INDEX

Explanations

references to quantities of items or groups, especially in contexts involving "two" or "three."

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

"Two" or "three" followed by a noun

np_acts-logits-general · gemini-2.0-flash

counting numbers followed by nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 among

-0.46

among

-0.36

 Among

-0.34

 another

-0.33

Among

-0.31

 exemplaires

-0.30

 amongst

-0.28

 rather

-0.27

 parmi

-0.27

archiviato

-0.27

POSITIVE LOGITS

 queſta

0.79

parsedMessage

0.76

$_(

0.75

 remaining

0.73

 myſelf

0.69

 vooz

0.69

remaining

0.69

 imagui

0.68

<unused79>

0.68

<unused32>

0.68

Activations Density 0.246%

references to quantities of items or groups, especially in contexts involving "two" or "three."

"Two" or "three" followed by a noun

counting numbers followed by nouns

No Comments

No Known Activations

references to quantities of items or groups, especially in contexts involving "two" or "three."

"Two" or "three" followed by a noun

counting numbers followed by nouns

No Comments

No Known Activations