INDEX

Explanations

numerical data and references to quantities or statistics

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

quantities and numbers

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

الحياه

-0.76

 Мексичка

-0.69

 autorytatywna

-0.68

 ब्रेकडाउन

-0.67

hyrchwyd

-0.64

 SwitchCompat

-0.63

IntoConstraints

-0.62

 Wiktionnaire

-0.60

RTLR

-0.60

 Италијани

-0.60

POSITIVE LOGITS

 number

0.33

 thousands

0.31

ste

0.29

fcntl

0.29

endphp

0.29

ste

0.29

 mass

0.29

 fread

0.28

sum

0.28

 untold

0.28

Activations Density 0.024%

numerical data and references to quantities or statistics

quantities and numbers

No Comments

No Known Activations