INDEX

Explanations

No Explanations Found

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

-0.39

Pat

-0.38

-0.36

Bar

-0.35

Pat

-0.35

-0.34

-0.33

-0.31

↵↵

-0.31

ok

-0.31

POSITIVE LOGITS

دانشنامهٔ

7.81

 ویکی‌پدی

1.33

 nakalista

1.24

 فريبيس

1.23

 للاسماء

1.17

 kasarigan

1.14

 يتيمه

1.07

 незавершена

1.06

adpleegd

1.03

 الرياضيه

1.01

Activations Density 0.000%

No Known Activations

This feature has no known activations.