INDEX

Explanations

phrases that describe the complexity and variability of systems or actions

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

across multiple entities

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

 autorytatywna

-0.40

AccessorTable

-0.35

 समीक्षाएं

-0.34

rrggbb

-0.33

AndEndTag

-0.33

Merkmale

-0.32

OGND

-0.31

 April

-0.30

 contro

-0.30

 instant

-0.29

POSITIVE LOGITS

 dAtA

0.57

 disparate

0.53

 inconsistent

0.52

 geïsole

0.52

 inconsistencies

0.51

ArrowToggle

0.51

 forskellige

0.50

 firebaseConfig

0.50

 Normdatei

0.49

arResult

0.49

Activations Density 0.073%

phrases that describe the complexity and variability of systems or actions

across multiple entities

No Comments

No Known Activations