INDEX

Explanations

instructions or commands related to software installation and usage

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

Followed by French/Spanish/Portuguese articles

np_acts-logits-general · gemini-2.0-flash

multilingual greetings and replies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 della

-0.95

 της

-0.80

Ms

-0.78

 DELLA

-0.76

jskiej

-0.71

 dalla

-0.70

Ms

-0.70

 této

-0.68

 उनकी

-0.67

Die

-0.67

POSITIVE LOGITS

el

0.83

het

0.74

 chi̍t

0.63

del

0.62

 propOrder

0.62

il

0.60

styleType

0.58

boyfriend

0.58

 este

0.56

ArgsConstructor

0.55

Activations Density 0.088%

instructions or commands related to software installation and usage

Followed by French/Spanish/Portuguese articles

multilingual greetings and replies

No Comments

No Known Activations