gpt-4o-2024-08-06 v3 Evaluation

Generated: 2026-09-20 11:20:30 UTC

Pairs100
Distinct lemmas100
Slope0.939
R^20.983
BLEU-442.5%
chrF++67.3%
METEOR73.7%
ROUGE-L74.8%
BERTScore95.4%
COMET78.0%
BLEURT71.6%

Profile version id: 2409

First translation using this version: 2026-07-16

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

Created: 2026-07-16 06:54:03.500756+10:00

Active: True

Notes: Model timeline evaluation: GPT-4o (2024-08-06) using prompt text from gpt-5.5 v3; approved-human-only for the 100-row Kappa corpus.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.939
Intercept1.884
R^20.983
P-value6.35e-88
Pearson r0.991
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.340
Mean absolute length residual3.626
Mean absolute percent error8.7%

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 = -0.328, R^2 = 0.108, p = 0.0009.
chrF++ vs passage length
chrF++: significant negative; r = -0.374, R^2 = 0.140, p = 0.0001.
METEOR vs passage length
METEOR: significant negative; r = -0.441, R^2 = 0.195, p = 4.29e-06.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.486, R^2 = 0.236, p = 3.04e-07.
BERTScore vs passage length
BERTScore: significant negative; r = -0.599, R^2 = 0.359, p = 4.54e-11.
COMET vs passage length
COMET: significant negative; r = -0.399, R^2 = 0.159, p = 3.93e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.662, R^2 = 0.438, p = 6.35e-14.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.348, R^2 = 0.121, p = 0.0004.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.346, R^2 = 0.120, p = 0.0004.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.347, R^2 = 0.120, p = 0.0004.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.343, R^2 = 0.118, p = 0.0005.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.328 0.108 0.0009 -0.00234 ok
chrF++ 100 significant negative -0.374 0.140 0.0001 -0.00158 ok
METEOR 100 significant negative -0.441 0.195 4.29e-06 -0.00227 ok
ROUGE-L 100 significant negative -0.486 0.236 3.04e-07 -0.00225 ok
BERTScore 100 significant negative -0.599 0.359 4.54e-11 -0.00050 ok
COMET 100 significant negative -0.399 0.159 3.93e-05 -0.00081 ok
BLEURT 100 significant negative -0.662 0.438 6.35e-14 -0.00248 ok
Trigram precision 100 significant negative -0.348 0.121 0.0004 -0.00301 ok
Trigram recall 100 significant negative -0.346 0.120 0.0004 -0.00292 ok
Trigram F1 100 significant negative -0.347 0.120 0.0004 -0.00295 ok
Trigram Jaccard 100 significant negative -0.343 0.118 0.0005 -0.00298 ok

