gpt-5.5_v4_mini_high v1 Evaluation

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

Pairs100
Distinct lemmas100
Slope0.892
R^20.985
BLEU-446.6%
chrF++70.1%
METEOR76.2%
ROUGE-L78.0%
BERTScore96.0%
COMET79.2%
BLEURT73.9%

Profile version id: 1766

First translation using this version: 2026-06-28

Total usable AI runs for this version: 101 runs across 101 lemmas

Created: 2026-06-19 08:17:46.611841+10:00

Active: True

Notes: Approved-human-only v4 mini reasoning trial using gpt-5.4-mini and Responses API reasoning.effort=high.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.892
Intercept2.412
R^20.985
P-value1.15e-91
Pearson r0.993
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words41.800
Mean absolute length residual2.985
Mean absolute percent error7.4%

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 = -0.367, R^2 = 0.135, p = 0.0002.
chrF++ vs passage length
chrF++: significant negative; r = -0.402, R^2 = 0.162, p = 3.31e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.476, R^2 = 0.226, p = 5.61e-07.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.510, R^2 = 0.260, p = 5.94e-08.
BERTScore vs passage length
BERTScore: significant negative; r = -0.574, R^2 = 0.329, p = 4.30e-10.
COMET vs passage length
COMET: significant negative; r = -0.420, R^2 = 0.176, p = 1.36e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.691, R^2 = 0.477, p = 1.86e-15.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.413, R^2 = 0.171, p = 1.92e-05.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.424, R^2 = 0.180, p = 1.09e-05.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.419, R^2 = 0.176, p = 1.42e-05.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.410, R^2 = 0.168, p = 2.31e-05.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.367 0.135 0.0002 -0.00252 ok
chrF++ 100 significant negative -0.402 0.162 3.31e-05 -0.00162 ok
METEOR 100 significant negative -0.476 0.226 5.61e-07 -0.00235 ok
ROUGE-L 100 significant negative -0.510 0.260 5.94e-08 -0.00225 ok
BERTScore 100 significant negative -0.574 0.329 4.30e-10 -0.00044 ok
COMET 100 significant negative -0.420 0.176 1.36e-05 -0.00081 ok
BLEURT 100 significant negative -0.691 0.477 1.86e-15 -0.00246 ok
Trigram precision 100 significant negative -0.413 0.171 1.92e-05 -0.00341 ok
Trigram recall 100 significant negative -0.424 0.180 1.09e-05 -0.00339 ok
Trigram F1 100 significant negative -0.419 0.176 1.42e-05 -0.00339 ok
Trigram Jaccard 100 significant negative -0.410 0.168 2.31e-05 -0.00337 ok

Translation Similarity Metrics

Mean BLEU-446.6%
Mean chrF++70.1%
Mean METEOR76.2%
Mean ROUGE-L78.0%
Mean BERTScore96.0%
Mean COMET79.2%
Mean BLEURT73.9%
Legacy smoothed corpus BLEU46.2%
Unigram F177.6%
Bigram F153.5%
Trigram precision41.0%
Trigram recall38.7%
Trigram F139.9%
Trigram Jaccard24.9%
4-gram F130.7%

