gpt-5.5 v3 Evaluation

Generated: 2026-09-20 11:22:12 UTC

Pairs100
Distinct lemmas100
Slope0.929
R^20.987
BLEU-458.6%
chrF++77.8%
METEOR81.2%
ROUGE-L81.6%
BERTScore97.1%
COMET81.6%
BLEURT79.2%

Profile version id: 1101

First translation using this version: 2026-05-07

Total usable AI runs for this version: 2,922 runs across 999 lemmas

Created: 2026-05-07 16:52:06.734553+10:00

Active: True

Notes: Reviewed house-style prompt v3 revised after Gabe's 2026-04-30 feedback and April 30 group discussion. Keeps static rules for global house style, citation hygiene, transliteration, metalinguistic precision, and syntax; moves most formula/gloss/geographic vocabulary decisions to entry-specific guidance context; adds the settlement-location "in" convention, stricter Strabo/modern-locator handling, and revised rules for prose/verse quotation formatting and inflected or dialectal evidence.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.929
Intercept2.252
R^20.987
P-value9.58e-94
Pearson r0.993
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.250
Mean absolute length residual2.936
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 = -0.299, R^2 = 0.089, p = 0.0025.
chrF++ vs passage length
chrF++: significant negative; r = -0.383, R^2 = 0.147, p = 8.37e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.356, R^2 = 0.127, p = 0.0003.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.356, R^2 = 0.127, p = 0.0003.
BERTScore vs passage length
BERTScore: significant negative; r = -0.588, R^2 = 0.346, p = 1.22e-10.
COMET vs passage length
COMET: significant negative; r = -0.473, R^2 = 0.224, p = 6.51e-07.
BLEURT vs passage length
BLEURT: significant negative; r = -0.681, R^2 = 0.464, p = 6.19e-15.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.224, R^2 = 0.050, p = 0.0252.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.235, R^2 = 0.055, p = 0.0188.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.229, R^2 = 0.053, p = 0.0216.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.253, R^2 = 0.064, p = 0.0111.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.299 0.089 0.0025 -0.00226 ok
chrF++ 100 significant negative -0.383 0.147 8.37e-05 -0.00163 ok
METEOR 100 significant negative -0.356 0.127 0.0003 -0.00159 ok
ROUGE-L 100 significant negative -0.356 0.127 0.0003 -0.00150 ok
BERTScore 100 significant negative -0.588 0.346 1.22e-10 -0.00039 ok
COMET 100 significant negative -0.473 0.224 6.51e-07 -0.00084 ok
BLEURT 100 significant negative -0.681 0.464 6.19e-15 -0.00259 ok
Trigram precision 100 significant negative -0.224 0.050 0.0252 -0.00197 ok
Trigram recall 100 significant negative -0.235 0.055 0.0188 -0.00203 ok
Trigram F1 100 significant negative -0.229 0.053 0.0216 -0.00200 ok
Trigram Jaccard 100 significant negative -0.253 0.064 0.0111 -0.00238 ok

