INDEX

Explanations

instances of the word "due" in various contexts

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

future events and deadlines

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 dependencia

-0.37

arum

-0.34

roid

-0.34

 betrug

-0.34

avez

-0.33

 implementación

-0.33

 köl

-0.32

 rask

-0.32

 préserver

-0.32

ESTON

-0.31

POSITIVE LOGITS

DockStyle

0.72

Cyfarwyddwr

0.68

 AssemblyCompany

0.66

⟬

0.59

存于互联网档案馆

0.56

 المعيارى

0.56

GEBURTSDATUM

0.55

 kaarangay

0.55

UpInside

0.53

 oprot

0.53

Activations Density 0.009%

instances of the word "due" in various contexts

future events and deadlines

No Comments

No Known Activations