claude_fable_5 v3 Evaluation

Generated: 2026-09-20 11:19:27 UTC

Pairs100
Distinct lemmas100
Slope0.958
R^20.990
BLEU-458.3%
chrF++77.5%
METEOR82.0%
ROUGE-L82.2%
BERTScore97.1%
COMET81.5%
BLEURT79.0%

Profile version id: 2017

First translation using this version: 2026-07-03

Total usable AI runs for this version: 100 runs across 100 lemmas

Created: 2026-07-02 10:20:24.063701+10:00

Active: True

Notes: External-only Claude workspace variant for Claude Fable 5; copied from legacy_scholarly v3 (source profile_version_id=1101); source_active=True; uses_guidance_context=True.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.958
Intercept1.156
R^20.990
P-value4.88e-99
Pearson r0.995
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.440
Mean absolute length residual2.591
Mean absolute percent error6.1%

Metric vs Passage Length

These plots test whether score changes with source passage length for this prompt version. The dashed line is an ordinary least-squares fit; the annotation reports n, r, R^2, p-value, and slope.

BLEU-4 vs passage length
BLEU-4: significant negative; r = -0.429, R^2 = 0.184, p = 8.27e-06.
chrF++ vs passage length
chrF++: significant negative; r = -0.481, R^2 = 0.231, p = 4.05e-07.
METEOR vs passage length
METEOR: significant negative; r = -0.452, R^2 = 0.204, p = 2.41e-06.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.461, R^2 = 0.213, p = 1.37e-06.
BERTScore vs passage length
BERTScore: significant negative; r = -0.656, R^2 = 0.431, p = 1.22e-13.
COMET vs passage length
COMET: significant negative; r = -0.504, R^2 = 0.254, p = 8.78e-08.
BLEURT vs passage length
BLEURT: significant negative; r = -0.764, R^2 = 0.583, p = 2.41e-20.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.330, R^2 = 0.109, p = 0.0008.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.347, R^2 = 0.120, p = 0.0004.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.338, R^2 = 0.114, p = 0.0006.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.339, R^2 = 0.115, p = 0.0006.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.429 0.184 8.27e-06 -0.00273 ok
chrF++ 100 significant negative -0.481 0.231 4.05e-07 -0.00181 ok
METEOR 100 significant negative -0.452 0.204 2.41e-06 -0.00179 ok
ROUGE-L 100 significant negative -0.461 0.213 1.37e-06 -0.00171 ok
BERTScore 100 significant negative -0.656 0.431 1.22e-13 -0.00041 ok
COMET 100 significant negative -0.504 0.254 8.78e-08 -0.00090 ok
BLEURT 100 significant negative -0.764 0.583 2.41e-20 -0.00262 ok
Trigram precision 100 significant negative -0.330 0.109 0.0008 -0.00253 ok
Trigram recall 100 significant negative -0.347 0.120 0.0004 -0.00258 ok
Trigram F1 100 significant negative -0.338 0.114 0.0006 -0.00255 ok
Trigram Jaccard 100 significant negative -0.339 0.115 0.0006 -0.00282 ok

