claude_sonnet_5 v3 Evaluation

Generated: 2026-09-20 11:19:52 UTC

Pairs100
Distinct lemmas100
Slope0.954
R^20.990
BLEU-454.7%
chrF++75.0%
METEOR81.6%
ROUGE-L82.0%
BERTScore96.8%
COMET81.0%
BLEURT77.6%

Profile version id: 2011

First translation using this version: 2026-07-05

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 Sonnet 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
Slope0.954
Intercept1.037
R^20.990
P-value7.98e-100
Pearson r0.995
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.160
Mean absolute length residual2.443
Mean absolute percent error6.3%

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 = -0.416, R^2 = 0.173, p = 1.67e-05.
chrF++ vs passage length
chrF++: significant negative; r = -0.455, R^2 = 0.207, p = 2.01e-06.
METEOR vs passage length
METEOR: significant negative; r = -0.520, R^2 = 0.270, p = 3.04e-08.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.527, R^2 = 0.277, p = 1.81e-08.
BERTScore vs passage length
BERTScore: significant negative; r = -0.640, R^2 = 0.409, p = 7.91e-13.
COMET vs passage length
COMET: significant negative; r = -0.517, R^2 = 0.267, p = 3.75e-08.
BLEURT vs passage length
BLEURT: significant negative; r = -0.740, R^2 = 0.548, p = 1.35e-18.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.404, R^2 = 0.163, p = 3.13e-05.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.406, R^2 = 0.165, p = 2.80e-05.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.405, R^2 = 0.164, p = 2.92e-05.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.407, R^2 = 0.165, p = 2.70e-05.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.416 0.173 1.67e-05 -0.00281 ok
chrF++ 100 significant negative -0.455 0.207 2.01e-06 -0.00177 ok
METEOR 100 significant negative -0.520 0.270 3.04e-08 -0.00213 ok
ROUGE-L 100 significant negative -0.527 0.277 1.81e-08 -0.00212 ok
BERTScore 100 significant negative -0.640 0.409 7.91e-13 -0.00043 ok
COMET 100 significant negative -0.517 0.267 3.75e-08 -0.00090 ok
BLEURT 100 significant negative -0.740 0.548 1.35e-18 -0.00246 ok
Trigram precision 100 significant negative -0.404 0.163 3.13e-05 -0.00345 ok
Trigram recall 100 significant negative -0.406 0.165 2.80e-05 -0.00341 ok
Trigram F1 100 significant negative -0.405 0.164 2.92e-05 -0.00342 ok
Trigram Jaccard 100 significant negative -0.407 0.165 2.70e-05 -0.00383 ok

Translation Similarity Metrics

Mean BLEU-454.7%
Mean chrF++75.0%
Mean METEOR81.6%
Mean ROUGE-L82.0%
Mean BERTScore96.8%
Mean COMET81.0%
Mean BLEURT77.6%
Legacy smoothed corpus BLEU54.7%
Unigram F181.1%
Bigram F160.8%
Trigram precision48.7%
Trigram recall47.5%
Trigram F148.1%
Trigram Jaccard31.7%
4-gram F139.5%

