INDEX

Explanations

descriptive phrases and terms relating to features and characteristics

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

function or action

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

 linkovi

-0.71

StructEnd

-0.68

 kasarigan

-0.65

 חיצוניים

-0.64

੝

-0.63

 الرياضيه

-0.63

MENAFN

-0.63

Aktualisiert

-0.62

 Photocase

-0.60

EndGlobalSection

-0.59

POSITIVE LOGITS

 ejecutivo

0.28

AnimationsModule

0.26

 مشين

0.26

usercontent

0.25

 muerta

0.24

EditorBrowsable

0.24

 estación

0.23

 inspección

0.23

 individuales

0.22

 huevo

0.22

Activations Density 0.033%

descriptive phrases and terms relating to features and characteristics

function or action

No Comments

No Known Activations