INDEX

Explanations

proper nouns, names, and titles

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.06

1:0.01

2:0.14

3:0.22

4:0.04

5:0.18

6:0.05

7:0.02

8:0.12

9:0.04

10:0.05

11:0.01

Negative Logits

 guiName

-1.43

agall

-1.32

apons

-1.29

inson

-1.26

inders

-1.26

ulkan

-1.25

cas

-1.24

aples

-1.24

zag

-1.22

hammad

-1.22

POSITIVE LOGITS

 trait

1.42

 overl

1.39

 resides

1.34

 shines

1.32

 namesake

1.27

 differs

1.27

 derives

1.26

 traits

1.26

 fortunes

1.25

 exceeds

1.25

Activations Density 0.034%

proper nouns, names, and titles

No Comments

No Known Activations