INDEX

Explanations

references to official documents or patents

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

patent documents

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

tagext

-0.43

 junger

-0.39

setFilter

-0.35

buttonBar

-0.33

<&

-0.33

 λε

-0.32

yyr

-0.32

 Verhaltens

-0.32

 invasion

-0.31

 Consejos

-0.31

POSITIVE LOGITS

 DOCUMENTS

2.70

 Documents

1.38

 documents

1.35

documents

1.28

Documents

1.27

 Document

1.16

 document

1.13

Document

1.09

document

1.05

 Dokumente

1.01

Activations Density 0.000%

references to official documents or patents

patent documents

No Comments

No Known Activations

references to official documents or patents

patent documents

No Comments

No Known Activations