INDEX

Explanations

the name "Andrea" in various contexts

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

proper nouns, specifically names of people

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.04

2:0.15

3:0.07

4:0.12

5:0.07

6:0.04

7:0.03

8:0.05

9:0.19

10:0.07

11:0.03

Negative Logits

 Starr

-1.13

Zup

-1.11

 Boots

-1.05

 Tablet

-1.04

 Jerusalem

-1.04

CLASS

-1.04

 largeDownload

-1.02

ItemThumbnailImage

-1.02

BuyableInstoreAndOnline

-1.01

 Everest

-1.00

POSITIVE LOGITS

nder

1.35

ğ

1.28

 breakdown

1.21

gio

1.18

iment

1.17

henko

1.13

ウス

1.12

 finish

1.11

ert

1.09

ili

1.09

Activations Density 0.002%

the name "Andrea" in various contexts

proper nouns, specifically names of people

No Comments

No Known Activations