INDEX

Explanations

names of programming or API-related elements

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

code parsing and processing

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

 arşivlendi

-0.37

YMMV

-0.33

 Others

-0.32

<tfoot>

-0.29

XmlAttribute

-0.29

################

-0.28

zar

-0.27

 ahli

-0.27

Menurutnya

-0.27

 Wide

-0.27

POSITIVE LOGITS

 kaarangay

0.72

 الرياضيه

0.70

WebVitals

0.65

 informée

0.65

MemoryWarning

0.65

Filmographie

0.64

rungsseite

0.63

Rhestr

0.60

complexContent

0.59

 يتيمه

0.59

Activations Density 0.063%

names of programming or API-related elements

code parsing and processing

No Comments

No Known Activations