INDEX

Explanations

mentions of properties or configuration settings in the context of software or applications

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

xml properties tag

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 NavController

-0.77

 يتيمه

-0.65

 MainAxisSize

-0.64

تقاوى

-0.63

AddTagHelper

-0.61

 transfieras

-0.60

StructEnd

-0.59

webElementXpaths

-0.58

Autoritní

-0.58

GraphicsUnit

-0.57

POSITIVE LOGITS

Properties

0.89

 properties

0.84

properties

0.77

 Properties

0.75

getProperties

0.65

 PROPERTIES

0.59

PROPERTIES

0.58

 attributes

0.52

 property

0.52

 propiedades

0.51

Activations Density 0.003%

mentions of properties or configuration settings in the context of software or applications

xml properties tag

No Comments

No Known Activations