INDEX

Explanations

programming-related syntax and functions

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

Appears before properties or database actions

np_acts-logits-general · gemini-2.0-flash

code referencing string literals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

__':

-0.72

__":

-0.66

 scolaires

-0.64

UserScript

-0.64

 متعلقه

-0.62

 soyez

-0.61

 infantiles

-0.61

stylers

-0.60

 MaterialApp

-0.59

invokeLater

-0.59

POSITIVE LOGITS

 cref

0.54

ref

0.48

($"

0.46

 nahilalakip

0.46

 string

0.42

ulture

0.41

 office

0.41

findpost

0.40

REF

0.40

endregion

0.39

Activations Density 0.049%

programming-related syntax and functions

Appears before properties or database actions

code referencing string literals

No Comments

No Known Activations