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

__":

-0.49

aroni

-0.44

 Wahl

-0.43

__':

-0.41

jokes

-0.40

 joking

-0.40

 Defensa

-0.39

 Emig

-0.39

 willed

-0.39

 linge

-0.39

POSITIVE LOGITS

HttpStatus

3.30

 HttpStatus

1.72

ResponseStatus

0.81

HttpStatusCode

0.71

rrggbb

0.69

SharedCtor

0.65

 الرياضيه

0.62

 propOrder

0.62

 GenerationType

0.59

HttpHeaders

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.