INDEX

Explanations

infrequent punctuation and connection words emphasizing relationships within text

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

axes and slopes

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

:✨

-0.67

новништво

-0.52

invokeLater

-0.51

 يتيمه

-0.50

bootstrapcdn

-0.49

HasForeignKey

-0.47

IsContent

-0.47

GEBURTSDATUM

-0.47

 المعيارى

-0.46

SequentialGroup

-0.46

POSITIVE LOGITS

XmlAccessType

0.40

calipsis

0.39

 axes

0.39

inSlope

0.38

TabIndex

0.38

</tfoot>

0.37

aarrggbb

0.36

 stops

0.36

 races

0.36

axe

0.35

Activations Density 0.171%

infrequent punctuation and connection words emphasizing relationships within text

axes and slopes

No Comments

No Known Activations