INDEX

Explanations

terms related to unique identifiers in a data context

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

UUID and ubiquitin words

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

 HomeController

-0.45

(\

-0.40

force

-0.39

return

-0.39

larg

-0.37

 whofe

-0.37

 stehenden

-0.37

ⓧ

-0.36

$("

-0.36

HomeController

-0.35

POSITIVE LOGITS

uuid

1.48

 uuid

1.45

 UUID

1.38

UUID

1.35

Uuid

1.17

randomUUID

1.13

 uuidv

1.03

NewGuid

0.72

 Guid

0.68

UniqueId

0.68

Activations Density 0.002%

terms related to unique identifiers in a data context

UUID and ubiquitin words

No Comments

No Known Activations