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 270 226 -17.292 35.5% 59.9% 58.9% 64.1% 91.8% 80.8% 55.2% reviewed/approved 7294 gpt-5.4-mini
2604 Καρχηδών 117 122 15.207 36.9% 65.6% 69.0% 66.4% 94.1% 79.9% 61.4% reviewed/approved 7309 gpt-5.4-mini
2603 Κάρυστος 175 171 12.462 41.3% 63.7% 64.1% 67.6% 94.3% 78.0% 62.9% reviewed/approved 7306 gpt-5.4-mini
2342 Κάμιρος 54 60 9.412 37.6% 70.6% 71.2% 73.7% 95.6% 79.0% 68.9% reviewed/approved 7270 gpt-5.4-mini
7260 Κυχρεῖος πάγος 61 66 9.167 48.3% 71.7% 77.4% 67.7% 96.3% 76.8% 64.0% reviewed/approved 7423 gpt-5.4-mini
2056 Καβειρία 112 94 -8.332 26.2% 64.0% 51.6% 63.1% 95.2% 79.3% 59.4% reviewed/approved 7145 gpt-5.4-mini
7251 Κύρτος 66 53 -8.293 38.8% 61.4% 64.1% 68.3% 94.3% 72.9% 64.3% reviewed/approved 7396 gpt-5.4-mini
7255 Κυτέριον 22 30 7.961 28.6% 60.4% 61.6% 61.5% 95.3% 75.6% 73.5% reviewed/approved 7408 gpt-5.4-mini
2605 Κάσιον 56 60 7.628 43.0% 71.1% 80.1% 70.7% 95.8% 77.9% 71.7% reviewed/approved 7312 gpt-5.4-mini
2119 Κάλλατις 65 68 7.599 20.8% 53.6% 60.7% 58.6% 93.3% 67.7% 58.9% reviewed/approved 7234 gpt-5.4-mini
2468 Καπετώλιον 138 118 -7.528 13.5% 47.3% 51.7% 53.9% 88.1% 67.4% 50.5% reviewed/approved 7285 gpt-5.4-mini
2081 Καλάθη 49 39 -7.127 33.1% 63.2% 62.6% 77.3% 95.0% 73.7% 68.3% reviewed/approved 7198 gpt-5.4-mini
3288 Κορώνεια 110 107 6.452 32.0% 64.8% 65.8% 71.6% 95.1% 75.9% 65.1% reviewed/approved 7357 gpt-5.4-mini
2477 Καρδαμύλη 43 47 6.226 44.2% 72.1% 86.8% 75.6% 95.2% 80.0% 77.2% reviewed/approved 7291 gpt-5.4-mini
7254 Κύτα 84 83 5.648 33.1% 62.1% 61.7% 67.1% 92.9% 77.2% 61.8% reviewed/approved 7405 gpt-5.4-mini
2115 Καλὴ ἀκτή 64 65 5.491 24.8% 54.2% 57.8% 57.4% 93.5% 69.8% 58.7% reviewed/approved 7222 gpt-5.4-mini
2346 Κάναι 77 76 4.893 51.3% 67.2% 78.9% 79.2% 95.2% 77.4% 63.3% reviewed/approved 7276 gpt-5.4-mini
7264 Κωλιάς 61 52 -4.833 32.5% 60.1% 54.4% 63.2% 94.7% 70.1% 54.3% reviewed/approved 7435 gpt-5.4-mini
2326 Καλλίαρος 47 40 -4.343 42.6% 67.8% 74.9% 80.5% 95.5% 71.1% 62.1% reviewed/approved 7240 gpt-5.4-mini
2116 Κάληρος 40 34 -4.098 30.7% 58.6% 68.3% 64.9% 92.8% 73.8% 70.5% reviewed/approved 7225 gpt-5.4-mini
7245 Κύρης 13 18 3.990 28.0% 60.1% 83.3% 77.4% 92.6% 76.1% 73.1% reviewed/approved 7378 gpt-5.4-mini
7248 Κύρου πόλις 32 27 -3.960 22.3% 62.1% 56.6% 67.8% 94.3% 77.0% 77.4% reviewed/approved 7387 gpt-5.4-mini
2455 Κάναστρον 60 52 -3.941 23.9% 56.9% 51.7% 58.9% 94.3% 72.1% 59.2% reviewed/approved 7279 gpt-5.4-mini
2802 Κάδρεμα 24 20 -3.823 21.9% 52.9% 51.0% 63.6% 96.6% 81.4% 75.6% reviewed/approved 7171 gpt-5.4-mini
2054 Καβαλίς 65 64 3.599 56.8% 70.9% 70.8% 76.0% 96.0% 77.1% 69.4% reviewed/approved 5763 gpt-5.4-mini
7259 Κύφος 54 54 3.412 45.2% 74.5% 85.9% 85.2% 96.9% 75.8% 75.0% reviewed/approved 7420 gpt-5.4-mini
2599 Καρπήσιοι 10 8 -3.333 23.1% 65.5% 70.6% 77.8% 97.1% 87.3% 78.2% reviewed/approved 7300 gpt-5.4-mini
2329 Κάλπη 57 50 -3.264 45.0% 72.6% 76.9% 81.5% 96.0% 74.9% 73.5% reviewed/approved 7249 gpt-5.4-mini
2624 Κατάβαθμος 20 17 -3.255 39.9% 70.6% 74.0% 75.7% 96.1% 81.0% 77.0% reviewed/approved 7333 gpt-5.4-mini
3496 Κοτιάειον 71 69 3.246 24.8% 51.8% 56.0% 58.6% 92.8% 70.2% 63.2% reviewed/approved 7360 gpt-5.4-mini

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: 2.894; cross-validated R^2: -0.115.

