INDEX

Explanations

programming constructs and types related to data handling and structure

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

class definition, type, mount

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

tvguidetime

-0.54

esModule

-0.42

 Administrativna

-0.41

fromnode

-0.37

tagHelperRunner

-0.37

 doGet

-0.36

 defaultstate

-0.35

localctx

-0.35

nloa

-0.34

!*\

-0.33

POSITIVE LOGITS

enderror

0.55

ListItemIcon

0.54

 autorytatywna

0.50

SuccessListener

0.50

RectangleBorder

0.49

CardHeader

0.49

mybatisplus

0.48

ListTile

0.48

EndGlobalSection

0.47

 HttpNotFound

0.47

Activations Density 0.150%

programming constructs and types related to data handling and structure

class definition, type, mount

No Comments

No Known Activations