Translation Similarity Metrics

Mean BLEU-458.3%
Mean chrF++77.5%
Mean METEOR82.0%
Mean ROUGE-L82.2%
Mean BERTScore97.1%
Mean COMET81.5%
Mean BLEURT79.0%
Legacy smoothed corpus BLEU56.0%
Unigram F182.2%
Bigram F161.8%
Trigram precision49.9%
Trigram recall49.1%
Trigram F149.5%
Trigram Jaccard32.9%
4-gram F140.3%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2604 Καρχηδών 117 129 15.788 34.1% 65.1% 73.0% 73.4% 94.4% 79.0% 66.1% reviewed/approved 6429 claude-fable-5
2484 Καρία 270 245 -14.747 44.4% 67.8% 66.8% 71.8% 93.3% 81.5% 55.4% reviewed/approved 6424 claude-fable-5
2603 Κάρυστος 175 183 14.239 44.2% 69.1% 75.0% 72.6% 94.6% 78.9% 56.0% reviewed/approved 6428 claude-fable-5
2468 Καπετώλιον 138 124 -9.324 32.3% 52.5% 56.5% 60.3% 91.5% 72.6% 53.5% reviewed/approved 6421 claude-fable-5
2328 Καλλίπολις 43 51 8.661 40.3% 68.0% 70.0% 63.8% 95.1% 75.5% 72.4% reviewed/approved 6408 claude-fable-5
2609 Κάσπειρος 135 139 8.549 47.0% 69.2% 74.0% 73.0% 92.2% 74.4% 66.8% reviewed/approved 6432 claude-fable-5
2342 Κάμιρος 54 59 6.126 55.7% 81.9% 85.3% 83.2% 97.8% 82.1% 78.0% reviewed/approved 6416 claude-fable-5
7255 Κυτέριον 22 28 5.774 52.4% 74.6% 70.6% 72.0% 96.5% 78.3% 79.2% reviewed/approved 6462 claude-fable-5
2059 Καδμεία 20 26 5.690 40.9% 76.9% 79.9% 78.3% 96.6% 82.4% 80.5% reviewed/approved 6380 claude-fable-5
3288 Κορώνεια 110 112 5.492 42.8% 68.6% 78.9% 73.5% 95.5% 76.8% 64.2% reviewed/approved 6445 claude-fable-5
2597 Καρπασία 112 103 -5.423 55.3% 77.6% 72.4% 75.3% 96.8% 78.4% 64.9% reviewed/approved 6425 claude-fable-5
2465 Κάνωπος 83 86 5.352 54.7% 76.2% 83.2% 82.8% 96.7% 78.4% 75.1% reviewed/approved 6420 claude-fable-5
2605 Κάσιον 56 60 5.211 51.4% 75.2% 84.9% 79.3% 96.3% 78.0% 71.8% reviewed/approved 6430 claude-fable-5
2602 Καρύανδα 34 29 -4.719 40.5% 66.5% 72.2% 73.0% 96.6% 79.9% 77.7% reviewed/approved 6427 claude-fable-5
7266 Κώμη 79 81 4.183 23.7% 53.8% 52.8% 57.5% 94.2% 74.0% 60.8% reviewed/approved 6473 claude-fable-5
7259 Κύφος 54 57 4.126 63.2% 84.3% 90.8% 84.7% 97.4% 77.6% 79.0% reviewed/approved 6466 claude-fable-5
2081 Καλάθη 49 44 -4.085 53.6% 73.4% 79.3% 86.0% 97.5% 75.8% 76.1% reviewed/approved 6392 claude-fable-5
2326 Καλλίαρος 47 50 3.830 70.7% 85.9% 94.4% 90.7% 98.2% 75.6% 79.7% reviewed/approved 6406 claude-fable-5
7261 Κύψελα 34 30 -3.719 77.1% 84.5% 89.2% 93.8% 98.5% 82.5% 83.5% reviewed/approved 6468 claude-fable-5
2455 Κάναστρον 60 55 -3.620 28.2% 61.2% 62.3% 64.3% 94.4% 72.9% 63.8% reviewed/approved 6419 claude-fable-5
2055 Καβασσός 109 102 -3.550 45.8% 69.3% 76.2% 73.9% 96.0% 79.4% 62.8% reviewed/approved 6376 claude-fable-5
2079 Καισάρεια 39 35 -3.508 57.5% 76.1% 87.9% 91.9% 97.6% 81.1% 82.3% reviewed/approved 6390 claude-fable-5
7250 Κυρταία 40 36 -3.465 48.0% 70.8% 73.8% 73.7% 97.2% 75.1% 77.2% reviewed/approved 6457 claude-fable-5
2115 Καλὴ ἀκτή 64 59 -3.451 37.2% 60.6% 70.7% 68.3% 95.7% 75.4% 68.1% reviewed/approved 6400 claude-fable-5
7267 Κωνώπη 64 59 -3.451 60.4% 77.0% 77.3% 82.9% 97.6% 79.6% 77.0% reviewed/approved 6474 claude-fable-5
7260 Κυχρεῖος πάγος 61 63 3.422 54.5% 75.4% 79.0% 69.4% 96.7% 78.3% 69.6% reviewed/approved 6467 claude-fable-5
7257 Κυτώνιον 25 22 -3.099 72.6% 80.5% 75.7% 80.9% 98.7% 85.4% 84.4% reviewed/approved 6464 claude-fable-5
3530 Κριώα 26 23 -3.057 48.6% 73.3% 70.0% 77.6% 96.9% 79.3% 78.5% reviewed/approved 6448 claude-fable-5
7247 Κύρνος 50 46 -3.043 44.4% 65.6% 68.7% 70.8% 96.5% 79.3% 80.5% reviewed/approved 6454 claude-fable-5
2628 Καταονία 24 27 2.859 27.7% 63.2% 57.0% 66.7% 95.6% 80.3% 77.9% reviewed/approved 6442 claude-fable-5

