INDEX

Explanations

key mathematical terms and constructs associated with formal proofs and results

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FunctionFlags

-0.48

+#+

-0.41

<tfoot>

-0.40

JAKARTA

-0.40

multirow

-0.40

modifier

-0.38

 Betis

-0.37

orianCalendar

-0.36

 mandatario

-0.36

موقع

-0.36

POSITIVE LOGITS

 autorytatywna

0.48

Jeografia

0.47

 sufficiente

0.47

Naissance

0.44

béco

0.44

Viungo

0.44

 Daß

0.44

 lecteurs

0.43

تقاوى

0.42

 sufficient

0.42

Activations Density 2.276%

key mathematical terms and constructs associated with formal proofs and results

No Comments

No Known Activations