INDEX

Explanations

words that express uncertainty or possibility

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

 carina

-0.59

Cip

-0.56

 Maire

-0.56

 Châ

-0.54

Tuc

-0.53

 Wär

-0.52

 Чар

-0.52

 Henry

-0.52

xampp

-0.50

 brim

-0.50

POSITIVE LOGITS

 perhaps

1.50

 possibly

1.48

 Possibly

1.42

 presumably

1.37

Possibly

1.36

possibly

1.35

 hopefully

1.34

perhaps

1.31

hopefully

1.28

 quizás

1.27

Activations Density 0.151%

words that express uncertainty or possibility

No Comments

No Known Activations