INDEX

Explanations

references to patent and non-patent documents

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

patent publication citations

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

 propOrder

-0.53

featureID

-0.47

 âmes

-0.45

 læng

-0.43

 actifs

-0.43

 retraite

-0.40

AntiForgeryToken

-0.40

ClientSize

-0.40

 Erskine

-0.40

 HasFactory

-0.40

POSITIVE LOGITS

 PUBLICATIONS

1.04

 ویکی‌پدی

0.71

RegressionTest

0.61

msgTypes

0.60

 EconPapers

0.52

adaptiveStyles

0.50

 publications

0.47

 препратки

0.47

SourceChecksum

0.46

脚注の使い方

0.45

Activations Density 0.018%

references to patent and non-patent documents

patent publication citations

No Comments

No Known Activations