Positive termCoefficientDocsMean presentMean absent
δε και 2.3880 8 4.055 -0.353
οι 1.9367 11 3.937 -0.487
ος 1.8525 6 4.777 -0.305
εστι δε και 1.7272 3 9.295 -0.287
εστι δε 1.6641 5 5.009 -0.264
απο 1.6581 27 1.837 -0.680
τη 1.5387 15 1.968 -0.347
εκληθη 1.4996 4 7.844 -0.327
εν τη 1.4978 7 4.066 -0.306
αφ ου 1.4382 5 4.205 -0.221
τε 1.2767 5 3.926 -0.207
λεγεται 1.2616 9 2.711 -0.268
δυο 1.2586 4 4.861 -0.203
αφ 1.1837 6 3.795 -0.242
στραβων 1.1586 15 2.363 -0.417
ου 1.1573 14 1.861 -0.303
των 1.1428 17 1.633 -0.335
δυναται 1.0937 3 4.586 -0.142
εκαλειτο δε 1.0827 3 3.459 -0.107
παιδος 1.0708 4 4.219 -0.176
Negative termCoefficientDocsMean presentMean absent
το -1.4427 70 0.032 -0.075
γαρ -1.2594 9 -0.495 0.049
εστι και -1.1835 20 -0.480 0.120
εν τω -1.1658 6 -2.419 0.154
τω περι -1.0827 4 -6.234 0.260
εν τω περι -1.0827 4 -6.234 0.260
εθνος -1.0719 8 -3.547 0.308
οικητωρ -0.9429 6 -4.128 0.264
μεν -0.9406 4 -7.626 0.318
ως το -0.8877 3 -5.761 0.178
και του -0.8634 4 -2.985 0.124
ουτως -0.8401 5 -4.373 0.230
τω -0.8169 16 -0.987 0.188
ηρωδιανος -0.7750 3 -8.584 0.265
δια του -0.7563 10 -0.790 0.088
πρωτη -0.7532 3 -7.460 0.231
εν πρωτη -0.7532 3 -7.460 0.231
θρακης -0.7234 5 -2.096 0.110
ως ελλανικος -0.7134 3 -6.656 0.206
λυκιας -0.7040 3 -1.650 0.051

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 1.425; cross-validated R^2: 0.030.

Positive termCoefficientDocsMean presentMean absent
την 2.8439 13 7.242 2.348
δε 2.7962 46 4.148 1.994
δε και 2.4071 8 8.679 2.489
εκαλειτο δε 2.2961 3 14.987 2.613
απο 2.1600 27 5.403 2.090
απο της 2.1275 6 9.695 2.556
τω 2.1111 16 5.151 2.572
γαρ 1.8880 9 6.439 2.643
εστι δε 1.8433 5 8.342 2.703
ος 1.8332 6 7.286 2.710
εκαλειτο 1.6757 7 7.468 2.647
εστι δε και 1.6201 3 9.295 2.789
εκαλειτο δε και 1.6047 3 14.987 2.613
αφ 1.5761 6 6.572 2.756
αφ ου 1.5756 5 7.538 2.745
εστι 1.5510 24 4.826 2.403
περι 1.3027 11 5.681 2.651
εν τω 1.2490 6 7.648 2.687
παρα 1.2294 7 7.092 2.675
λεγεται 1.2226 9 6.752 2.612
Negative termCoefficientDocsMean presentMean absent
τον -1.7696 10 2.363 3.054
καππαδοκιας -1.4280 3 1.013 3.046
προς τη -1.4034 4 1.169 3.060
μια -1.2585 3 0.984 3.046
νησος -1.2136 9 1.847 3.097
πολυβιος -1.2073 5 0.806 3.099
τους -1.1255 4 1.590 3.043
το εθνικον -1.0443 56 2.835 3.175
εθνικον -1.0443 56 2.835 3.175
φασι -0.8646 8 3.754 2.918
προς -0.8434 15 3.337 2.922
ειρηται -0.8174 3 2.131 3.011
εδει -0.8021 3 1.631 3.026
και νησος -0.7801 3 1.193 3.040
παυσανιας -0.7645 4 2.613 3.000
και το -0.7116 9 3.806 2.903
χωρα -0.6915 6 4.068 2.915
τριτω -0.6746 3 1.432 3.033
τινες -0.6269 6 2.023 3.046
στραβων -0.6158 15 3.615 2.873

AI English terms predicting large absolute residuals

Features: 317; ridge alpha: 2.894; cross-validated R^2: -0.086.

