INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RTDA

-0.46

neſs

-0.44

DB

-0.43

Билгалдахарш

-0.41

Db

-0.41

ſelf

-0.40

DIS

-0.39

NTT

-0.38

iress

-0.38

HasAnnotation

-0.38

POSITIVE LOGITS

ParallelGroup

3.05

StoryboardSegue

0.71

httphttps

0.71

CppMethod

0.62

 consultato

0.55

ReusableCell

0.54

AccessorTable

0.53

PerformLayout

0.51

 AssemblyTitle

0.51

ValueStyle

0.50

Activations Density 0.000%

No Known Activations

This feature has no known activations.