Residual Predictors

Positive signed-residual terms are associated with AI translations longer than expected from the human length; negative signed-residual terms are associated with shorter AI translations. Absolute-residual models identify terms associated with larger AI-human length divergence.

Greek source terms predicting signed length residuals

Features: 176; ridge alpha: 2.894; cross-validated R^2: -0.076.

Positive termCoefficientDocsMean presentMean absent
οι 2.4559 11 4.522 -0.559
τε 1.9033 5 4.354 -0.229
δε και 1.8003 8 2.864 -0.249
απο 1.7207 27 1.770 -0.655
εστι δε και 1.4680 3 7.314 -0.226
εκαλειτο δε 1.3369 3 5.093 -0.158
αφ 1.2872 6 3.481 -0.222
πολις 1.2852 68 0.335 -0.711
και το 1.2734 9 3.034 -0.300
τη 1.2423 15 1.759 -0.310
ου 1.0571 14 1.769 -0.288
εν τη 1.0415 7 2.879 -0.217
εκαλειτο 1.0231 7 1.804 -0.136
εστι δε 1.0052 5 3.414 -0.180
παιδος 0.9784 4 3.506 -0.146
εν 0.9314 37 0.911 -0.535
αφ ου 0.9195 5 3.039 -0.160
ος 0.8640 6 1.640 -0.105
τοπος 0.8566 4 5.176 -0.216
πολις προς 0.8345 3 5.999 -0.186
Negative termCoefficientDocsMean presentMean absent
εστι και -1.4573 20 -0.641 0.160
γαρ -1.2667 9 -0.032 0.003
το -1.1967 70 -0.320 0.747
δια -1.1690 16 0.113 -0.022
πλησιον -1.0485 8 -1.731 0.151
φησιν -0.9426 10 -2.072 0.230
οικητωρ -0.9308 6 -2.939 0.188
το εθνικον -0.8878 56 -0.504 0.642
εθνικον -0.8878 56 -0.504 0.642
ζευς -0.8495 3 -8.658 0.268
οτι -0.8099 5 -1.803 0.095
ην -0.7923 13 -0.432 0.065
δε -0.7724 46 0.098 -0.084
εστι -0.7660 24 -0.052 0.016
θρακης -0.7343 5 -2.178 0.115
τους -0.7327 4 -1.994 0.083
θεοπομπος -0.7286 5 -2.353 0.124
το θηλυκον -0.7274 6 -2.263 0.144
χωρα -0.6998 6 -3.165 0.202
κατα -0.6852 10 -0.430 0.048

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 4.125; cross-validated R^2: -0.162.

