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

setVerticalGroup

-0.56

ureusement

-0.48

userManager

-0.43

createObject

-0.43

énario

-0.42

UniformLocation

-0.42

ulemon

-0.42

tease

-0.41

 considérons

-0.41

ilever

-0.41

POSITIVE LOGITS

########.

0.71

StructEnd

0.63

 Numerade

0.56

 يتيمه

0.55

 autorytatywna

0.53

 SYLLABLE

0.51

Autoritní

0.48

 мәкал

0.48



0.48

rungsseite

0.47

Activations Density 0.000%

No Known Activations

This feature has no known activations.