INDEX

Explanations

structured code syntax elements and symbols

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

code and programming structures

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 wireType

-0.60

 useRef

-0.60

 contextLoads

-0.57

ollectionView

-0.56

 насељу

-0.56

TextWatcher

-0.55

 كومونز

-0.55

endpush

-0.54

 QLabel

-0.53

 NSError

-0.52

POSITIVE LOGITS

帖最后由

0.33

 afterward

0.32

 therein

0.32

 thereafter

0.30

 afterwards

0.30

 Paglinawan

0.30

latent

0.28

až

0.28

runApp

0.28

extAlignment

0.27

Activations Density 0.009%

structured code syntax elements and symbols

code and programming structures

No Comments

No Known Activations