gpt-5.2 v3 Evaluation

Generated: 2026-09-20 11:21:34 UTC

Pairs100
Distinct lemmas100
Slope0.979
R^20.990
BLEU-446.6%
chrF++69.9%
METEOR77.2%
ROUGE-L77.5%
BERTScore95.7%
COMET79.0%
BLEURT73.6%

Profile version id: 2330

First translation using this version: 2026-07-15

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

Created: 2026-07-15 12:28:27.643960+10:00

Active: True

Notes: Model timeline evaluation: GPT-5.2 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
Slope0.979
Intercept0.716
R^20.990
P-value4.08e-100
Pearson r0.995
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.960
Mean absolute length residual2.656
Mean absolute percent error7.0%

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 = -0.308, R^2 = 0.095, p = 0.0018.
chrF++ vs passage length
chrF++: significant negative; r = -0.331, R^2 = 0.110, p = 0.0008.
METEOR vs passage length
METEOR: significant negative; r = -0.360, R^2 = 0.130, p = 0.0002.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.342, R^2 = 0.117, p = 0.0005.
BERTScore vs passage length
BERTScore: significant negative; r = -0.587, R^2 = 0.344, p = 1.39e-10.
COMET vs passage length
COMET: significant negative; r = -0.406, R^2 = 0.164, p = 2.84e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.705, R^2 = 0.498, p = 2.55e-16.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.272, R^2 = 0.074, p = 0.0062.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.277, R^2 = 0.077, p = 0.0053.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.274, R^2 = 0.075, p = 0.0058.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.286, R^2 = 0.082, p = 0.0040.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.308 0.095 0.0018 -0.00204 ok
chrF++ 100 significant negative -0.331 0.110 0.0008 -0.00136 ok
METEOR 100 significant negative -0.360 0.130 0.0002 -0.00161 ok
ROUGE-L 100 significant negative -0.342 0.117 0.0005 -0.00144 ok
BERTScore 100 significant negative -0.587 0.344 1.39e-10 -0.00044 ok
COMET 100 significant negative -0.406 0.164 2.84e-05 -0.00075 ok
BLEURT 100 significant negative -0.705 0.498 2.55e-16 -0.00248 ok
Trigram precision 100 significant negative -0.272 0.074 0.0062 -0.00212 ok
Trigram recall 100 significant negative -0.277 0.077 0.0053 -0.00209 ok
Trigram F1 100 significant negative -0.274 0.075 0.0058 -0.00209 ok
Trigram Jaccard 100 significant negative -0.286 0.082 0.0040 -0.00226 ok