Translation Similarity Metrics

Mean BLEU-442.5%
Mean chrF++67.3%
Mean METEOR73.7%
Mean ROUGE-L74.8%
Mean BERTScore95.4%
Mean COMET78.0%
Mean BLEURT71.6%
Legacy smoothed corpus BLEU43.8%
Unigram F175.5%
Bigram F150.1%
Trigram precision36.8%
Trigram recall36.1%
Trigram F136.5%
Trigram Jaccard22.3%
4-gram F127.7%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
3288 Κορώνεια 110 123 17.828 29.5% 62.9% 61.6% 63.2% 93.6% 76.3% 62.0% reviewed/approved 12746 gpt-4o-2024-08-06
2609 Κάσπειρος 135 116 -12.647 34.2% 57.6% 57.9% 62.2% 89.3% 71.6% 60.9% reviewed/approved 12707 gpt-4o-2024-08-06
7254 Κύτα 84 93 12.241 43.0% 66.3% 74.6% 68.9% 93.7% 76.4% 64.4% reviewed/approved 12794 gpt-4o-2024-08-06
7260 Κυχρεῖος πάγος 61 71 11.838 42.3% 68.8% 73.2% 66.7% 95.7% 76.0% 62.7% reviewed/approved 12812 gpt-4o-2024-08-06
2605 Κάσιον 56 66 11.533 33.6% 65.2% 71.7% 65.6% 94.0% 74.1% 67.1% reviewed/approved 12701 gpt-4o-2024-08-06
2455 Κάναστρον 60 48 -10.223 19.5% 51.8% 54.3% 66.7% 93.0% 65.6% 57.6% reviewed/approved 12668 gpt-4o-2024-08-06
2465 Κάνωπος 83 90 10.180 44.6% 70.1% 77.5% 78.6% 94.6% 75.4% 64.9% reviewed/approved 12671 gpt-4o-2024-08-06
7255 Κυτέριον 22 32 9.459 29.0% 61.1% 57.4% 59.3% 94.6% 73.0% 66.7% reviewed/approved 12797 gpt-4o-2024-08-06
2604 Καρχηδών 117 121 9.255 23.7% 57.6% 62.2% 62.8% 91.9% 76.2% 54.8% reviewed/approved 12698 gpt-4o-2024-08-06
2116 Κάληρος 40 31 -8.443 11.6% 44.4% 45.8% 53.5% 92.5% 69.9% 64.1% reviewed/approved 12614 gpt-4o-2024-08-06
7257 Κυτώνιον 25 17 -8.358 31.6% 53.3% 56.5% 66.7% 94.0% 75.2% 57.6% reviewed/approved 12803 gpt-4o-2024-08-06
2119 Κάλλατις 65 71 8.082 24.3% 55.1% 63.3% 58.8% 93.3% 69.7% 58.7% reviewed/approved 12623 gpt-4o-2024-08-06
7259 Κύφος 54 60 7.411 31.4% 64.4% 68.8% 63.2% 93.9% 72.2% 59.7% reviewed/approved 12809 gpt-4o-2024-08-06
2055 Καβασσός 109 97 -7.233 20.8% 55.5% 53.9% 55.3% 93.2% 77.6% 55.0% reviewed/approved 12539 gpt-4o-2024-08-06
2056 Καβειρία 112 100 -7.050 25.7% 60.1% 60.3% 67.0% 94.3% 74.0% 59.8% reviewed/approved 12542 gpt-4o-2024-08-06
2115 Καλὴ ἀκτή 64 55 -6.979 10.8% 41.2% 45.9% 52.1% 90.9% 65.6% 46.4% reviewed/approved 12611 gpt-4o-2024-08-06
7247 Κύρνος 50 42 -6.833 37.7% 59.5% 66.2% 65.2% 95.8% 75.3% 73.9% reviewed/approved 12773 gpt-4o-2024-08-06
2477 Καρδαμύλη 43 49 6.740 23.1% 61.9% 78.2% 69.6% 95.1% 78.6% 74.7% reviewed/approved 12680 gpt-4o-2024-08-06
3496 Κοτιάειον 71 75 6.448 29.3% 53.0% 62.5% 57.5% 93.4% 74.7% 66.4% reviewed/approved 12749 gpt-4o-2024-08-06
2484 Καρία 270 249 -6.410 22.9% 54.2% 52.7% 54.3% 89.2% 77.6% 45.4% reviewed/approved 12683 gpt-4o-2024-08-06
2626 Κατάνη 95 85 -6.087 36.2% 59.7% 62.4% 67.8% 93.2% 73.1% 63.4% reviewed/approved 12731 gpt-4o-2024-08-06
2331 Κάλυδνα 43 48 5.740 50.9% 76.3% 85.2% 81.3% 95.7% 80.0% 77.1% reviewed/approved 12644 gpt-4o-2024-08-06
2342 Κάμιρος 54 58 5.411 39.1% 65.2% 72.5% 71.4% 94.9% 76.9% 70.3% reviewed/approved 12659 gpt-4o-2024-08-06
2470 Καππαδοκία 68 71 5.265 17.3% 55.8% 74.0% 70.5% 95.3% 76.2% 68.7% reviewed/approved 12677 gpt-4o-2024-08-06
2597 Καρπασία 112 112 4.950 37.1% 66.2% 62.0% 66.1% 94.4% 76.2% 59.8% reviewed/approved 12686 gpt-4o-2024-08-06
7251 Κύρτος 66 59 -4.857 40.4% 56.9% 74.8% 75.2% 92.6% 66.6% 61.7% reviewed/approved 12785 gpt-4o-2024-08-06
7264 Κωλιάς 61 64 4.838 45.4% 68.5% 68.6% 72.0% 95.8% 72.6% 64.6% reviewed/approved 12824 gpt-4o-2024-08-06
2621 Κάστνιον 25 30 4.642 43.7% 72.7% 75.2% 72.7% 97.2% 80.1% 75.9% reviewed/approved 12713 gpt-4o-2024-08-06
2620 Κάσταξ 20 25 4.336 37.9% 62.4% 65.4% 71.1% 95.1% 75.8% 73.8% reviewed/approved 12710 gpt-4o-2024-08-06
3258 Κεκρυφάλεια 26 22 -4.297 48.2% 71.2% 77.1% 83.3% 98.0% 78.8% 77.8% reviewed/approved 12743 gpt-4o-2024-08-06

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: 4.125; cross-validated R^2: -0.084.

