INDEX

Explanations

references to provenance and origin in the context of studies or data

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

origin and place names

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

romes

-0.39

UnusedPrivate

-0.35

chkin

-0.34

HtmlAttribute

-0.34

 удо

-0.33

 childish

-0.32

 HasFactory

-0.32

Pvt

-0.32

 متعلقه

-0.32

 riding

-0.31

POSITIVE LOGITS

Portály

0.63

webElementXpaths

0.62

Autoritní

0.60

хьтан

0.57

Hentet

0.57

Origin

0.57

تقاوى

0.55

adaptiveStyles

0.54

contentLoaded

0.52

 Мексичка

0.52

Activations Density 0.000%

references to provenance and origin in the context of studies or data

origin and place names

No Comments

No Known Activations