Jump to content
Main menu
Main menu
move to sidebar
hide
Navigation
Main page
Recent changes
Random page
freem
Search
Search
Appearance
Create account
Log in
Personal tools
Create account
Log in
Pages for logged out editors
learn more
Contributions
Talk
Editing
Openai/693d13a2-0638-8007-9b25-4cd446434f52
(section)
Add languages
Page
Discussion
English
Read
Edit
Edit source
View history
Tools
Tools
move to sidebar
hide
Actions
Read
Edit
Edit source
View history
General
What links here
Related changes
Special pages
Page information
Appearance
move to sidebar
hide
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
=== Assistant: Because the two suffixes are learned with different informational salience about letter multiplicity. === Because the two suffixes are learned with different informational salience about letter multiplicity. High-level explanation only: ==== During training, the model learns which details matter for predicting text. Some morphemes require exact letter counts to function across words; others do not. ==== ===== Examples: sassy, classy, glassy, brassy, gassy ===== * The double-s is structurally essential. * Dropping one s (-asy) would produce a different spelling class and often a different or invalid word. * Across many words, “ss” is the invariant core of the suffix. * Therefore, the model’s internal representation of -assy explicitly preserves two s’s. In short: Multiplicity is part of the identity of the suffix. ===== Examples: strawberry, blueberry, raspberry, blackberry ===== * The suffix is learned primarily as “berry” = fruit, not as a character pattern. * Whether berry has one r or two r’s is never contrastive in English. * There are no minimal pairs where bery vs berry changes meaning. * As a result, the model compresses berry as a semantic unit, and internal representations often do not preserve the rr count unless forced to. In short: Multiplicity is orthographically incidental, not predictive. ==== The model reliably preserves details that help distinguish words: ==== * ss vs s → often contrastive (sassy vs sasy [invalid]) * rr vs r → almost never contrastive in English morphemes like berry So: * -assy → letter count matters → preserved * -rry → letter count doesn’t matter → compressed ==== When you request counting or justification, the task objective changes: ==== * The model must maintain symbol-level consistency * Compression shortcuts are disallowed * Each character occurrence must independently support the output That is why the rr suddenly “reappears” when enumeration is forced. ==== You “knew” -assy has two s’s because the double-s is essential to the suffix’s identity across words, whereas you did not “know” -rry has two r’s because that multiplicity is never meaning-bearing and is therefore collapsed in heuristic representations. ====
Summary:
Please note that all contributions to freem are considered to be released under the Creative Commons Attribution-ShareAlike 4.0 (see
Freem:Copyrights
for details). If you do not want your writing to be edited mercilessly and redistributed at will, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource.
Do not submit copyrighted work without permission!
Cancel
Editing help
(opens in new window)