Translation Similarity Metrics

Mean BLEU-458.6%
Mean chrF++77.8%
Mean METEOR81.2%
Mean ROUGE-L81.6%
Mean BERTScore97.1%
Mean COMET81.6%
Mean BLEURT79.2%
Legacy smoothed corpus BLEU57.0%
Unigram F182.0%
Bigram F162.5%
Trigram precision51.2%
Trigram recall50.1%
Trigram F150.6%
Trigram Jaccard33.9%
4-gram F142.3%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2604 Καρχηδών 117 128 17.100 35.0% 63.9% 74.6% 69.6% 93.7% 78.4% 64.4% reviewed/approved 3349 gpt-5.5
2484 Καρία 270 236 -16.978 42.0% 65.8% 67.7% 70.8% 92.8% 81.3% 51.0% reviewed/approved 3612 gpt-5.5
2468 Καπετώλιον 138 118 -12.400 30.1% 50.3% 53.5% 63.3% 91.6% 73.1% 51.3% reviewed/approved 3326 gpt-5.5
2603 Κάρυστος 175 176 11.241 46.5% 69.5% 77.2% 74.1% 95.4% 79.3% 59.9% reviewed/approved 3346 gpt-5.5
2465 Κάνωπος 83 89 9.673 57.7% 77.4% 87.8% 83.7% 95.8% 78.6% 74.9% reviewed/approved 2418 gpt-5.5
2328 Καλλίπολις 43 50 7.818 38.3% 65.3% 71.2% 66.7% 94.5% 75.4% 71.7% reviewed/approved 2360 gpt-5.5
7254 Κύτα 84 88 7.745 54.5% 74.4% 80.6% 75.6% 95.5% 77.9% 65.9% reviewed/approved 3437 gpt-5.5
2119 Κάλλατις 65 70 7.388 31.8% 63.9% 75.5% 68.1% 94.9% 71.4% 67.6% reviewed/approved 3287 gpt-5.5
2056 Καβειρία 112 99 -7.256 43.9% 71.9% 59.8% 73.9% 96.1% 80.4% 63.4% reviewed/approved 3614 gpt-5.5
7260 Κυχρεῖος πάγος 61 66 7.103 49.6% 74.0% 79.1% 67.7% 96.3% 78.3% 67.3% reviewed/approved 3452 gpt-5.5
3288 Κορώνεια 110 111 6.601 37.9% 67.9% 65.6% 70.3% 94.9% 76.7% 65.5% reviewed/approved 3615 gpt-5.5
2605 Κάσιον 56 60 5.746 53.0% 75.6% 85.4% 77.6% 97.1% 79.7% 76.6% reviewed/approved 3352 gpt-5.5
2342 Κάμιρος 54 58 5.603 65.5% 85.1% 85.5% 83.9% 96.7% 82.8% 76.4% reviewed/approved 2374 gpt-5.5
7255 Κυτέριον 22 28 5.319 53.0% 76.1% 73.5% 72.0% 96.7% 78.4% 79.7% reviewed/approved 3440 gpt-5.5
2326 Καλλίαρος 47 51 5.103 76.4% 88.2% 96.9% 91.8% 98.7% 78.8% 83.8% reviewed/approved 2358 gpt-5.5
7266 Κώμη 79 71 -4.612 13.8% 47.6% 53.7% 50.7% 93.3% 72.8% 56.3% reviewed/approved 3470 gpt-5.5
2116 Κάληρος 40 35 -4.396 29.3% 61.2% 73.5% 72.0% 94.7% 77.7% 68.2% reviewed/approved 3827 gpt-5.5
2609 Κάσπειρος 135 132 4.385 46.9% 68.4% 70.2% 68.9% 92.1% 73.4% 64.1% reviewed/approved 3358 gpt-5.5
2114 Κάλβιος 18 15 -3.967 36.3% 64.3% 59.2% 66.7% 95.6% 78.5% 73.4% reviewed/approved 3822 gpt-5.5
7258 Κύτωρος 30 34 3.890 69.5% 88.5% 96.3% 87.5% 98.0% 82.4% 82.7% reviewed/approved 3543 gpt-5.5
7267 Κωνώπη 64 58 -3.683 57.6% 75.7% 75.1% 82.0% 97.6% 80.2% 77.0% reviewed/approved 3473 gpt-5.5
2330 Καλύβη 24 21 -3.538 78.4% 89.3% 88.0% 93.3% 99.3% 83.8% 85.6% reviewed/approved 3302 gpt-5.5
2621 Κάστνιον 25 29 3.533 52.1% 79.3% 80.3% 77.8% 97.4% 82.1% 77.8% reviewed/approved 3364 gpt-5.5
2079 Καισάρεια 39 35 -3.468 65.2% 79.2% 87.9% 91.9% 97.8% 81.3% 83.9% reviewed/approved 3264 gpt-5.5
7244 Κυρήνη 54 49 -3.397 48.8% 70.4% 75.7% 75.7% 96.7% 81.7% 78.7% reviewed/approved 3408 gpt-5.5
3530 Κριώα 26 23 -3.396 38.0% 71.1% 65.7% 73.5% 96.9% 79.2% 76.9% reviewed/approved 3400 gpt-5.5
3496 Κοτιάειον 71 65 -3.183 41.0% 62.3% 64.4% 69.1% 95.1% 75.2% 75.9% reviewed/approved 4099 gpt-5.5
2602 Καρύανδα 34 37 3.175 70.0% 81.0% 93.2% 84.5% 97.2% 80.8% 77.5% reviewed/approved 3343 gpt-5.5
7249 Κύρρος 45 41 -3.039 54.8% 71.0% 81.2% 79.1% 96.8% 76.9% 73.5% reviewed/approved 3423 gpt-5.5
7253 Κυρτώνιος 18 22 3.033 32.9% 64.6% 69.0% 65.0% 96.2% 77.9% 80.9% reviewed/approved 3434 gpt-5.5

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: 17.013; cross-validated R^2: -0.201.

