INDEX

Explanations

programming-related terms and identifiers

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

Code, packages, and programming

np_acts-logits-general · gemini-2.0-flash

software library and package names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

bootstrapcdn

-0.79

TagMode

-0.76

photobucket

-0.73

EndGlobalSection

-0.68

 مشين

-0.68

fjspx

-0.66

 CreateTagHelper

-0.65

 nemlig

-0.65

AddHtmlAttribute

-0.64

isContained

-0.64

POSITIVE LOGITS

 mercies

0.65

byte

0.61

tech

0.57

 Techno

0.55

 techno

0.53

 technologies

0.53

 hoods

0.52

 techs

0.51

 craters

0.50

 disinfection

0.50

Activations Density 0.169%

programming-related terms and identifiers

Code, packages, and programming

software library and package names

No Comments

No Known Activations