INDEX

Explanations

mentions of recent events or studies

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

recent paper, report, study, findings, case, work, developments

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

 MainAxisSize

-0.53

DockStyle

-0.51

awtextra

-0.50

 snippetHide

-0.49

Geplaatst

-0.49



-0.48

Autoritní

-0.47

 BorderSide

-0.45

windowFixed

-0.45

WriteLiteral

-0.45

POSITIVE LOGITS

 recent

0.62

recent

0.56

出版年

0.52

 reciente

0.52

 recently

0.51

InjectAttribute

0.50

 recente

0.50

 RECENT

0.49

recently

0.49

 récente

0.49

Activations Density 0.010%

mentions of recent events or studies

recent paper, report, study, findings, case, work, developments

No Comments

No Known Activations