INDEX

Explanations

symbols and notations relevant to programming and mathematics

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

mathematical or code expressions

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

AndEndTag

-0.65

يكب

-0.61

EndContext

-0.57

antiation

-0.57

GOTREF

-0.56

IndentedString

-0.52

AntiForgeryToken

-0.51

gameserver

-0.51

tonode

-0.49

 betweenstory

-0.49

POSITIVE LOGITS

 NSCoder

0.35

Dichloroethene

0.35

ScopeManager

0.33

gehenden

0.32

GEBURTSDATUM

0.32

 fehlen

0.30

 companhia

0.30

Dichloropropane

0.30

帖最后由

0.29

enumi

0.29

Activations Density 0.369%

symbols and notations relevant to programming and mathematics

mathematical or code expressions

No Comments

No Known Activations