Translation Similarity Metrics

Mean BLEU-446.6%
Mean chrF++69.9%
Mean METEOR77.2%
Mean ROUGE-L77.5%
Mean BERTScore95.7%
Mean COMET79.0%
Mean BLEURT73.6%
Legacy smoothed corpus BLEU49.5%
Unigram F178.9%
Bigram F155.3%
Trigram precision42.1%
Trigram recall41.9%
Trigram F142.0%
Trigram Jaccard26.6%
4-gram F133.0%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2484 Καρία 270 251 -14.174 34.6% 61.0% 61.8% 66.0% 91.6% 79.4% 50.9% reviewed/approved 11484 gpt-5.2-2025-12-11
2604 Καρχηδών 117 128 12.685 29.8% 60.6% 72.0% 68.0% 93.1% 77.7% 60.6% reviewed/approved 11499 gpt-5.2-2025-12-11
2603 Κάρυστος 175 184 11.876 29.4% 61.4% 73.3% 70.8% 92.5% 77.3% 58.6% reviewed/approved 11496 gpt-5.2-2025-12-11
7254 Κύτα 84 94 11.008 38.4% 67.4% 70.8% 68.5% 93.1% 76.8% 59.3% reviewed/approved 11593 gpt-5.2-2025-12-11
2056 Καβειρία 112 100 -10.417 38.4% 67.6% 71.4% 72.6% 94.1% 77.1% 56.0% reviewed/approved 11344 gpt-5.2-2025-12-11
2609 Κάσπειρος 135 142 9.055 42.4% 67.1% 74.7% 72.2% 92.1% 73.5% 64.6% reviewed/approved 11508 gpt-5.2-2025-12-11
2477 Καρδαμύλη 43 50 7.166 27.5% 65.4% 75.3% 66.7% 94.6% 77.9% 71.3% reviewed/approved 11481 gpt-5.2-2025-12-11
7255 Κυτέριον 22 29 6.735 29.0% 64.0% 67.5% 66.7% 95.3% 74.8% 73.3% reviewed/approved 11596 gpt-5.2-2025-12-11
3288 Κορώνεια 110 115 6.542 39.7% 69.3% 64.6% 75.2% 94.6% 77.6% 64.7% reviewed/approved 11547 gpt-5.2-2025-12-11
2455 Κάναστρον 60 54 -5.485 23.5% 58.4% 65.5% 68.4% 94.1% 71.4% 60.6% reviewed/approved 11469 gpt-5.2-2025-12-11
7251 Κύρτος 66 60 -5.361 32.8% 56.6% 69.3% 69.3% 93.7% 72.3% 64.9% reviewed/approved 11584 gpt-5.2-2025-12-11
7257 Κυτώνιον 25 20 -5.203 55.8% 76.3% 76.8% 80.0% 96.9% 83.4% 81.7% reviewed/approved 11602 gpt-5.2-2025-12-11
2626 Κατάνη 95 89 -4.766 36.5% 57.9% 64.7% 65.2% 92.2% 71.2% 58.0% reviewed/approved 11532 gpt-5.2-2025-12-11
2326 Καλλίαρος 47 42 -4.751 47.2% 72.1% 79.7% 83.1% 96.6% 73.2% 72.0% reviewed/approved 11430 gpt-5.2-2025-12-11
2081 Καλάθη 49 44 -4.710 39.8% 69.2% 80.3% 83.9% 95.4% 74.6% 68.8% reviewed/approved 11388 gpt-5.2-2025-12-11
2059 Καδμεία 20 25 4.694 22.0% 58.5% 65.3% 62.2% 94.3% 78.3% 72.9% reviewed/approved 11353 gpt-5.2-2025-12-11
7253 Κυρτώνιος 18 23 4.653 38.1% 66.4% 89.9% 78.0% 97.4% 82.4% 82.7% reviewed/approved 11590 gpt-5.2-2025-12-11
7244 Κυρήνη 54 49 -4.608 33.8% 60.9% 61.9% 68.0% 94.3% 74.2% 61.1% reviewed/approved 11564 gpt-5.2-2025-12-11
7267 Κωνώπη 64 59 -4.403 56.2% 73.7% 73.2% 79.7% 97.3% 78.4% 76.1% reviewed/approved 11632 gpt-5.2-2025-12-11
2342 Κάμιρος 54 58 4.392 48.0% 76.3% 84.8% 75.0% 96.7% 80.5% 69.3% reviewed/approved 11460 gpt-5.2-2025-12-11
2328 Καλλίπολις 43 47 4.166 33.2% 60.3% 64.7% 66.7% 93.8% 74.1% 68.4% reviewed/approved 11436 gpt-5.2-2025-12-11
2057 Καβελλιών 34 30 -4.018 62.2% 81.5% 83.3% 87.5% 98.4% 82.7% 80.3% reviewed/approved 11347 gpt-5.2-2025-12-11
7248 Κύρου πόλις 32 36 3.941 28.8% 65.7% 67.1% 67.6% 94.9% 78.7% 77.2% reviewed/approved 11576 gpt-5.2-2025-12-11
2079 Καισάρεια 39 35 -3.916 69.9% 84.1% 87.9% 91.9% 98.3% 83.0% 84.2% reviewed/approved 11382 gpt-5.2-2025-12-11
2628 Καταονία 24 28 3.776 13.8% 49.6% 52.4% 53.8% 94.7% 78.6% 75.4% reviewed/approved 11538 gpt-5.2-2025-12-11
2080 Καλαβρία 18 15 -3.347 22.3% 54.9% 45.1% 60.6% 93.8% 79.3% 73.9% reviewed/approved 11385 gpt-5.2-2025-12-11
2078 Καινύς 22 19 -3.265 62.3% 69.6% 73.2% 73.2% 95.7% 83.5% 76.3% reviewed/approved 11377 gpt-5.2-2025-12-11
2602 Καρύανδα 34 31 -3.018 48.5% 71.9% 81.7% 76.9% 96.8% 79.7% 74.9% reviewed/approved 11493 gpt-5.2-2025-12-11
7261 Κύψελα 34 31 -3.018 56.1% 75.0% 81.7% 83.1% 97.1% 80.3% 78.3% reviewed/approved 11614 gpt-5.2-2025-12-11
2087 Καλαύρεια 30 33 2.900 53.7% 76.6% 92.0% 85.7% 96.6% 77.8% 75.5% reviewed/approved 11406 gpt-5.2-2025-12-11

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: 1.000; cross-validated R^2: 0.007.