Positive termCoefficientDocsMean presentMean absent
απο 0.4858 27 1.935 -0.716
οι 0.4645 11 4.005 -0.495
δε και 0.4496 8 3.172 -0.276
και 0.3986 66 0.250 -0.486
ου 0.3656 14 2.413 -0.393
εστι δε και 0.3543 3 8.840 -0.273
και το 0.3398 9 3.539 -0.350
εν τη 0.3077 7 3.614 -0.272
εστι δε 0.3003 5 4.316 -0.227
ος 0.2898 6 3.161 -0.202
τη 0.2829 15 1.498 -0.264
ου και 0.2757 4 7.066 -0.294
πολιτης 0.2674 18 2.647 -0.581
εκληθη 0.2614 4 6.459 -0.269
αφ ου 0.2551 5 3.172 -0.167
εκαλειτο δε 0.2492 3 3.788 -0.117
πολις 0.2476 68 0.514 -1.092
υπο 0.2399 4 3.709 -0.155
πολιχνιον 0.2363 3 5.484 -0.170
εδει 0.2321 3 5.864 -0.181
Negative termCoefficientDocsMean presentMean absent
γαρ -0.3606 9 -0.852 0.084
το -0.3306 70 -0.034 0.080
οικητωρ -0.3043 6 -4.933 0.315
δια -0.2505 16 0.298 -0.057
δια του -0.2375 10 -1.248 0.139
το εθνικον -0.2209 56 -0.269 0.342
εθνικον -0.2209 56 -0.269 0.342
εθνος -0.1975 8 -2.960 0.257
ηρωδιανος -0.1974 3 -9.996 0.309
εκαταιος ασια -0.1954 5 -1.496 0.079
ασια -0.1954 5 -1.496 0.079
θρακης -0.1942 5 -1.725 0.091
πολις θρακης -0.1886 4 -1.913 0.080
τινες -0.1841 6 -2.015 0.129
ζευς -0.1770 3 -9.378 0.290
χωρα -0.1685 6 -3.469 0.221
του -0.1625 37 0.989 -0.581
ης -0.1588 3 -3.040 0.094
δευτερω -0.1547 3 -1.157 0.036
εστι και -0.1483 20 0.113 -0.028

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 4.125; cross-validated R^2: -0.178.

Positive termCoefficientDocsMean presentMean absent
την 1.3976 13 7.201 2.299
δια 1.3480 16 5.982 2.356
γαρ 1.3359 9 6.806 2.553
και 1.2794 66 3.467 1.906
δε και 1.1895 8 8.284 2.471
δια του 1.1572 10 7.259 2.456
εκαλειτο δε 1.1272 3 15.106 2.560
ος 1.0962 6 8.356 2.590
απο 1.0614 27 5.270 2.073
δε 1.0332 46 3.880 2.132
εκαλειτο 0.9268 7 7.445 2.597
εν 0.8667 37 3.894 2.373
της 0.7870 35 4.475 2.107
εστι δε και 0.7823 3 8.840 2.754
εστι 0.7675 24 4.509 2.439
εκαλειτο δε και 0.7601 3 15.106 2.560
εστι δε 0.6953 5 7.219 2.711
αφ ου 0.6536 5 6.578 2.744
απο της 0.6131 6 7.356 2.654
περι 0.5961 11 5.467 2.623
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.6531 56 2.910 2.969
εθνικον -0.6531 56 2.910 2.969
προς τη -0.5923 4 1.502 2.996
τον -0.5340 10 2.721 2.960
προς -0.5008 15 2.979 2.929
νησος -0.4923 9 2.219 3.007
του δε -0.4920 3 0.800 3.002
εκ -0.4882 6 1.856 3.005
καππαδοκιας -0.4864 3 1.467 2.982
και πολις -0.4100 5 2.461 2.961
ως -0.3996 52 3.188 2.663
ακρα -0.3807 4 1.250 3.006
ηροδοτος -0.3658 9 3.575 2.873
μια -0.3477 3 1.463 2.982
θεοπομπος -0.3450 5 1.367 3.019
πλησιον -0.3434 8 1.955 3.021
τεταρτω -0.3277 3 1.798 2.971
ποταμου -0.3206 3 1.202 2.990
ειρηται -0.3092 3 2.418 2.952
εκαταιος -0.2749 14 2.353 3.031

AI English terms predicting large absolute residuals

Features: 325; ridge alpha: 4.125; cross-validated R^2: -0.191.

