INDEX

Explanations

structures related to code or programming constructs

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

double digit 5

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

UserScript

-0.56

awaiter

-0.54

Geplaatst

-0.53

RegressionTest

-0.51

complexType

-0.49

 <<<<<<<<<<<<<<

-0.49

 المعيارى

-0.48

 autorytatywna

-0.46

frameCount

-0.46

transQ

-0.45

POSITIVE LOGITS

WireFormatLite

0.47

 Vuitton

0.39

 recic

0.39

 raspberries

0.38

 TextAppearance

0.38

aarrggbb

0.37

 broom

0.37

umerate

0.37

ruct

0.37

 Chattanooga

0.36

Activations Density 0.000%

structures related to code or programming constructs

double digit 5

No Comments

No Known Activations