INDEX

Explanations

specific elements related to patent details

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

patent numbers

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

 Drittan

-0.55

Carriera

-0.47

 Angry

-0.46

afficheront

-0.46

 estekak

-0.46

 prioriti

-0.45

 Вікі

-0.44

 Monfieur

-0.44

NegativeButton

-0.44

 bastion

-0.44

POSITIVE LOGITS

 kasarigan

0.40

EDEFAULT

0.37

IntoConstraints

0.36

 erhielten

0.35

MathML

0.35

ItemBackground

0.34

 Clau

0.34

 getNumber

0.33

CMP

0.32

 erhielt

0.31

Activations Density 0.043%

specific elements related to patent details

patent numbers

No Comments

No Known Activations