INDEX

Explanations

underscore-prefixed terms or structural elements in a document

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

hexadecimal numbers and codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

PreExecute

-0.56

 mergeFrom

-0.47

NameInMap

-0.45

лтемелер

-0.43

 menit

-0.41

 mView

-0.41

 mouseY

-0.40

SequentialGroup

-0.40

 mucosal

-0.40

SizeMode

-0.39

POSITIVE LOGITS

ंदीखरीदारी

0.43

 cassa

0.39

Pad

0.39

Obi

0.37

oldt

0.36

vability

0.35

Pad

0.35

:✨

0.35

 giấy

0.35

 Ordnung

0.34

Activations Density 0.021%

underscore-prefixed terms or structural elements in a document

hexadecimal numbers and codes

No Comments

No Known Activations

underscore-prefixed terms or structural elements in a document

hexadecimal numbers and codes

No Comments

No Known Activations