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 117 132 19.335 33.9% 64.9% 72.5% 70.9% 94.1% 79.0% 63.3% reviewed/approved 6615 claude-sonnet-5
2342 Κάμιρος 54 64 11.442 37.5% 71.4% 66.8% 72.9% 95.8% 78.4% 69.7% reviewed/approved 6602 claude-sonnet-5
2056 Καβειρία 112 97 -10.895 37.1% 68.5% 67.7% 70.8% 95.7% 79.8% 61.1% reviewed/approved 6563 claude-sonnet-5
2484 Καρία 270 249 -9.641 39.2% 63.9% 61.9% 65.9% 92.1% 81.3% 54.7% reviewed/approved 6610 claude-sonnet-5
7266 Κώμη 79 67 -9.410 19.9% 47.9% 43.9% 49.3% 92.6% 73.5% 57.5% reviewed/approved 6659 claude-sonnet-5
2470 Καππαδοκία 68 73 7.085 20.2% 60.7% 75.8% 69.5% 95.9% 77.1% 72.0% reviewed/approved 6608 claude-sonnet-5
7255 Κυτέριον 22 29 6.973 57.8% 78.1% 73.2% 74.5% 97.3% 79.8% 82.3% reviewed/approved 6648 claude-sonnet-5
2328 Καλλίπολις 43 49 6.937 55.0% 77.2% 78.2% 78.3% 96.1% 76.9% 76.0% reviewed/approved 6594 claude-sonnet-5
7260 Κυχρεῖος πάγος 61 66 6.764 53.4% 74.3% 80.0% 74.0% 96.6% 77.4% 66.1% reviewed/approved 6653 claude-sonnet-5
2059 Καδμεία 20 26 5.881 35.0% 69.1% 79.9% 78.3% 96.1% 81.6% 76.9% reviewed/approved 6566 claude-sonnet-5
2055 Καβασσός 109 110 4.967 33.0% 64.9% 68.5% 65.8% 95.2% 77.8% 61.1% reviewed/approved 6562 claude-sonnet-5
7253 Κυρτώνιος 18 23 4.789 28.3% 57.4% 72.7% 63.4% 96.1% 76.1% 78.6% reviewed/approved 6646 claude-sonnet-5
2605 Κάσιον 56 59 4.534 55.4% 76.3% 83.4% 78.3% 96.8% 77.8% 73.8% reviewed/approved 6616 claude-sonnet-5
2455 Κάναστρον 60 54 -4.282 32.9% 64.6% 66.6% 71.9% 94.8% 73.6% 65.8% reviewed/approved 6605 claude-sonnet-5
2597 Καρπασία 112 112 4.105 60.7% 79.2% 85.5% 81.2% 97.1% 78.1% 71.8% reviewed/approved 6611 claude-sonnet-5
2119 Κάλλατις 65 59 -4.053 27.3% 56.6% 65.6% 64.5% 95.1% 68.9% 59.9% reviewed/approved 6590 claude-sonnet-5
3530 Κριώα 26 22 -3.843 41.1% 68.8% 70.3% 79.2% 96.4% 78.9% 76.5% reviewed/approved 6634 claude-sonnet-5
7258 Κύτωρος 30 26 -3.660 59.6% 79.3% 81.8% 85.7% 97.5% 81.4% 82.7% reviewed/approved 6651 claude-sonnet-5
7248 Κύρου πόλις 32 35 3.432 26.8% 63.6% 67.3% 65.7% 95.9% 81.5% 79.3% reviewed/approved 6641 claude-sonnet-5
2607 Κάσος 57 52 -3.420 37.9% 65.1% 74.7% 77.1% 95.6% 78.1% 75.4% reviewed/approved 6617 claude-sonnet-5
7264 Κωλιάς 61 56 -3.236 41.2% 64.0% 66.5% 70.1% 95.5% 73.5% 63.2% reviewed/approved 6657 claude-sonnet-5
2080 Καλαβρία 18 15 -3.211 43.3% 75.6% 69.8% 78.8% 96.9% 86.4% 86.5% reviewed/approved 6577 claude-sonnet-5
2114 Κάλβιος 18 15 -3.211 29.4% 52.9% 63.6% 66.7% 93.6% 74.5% 73.9% reviewed/approved 6585 claude-sonnet-5
2326 Καλλίαρος 47 49 3.121 54.9% 79.3% 95.0% 91.7% 97.7% 74.5% 78.1% reviewed/approved 6592 claude-sonnet-5
7267 Κωνώπη 64 59 -3.099 61.1% 76.9% 80.5% 84.6% 97.7% 79.7% 75.7% reviewed/approved 6660 claude-sonnet-5
3288 Κορώνεια 110 109 3.013 43.2% 66.7% 70.3% 72.7% 95.2% 77.1% 62.6% reviewed/approved 6631 claude-sonnet-5
7251 Κύρτος 66 61 -3.007 31.9% 58.6% 69.1% 68.8% 92.9% 72.7% 66.8% reviewed/approved 6644 claude-sonnet-5
2603 Κάρυστος 175 171 2.997 43.3% 66.0% 72.5% 67.6% 94.9% 78.2% 59.9% reviewed/approved 6614 claude-sonnet-5
2331 Κάλυδνα 43 45 2.937 50.9% 75.0% 94.2% 93.2% 97.1% 81.8% 80.1% reviewed/approved 6597 claude-sonnet-5
7257 Κυτώνιον 25 22 -2.889 75.1% 80.0% 84.7% 89.4% 97.6% 81.1% 77.6% reviewed/approved 6650 claude-sonnet-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: 11.938; cross-validated R^2: -0.100.

