INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

 iſt

-0.78

帖最后由

-0.73

WriteTagHelper

-0.68

ſelf

-0.66

 näky

-0.66

 getItemCount

-0.63

Enllaços

-0.63

 ―――――

-0.63

Efq

-0.62

zegovina

-0.62

POSITIVE LOGITS

uxxxx

0.66

fjspx

0.60

esModule

0.60

RegressionTest

0.60

mpotent

0.57

 viewDidLoad

0.57

httphttps

0.54

 TestBed

0.54

XmlAccessType

0.53

SpringBootTest

0.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.