Positive termCoefficientDocsMean presentMean absent
of 1.7517 66 3.560 1.867
the 1.7323 93 3.090 1.587
was 1.3551 16 6.227 2.367
and 1.3336 58 3.881 1.747
there 1.0711 31 4.653 2.235
after 1.0454 22 5.273 2.339
too 0.9578 10 6.656 2.577
on 0.9469 16 4.910 2.618
is also 0.9405 31 4.487 2.309
there is 0.9214 25 4.882 2.352
renowned 0.9213 3 10.971 2.738
because of 0.9067 3 11.329 2.726
is 0.8522 89 3.081 2.205
there is also 0.8211 24 4.738 2.431
also 0.8034 42 4.074 2.195
it 0.8016 36 4.664 2.040
most 0.8001 4 11.289 2.639
place 0.7979 6 6.126 2.784
diphthong 0.7973 3 8.829 2.804
while 0.7620 3 13.610 2.656
Negative termCoefficientDocsMean presentMean absent
they -0.7266 7 1.504 3.096
the ethnonym -0.7015 57 2.796 3.235
ethnonym -0.6334 58 2.891 3.114
the ethnonym is -0.6317 55 2.872 3.122
one of the -0.6153 3 0.984 3.046
one of -0.6153 3 0.984 3.046
and the -0.5765 20 3.670 2.813
ethnonym is -0.5629 56 2.969 3.004
an island -0.5502 10 1.723 3.125
some -0.5258 9 2.233 3.059
book the ethnonym -0.4744 7 1.169 3.121
book -0.4586 40 3.006 2.971
been -0.4526 3 0.595 3.058
inhabitant is -0.4389 6 1.555 3.076
are called -0.4385 3 1.546 3.029
polybios book -0.4334 5 0.806 3.099
polybios -0.4334 5 0.806 3.099
strabo book -0.4312 9 2.213 3.061
book the -0.4297 11 2.744 3.014
island -0.4270 14 2.614 3.045

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 4.125; cross-validated R^2: -0.122.

Positive termCoefficientDocsMean presentMean absent
the 1.4529 90 3.115 1.812
and 1.1153 52 4.014 1.870
was 1.1047 20 5.070 2.463
there 0.8104 34 4.480 2.214
after 0.7745 22 5.157 2.372
is with 0.7686 4 10.124 2.687
from the 0.7617 13 6.522 2.456
called 0.7521 21 5.303 2.368
on 0.7516 16 5.197 2.563
there is 0.7264 24 4.712 2.439
also 0.7200 43 4.035 2.192
with 0.7050 16 4.477 2.700
of 0.6972 66 3.508 1.968
is also 0.6442 31 4.299 2.394
to 0.6439 31 4.161 2.456
is 0.6361 87 3.083 2.327
because 0.6241 12 6.370 2.523
their 0.5902 3 14.987 2.613
there is also 0.5837 22 4.812 2.469
who 0.5828 13 5.201 2.653
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.6092 52 2.621 3.379
ethnonym is -0.6092 52 2.621 3.379
kappadokia -0.5906 4 1.490 3.047
an -0.5592 22 2.365 3.159
a city in -0.5082 40 3.096 2.911
ethnonym -0.4519 57 2.920 3.070
city in -0.4365 41 3.206 2.831
the ethnonym -0.4234 56 2.960 3.015
one of the -0.4146 4 1.280 3.056
book -0.3980 39 2.848 3.072
polybios book -0.3807 5 0.806 3.099
polybios -0.3807 5 0.806 3.099
they -0.3755 12 3.249 2.948
some -0.3621 10 2.192 3.073
which -0.3419 20 3.139 2.946
an island -0.3353 9 1.730 3.109
is a city -0.3344 3 1.284 3.037
and the -0.3321 15 3.320 2.925
territory -0.2980 5 1.849 3.044
but -0.2978 12 2.516 3.049

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.