Positive termCoefficientDocsMean presentMean absent
δε 1.4164 46 3.625 1.710
εν 1.3182 37 3.783 1.891
και 1.2757 66 3.226 1.359
γαρ 1.2716 9 6.873 2.168
εκαλειτο δε 1.1491 3 14.925 2.210
την 1.0950 13 6.160 2.058
δια 1.0792 16 5.563 2.025
δε και 0.9960 8 7.428 2.171
εκαλειτο 0.8417 7 7.451 2.225
τω 0.8297 16 4.784 2.173
τη 0.8282 15 4.747 2.211
εκαλειτο δε και 0.7983 3 14.925 2.210
τα 0.7856 12 6.018 2.124
αφ 0.7374 6 5.763 2.389
δια του 0.7179 10 5.992 2.213
κατα 0.6759 10 5.759 2.239
οι 0.6475 11 4.773 2.321
εστι 0.6092 24 4.101 2.114
εστι δε και 0.5992 3 7.314 2.445
τε 0.5986 5 4.354 2.498
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.6802 56 2.591 2.591
εθνικον -0.6802 56 2.591 2.591
εθνος -0.5313 8 2.609 2.590
ως -0.4880 52 2.811 2.353
προς τη -0.4575 4 1.468 2.638
νησος -0.3808 9 2.277 2.622
μια -0.3760 3 1.197 2.634
εκαταιος ασια -0.3707 5 0.825 2.684
ασια -0.3707 5 0.825 2.684
παυσανιας -0.3643 4 1.081 2.654
ηροδοτος -0.3573 9 3.239 2.527
στραβων -0.3339 15 3.067 2.507
του δε -0.3190 3 1.033 2.639
οτι -0.3107 5 2.361 2.603
τινες -0.3091 6 1.768 2.644
τεταρτω -0.2966 3 1.639 2.621
ειρηται -0.2896 3 1.821 2.615
πολιτης -0.2830 18 3.281 2.440
και πολις -0.2727 5 2.451 2.598
δε οτι -0.2617 3 1.413 2.628

AI English terms predicting large absolute residuals

Features: 343; ridge alpha: 4.125; cross-validated R^2: -0.182.

Positive termCoefficientDocsMean presentMean absent
the 1.5963 95 2.669 1.104
with 1.0952 15 5.314 2.111
and 1.0617 58 3.413 1.456
on 1.0405 15 4.447 2.264
of 0.9728 64 3.156 1.587
it 0.8546 37 4.002 1.763
for 0.7496 12 5.697 2.168
on the 0.7492 7 4.821 2.423
their 0.7089 4 10.507 2.261
after 0.6907 21 4.661 2.041
by 0.6815 12 6.114 2.111
used 0.6498 9 7.053 2.150
be called 0.6438 6 8.674 2.203
to be called 0.6438 6 8.674 2.203
to 0.6250 24 4.111 2.111
is 0.6092 91 2.741 1.072
all 0.5679 5 7.490 2.333
are 0.5545 22 4.772 1.976
called 0.5437 24 4.207 2.081
among 0.5422 5 5.828 2.421
Negative termCoefficientDocsMean presentMean absent
ethnonym is -0.6694 53 2.310 2.908
the ethnonym is -0.6605 52 2.327 2.877
ethnonym -0.5298 58 2.569 2.621
the ethnonym -0.5265 56 2.572 2.615
a city in -0.3930 38 2.630 2.567
as in -0.3726 29 2.527 2.617
city in -0.3604 39 2.696 2.524
strabo book -0.3504 9 1.844 2.665
a citizen is -0.3465 13 2.103 2.664
a citizen -0.3465 13 2.103 2.664
island -0.3308 14 2.396 2.623
one of -0.3298 3 0.906 2.643
herodotos -0.3284 7 2.745 2.580
pausanias -0.3081 4 1.081 2.654
is called -0.3059 3 0.897 2.644
people -0.2997 13 3.045 2.523
asia -0.2972 6 0.925 2.697
about -0.2961 3 0.572 2.654
one -0.2811 14 2.579 2.593
in his asia -0.2711 5 0.825 2.684

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 4.125; cross-validated R^2: -0.207.

