INDEX

Explanations

patterns related to numerical data or code syntax

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

Code or numerical series

np_acts-logits-general · gemini-2.0-flash

numbers and apostrophes

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

oredCriteria

-0.99

 "..\..\..\

-0.91

unknownFields

-0.90

queryInterface

-0.83

SOUNDBITE

-0.81

bitField

-0.78

 ویکی‌پدیا

-0.77

 nahilalakip

-0.77

pushFollow

-0.75

 "..\..\

-0.74

POSITIVE LOGITS

0.50

enumi

0.49

あら

0.48

CHRON

0.47

begin

0.44

 Rever

0.43

toxicity

0.43

’

0.42

mathrm

0.42

 مشين

0.42

Activations Density 2.204%

patterns related to numerical data or code syntax

Code or numerical series

numbers and apostrophes

No Comments

No Known Activations