INDEX

Explanations

the word "do" in various contexts of questioning or asserting actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itſelf

-0.85

 للمعارف

-0.83

TMP

-0.78

 Мексичка

-0.74

 nguyễn

-0.73

)}.

-0.71

 CHAPITRE

-0.71

 Cuer

-0.69

 himſelf

-0.69

 eſſ

-0.69

POSITIVE LOGITS

Do

2.30

Do

2.29

do

2.15

DO

1.94

do

1.93

DO

1.74

 Does

1.37

Does

1.36

 does

1.34

does

1.19

Activations Density 0.105%

the word "do" in various contexts of questioning or asserting actions

No Comments

No Known Activations