INDEX

Explanations

keystroke and authentication-related keywords and phrases

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

Code, file paths, or commands

np_acts-logits-general · gemini-2.0-flash

linux commands and system contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ArgsConstructor

-0.64

PerformLayout

-0.58

StoryboardSegue

-0.58

Géographie

-0.58

mvc

-0.53

//

-0.52

RectangleBorder

-0.50

gemeente

-0.48

mittances

-0.48

ExtendWith

-0.47

POSITIVE LOGITS

 Linux

0.72

 linux

0.66

 kaarangay

0.64

linux

0.62

 PagesJaunes

0.61

 Debian

0.59

Linux

0.58

HomeAsUpEnabled

0.57

PasswordEncoder

0.55

 Geraadpleegd

0.55

Activations Density 1.594%

keystroke and authentication-related keywords and phrases

Code, file paths, or commands

linux commands and system contexts

No Comments

No Known Activations