Positive termCoefficientDocsMean presentMean absent
the 1.5813 90 2.758 1.093
with 1.2335 16 4.989 2.134
and 1.1708 52 3.602 1.496
on 0.9066 16 4.406 2.245
of 0.8818 66 3.266 1.281
is with 0.8439 4 10.564 2.259
after 0.7092 22 4.896 1.941
who 0.6911 13 5.258 2.193
says 0.6828 17 5.369 2.022
their 0.6429 3 14.925 2.210
situated 0.6107 4 9.218 2.315
when 0.6000 7 6.986 2.260
there 0.5952 34 3.774 1.982
there is 0.5860 24 4.282 2.057
for 0.5770 11 5.287 2.258
usage 0.5718 4 6.648 2.422
on the 0.5256 10 4.634 2.364
diphthong 0.5163 3 9.831 2.367
that 0.5114 29 3.623 2.170
to 0.5077 31 3.482 2.191
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.7192 52 2.320 2.885
ethnonym is -0.7192 52 2.320 2.885
the ethnonym -0.5547 56 2.585 2.598
ethnonym -0.5487 57 2.602 2.576
as in -0.4798 29 2.663 2.562
an -0.4582 22 2.364 2.655
a city in -0.4537 40 2.527 2.634
people -0.4445 9 2.451 2.605
city in -0.4140 41 2.593 2.590
a people -0.3373 5 1.034 2.673
some -0.3183 10 1.861 2.672
asia -0.3172 6 0.925 2.697
herodotos -0.3092 9 3.239 2.527
in his asia -0.2950 5 0.825 2.684
his asia -0.2950 5 0.825 2.684
crete -0.2908 3 0.774 2.647
one of the -0.2681 4 1.433 2.639
citizen is a -0.2672 15 2.466 2.613
citizen is -0.2672 15 2.466 2.613
a citizen is -0.2672 15 2.466 2.613

Prompt Text

Show prompt text
You are an expert classical philologist and translator specialising in Stephanos of Byzantium's Ethnika.

Goal
- Produce a clear, scholarly English translation in the established reviewed house style.
- Stay close to the Greek while making Stephanos' compressed reference-work prose intelligible in English.
- Translate the Greek main text of the entry, not the modern apparatus, editorial locators, or later bibliographic explanation.

Priority order
- Follow the source Greek first.
- Apply the output and formatting rules below.
- Apply any entry-specific Additional translation context supplied with the entry, but only where the cited Greek evidence is relevant.
- Do not force a rule, gloss, or proper-noun convention when the local syntax or context contradicts it.

Output rules (required)
- Respond ONLY by calling the submit_translation tool with a single string field: {"translation": "..."}.
- The translation text must contain only the translation.
- Do not include analysis, commentary, alternative translations, uncertainty notes, or Markdown outside the translation itself.

A) Formatting + spelling
- Use Australasian spelling and punctuation conventions.
- Preserve paragraphing and line-breaks when they are present in the Greek source.
- For verse quotations, preserve line segmentation in English where the Greek source marks or implies individual verse lines.
- For lengthy prose quotations that function as a quoted block, allow paragraph/quotation separation so the quotation remains readable. Do not create stanza-like line breaks for ordinary short prose snippets.
- Use single quotes for quoted forms, cited snippets, translated quotations, appellations derived from or pertaining to the headword (such as ethnonyms and similar gentilics, but not divine epithets), and Greek forms rendered in Latin letters: '...'.
- Avoid double quotes.
- Use *italics* for titles of ancient works when they are actual work titles: *Cypriaka*, *Bassarika*, *Europe*, *Periegesis*, *Geography*.
- Do not italicise descriptive phrases that are not work titles, such as an epitome of another author's books.

B) Opening / structure
- Let the Greek structure control the opening. Do not force every entry into a single template.
- When the entry opens with the headword and a brief identification, normally begin with the headword transliterated into Latin letters without Greek diacritics, followed by the identification.
- Typical form where appropriate: Headword: ...
- Appositive openings like 'Karia, the country.' are acceptable when the entry is of that type.
- If the headword line gives both nominative and genitive, render it compactly as 'X, Y: ...' unless a grammatical tag is needed for clarity.
- Keep numbered or lettered sequences of homonymous places as inline numbered items like (2) ... (3) ..., matching the Greek's structure.
- When Greek ἔστι καὶ ἄλλη / ἕτερος introduces another place with the same name, use 'There is also another ...' when that is needed for sense.
- For appositional settlement descriptions of the type X (proper noun) + Y (settlement noun) + Z (genitive territory), translate the territory as location: 'X, a city/town/village in Z', not 'of Z', unless the Greek clearly expresses possession, origin, descent, or another non-locative relationship.

