INDEX

Explanations

terms related to access and identification in a security or transactional context

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

cards, passes, and identification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LookAnd

-0.72

 AssemblyProduct

-0.55

الدراسه

-0.53

XmlAccessorType

-0.52

/**

-0.50

Pyx

-0.49

 kiin

-0.49

 '\\;'

-0.49

انيف

-0.48

addContainerGap

-0.48

POSITIVE LOGITS

 card

0.54

 pass

0.51

 permit

0.47

 tickets

0.46

 ticket

0.46

 passes

0.45

 badge

0.44

 Passes

0.44

card

0.43

 Pass

0.43

Activations Density 0.033%

terms related to access and identification in a security or transactional context

cards, passes, and identification

No Comments

No Known Activations