INDEX

Explanations

programming comments and annotations related to autogenerated code in a software context

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

Code snippets or programming-related text

np_acts-logits-general · gemini-2.0-flash

Javadoc comments

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 elevación

-0.34

 elevation

-0.34

 bénéfices

-0.32

 suspensión

-0.31

inf

-0.31

 beliau

-0.30

 navideñas

-0.30

 nobly

-0.29

Seattle

-0.29

 Seattle

-0.29

POSITIVE LOGITS

OGND

0.89

ftagPool

0.71

الحياه

0.71

 autorytatywna

0.66

XmlAccessorType

0.65

awtextra

0.65

 propOrder

0.63

GraphicsUnit

0.62

 оригіналу

0.60

Javadoc

0.59

Activations Density 0.034%

programming comments and annotations related to autogenerated code in a software context

Code snippets or programming-related text

Javadoc comments

No Comments

No Known Activations