gpt-5.5_v4_mini v1 Evaluation

Generated: 2026-09-20 11:22:16 UTC

Pairs100
Distinct lemmas100
Slope0.937
R^20.989
BLEU-442.4%
chrF++67.0%
METEOR72.8%
ROUGE-L74.6%
BERTScore95.3%
COMET77.6%
BLEURT71.5%

Profile version id: 1747

First translation using this version: 2026-06-25

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

Created: 2026-06-19 00:36:32.588791+10:00

Active: True

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

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.937
Intercept1.365
R^20.989
P-value2.13e-97
Pearson r0.994
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words42.720
Mean absolute length residual2.833
Mean absolute percent error7.1%

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 = -0.387, R^2 = 0.149, p = 7.11e-05.
chrF++ vs passage length
chrF++: significant negative; r = -0.414, R^2 = 0.172, p = 1.81e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.501, R^2 = 0.251, p = 1.13e-07.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.495, R^2 = 0.245, p = 1.65e-07.
BERTScore vs passage length
BERTScore: significant negative; r = -0.589, R^2 = 0.347, p = 1.16e-10.
COMET vs passage length
COMET: significant negative; r = -0.414, R^2 = 0.172, p = 1.83e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.722, R^2 = 0.521, p = 2.29e-17.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.368, R^2 = 0.136, p = 0.0002.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.373, R^2 = 0.139, p = 0.0001.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.371, R^2 = 0.137, p = 0.0001.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.363, R^2 = 0.131, p = 0.0002.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.387 0.149 7.11e-05 -0.00260 ok
chrF++ 100 significant negative -0.414 0.172 1.81e-05 -0.00179 ok
METEOR 100 significant negative -0.501 0.251 1.13e-07 -0.00264 ok
ROUGE-L 100 significant negative -0.495 0.245 1.65e-07 -0.00232 ok
BERTScore 100 significant negative -0.589 0.347 1.16e-10 -0.00051 ok
COMET 100 significant negative -0.414 0.172 1.83e-05 -0.00092 ok
BLEURT 100 significant negative -0.722 0.521 2.29e-17 -0.00274 ok
Trigram precision 100 significant negative -0.368 0.136 0.0002 -0.00293 ok
Trigram recall 100 significant negative -0.373 0.139 0.0001 -0.00294 ok
Trigram F1 100 significant negative -0.371 0.137 0.0001 -0.00292 ok
Trigram Jaccard 100 significant negative -0.363 0.131 0.0002 -0.00276 ok

