INDEX

Explanations

cyrillic characters, possibly used in greetings or other common phrases, and words related to website navigation

oai_token-act-pair · gemini-2.0-flash

Non-English, code, acknowledgments

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 سكانية

-0.60

SharedCtor

-0.60

pography

-0.57

новниш

-0.56

soever

-0.52

enderror

-0.52

IndentedString

-0.52

Personendaten

-0.51

Попис

-0.49

romeda

-0.49

POSITIVE LOGITS

 élas

0.71

 démocr

0.69

 réfrig

0.69

resti

0.68

 présidenti

0.63

 rayures

0.63

 jouets

0.62

 étan

0.62

 själ

0.60

 adhé

0.60

Activations Density 1.449%

cyrillic characters, possibly used in greetings or other common phrases, and words related to website navigation

Non-English, code, acknowledgments

No Comments

No Known Activations