gpt-4o-2024-05-13 v3 Evaluation

Generated: 2026-09-20 11:20:18 UTC

Pairs100
Distinct lemmas100
Slope0.880
R^20.913
BLEU-441.6%
chrF++67.0%
METEOR72.9%
ROUGE-L74.5%
BERTScore95.3%
COMET78.1%
BLEURT71.9%

Profile version id: 2406

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 (2024-05-13) 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
Slope0.880
Intercept3.621
R^20.913
P-value9.63e-54
Pearson r0.955
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words42.460
Mean absolute length residual4.427
Mean absolute percent error8.4%

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 = -0.318, R^2 = 0.101, p = 0.0013.
chrF++ vs passage length
chrF++: significant negative; r = -0.381, R^2 = 0.145, p = 9.16e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.458, R^2 = 0.210, p = 1.65e-06.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.452, R^2 = 0.204, p = 2.38e-06.
BERTScore vs passage length
BERTScore: significant negative; r = -0.529, R^2 = 0.280, p = 1.48e-08.
COMET vs passage length
COMET: significant negative; r = -0.410, R^2 = 0.168, p = 2.26e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.713, R^2 = 0.508, p = 9.16e-17.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.307, R^2 = 0.094, p = 0.0019.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.329, R^2 = 0.108, p = 0.0008.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.321, R^2 = 0.103, p = 0.0011.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.321, R^2 = 0.103, p = 0.0011.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.318 0.101 0.0013 -0.00223 ok
chrF++ 100 significant negative -0.381 0.145 9.16e-05 -0.00175 ok
METEOR 100 significant negative -0.458 0.210 1.65e-06 -0.00256 ok
ROUGE-L 100 significant negative -0.452 0.204 2.38e-06 -0.00229 ok
BERTScore 100 significant negative -0.529 0.280 1.48e-08 -0.00047 ok
COMET 100 significant negative -0.410 0.168 2.26e-05 -0.00091 ok
BLEURT 100 significant negative -0.713 0.508 9.16e-17 -0.00274 ok
Trigram precision 100 significant negative -0.307 0.094 0.0019 -0.00280 ok
Trigram recall 100 significant negative -0.329 0.108 0.0008 -0.00295 ok
Trigram F1 100 significant negative -0.321 0.103 0.0011 -0.00290 ok
Trigram Jaccard 100 significant negative -0.321 0.103 0.0011 -0.00295 ok