C) Transliteration + naming
- Do NOT use macrons or acute accents in ordinary transliteration: Karystos, not Kárystos.
- Prefer Greek-form transliteration when Stephanos is discussing Greek forms, derivation, spelling, ethnonyms, or other metalinguistic material.
- Use these general correspondences unless an established house-style exception is supplied: η = e, κ = k, φ = ph, ῥ = rh, ῤ = r, υ = y, χ = ch.
- Use Greek-style transliteration of diphthongs, especially ai for αι. Do not drift into Latinised forms such as ae or e where the Greek form matters.
- Use forms such as Karchedon, Chalkedon, Kapetolion, Kyrene, Kyrtos, Kappadokia, and Palaistine in Greek-form contexts.
- Avoid Latinised forms such as Carthage, Chalcedon, Capitolium, Cyrene, Cyrtos, Cappadocia, and Palestine in Greek-form contexts.
- Use established conventional English forms only when they are house-style exceptions or clearly ordinary geography, for example Rome, Cyprus, Egypt, Syria, India, Sicily, Thebes, Peloponnese, the Black Sea, the Red Sea, and the Persian Gulf.
- Treat conventional English-name exceptions as a conservative whitelist. Do not expand it by analogy without entry-specific guidance.
- Keep Greek-form transliteration for most ethnonyms and derived adjectives, and put the cited form in single quotes.
- Do not Latinise river, people, or place names merely because a familiar Latin or English form exists.

D) Citations (authors/works/books)
- Translate only the citation information that is present in the original Greek words of the lemma entry.
- Ignore modern/editorial locator codes and apparatus-like add-ons: omit RE/SH numbers, FGrHist numbers, fragment numbers, chapter/section locators, column references, and parenthetical references such as (12,1,2 [C 533,12]), (Op. 344), or (A. R. 2,403).
- Do not infer a book number from a modern parenthetical reference.
- If the Greek says only 'Strabo' plus a modern parenthetical locator, translate 'Strabo', not 'Strabo, book 12'.
- Use 'as per X' only where the Greek itself has ὡς + author/name or an equivalent example-marker construction that supports that rendering.
- If the Greek explicitly gives a Greek book numeral, translate it: 'Ktesias in book 3 of his *Persika*'.
- Keep citations compact, mirroring the source's incompleteness:
  - Author + work title: 'Hellanikos in his *Cypriaka*'
  - Author + book: 'Pausanias, book 9.'
  - Author + work + book: 'Dionysios in book 3 of his *Bassarika*: ...'
- Work-title guidance:
  - Γεωγραφούμενα / γεωγραφικά in citation contexts -> *Geography*, not *Geographoumena*, when this is the established house treatment for the cited work.
  - Περιήγησις / Περίοδος should normally preserve the transmitted title as *Periegesis* when that is what Stephanos gives.
  - If the Greek says an author is cited in an epitome of the eleven books, do not italicise *Epitome* as a work title; translate 'in an epitome of his eleven books'.
- If a citation is present in the source as bracketed main text and later syntax depends on it, translate it in brackets rather than omitting it.
- Do not add external citation labels like Iliad, Works and Days, or Argonautica unless the Greek itself names them.

E) Entry-specific guidance context
- Additional translation context may be supplied for the current entry.
- This context can include matched formula rules, glossary rules, proper-noun conventions, contextual vocabulary bias, and external source passages.
- Apply required or replacement guidance strongly when the evidence matches the current Greek.
- Apply formula and gloss guidance according to the local syntax; do not make it mechanical if the context requires a different construction.
- Apply proper-noun guidance for consistency where supplied, but do not overwrite the Greek's grammatical point.
- Apply contextual vocabulary bias as a preference, not as a one-to-one replacement. It should steer the wording when the stated semantic context is present.
- Use external source passages only for quoted or allusive material. Do not replace Stephanos' own wording with the external source translation.
- Do not mention which guidance rules fired in the translation output. The review interface should surface that separately.

