INDEX

Explanations

specific structured data formats or programming constructs

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

Code snippets and numerals

np_acts-logits-general · gemini-2.0-flash

52

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 defaultstate

-0.47

-0.42

 fieltro

-0.40

 rojos

-0.39

 amarillas

-0.38

 dapur

-0.38

 comien

-0.37

 zapato

-0.36

 Drawer

-0.36

 rodilla

-0.36

POSITIVE LOGITS

 iNdEx

0.50

HasForeignKey

0.50

RefNanny

0.49

 ویکی‌پدی

0.45

новниш

0.45

 utafitiHapana

0.45

UnusedPrivate

0.44

oredCriteria

0.43

 تانيه

0.43

ьажоргаш

0.43

Activations Density 2.057%

specific structured data formats or programming constructs

Code snippets and numerals

52

No Comments

No Known Activations