INDEX

Explanations

phrases indicating a strong personal belief or intention

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

0:0.11

1:0.07

2:0.06

3:0.08

4:0.11

5:0.07

6:0.09

7:0.06

8:0.10

9:0.11

10:0.07

11:0.03

Negative Logits

roma

-1.42

escription

-1.39

 mediation

-1.30

earch

-1.30

ascript

-1.30

DragonMagazine

-1.28

EUR

-1.21

translation

-1.19

 inquest

-1.18

 Examination

-1.18

POSITIVE LOGITS

 Dems

1.41

 Vote

1.39

 McCain

1.27

 Stalin

1.26

Ron

1.26

 Hillary

1.25

Jim

1.24

 dodged

1.24

 Democracy

1.24

..."

1.24

Activations Density 0.002%

phrases indicating a strong personal belief or intention

No Comments

No Known Activations