INDEX

Explanations

programming-related terms and operations

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

C register access

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

 BoxFit

-0.72

tagHelperRunner

-0.68

 arşivlendi

-0.68

 invokingState

-0.67

Vidite

-0.67

AddTagHelper

-0.64

__(/*!

-0.63

LookAnd

-0.63

ValueStyle

-0.61

__':

-0.61

POSITIVE LOGITS

MIDDLEWARE

0.36

 perto

0.34

ładka

0.32

 Provisions

0.31

 Personnel

0.31

 continuously

0.30

 Contents

0.29

 fonctionnement

0.29

 setting

0.29

LayoutStyle

0.29

Activations Density 0.070%

programming-related terms and operations

C register access

No Comments

No Known Activations