Positive termCoefficientDocsMean presentMean absent
οι 6.1164 11 5.171 -0.639
εστι δε και 3.5097 3 8.798 -0.272
τε 3.1775 5 3.995 -0.210
δε και 3.1013 8 3.044 -0.265
ος 3.0311 6 4.461 -0.285
πολις 2.7032 68 0.257 -0.546
εκαλειτο 2.4050 7 2.481 -0.187
εστι δε 2.3866 5 3.115 -0.164
ου και 2.3545 4 6.779 -0.282
δυναται 2.1708 3 4.599 -0.142
εκαλειτο δε 2.1238 3 3.463 -0.107
λεγεται 2.1129 9 2.120 -0.210
τη 2.0878 15 1.364 -0.241
της χωρας 2.0175 3 3.164 -0.098
υπο 1.9431 4 3.032 -0.126
χωρας 1.9035 4 2.312 -0.096
ου 1.8264 14 1.556 -0.253
πολις προς 1.8056 3 5.816 -0.180
δυο 1.7467 4 3.216 -0.134
ομηρος 1.6865 3 4.027 -0.125
Negative termCoefficientDocsMean presentMean absent
το -2.5412 70 -0.143 0.335
ει -2.4633 4 -6.034 0.251
μεν -2.3770 4 -7.390 0.308
παρα το -2.3297 3 -5.190 0.161
περι -2.0526 11 -1.804 0.223
πρωτη -2.0053 3 -7.648 0.237
εν πρωτη -2.0053 3 -7.648 0.237
εστι και -1.8877 20 -0.431 0.108
και -1.7032 66 0.106 -0.205
οικητωρ -1.6651 6 -3.902 0.249
κατα -1.6460 10 -0.501 0.056
παρα -1.6002 7 -2.920 0.220
θρακης -1.5732 5 -2.447 0.129
τους -1.5710 4 -1.037 0.043
ως το -1.4613 3 -5.086 0.157
απο της -1.4343 6 -1.124 0.072
δια -1.4222 16 0.586 -0.112
χωρα -1.3421 6 -3.268 0.209
του ποντου -1.3345 3 -2.840 0.088
ποντου -1.3345 3 -2.840 0.088

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 1.425; cross-validated R^2: 0.067.

Positive termCoefficientDocsMean presentMean absent
τω 2.1990 16 4.388 2.326
της 2.1731 35 4.240 1.803
οι 2.0958 11 5.171 2.345
δε και 2.0051 8 7.566 2.229
εκαλειτο δε 1.9873 3 12.912 2.339
εστι 1.9053 24 4.721 2.004
δε 1.9037 46 3.591 1.859
την 1.8230 13 6.162 2.132
τε 1.7696 5 4.832 2.542
παρα 1.5806 7 6.837 2.342
εστι δε και 1.5800 3 8.798 2.466
και 1.5417 66 3.244 1.516
κατα 1.4479 10 5.149 2.379
εκαλειτο δε και 1.4371 3 12.912 2.339
τη 1.4275 15 4.347 2.358
εστι δε 1.4065 5 7.443 2.404
εκαλειτο 1.4009 7 6.531 2.365
και αλλη 1.2837 3 9.487 2.445
αλλη 1.2602 4 8.267 2.422
το θηλυκον 1.1738 6 6.594 2.405
Negative termCoefficientDocsMean presentMean absent
νησος -1.2618 9 1.268 2.793
μια -1.1632 3 0.912 2.710
εκ -1.1244 6 1.857 2.707
και πολις -1.0673 5 1.837 2.699
και το -1.0458 9 2.761 2.646
τεταρτω -1.0301 3 0.475 2.724
ακρα -0.9771 4 0.773 2.735
στραβων -0.8908 15 2.648 2.658
ειρηται -0.8658 3 1.381 2.696
τοπικα -0.8456 3 1.056 2.706
τα τοπικα -0.8456 3 1.056 2.706
τινες -0.8260 6 1.314 2.742
δε απο -0.8167 3 1.049 2.706
παιδος -0.8041 4 1.713 2.695
και νησος -0.6968 3 0.941 2.709
και το κτητικον -0.6873 4 1.294 2.713
το εθνικον -0.6709 56 2.775 2.505
εθνικον -0.6709 56 2.775 2.505
ως -0.6584 52 2.834 2.464
αιγυπτου -0.6581 3 2.448 2.663

AI English terms predicting large absolute residuals

Features: 326; ridge alpha: 1.425; cross-validated R^2: -0.037.

