INDEX

Explanations

phrases related to legal or religious contexts featuring auxiliary verbs like "can" or "must" along with "to be"

oai_token-act-pair · gemini-2.0-flash

Code, languages, legal documents

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 أعلام

-0.52

 للمعارف

-0.50

 crossed

-0.47

NewGuid

-0.47

 التح

-0.47

ίο

-0.45

Ze

-0.45

#![

-0.44

ValueGeneration

-0.44

ianto

-0.44

POSITIVE LOGITS

thâu

0.68

帖最后由

0.65

 Italijanski

0.63

 MyBatis

0.63

 محفوظة

0.60

IVEREF

0.60

 Scrolls

0.59

ergies

0.59

 Wikimedijinoj

0.57

 متعلقه

0.56

Activations Density 0.235%

phrases related to legal or religious contexts featuring auxiliary verbs like "can" or "must" along with "to be"

Code, languages, legal documents

No Comments

No Known Activations