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

MLLoader

-0.63

+#+#

-0.63

transQ

-0.62

BASEPATH

-0.59

RenderAtEndOf

-0.59

 Meksiku

-0.56

verifyException

-0.56



-0.54

 betweenstory

-0.53

ModelAdmin

-0.52

POSITIVE LOGITS

IUrlHelper

0.42

Хьажоргаш

0.36

ViewFeatures

0.35

utriche

0.32

Források

0.32

BeginInit

0.31

 ExecuteAsync

0.31

 nahilalakip

0.31

Rhestr

0.30

Искәрмәләр

0.30

Activations Density 0.000%

No Known Activations

This feature has no known activations.