Translation Similarity Metrics

Mean BLEU-441.6%
Mean chrF++67.0%
Mean METEOR72.9%
Mean ROUGE-L74.5%
Mean BERTScore95.3%
Mean COMET78.1%
Mean BLEURT71.9%
Legacy smoothed corpus BLEU43.0%
Unigram F174.8%
Bigram F149.9%
Trigram precision37.0%
Trigram recall35.5%
Trigram F136.2%
Trigram Jaccard22.1%
4-gram F127.2%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2468 Καπετώλιον 138 32 -93.020 0.3% 14.4% 12.9% 20.0% 82.4% 47.1% 37.4% reviewed/approved 13573 gpt-4o-2024-05-13
3288 Κορώνεια 110 126 25.612 32.4% 63.6% 63.7% 65.0% 92.5% 77.1% 61.1% reviewed/approved 13645 gpt-4o-2024-05-13
2603 Κάρυστος 175 176 18.432 32.6% 62.3% 64.6% 63.8% 94.1% 77.7% 60.2% reviewed/approved 13594 gpt-4o-2024-05-13
2604 Καρχηδών 117 123 16.454 35.3% 63.4% 65.1% 71.4% 93.5% 78.6% 60.6% reviewed/approved 13597 gpt-4o-2024-05-13
2119 Κάλλατις 65 74 13.198 18.6% 52.6% 56.2% 56.1% 91.8% 67.4% 53.0% reviewed/approved 13522 gpt-4o-2024-05-13
7254 Κύτα 84 90 12.484 39.0% 66.6% 65.3% 67.8% 93.0% 78.1% 69.2% reviewed/approved 13693 gpt-4o-2024-05-13
2326 Καλλίαρος 47 57 12.033 37.4% 68.1% 74.0% 73.1% 94.8% 71.4% 64.4% reviewed/approved 13528 gpt-4o-2024-05-13
7258 Κύτωρος 30 39 8.988 34.3% 70.3% 67.7% 66.7% 94.8% 76.3% 72.4% reviewed/approved 13705 gpt-4o-2024-05-13
7260 Κυχρεῖος πάγος 61 66 8.717 40.2% 65.4% 72.5% 64.6% 95.2% 75.1% 63.6% reviewed/approved 13711 gpt-4o-2024-05-13
2342 Κάμιρος 54 58 6.875 33.7% 66.1% 79.1% 76.8% 95.3% 77.3% 69.0% reviewed/approved 13558 gpt-4o-2024-05-13
2328 Καλλίπολις 43 48 6.552 42.8% 66.6% 73.3% 68.1% 94.6% 74.5% 69.0% reviewed/approved 13534 gpt-4o-2024-05-13
2055 Καβασσός 109 93 -6.508 33.5% 58.1% 60.7% 65.3% 94.1% 75.5% 51.5% reviewed/approved 13438 gpt-4o-2024-05-13
2465 Κάνωπος 83 83 6.364 39.3% 62.9% 81.3% 78.3% 94.2% 76.3% 68.1% reviewed/approved 13570 gpt-4o-2024-05-13
7248 Κύρου πόλις 32 38 6.228 10.2% 55.2% 54.9% 57.1% 93.0% 75.9% 77.7% reviewed/approved 13675 gpt-4o-2024-05-13
7255 Κυτέριον 22 29 6.025 30.6% 62.5% 64.7% 66.7% 94.8% 72.9% 68.3% reviewed/approved 13696 gpt-4o-2024-05-13
7266 Κώμη 79 79 5.883 14.1% 46.5% 47.8% 41.8% 92.3% 73.4% 56.1% reviewed/approved 13729 gpt-4o-2024-05-13
7259 Κύφος 54 57 5.875 27.6% 63.7% 67.5% 70.3% 94.7% 73.4% 67.0% reviewed/approved 13708 gpt-4o-2024-05-13
2484 Καρία 270 247 5.860 28.3% 57.8% 53.5% 58.8% 90.5% 78.5% 40.7% reviewed/approved 13582 gpt-4o-2024-05-13
7249 Κύρρος 45 49 5.792 40.5% 62.8% 75.8% 70.2% 94.7% 74.3% 61.8% reviewed/approved 13678 gpt-4o-2024-05-13
7251 Κύρτος 66 56 -5.681 28.6% 48.8% 63.3% 62.9% 92.3% 64.3% 63.1% reviewed/approved 13684 gpt-4o-2024-05-13
7247 Κύρνος 50 42 -5.606 16.4% 48.1% 54.4% 52.2% 93.2% 75.0% 69.0% reviewed/approved 13672 gpt-4o-2024-05-13
2470 Καππαδοκία 68 69 5.559 24.2% 64.3% 77.1% 75.9% 96.4% 77.5% 70.8% reviewed/approved 13576 gpt-4o-2024-05-13
2602 Καρύανδα 34 28 -5.531 30.0% 63.8% 74.9% 77.4% 96.8% 79.9% 76.1% reviewed/approved 13591 gpt-4o-2024-05-13
7257 Κυτώνιον 25 21 -4.614 66.1% 75.8% 80.7% 87.0% 96.9% 78.7% 63.9% reviewed/approved 13702 gpt-4o-2024-05-13
2077 Καιρή 17 14 -4.576 44.7% 61.5% 53.4% 64.5% 94.4% 77.0% 68.0% reviewed/approved 13477 gpt-4o-2024-05-13
7253 Κυρτώνιος 18 15 -4.456 43.3% 69.8% 78.2% 84.8% 97.2% 79.8% 80.8% reviewed/approved 13690 gpt-4o-2024-05-13
2599 Καρπήσιοι 10 8 -4.418 17.5% 54.9% 68.6% 77.8% 95.4% 86.9% 78.8% reviewed/approved 13588 gpt-4o-2024-05-13
2621 Κάστνιον 25 30 4.386 41.9% 72.2% 75.2% 76.4% 96.9% 79.9% 75.8% reviewed/approved 13612 gpt-4o-2024-05-13
2627 Κατάνειρα 21 18 -4.095 23.6% 63.8% 70.4% 82.1% 95.5% 73.7% 71.4% reviewed/approved 13627 gpt-4o-2024-05-13
2075 Καινίνη 15 13 -3.817 42.6% 69.2% 73.6% 78.6% 96.7% 83.2% 75.8% reviewed/approved 13468 gpt-4o-2024-05-13

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: -0.120.

