INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.10

1:0.12

2:0.11

3:0.15

4:0.13

5:0.08

6:0.16

7:0.12

Negative Logits

fromnode

-0.34

TypedDataSet

-0.34

 MutableLiveData

-0.30

 AssemblyTitle

-0.29

PreferredItem

-0.28

<h6>

-0.28

AutoScale

-0.28

RegressionTest

-0.28

 resourceCulture

-0.28

aarrggbb

-0.27

POSITIVE LOGITS

Gön

0.40

 autorytatywna

0.37

 vigueur

0.36

thâu

0.36

ⓧ

0.35

 Becher

0.33

 fidé

0.33

becher

0.31

 contemporanea

0.31

rials

0.30

Activations Density 0.000%

No Known Activations

This feature has no known activations.