INDEX

Explanations

mathematical expressions and symbols related to tensor products and transformations

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

tensor product or multiplication

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 ſtate

-0.57

 myſelf

-0.57

 houſe

-0.54

 pleaſure

-0.52

 juſt

-0.52

WebElementEntity

-0.50

 ftate

-0.49

Telegram

-0.48

 fometimes

-0.48

localctx

-0.47

POSITIVE LOGITS

 cross

0.41

 المعيارى

0.41

HostException

0.40

"]();

0.40

bung

0.40

awtextra

0.38

rmi

0.38

"}")

0.38

avx

0.38

uesia

0.37

Activations Density 0.196%

mathematical expressions and symbols related to tensor products and transformations

tensor product or multiplication

No Comments

No Known Activations