INDEX

Explanations

terms related to champions and championship events

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

champion or winner

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

Personensuche

-0.56

 utafitiHapana

-0.52

makeConstraints

-0.52

 Wiktionnaire

-0.51

Carriera

-0.51

 Taktlose

-0.51

elemField

-0.49

Carrière

-0.49

vägen

-0.47

forRoot

-0.46

POSITIVE LOGITS

 برانيه

0.40

 scoreless

0.36

 vainqueur

0.36

 previous

0.35

ModelSerializer

0.35

 býval

0.34

 recent

0.34

谷川

0.33

 reelection

0.33

 precedente

0.33

Activations Density 0.007%

terms related to champions and championship events

champion or winner

No Comments

No Known Activations