Positive termCoefficientDocsMean presentMean absent
και 2.1276 66 0.497 -0.965
απο 1.6490 27 5.536 -2.048
δε και 1.0319 8 9.515 -0.827
ου 0.9176 14 4.568 -0.744
τη 0.8531 15 3.990 -0.704
οι 0.8217 11 6.323 -0.782
στραβων 0.8141 15 4.607 -0.813
πολιτης 0.7938 18 4.969 -1.091
εστι 0.7708 24 3.099 -0.979
πολις 0.7219 68 1.210 -2.571
παρα 0.7060 7 5.867 -0.442
την 0.7052 13 -1.065 0.159
εστι δε και 0.6812 3 14.539 -0.450
θηλυκον 0.6462 12 4.248 -0.579
βοιωτιας 0.6449 3 8.030 -0.248
λεγεται 0.5995 9 6.683 -0.661
λεγεται και 0.5959 8 7.324 -0.637
της 0.5786 35 -0.139 0.075
εν τη 0.5636 7 4.742 -0.357
εκαλειτο δε 0.5520 3 13.582 -0.420
Negative termCoefficientDocsMean presentMean absent
γαρ -3.0897 9 -6.495 0.642
δια του -1.9001 10 -5.435 0.604
του -1.6491 37 0.922 -0.542
δια -1.4375 16 -2.176 0.414
ζευς -1.0701 3 -28.839 0.892
αι -0.9961 3 -29.253 0.905
ηρωδιανος -0.9726 3 -28.671 0.887
ωστε -0.9401 3 -25.067 0.775
δε της -0.8758 6 -13.640 0.871
εξ -0.7784 3 -25.383 0.785
φησιν -0.7368 10 -7.211 0.801
τα -0.6719 12 -6.346 0.865
ειναι -0.6372 8 -9.400 0.817
το -0.6083 70 -0.259 0.603
τω -0.6034 16 -4.842 0.922
θεοπομπος -0.5321 5 -3.119 0.164
το εθνικον -0.4917 56 -1.081 1.376
εθνικον -0.4917 56 -1.081 1.376
περι -0.4692 11 -6.726 0.831
δε -0.4553 46 0.296 -0.252

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 5.878; cross-validated R^2: -0.585.

Positive termCoefficientDocsMean presentMean absent
γαρ 6.1791 9 16.801 3.203
του 3.9217 37 7.824 2.432
δια του 3.6619 10 13.844 3.380
δια 3.0076 16 10.905 3.193
την 2.7985 13 15.040 2.841
εις 2.2960 7 23.165 3.016
ος 2.1968 6 23.026 3.240
δε 2.1299 46 6.307 2.825
εξ 2.1252 3 36.630 3.431
εν 2.0142 37 7.340 2.716
ωστε 1.8608 3 37.355 3.408
ειναι 1.8538 8 16.883 3.344
αι 1.8519 3 33.836 3.517
ζευς 1.8475 3 33.174 3.538
ηρωδιανος 1.6871 3 33.342 3.533
δε της 1.6641 6 20.600 3.394
περι 1.3399 11 13.525 3.302
τα 1.0310 12 11.753 3.428
της 0.9718 35 7.338 2.859
τω 0.9568 16 9.871 3.390
Negative termCoefficientDocsMean presentMean absent
ως -1.4590 52 3.474 5.459
το εθνικον -0.9132 56 4.779 3.979
εθνικον -0.9132 56 4.779 3.979
προς τη -0.8798 4 0.916 4.573
οικητωρ -0.7992 6 2.782 4.532
ηροδοτος -0.7950 9 2.561 4.611
τον -0.7661 10 4.247 4.447
τους -0.7124 4 2.217 4.519
τινες -0.6943 6 1.789 4.595
νησος -0.6795 9 2.501 4.617
πολις -0.5984 68 3.839 5.676
ην -0.5934 13 3.660 4.541
δευτερω -0.5690 3 1.422 4.520
εκ -0.5518 6 3.204 4.505
και πολις -0.5455 5 1.904 4.560
τινες δε -0.5353 3 1.623 4.514
και του -0.4975 4 3.097 4.482
καππαδοκιας -0.4947 3 3.057 4.469
εκαταιος -0.4908 14 4.586 4.401
λυκιας -0.4847 3 2.899 4.474

AI English terms predicting large absolute residuals

Features: 317; ridge alpha: 24.245; cross-validated R^2: -0.754.

