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

IonicModule

-0.67

ſammen

-0.58

bcryptjs

-0.56

geschlossen

-0.55

windowFixed

-0.54

kijken

-0.50

 nigdy

-0.49

fromnode

-0.49

TestingModule

-0.49

Попис

-0.49

POSITIVE LOGITS

帖最后由

0.54

httphttps

0.52

Personendaten

0.45

Personensuche

0.39

enumi

0.36

 يتيمه

0.35

 незавершена

0.35

 kasarigan

0.35

0.34

 cherchés

0.34

Activations Density 0.000%

No Known Activations

This feature has no known activations.