INDEX

Explanations

references to patents and related legal documents

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

patent application and law

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

 يتيمه

-0.63

kreises

-0.49

RemoveField

-0.49

 referenties

-0.48

sli

-0.48

blouse

-0.47

Aholisi

-0.47

extAlignment

-0.47

hs

-0.46

__":

-0.46

POSITIVE LOGITS

 Patent

1.15

 patent

1.13

Patent

1.05

patent

0.98

 patents

0.97

 PATENT

0.96

 Patents

0.91

 patente

0.82

 patented

0.74

 INVENTION

0.70

Activations Density 0.018%

references to patents and related legal documents

patent application and law

No Comments

No Known Activations