Translation Similarity Metrics

Mean BLEU-442.4%
Mean chrF++67.0%
Mean METEOR72.8%
Mean ROUGE-L74.6%
Mean BERTScore95.3%
Mean COMET77.6%
Mean BLEURT71.5%
Legacy smoothed corpus BLEU42.6%
Unigram F174.6%
Bigram F149.5%
Trigram precision36.3%
Trigram recall35.1%
Trigram F135.7%
Trigram Jaccard21.7%
4-gram F126.6%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2603 Κάρυστος 175 177 11.713 31.7% 62.0% 59.4% 63.6% 93.3% 74.9% 53.2% reviewed/approved 7305 gpt-5.4-mini
2056 Καβειρία 112 95 -11.275 24.1% 54.3% 46.3% 49.3% 93.2% 73.1% 53.7% reviewed/approved 7153 gpt-5.4-mini
2604 Καρχηδών 117 122 11.042 28.2% 59.0% 62.8% 64.7% 93.1% 76.5% 56.3% reviewed/approved 7308 gpt-5.4-mini
2609 Κάσπειρος 135 138 10.181 31.7% 59.5% 53.1% 57.1% 88.9% 71.8% 54.3% reviewed/approved 7317 gpt-5.4-mini
2597 Καρπασία 112 97 -9.275 31.0% 56.2% 57.3% 66.0% 92.7% 71.2% 55.4% reviewed/approved 7296 gpt-5.4-mini
2342 Κάμιρος 54 61 9.054 37.5% 67.3% 68.6% 69.6% 94.7% 74.9% 61.1% reviewed/approved 7269 gpt-5.4-mini
2116 Κάληρος 40 30 -8.833 25.1% 56.5% 62.5% 62.9% 93.6% 72.9% 68.7% reviewed/approved 7224 gpt-5.4-mini
2484 Καρία 270 246 -8.273 23.3% 54.0% 54.3% 59.3% 89.8% 78.0% 48.1% reviewed/approved 7293 gpt-5.4-mini
2326 Καλλίαρος 47 53 7.610 37.2% 66.6% 76.0% 74.0% 94.7% 69.2% 58.5% reviewed/approved 7239 gpt-5.4-mini
2455 Κάναστρον 60 50 -7.567 22.6% 57.9% 57.4% 60.0% 94.6% 72.4% 60.4% reviewed/approved 7278 gpt-5.4-mini
2477 Καρδαμύλη 43 49 7.357 28.7% 62.8% 82.0% 69.6% 95.1% 78.9% 73.8% reviewed/approved 7290 gpt-5.4-mini
7260 Κυχρεῖος πάγος 61 65 6.497 48.2% 71.3% 75.1% 68.3% 96.2% 77.7% 60.3% reviewed/approved 7422 gpt-5.4-mini
7255 Κυτέριον 22 28 6.028 28.2% 61.0% 62.0% 64.0% 94.7% 72.8% 65.4% reviewed/approved 7407 gpt-5.4-mini
2055 Καβασσός 109 98 -5.465 30.1% 61.6% 63.9% 62.8% 94.6% 77.5% 56.2% reviewed/approved 6082 gpt-5.4-mini
2605 Κάσιον 56 59 5.180 23.9% 56.1% 72.6% 67.8% 93.5% 73.3% 69.0% reviewed/approved 7311 gpt-5.4-mini
7254 Κύτα 84 85 4.953 30.9% 59.1% 54.2% 60.0% 92.1% 73.0% 58.2% reviewed/approved 7404 gpt-5.4-mini
2325 Καλλίαι 24 19 -4.846 30.7% 60.9% 65.0% 69.8% 95.2% 79.7% 64.2% reviewed/approved 7236 gpt-5.4-mini
7253 Κυρτώνιος 18 23 4.775 32.9% 66.1% 74.9% 63.4% 95.5% 76.5% 79.5% reviewed/approved 7401 gpt-5.4-mini
2119 Κάλλατις 65 67 4.750 23.7% 53.6% 57.8% 56.1% 92.9% 65.7% 53.6% reviewed/approved 7233 gpt-5.4-mini
2346 Κάναι 77 78 4.510 52.1% 68.1% 74.3% 79.5% 94.9% 76.1% 62.4% reviewed/approved 7275 gpt-5.4-mini
7264 Κωλιάς 61 54 -4.503 10.4% 38.2% 32.5% 44.8% 91.6% 62.6% 47.7% reviewed/approved 7434 gpt-5.4-mini
7266 Κώμη 79 71 -4.364 12.2% 42.9% 39.3% 45.3% 91.8% 70.0% 48.8% reviewed/approved 7440 gpt-5.4-mini
7251 Κύρτος 66 59 -4.187 38.5% 60.1% 68.3% 72.0% 92.4% 71.4% 61.9% reviewed/approved 7395 gpt-5.4-mini
2468 Καπετώλιον 138 127 -3.629 18.1% 46.4% 46.5% 49.1% 87.0% 64.0% 46.9% reviewed/approved 7284 gpt-5.4-mini
2626 Κατάνη 95 87 -3.351 37.2% 61.5% 58.3% 68.1% 93.8% 74.2% 64.6% reviewed/approved 7338 gpt-5.4-mini
2607 Κάσος 57 58 3.243 35.4% 60.2% 69.3% 71.3% 93.9% 76.2% 69.2% reviewed/approved 7314 gpt-5.4-mini
7261 Κύψελα 34 30 -3.213 54.2% 72.1% 78.9% 81.2% 97.7% 80.4% 70.5% reviewed/approved 7425 gpt-5.4-mini
7247 Κύρνος 50 45 -3.200 31.9% 58.4% 68.8% 65.3% 95.0% 76.9% 74.8% reviewed/approved 7383 gpt-5.4-mini
2628 Καταονία 24 27 3.154 5.7% 48.7% 54.3% 54.9% 93.3% 78.0% 70.7% reviewed/approved 7344 gpt-5.4-mini
2078 Καινύς 22 19 -2.972 59.1% 68.7% 73.2% 78.0% 95.5% 82.4% 79.2% reviewed/approved 7188 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: -0.053.

