INDEX

Explanations

programming-related comments and annotations

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

android manifest activity

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

 Región

-0.36

 fallen

-0.35

 Feier

-0.34

 Stär

-0.34

-0.32

unjang

-0.32

 fuertes

-0.31

 hatır

-0.30

 الأق

-0.30

 globally

-0.30

POSITIVE LOGITS

 nahilalakip

0.69

 ligiloj

0.69

Tikang

0.69

Personensuche

0.68

 lenker

0.67

parsedMessage

0.66

تقاوى

0.66

IndentedString

0.65

 EconPapers

0.65

verwijspagina

0.64

Activations Density 0.000%

programming-related comments and annotations

android manifest activity

No Comments

No Known Activations