

-1.15

llor

-1.14

ndra

-1.14

ache

-1.13

 dearly

-1.12

 Fant

-1.12

 cannibal

-1.11

POSITIVE LOGITS

taboola

1.46

────

1.41

 confines

1.38

CHAPTER

1.36

Armor

1.35

ropolitan

1.35

Characters

1.34

────────

1.33

advertisement

1.33

 encl

1.33

Activations Density 0.002%

phrases related to geographical locations or descriptions of physical spaces

No Comments

No Known Activations