INDEX

Explanations

references to geometrical concepts and movement directions

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

polygon locations

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

exitRule

-0.44

PreInfinity

-0.38

BagConstraints

-0.37

Personendaten

-0.35

 Dependent

-0.34

 Chwiliwch

-0.34

Function

-0.33

 funzioni

-0.33

convertTo

-0.32

 functions

-0.32

POSITIVE LOGITS

IContainer

0.56

TagMode

0.54

 initComponents

0.52

 OnTrigger

0.52

AnchorStyles

0.52

 HasFactory

0.51

 مشين

0.49

 AssemblyProduct

0.49

AndEndTag

0.48

 PagesJaunes

0.47

Activations Density 0.057%

references to geometrical concepts and movement directions

polygon locations

No Comments

No Known Activations