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

 Савезне

-0.71

AddTagHelper

-0.70

DotNetBar

-0.68

 متعلقه

-0.68

 Majefty

-0.65

 désol

-0.63

 leaſt

-0.62

 pleaſure

-0.62

 ſeveral

-0.62

 dAtA

-0.61

POSITIVE LOGITS

 Référence

0.50

ppuden

0.47

ibatis

0.47

اریخ

0.45

이고

0.44

مراجع

0.43

__]

0.42

 versus

0.42

 Artik

0.42

Cecil

0.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.