INDEX

Explanations

identifiers and characters that indicate programming syntax

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

C preprocessor macros using shift operators

np_acts-logits-general · gemini-2.0-flash

V_ macros

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

 autorytatywna

-0.64

 дописавши

-0.56

WebElementEntity

-0.54

migrationBuilder

-0.53

Hentet

-0.51

UnusedPrivate

-0.48

viewDidLoad

-0.48

UserScript

-0.47

مصادر

-0.47

rmtree

-0.46

POSITIVE LOGITS

 Exacts

0.62

 onOptions

0.61

 Exactos

0.55

 Elisha

0.54

 canina

0.53

 pleaſure

0.53

enumi

0.53

0.52

berra

0.51

 reafon

0.50

Activations Density 0.779%

identifiers and characters that indicate programming syntax

C preprocessor macros using shift operators

V_ macros

No Comments

No Known Activations