INDEX

Explanations

phrases related to actions and states involving "to be" and "to"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

phrases related to functionalities and capabilities of devices

oai_token-act-pair · gpt-4o-mini Triggered by @bot

to verb infinitives

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 позволяют

-0.53

 aiutare

-0.52

legate

-0.52

 néglig

-0.51

 sprawie

-0.50

 permiten

-0.50

inspira

-0.49

evo

-0.48

 mivel

-0.47

 permettra

-0.47

POSITIVE LOGITS

MigrationBuilder

0.81

typeorm

0.77

 للمعارف

0.74

verwijspagina

0.71

 ویکی‌پدیای

0.68

évaluateur

0.67

 lenker

0.67

Referanser

0.66

 <=",

0.66

 autorytatywna

0.65

Activations Density 0.098%

phrases related to actions and states involving "to be" and "to"

phrases related to functionalities and capabilities of devices

to verb infinitives

No Comments

No Known Activations