INDEX

Explanations

specific categories or classifications within a given text

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

proper nouns after outrageous

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

verwijspagina

-0.40

ckså

-0.40

 зави

-0.38

exitRule

-0.37

Serviço

-0.37

ihnach

-0.37

miyorum

-0.36

 Schluss

-0.36

elett

-0.36

verhält

-0.35

POSITIVE LOGITS

Portale

0.76

पया

0.56

المناصب

0.55

最快更新

0.54

AccessorTable

0.53

GEBURTSDATUM

0.51

Hentet

0.50

mergeFrom

0.50

 виправивши

0.49

tagHelperRunner

0.49

Activations Density 0.003%

specific categories or classifications within a given text

proper nouns after outrageous

No Comments

No Known Activations