Positive termCoefficientDocsMean presentMean absent
δε και 0.6478 8 2.988 -0.260
ος 0.5787 6 5.421 -0.346
δε 0.5466 46 0.591 -0.504
εστι δε και 0.4499 3 7.762 -0.240
εκαλειτο 0.3897 7 1.768 -0.133
πολις 0.3701 68 0.396 -0.842
εκαλειτο δε 0.3686 3 4.230 -0.131
οι 0.3640 11 1.714 -0.212
πολιτης 0.3415 18 2.319 -0.509
εστι δε 0.3385 5 2.858 -0.150
φησιν 0.3240 10 2.308 -0.256
ομηρος 0.3158 3 5.782 -0.179
της χωρας 0.3144 3 3.521 -0.109
δυναται 0.3037 3 4.985 -0.154
εν τη 0.3011 7 2.316 -0.174
χωρας 0.2989 4 2.645 -0.110
τη 0.2738 15 1.141 -0.201
παιδος 0.2557 4 3.291 -0.137
την 0.2360 13 0.960 -0.143
φησι 0.2331 12 1.680 -0.229
Negative termCoefficientDocsMean presentMean absent
το -0.5885 70 -0.273 0.637
οικητωρ -0.4275 6 -4.557 0.291
εστι και -0.4253 20 -1.090 0.273
ει -0.3839 4 -6.847 0.285
το εθνικον -0.3076 56 -0.515 0.656
εθνικον -0.3076 56 -0.515 0.656
θηλυκον -0.2785 12 -1.278 0.174
κωμη -0.2645 3 -2.190 0.068
τα -0.2510 12 -1.027 0.140
εθνος -0.2375 8 -3.080 0.268
πλησιον -0.2371 8 -1.093 0.095
γαρ -0.2086 9 -1.349 0.133
εστι -0.2051 24 -0.439 0.139
το θηλυκον -0.2043 6 -3.279 0.209
φασι -0.2019 8 -0.385 0.033
παυσανιας -0.1910 4 -3.313 0.138
του ποντου -0.1705 3 -2.411 0.075
ποντου -0.1705 3 -2.411 0.075
τινες δε -0.1605 3 -1.944 0.060
μεν -0.1600 4 -5.275 0.220

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 4.125; cross-validated R^2: -0.080.

Positive termCoefficientDocsMean presentMean absent
και 2.7070 66 3.121 1.126
δε και 1.1873 8 5.960 2.137
εκαλειτο δε 1.0000 3 10.658 2.188
εκαλειτο 0.9396 7 6.032 2.172
δε 0.8896 46 3.188 1.807
ος 0.8013 6 5.814 2.227
εστι δε 0.7889 5 7.216 2.191
την 0.7782 13 4.996 2.061
αφ 0.7509 6 5.411 2.253
εστι δε και 0.7447 3 7.762 2.278
απο 0.7223 27 4.509 1.678
λεγεται και 0.5916 8 5.748 2.155
εκαλειτο δε και 0.5432 3 10.658 2.188
της χωρας 0.5182 3 4.321 2.384
ης 0.5156 3 5.916 2.335
εν τη 0.4985 7 3.817 2.339
φησιν 0.4907 10 5.450 2.108
κατα 0.4601 10 5.186 2.138
λεγεται 0.4589 9 5.214 2.168
ει 0.4510 4 6.847 2.259
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.8394 56 2.088 2.894
εθνικον -0.8394 56 2.088 2.894
στραβων -0.5427 15 2.042 2.513
νησος -0.4319 9 1.725 2.514
τεταρτω -0.4091 3 1.019 2.487
εκαταιος ασια -0.3920 5 0.686 2.535
ασια -0.3920 5 0.686 2.535
εστι και -0.3912 20 2.530 2.421
εκ -0.3673 6 1.855 2.480
πολις -0.3457 68 2.500 2.320
εκαταιος -0.3437 14 1.905 2.530
προς τη -0.3283 4 1.545 2.480
πλησιον -0.3273 8 1.564 2.519
και πολις -0.3243 5 2.484 2.440
προς -0.3226 15 2.571 2.420
ωστε -0.3225 3 1.136 2.483
το -0.3209 70 2.339 2.684
εξ -0.3118 3 0.863 2.491
οτι -0.3061 5 2.054 2.463
πολις θρακης -0.3051 4 0.647 2.517

AI English terms predicting large absolute residuals

Features: 327; ridge alpha: 4.125; cross-validated R^2: -0.124.