Positive termCoefficientDocsMean presentMean absent
απο 1.9994 27 3.226 -1.193
την 1.9448 13 1.911 -0.286
ος 1.8176 6 7.264 -0.464
πολιτης 1.6686 18 4.488 -0.985
δε και 1.6392 8 4.337 -0.377
εστι δε και 1.4263 3 11.274 -0.349
και το 1.4081 9 3.512 -0.347
εν τη 1.2779 7 4.339 -0.327
ου 1.2208 14 1.958 -0.319
και 1.1387 66 0.413 -0.802
παιδος 1.1193 4 5.377 -0.224
εις 1.0935 7 3.402 -0.256
δυναται 1.0627 3 6.684 -0.207
βοιωτιας 1.0519 3 3.847 -0.119
ου και 1.0490 4 6.052 -0.252
του 1.0149 37 1.770 -1.040
ποσειδωνος 1.0009 3 7.655 -0.237
απο του 0.9131 9 3.513 -0.347
εκληθη 0.9122 4 4.961 -0.207
λεγεται 0.8973 9 2.461 -0.243
Negative termCoefficientDocsMean presentMean absent
τα -1.9038 12 -3.363 0.459
τω -1.5429 16 -1.627 0.310
θεοπομπος -1.3013 5 -2.392 0.126
το -1.2989 70 -0.023 0.053
γαρ -1.0514 9 -2.658 0.263
ουτως -1.0401 5 -4.681 0.246
μετα -0.9945 3 -8.381 0.259
αλλ -0.9840 3 -5.868 0.181
τε -0.9548 5 0.007 -3.62e-04
μεταξυ -0.8461 5 -0.291 0.015
δια το -0.8013 3 -4.446 0.137
εν τω -0.7904 6 -3.020 0.193
ως το -0.7598 3 -6.380 0.197
εκαλειτο -0.7524 7 -1.928 0.145
τινες -0.7414 6 -1.927 0.123
θρακης -0.7313 5 -3.495 0.184
το εθνικον -0.7281 56 -0.608 0.774
εθνικον -0.7281 56 -0.608 0.774
παρα το -0.7081 3 -3.348 0.104
εκει -0.7029 3 -4.405 0.136

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 1.000; cross-validated R^2: -0.010.

Positive termCoefficientDocsMean presentMean absent
την 4.6125 13 6.888 3.139
ου 2.7735 14 6.510 3.157
ουτως 2.6640 5 8.754 3.356
δυναται 2.6606 3 11.507 3.382
απο 2.6319 27 6.260 2.652
ος 2.4819 6 7.752 3.363
δε 2.4341 46 5.235 2.256
παρα 2.4283 7 7.567 3.329
μεταξυ 2.3277 5 7.423 3.426
τε 2.2706 5 5.891 3.507
αλλ 2.2396 3 9.378 3.448
δε και 2.1908 8 7.617 3.279
εστι δε 2.1656 5 9.570 3.313
και 2.1475 66 4.392 2.139
τα 2.1180 12 6.092 3.290
εκαταιος ευρωπη 1.9137 5 8.220 3.384
ευρωπη 1.9137 5 8.220 3.384
εστι δε και 1.8944 3 11.274 3.390
εδει 1.7176 3 9.156 3.455
αφ ου 1.6954 5 7.164 3.440
Negative termCoefficientDocsMean presentMean absent
πολις και -1.8143 6 2.688 3.686
κατα -1.6139 10 3.267 3.666
πορρω -1.5298 3 0.639 3.718
ου πορρω -1.5298 3 0.639 3.718
προς τη -1.5063 4 1.598 3.711
οικητωρ -1.4556 6 2.593 3.692
του -1.4103 37 4.810 2.931
ην -1.3913 13 3.547 3.638
προς -1.3567 15 4.005 3.559
δευτερω -1.3157 3 1.555 3.690
του δε -1.2956 3 2.675 3.655
αυτην -1.2538 9 3.083 3.680
τους -1.2377 4 3.283 3.640
παυσανιας -1.1783 4 2.892 3.657
και του -0.9988 4 2.955 3.654
φησι το -0.9536 3 0.819 3.713
και θηλυκως -0.9443 5 3.196 3.649
θηλυκως -0.9443 5 3.196 3.649
μια -0.9428 3 1.982 3.677
τεταρτω -0.9387 3 3.175 3.640

AI English terms predicting large absolute residuals

Features: 321; ridge alpha: 1.425; cross-validated R^2: -0.142.

