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

 виправивши

-0.58

writeField

-0.52

ValueStyle

-0.49

 Infórmanos

-0.49

-0.48

routeProvider

-0.46

 Midi

-0.46

Insn

-0.44

agog

-0.43

 mach

-0.43

POSITIVE LOGITS

SupportActionBar

1.52

 miniaturka

0.44

Personendaten

0.42

PerformLayout

0.41

 salonu

0.41

InputBorder

0.39

rungsseite

0.38

SharedCtor

0.38

 cenar

0.37

 EconPapers

0.36

Activations Density 0.000%

No Known Activations

This feature has no known activations.