INDEX

Explanations

specific terms and categories related to program structures and participant roles

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

structured metadata or template-like information with colons and asterisks, such as program names, settings, goals, and key components.

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

 laſſen

-0.55

genodigd

-0.55

ftagPool

-0.53

 müſſen

-0.49

EndContext

-0.49

exitRule

-0.48

jenige

-0.47

aronder

-0.46

存于互联网档案馆

-0.46

makeConstraints

-0.45

POSITIVE LOGITS

 nahilalakip

0.53

Autoritní

0.46

Hentet

0.45

">//

0.44

0.43

Становништво

0.42

 VIAF

0.42

 tartalomajánló

0.42

Erstellt

0.40

Activations Density 0.345%

specific terms and categories related to program structures and participant roles

structured metadata or template-like information with colons and asterisks, such as program names, settings, goals, and key components.

No Comments

No Known Activations