Positive termCoefficientDocsMean presentMean absent
and 3.3926 54 4.922 2.105
among 2.2281 5 10.091 3.286
among the 2.0071 4 10.906 3.323
say 1.8960 10 7.272 3.221
europe 1.8851 6 8.890 3.290
the 1.8658 94 3.697 2.513
this 1.6456 10 6.490 3.308
of 1.6134 64 4.192 2.620
after 1.5789 20 6.395 2.934
was 1.4795 16 5.719 3.228
to 1.4591 21 5.274 3.188
the island 1.4241 3 8.655 3.471
is a 1.3851 20 5.593 3.134
it 1.3129 42 5.215 2.476
between 1.3062 5 7.423 3.426
was also called 1.3060 3 8.433 3.477
there 1.2722 31 5.434 2.814
boiotia 1.2244 3 8.547 3.474
who 1.1595 12 6.597 3.221
and there 1.1502 3 10.694 3.408
Negative termCoefficientDocsMean presentMean absent
also a city -1.2179 6 2.462 3.700
inhabitant -1.1631 6 2.810 3.678
call -1.0529 4 2.019 3.693
of the -0.8940 33 3.946 3.469
a city and -0.8865 4 2.074 3.691
in his asia -0.8512 4 1.064 3.733
his asia -0.8512 4 1.064 3.733
not far from -0.8109 3 0.639 3.718
not far -0.8109 3 0.639 3.718
far from -0.8109 3 0.639 3.718
far -0.8109 3 0.639 3.718
which is -0.8094 6 2.872 3.674
according to -0.8037 6 3.574 3.629
according -0.8037 6 3.574 3.629
for the -0.7760 6 3.469 3.636
town -0.7535 4 2.943 3.655
strabo -0.7325 15 3.941 3.571
it used to -0.7266 3 3.317 3.636
it used -0.7266 3 3.317 3.636
city in -0.7156 44 3.681 3.583

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 1.000; cross-validated R^2: -0.081.

Positive termCoefficientDocsMean presentMean absent
who 3.3369 13 7.507 3.046
and 2.8433 52 4.793 2.362
the 2.6448 90 3.870 1.434
this 2.4386 15 7.011 3.029
has 2.1941 6 8.042 3.344
europe 2.1580 6 8.890 3.290
out 1.9802 3 9.774 3.436
boiotia 1.9289 3 8.547 3.474
her 1.8952 3 10.976 3.399
be 1.8824 15 6.374 3.141
as 1.8505 60 4.075 2.953
among 1.7079 4 6.966 3.487
his europe 1.6321 5 8.220 3.384
in his europe 1.6321 5 8.220 3.384
was named 1.6072 3 9.192 3.454
like 1.5803 5 7.069 3.445
because 1.5613 12 5.901 3.316
between 1.5586 6 7.063 3.407
of 1.5524 66 4.179 2.552
all 1.5345 3 10.738 3.406
Negative termCoefficientDocsMean presentMean absent
either -1.5580 6 3.374 3.642
they are -1.4660 7 2.465 3.713
one of -1.4465 5 2.585 3.681
just as -1.4042 5 1.551 3.735
are -1.2838 25 4.009 3.498
town -1.1789 3 1.230 3.700
to the -1.1756 10 3.438 3.647
it used to -1.0811 3 2.582 3.658
it used -1.0811 3 2.582 3.658
and a -1.0793 3 1.500 3.692
one of the -1.0192 4 3.008 3.652
which -1.0158 20 4.054 3.519
with a -1.0154 4 2.356 3.679
says that -0.9891 12 3.256 3.677
the ethnonym is -0.9598 52 3.451 3.816
ethnonym is -0.9598 52 3.451 3.816
not far from -0.9346 3 0.639 3.718
not far -0.9346 3 0.639 3.718
far from -0.9346 3 0.639 3.718
far -0.9346 3 0.639 3.718

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.