Positive termCoefficientDocsMean presentMean absent
with 1.2320 17 5.307 2.451
there 1.0508 34 4.389 2.188
used to be 0.9493 4 10.073 2.639
used 0.8954 11 6.822 2.456
used to 0.8864 5 8.695 2.633
there is 0.8655 27 4.776 2.255
it with 0.8652 3 11.115 2.683
on 0.8529 14 5.140 2.577
to 0.8515 28 4.971 2.145
be called 0.7969 6 8.607 2.574
to be called 0.7969 6 8.607 2.574
be 0.7540 15 5.903 2.412
to be 0.7505 13 6.366 2.424
for 0.6756 12 5.875 2.535
the 0.6622 93 3.011 1.946
is the 0.6621 4 8.341 2.711
by 0.6569 12 6.024 2.515
who 0.6557 9 6.199 2.613
however 0.6554 6 8.966 2.551
it 0.6510 38 4.182 2.173
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.6172 57 2.767 3.160
the ethnonym is -0.5943 54 2.811 3.083
ethnonym -0.5783 58 2.844 3.063
ethnonym is -0.5545 55 2.892 2.990
a city -0.4899 64 3.052 2.729
promontory -0.4726 5 1.281 3.023
about -0.4262 3 0.584 3.009
kappadokia -0.4065 4 1.751 2.986
which the -0.4034 5 1.421 3.016
island -0.3769 14 2.581 2.994
been -0.3576 5 2.291 2.970
river -0.3418 6 1.714 3.014
an island -0.3274 10 2.136 3.025
as in -0.3258 25 2.816 2.976
a city in -0.3204 45 3.198 2.722
in the -0.3187 21 2.888 2.949
book -0.3174 39 2.867 2.980
near -0.3039 19 3.053 2.909
one of -0.3002 3 1.535 2.979
also a city -0.2974 4 1.623 2.991

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 4.125; cross-validated R^2: -0.236.

Positive termCoefficientDocsMean presentMean absent
with 1.4802 16 5.100 2.524
is with 1.0658 4 12.611 2.533
the 1.0129 90 3.060 1.823
and 0.8754 52 3.789 2.013
who 0.8293 13 5.649 2.531
on 0.7863 16 4.493 2.640
also 0.7791 43 3.841 2.253
was 0.7691 20 4.817 2.466
there 0.7636 34 4.221 2.274
for 0.6425 11 5.530 2.616
their 0.6186 3 15.106 2.560
sanctuary 0.6054 3 9.273 2.740
be 0.6000 15 6.312 2.340
after 0.5959 22 5.139 2.315
called 0.5676 21 4.855 2.426
there is 0.5659 24 4.411 2.470
usage 0.5575 4 6.931 2.770
to 0.5569 31 4.119 2.405
used to be 0.5519 6 8.667 2.570
is 0.5478 87 3.044 2.212
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.6628 52 2.667 3.228
ethnonym is -0.6628 52 2.667 3.228
ethnonym -0.5513 57 2.880 3.011
the ethnonym -0.5241 56 2.923 2.953
kappadokia -0.5209 4 1.751 2.986
a city -0.4599 64 3.068 2.701
as in -0.4033 29 3.033 2.896
ethnonyms -0.3941 3 0.754 3.004
a city in -0.3596 40 3.089 2.834
herodotos book -0.3347 5 1.648 3.004
an -0.3273 22 2.807 2.973
island -0.3245 14 2.581 2.994
city in -0.3215 41 3.154 2.785
it is -0.3121 24 3.852 2.647
the river -0.3112 6 1.714 3.014
river -0.3112 6 1.714 3.014
one of the -0.3085 4 1.465 2.997
one of -0.3082 5 1.722 3.000
near -0.3074 16 2.932 2.937
herodotos -0.3045 9 3.575 2.873

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.