INDEX

Explanations

numeric results and findings within scientific research

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

Enumerated items, often starting with "1" or "i"

np_acts-logits-general · gemini-2.0-flash

numbered or lettered list items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 pinulongan

-0.94

StructEnd

-0.77

 незавершена

-0.76

 Wiktionnaire

-0.72

mybatisplus

-0.67

)}}{

-0.66

GOTREF

-0.65

 Roskov

-0.64

Datuak

-0.64

таратура

-0.64

POSITIVE LOGITS

 TextAppearance

0.50

 pata

0.47

šče

0.44

청

0.44

landır

0.44

zeiti

0.41

uous

0.41

canActivate

0.41

arrête

0.41

 diagnostica

0.41

Activations Density 0.247%

numeric results and findings within scientific research

Enumerated items, often starting with "1" or "i"

numbered or lettered list items

No Comments

No Known Activations