INDEX

Explanations

terms and phrases that describe concepts or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 extranjera

-0.40

 votación

-0.40

 muchacha

-0.40

 oposición

-0.40

 colecciones

-0.39

 competición

-0.39

 documentación

-0.38

 cópia

-0.38

 mariée

-0.38

 plazos

-0.37

POSITIVE LOGITS

‘

0.89

]")]

0.88

 Савезне

0.88

richTextPanel

0.87

0.86

المناصب

0.85

Personensuche

0.83

NameInMap

0.83

所謂

0.83

AutoScaleMode

0.83

Activations Density 1.043%

terms and phrases that describe concepts or classifications

No Comments

No Known Activations

terms and phrases that describe concepts or classifications

No Comments

No Known Activations