INDEX

Explanations

commands or functions related to adding or updating elements in a configuration system

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

code constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 africaine

-0.71

 vectorielle

-0.65

 africain

-0.63

 asiatique

-0.61

 Constitucional

-0.59

 mijne

-0.58

 Londra

-0.57

routeProvider

-0.57

 étrangère

-0.56

 EClass

-0.56

POSITIVE LOGITS

UnusedPrivate

0.73

 ujednoznacz

0.58

 متعلقه

0.58

 referrerpolicy

0.58

AlterField

0.56

">//

0.56

brokes

0.56

openqa

0.56

 TEMPLATE

0.55

nó

0.55

Activations Density 0.003%

commands or functions related to adding or updating elements in a configuration system

code constructs

No Known Activations