INDEX

Explanations

expressions related to participation and interactivity in activities or events

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

certain phrases like "to be"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 HttpNotFound

-0.65

BagConstraints

-0.52

WebVitals

-0.51

 للاسماء

-0.51

AutoresizingMask

-0.50

parsedMessage

-0.49

 nahilalakip

-0.49

DataAnnotations

-0.49

 wherefore

-0.48

Bands

-0.48

POSITIVE LOGITS

 EconPapers

0.54

GEBURTSDATUM

0.39

rambled

0.33



0.32

 зро

0.31

uth

0.31

最快更新

0.30

httphttps

0.30

 Spez

0.29

ători

0.29

Activations Density 0.054%

expressions related to participation and interactivity in activities or events

certain phrases like "to be"

No Comments

No Known Activations