Positive termCoefficientDocsMean presentMean absent
the 2.6866 93 2.765 1.210
is the 1.9758 3 9.165 2.455
among 1.8693 6 6.799 2.392
also 1.7798 46 3.457 1.974
to 1.6756 20 4.254 2.257
usage 1.6653 4 7.376 2.460
of 1.5302 66 3.121 1.753
among the 1.5056 5 7.820 2.384
there 1.4966 33 4.179 1.906
he 1.4132 4 10.101 2.346
too 1.4084 10 6.405 2.240
there is 1.3684 24 4.346 2.123
is also 1.3520 33 3.645 2.169
their 1.3038 3 8.714 2.469
derives 1.2042 3 7.237 2.514
derives from 1.2042 3 7.237 2.514
from 1.1730 43 3.233 2.221
there is also 1.1424 24 4.346 2.123
this 1.1329 10 5.854 2.301
when 1.0685 6 7.742 2.331
Negative termCoefficientDocsMean presentMean absent
island -1.1015 14 1.423 2.857
one of -0.9118 4 1.409 2.708
one of the -0.8883 3 0.912 2.710
an island -0.8046 10 1.159 2.822
written with -0.7651 7 1.119 2.772
written -0.7651 7 1.119 2.772
an -0.7587 17 1.815 2.828
but it -0.7308 4 0.995 2.725
kappadokia -0.6585 4 2.223 2.674
and a city -0.6400 4 1.479 2.705
and the feminine -0.6318 6 2.827 2.645
terms -0.6165 3 1.056 2.706
locative terms -0.6165 3 1.056 2.706
locative -0.6165 3 1.056 2.706
river -0.6100 6 2.035 2.696
a city in -0.6013 38 2.625 2.675
the ethnonym is -0.5950 56 2.788 2.489
ethnonym is -0.5950 56 2.788 2.489
one -0.5892 13 2.501 2.679
strabo -0.5885 15 2.648 2.658

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 2.031; cross-validated R^2: -0.048.

Positive termCoefficientDocsMean presentMean absent
the 2.2586 90 2.803 1.331
among 1.4394 4 8.584 2.409
also 1.3306 43 3.584 1.956
there 1.2450 34 3.865 2.034
among the 1.2027 4 8.584 2.409
of 1.1599 66 3.150 1.697
there is also 1.0768 22 4.685 2.084
for 1.0634 11 5.538 2.300
there is 1.0606 24 4.372 2.114
usage 1.0507 4 6.864 2.481
to 1.0148 31 3.733 2.172
is also 0.9861 31 3.959 2.071
is with 0.9681 4 7.157 2.469
this 0.9600 15 5.475 2.159
who 0.9594 13 4.360 2.402
local 0.9458 6 5.766 2.458
came 0.9152 4 8.835 2.399
their 0.8935 3 12.912 2.339
another 0.8887 7 6.387 2.375
are 0.8239 25 3.843 2.261
Negative termCoefficientDocsMean presentMean absent
an -0.8553 22 1.896 2.871
the ethnonym is -0.8110 52 2.424 2.908
ethnonym is -0.8110 52 2.424 2.908
island -0.7211 14 1.423 2.857
and the -0.6904 15 2.767 2.637
kappadokia -0.6855 4 2.223 2.674
like -0.6788 5 0.988 2.744
is a city -0.6161 3 0.996 2.708
promontory -0.6095 3 0.410 2.726
an island -0.5815 9 1.223 2.798
sanctuary -0.5231 3 0.854 2.712
strabo book -0.5039 10 2.381 2.687
which the -0.4992 3 0.504 2.723
city in -0.4838 41 3.039 2.390
now -0.4746 4 1.583 2.701
one of the -0.4577 4 1.876 2.689
a city in -0.4553 40 3.079 2.374
book -0.4548 39 2.668 2.649
ethnonym -0.4436 57 2.800 2.466
one -0.4373 12 2.657 2.656

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.