Positive termCoefficientDocsMean presentMean absent
about 1.1935 3 31.392 3.593
called the 1.1700 4 24.886 3.574
that 0.9868 13 13.588 3.058
says 0.9750 8 14.824 3.523
to the 0.9589 10 13.142 3.458
the 0.9292 93 4.618 1.887
name 0.8035 13 10.913 3.458
to 0.7785 22 9.069 3.118
with 0.6993 17 9.046 3.481
called 0.6737 34 8.092 2.539
an 0.6224 17 8.654 3.561
of the 0.6160 28 7.872 3.087
of 0.4925 66 5.492 2.358
it 0.4814 42 7.113 2.482
and 0.3513 52 6.131 2.580
form 0.2655 17 7.317 3.835
from the form 0.2207 4 10.970 4.154
the form 0.2207 4 10.970 4.154
the son 0.1613 13 9.148 3.721
the son of 0.1613 13 9.148 3.721
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.3812 56 3.234 5.945
ethnonym -0.3785 56 3.234 5.945
the ethnonym is -0.3732 53 3.038 5.993
ethnonym is -0.3732 53 3.038 5.993
is -0.3667 89 3.646 10.741
as -0.3027 60 3.759 5.429
a city -0.2946 66 3.936 5.380
just as -0.2782 13 1.706 4.833
just -0.2782 13 1.706 4.833
book -0.2653 41 3.067 5.372
a city in -0.2638 41 3.718 4.919
city in -0.2346 42 3.656 4.985
in the -0.2285 20 3.478 4.664
as in -0.1820 20 3.405 4.682
island -0.1817 14 2.720 4.705
in his -0.1754 24 3.865 4.604
people -0.1752 12 3.254 4.587
an island -0.1727 9 2.155 4.651
book the -0.1714 11 2.572 4.656
his -0.1674 34 4.017 4.638

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 5.878; cross-validated R^2: -0.621.

Positive termCoefficientDocsMean presentMean absent
with 6.2457 16 9.676 3.427
the 2.6693 90 4.706 1.918
on the 2.5634 10 12.168 3.567
sanctuary 2.5494 3 33.500 3.528
forms 2.4567 4 26.647 3.501
on 1.9905 16 9.176 3.522
is 1.7397 87 4.678 2.746
from 1.7344 37 7.148 2.828
it 1.6464 37 6.953 2.943
that 1.5639 29 7.850 3.029
at 1.4732 6 22.087 3.300
has 1.4056 6 23.612 3.202
and 1.3346 52 6.157 2.552
long 1.3127 3 34.921 3.484
this is 1.2153 4 26.658 3.500
diphthong 1.1836 3 33.342 3.532
is with 1.1575 4 28.948 3.405
zeus 1.1549 5 23.814 3.406
herodian 1.1429 3 33.342 3.533
so 1.1250 5 23.823 3.406
Negative termCoefficientDocsMean presentMean absent
as -1.4961 60 3.557 5.731
as in -0.8651 29 3.338 4.871
the ethnonym is -0.7509 52 4.731 4.098
ethnonym is -0.7509 52 4.731 4.098
an -0.7214 22 2.840 4.874
near -0.7091 16 3.355 4.631
the ethnonym -0.6941 56 4.876 3.855
ethnonym -0.6817 57 4.905 3.793
some -0.6378 10 3.079 4.577
island -0.6342 14 2.720 4.705
his -0.6328 34 3.725 4.788
herodotos -0.5576 9 2.561 4.611
book -0.5365 39 3.311 5.140
or after -0.5057 3 0.485 4.549
city in -0.4771 41 3.759 4.891
say that -0.4692 9 3.173 4.551
an island -0.4483 9 2.650 4.602
in his -0.4418 19 4.407 4.431
a city in -0.4309 40 3.825 4.828
ethnonyms -0.4273 3 2.773 4.478

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.