INDEX

Explanations

elements related to coding or programming constructs

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

technical module names

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

 ProtoMessage

-0.45

UnusedPrivate

-0.44

AsUp

-0.42

msgTypes

-0.41

 hObject

-0.41

 initComponents

-0.40

PositiveButton

-0.40

PFC

-0.39

وروبا

-0.39

 AssemblyCulture

-0.38

POSITIVE LOGITS

 BrowserModule

0.55

 CommonModule

0.46

 AssemblyTitle

0.46

AndroidJUnit

0.43

+#+

0.42

portál

0.40

Ɓ

0.40

dafx

0.40

LookAndFeel

0.39

 Phal

0.39

Activations Density 0.000%

elements related to coding or programming constructs

technical module names

No Comments

No Known Activations