INDEX

Explanations

references to protective layers or coverings in various contexts

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

covering or topping surfaces

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



-0.55

TypedDataSet

-0.49

httphttps

-0.48

RTLR

-0.48

fvar

-0.48

 żół

-0.47

 médicas

-0.43

 zijne

-0.43

 zoude

-0.43

 børn

-0.42

POSITIVE LOGITS

IntoConstraints

0.50

 المعيارى

0.45

 сверху

0.43

 שוליים

0.43

MethodManager

0.41

をかける

0.41

 cover

0.41

 Biôgrafia

0.40

lid

0.39

 topping

0.39

Activations Density 0.191%

references to protective layers or coverings in various contexts

covering or topping surfaces

No Comments

No Known Activations