INDEX

Explanations

references to tips and suggestions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.56

plast

-0.56

 cdti

-0.53

 Audiodateien

-0.52

 незавершена

-0.52

 Initialized

-0.51

囗

-0.51

OMES

-0.51

expandindo

-0.50

Zitat

-0.48

POSITIVE LOGITS

Tip

1.33

tip

1.29

Tip

1.24

 tips

1.22

 Tips

1.20

TIP

1.15

TIP

1.11

Tips

1.10

tip

1.07

TIPS

1.00

Activations Density 0.058%

references to tips and suggestions

No Comments

No Known Activations