F) Philological, grammatical, and orthographic discussion
- When the entry discusses spelling, letters, diphthongs, accents, stress, case, gender, number, dialect, or morphology, preserve the technical claim precisely.
- Preserve Greek letters or digraphs in Greek script when the letter or digraph itself is under discussion, for example ι, ει, οι, υ.
- Prefer transliteration for cited word-forms unless the point depends on Greek spelling, accentuation, a specific letter, or a visible dialectal/prosodic distinction.
- If accentuation or prosody is the topic:
  - τόνος -> stress or accentuation, according to context.
  - ὀξύς describing a word's τόνος -> oxytone, not merely 'acute'.
  - βαρύς describing a word's τόνος -> barytone, not merely 'grave'.
  - Keep the form in a representation that makes the contrast visible. Use Greek script where the Greek spelling or accent is the point.
- When Stephanos cites an inflected form, dialectal form, or poetic form as evidence for another form, do not normalise away the form being demonstrated.
- If the cited form is inflected or dialectal, translate using the English lemma derived from the appropriate Greek nominative where that is needed for consistency. For example, treat Κυταιέος as evidence for Kytaieus from Κυταιεύς, not as a different nominative such as Kytaios.
- Use brief grammatical tags only when they are necessary to make the philological point visible:
  - case: (nom.), (acc.), (gen.), (dat.), (voc.)
  - gender: (m.), (f.), (n.)
  - number: (sing.), (pl.)

G) Comparisons / derivational analogies
- For ὡς + X where X is an analogous form or etymologically related noun, use a form like "(as in 'X')" where this makes the analogy clear.
- For fuller derivational analogies, keep the relation explicit:
  - as 'Byzantios' is from the name Byzantion
  - as in 'Lykaones', from Lykaonia
- Avoid bare 'like X' when the point is morphological analogy.

H) ἀπό, ἀφ' οὗ, and naming/derivation formulae
- If ἀπό + genitive gives an eponym or naming source, prefer 'after X' or 'named after X' rather than a flat 'from X'.
- In entries with temporal naming contrast, ἐκαλεῖτο X ... κέκληται Y -> 'It used to be called X; it is now called/named Y.'
- For ἀφ' οὗ:
  - With a masculine personal antecedent in a naming context, translate 'after whom'.
  - With an explicit neuter antecedent, translate 'from which'.
  - With a dropped neuter antecedent or Stephanos' example-marker usage, translate naturally according to the syntax; use 'as per' only when the Greek construction supports an evidentiary/example sense.
- Avoid archaic 'whence' unless it is clearly the most natural scholarly English.

I) Syntax and restraint
- Translate directly; do not add background explanation, modern bibliography, or parenthetical glosses unless Stephanos gives them.
- Do not add glosses like (synoikia), (kalathos), or explanatory etymologies unless the Greek explicitly defines the term.
- Avoid gratuitous adversatives for δέ.
- Use 'however' only for a real contrast, such as a correction, opposition, exception, or contrastive second item that English needs to mark.
- Avoid 'on the other hand' unless the Greek contrast is strong and English needs it.
- Convert participles into natural English when strict literalness obscures the structure.
- Circumstantial participles may need 'because', 'as', 'when', or a relative clause.
- Preserve uncertainty when the Greek is uncertain, corrupt, or genuinely ambiguous; do not invent a resolution.
- If the source includes bracketed or braced text as part of the lemma main Greek text of the Meineke edition rather than apparatus, translate it and keep brackets if that best reflects the source.
- If the source gives indirect speech under φασί, do not automatically turn it into a formal quotation unless the source punctuation or reference-work style calls for it.

Now translate the provided entry accordingly, and submit it via submit_translation.