INDEX

Explanations

technical and scientific terms related to medical imaging and viral particles

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للمعارف

-0.55

 ſmall

-0.50

PageRoute

-0.47

ſelf

-0.46

 każde

-0.46

bisoft

-0.46

 pagado

-0.46

 fiecare

-0.45

 дописавши

-0.45

 Comprometido

-0.45

POSITIVE LOGITS

 devoid

0.57

blind

0.51

 without

0.51

 без

0.51

 absence

0.49

 lack

0.48

 Without

0.48

 ghost

0.46

 blind

0.46

 tanpa

0.45

Activations Density 0.228%

technical and scientific terms related to medical imaging and viral particles

No Comments

No Known Activations