INDEX

Explanations

terms related to programming and code structures

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

Code, file paths, or imported modules

np_acts-logits-general · gemini-2.0-flash

Russian words and programming contexts

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

 nahilalakip

-0.65

ंदीखरीदारी

-0.61

НОЙ

-0.59

memberof

-0.56

amerikanischer

-0.56

وردار

-0.56

Зно

-0.55

Personendaten

-0.54

ută

-0.53

НЫХ

-0.51

POSITIVE LOGITS

 Yandex

0.78

yandex

0.73

№

0.68

 Moscow

0.66

 peculiarities

0.64

 Alexey

0.64

 Andrey

0.63

 russian

0.62

 analogs

0.62

 Russian

0.61

Activations Density 0.053%

terms related to programming and code structures

Code, file paths, or imported modules

Russian words and programming contexts

No Comments

No Known Activations