INDEX

Explanations

articles and prepositions in text

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

articles followed by specific nouns

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

 sorpresas

-0.49

<h1>

-0.43

 pouvoit

-0.42

 navideño

-0.37

 työn

-0.37

 Abonnez

-0.37

 Privacidad

-0.36

 journées

-0.36

 eccell

-0.36

 sério

-0.36

POSITIVE LOGITS

 Administrativna

0.75

ftagPool

0.73

ValueStyle

0.66

 المعيارى

0.65

GOTREF

0.65

 wireType

0.64

 <=",

0.63

CodedInputStream

0.63

NameInMap

0.61

RTLD

0.60

Activations Density 0.042%

articles and prepositions in text

articles followed by specific nouns

No Comments

No Known Activations