INDEX

Explanations

programming-related keywords and error messages

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Code-related terms and file paths

np_acts-logits-general · gemini-2.0-flash

module

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 propOrder

-0.50

XmlRootElement

-0.47

doctype

-0.46

 unknownFields

-0.44

endphp

-0.43

XMLSchema

-0.42

 SqlCommand

-0.41

"?>

-0.40

 Stork

-0.40

 تعدى

-0.40

POSITIVE LOGITS

 module

0.54

 exportación

0.52

 MODULE

0.52

 modules

0.51

 módulos

0.51

webpack

0.51

模块

0.49

 Module

0.49

module

0.48

 módulo

0.48

Activations Density 0.219%

programming-related keywords and error messages

Code-related terms and file paths

module

No Known Activations