Positive termCoefficientDocsMean presentMean absent
and 2.0424 54 3.297 1.439
called 1.0084 26 4.343 1.775
used 0.7933 8 6.243 2.112
be called 0.7865 4 8.399 2.194
to be called 0.7865 4 8.399 2.194
used to 0.7865 4 8.399 2.194
used to be 0.7865 4 8.399 2.194
to 0.7490 25 3.322 2.149
after 0.7194 23 4.595 1.800
also 0.6246 41 3.686 1.578
while 0.6201 4 10.719 2.098
be 0.6192 13 4.531 2.131
he 0.5837 3 9.282 2.231
named 0.5740 22 4.528 1.854
of 0.5646 67 2.840 1.635
is also 0.4817 32 3.657 1.871
on 0.4816 14 3.913 2.203
as per 0.4739 25 4.528 1.747
per 0.4739 25 4.528 1.747
to be 0.4713 10 4.580 2.205
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.6707 52 1.915 3.014
the ethnonym -0.6284 57 2.000 3.029
ethnonym is -0.5906 53 2.084 2.847
a city in -0.5545 42 2.361 2.502
ethnonym -0.5529 58 2.153 2.842
city in -0.5294 43 2.411 2.466
an -0.5026 22 2.251 2.497
book -0.4908 39 2.004 2.723
a city -0.4508 61 2.481 2.383
that -0.3917 16 2.136 2.501
is from -0.3807 12 2.585 2.423
or -0.3722 14 2.581 2.420
as in -0.3693 28 2.360 2.475
in -0.3655 86 2.396 2.729
strabo book -0.3611 10 1.515 2.546
people -0.3332 11 2.621 2.420
in his asia -0.3228 5 0.686 2.535
his asia -0.3228 5 0.686 2.535
sea -0.3209 7 1.387 2.522
strabo -0.3167 15 2.042 2.513

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 2.894; cross-validated R^2: -0.104.

Positive termCoefficientDocsMean presentMean absent
and 2.5828 52 3.403 1.402
also 1.4805 43 3.689 1.502
was 1.2672 20 4.097 2.029
after 1.0074 22 4.863 1.760
too 0.9940 3 11.238 2.171
are 0.9183 25 3.943 1.942
for 0.8979 11 4.385 2.203
is with 0.8818 4 8.121 2.206
says 0.8052 17 4.321 2.058
are also 0.8051 4 7.061 2.250
who was 0.7896 3 10.355 2.198
local 0.7795 6 5.889 2.223
were 0.7571 4 6.903 2.257
usage 0.7517 4 5.651 2.309
was also 0.7234 3 9.839 2.214
also said 0.6998 6 6.959 2.154
said 0.6998 6 6.959 2.154
it was 0.6994 4 8.266 2.200
came 0.6752 4 8.371 2.196
called 0.6748 21 3.729 2.100
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.8062 52 1.893 3.037
ethnonym is -0.8062 52 1.893 3.037
as in -0.6232 29 2.199 2.542
the ethnonym -0.5975 56 2.139 2.828
ethnonym -0.5710 57 2.189 2.779
name -0.5489 7 1.304 2.528
a city in -0.5204 40 2.477 2.420
city in -0.4888 41 2.527 2.384
that -0.4848 29 2.226 2.531
book -0.4795 39 2.004 2.723
or -0.4504 18 2.600 2.408
strabo book -0.4480 10 1.515 2.546
one -0.4368 12 2.077 2.492
strabo -0.4341 15 2.042 2.513
just as -0.4319 5 0.559 2.542
the city -0.4277 6 1.628 2.495
either -0.4273 6 2.882 2.415
in his asia -0.4265 5 0.686 2.535
his asia -0.4265 5 0.686 2.535
that the -0.3750 6 1.469 2.505

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.