Positive termCoefficientDocsMean presentMean absent
τε 2.4972 5 5.094 -0.268
οι 2.2549 11 3.509 -0.434
δε και 1.9665 8 3.594 -0.313
απο 1.5751 27 1.074 -0.397
στραβων 1.5023 15 2.498 -0.441
ος 1.4292 6 3.655 -0.233
τη 1.3766 15 1.801 -0.318
εστι δε και 1.3224 3 6.333 -0.196
λεγεται 1.2620 9 2.864 -0.283
πολις 1.1845 68 0.433 -0.919
πολις και 1.1457 6 4.170 -0.266
εκληθη 1.1220 4 5.921 -0.247
αυτην 1.0962 9 3.082 -0.305
των 1.0747 17 1.775 -0.364
εκαλειτο δε 1.0383 3 4.828 -0.149
παιδος 0.9325 4 3.500 -0.146
πολις προς 0.9315 3 5.695 -0.176
πολιτης 0.9238 18 1.857 -0.408
λεγεται και 0.9134 8 2.303 -0.200
ου 0.9089 14 1.373 -0.224
Negative termCoefficientDocsMean presentMean absent
οτι -1.7710 5 -3.242 0.171
παυσανιας -1.0514 4 -4.393 0.183
το -1.0436 70 -0.002 0.004
ως το -1.0128 3 -6.975 0.216
τον -1.0088 10 -1.410 0.157
εθνος -1.0063 8 -2.866 0.249
νησος -0.9738 9 -1.951 0.193
θρακης -0.9046 5 -2.501 0.132
εστι και -0.8834 20 -0.775 0.194
το κτητικον -0.8454 7 -3.794 0.286
κτητικον -0.8454 7 -3.794 0.286
ης -0.8212 3 -3.630 0.112
μεν -0.8036 4 -5.670 0.236
ει -0.7990 4 -3.863 0.161
φησιν -0.7984 10 -0.898 0.100
μεταξυ -0.7905 5 -2.045 0.108
δια -0.7653 16 0.063 -0.012
οικητωρ -0.7370 6 -3.130 0.200
και -0.7349 66 0.061 -0.119
δια του -0.6653 10 -0.752 0.084

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 5.878; cross-validated R^2: -0.522.

Positive termCoefficientDocsMean presentMean absent
και 1.0808 66 3.576 1.390
τε 0.9578 5 6.840 2.622
δε 0.9295 46 3.943 1.888
εκαλειτο 0.9236 7 7.158 2.507
εν 0.7632 37 3.704 2.321
απο 0.7254 27 4.834 2.093
αφ 0.7152 6 7.349 2.545
φησιν 0.7091 10 5.944 2.487
δε και 0.6868 8 6.386 2.524
αφ ου 0.6676 5 8.439 2.538
τα 0.6449 12 5.565 2.460
την 0.6380 13 5.896 2.375
οτι 0.5298 5 5.341 2.701
εκαλειτο δε 0.5036 3 10.343 2.601
της 0.4770 35 3.869 2.275
τω 0.4709 16 3.987 2.613
γαρ 0.4593 9 5.933 2.526
αυτην 0.4454 9 5.143 2.604
ομηρος 0.4341 3 7.292 2.695
ουτως 0.4333 5 6.350 2.648
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.6523 56 2.838 2.826
εθνικον -0.6523 56 2.838 2.826
τους -0.4640 4 1.099 2.905
πολις -0.4553 68 3.067 2.335
τεταρτω -0.4528 3 0.592 2.902
δευτερω -0.4395 3 1.033 2.889
πολυβιος -0.2887 5 1.446 2.906
οικητωρ -0.2448 6 3.130 2.814
εκαταιος ασια -0.2381 5 1.708 2.892
ασια -0.2381 5 1.708 2.892
πορρω -0.2348 3 1.267 2.881
ου πορρω -0.2348 3 1.267 2.881
φασι -0.2303 8 3.018 2.817
πολις θρακης -0.2274 4 1.294 2.897
δευτερα -0.2136 3 0.987 2.890
ειρηται -0.2021 3 1.864 2.863
εκ -0.2017 6 2.558 2.850
πελοποννησου -0.1957 3 1.225 2.883
τριτη -0.1918 3 1.379 2.878
λυκιας -0.1883 3 2.362 2.847

AI English terms predicting large absolute residuals

Features: 310; ridge alpha: 5.878; cross-validated R^2: -0.507.

Positive termCoefficientDocsMean presentMean absent
and 0.9637 58 3.642 1.715
the 0.8743 92 2.983 1.106
says 0.7290 17 5.170 2.354
there 0.6336 31 4.575 2.050
are 0.6203 23 4.712 2.271
was 0.6061 18 5.142 2.326
too 0.5947 8 7.757 2.405
who 0.5002 9 6.225 2.497
whom 0.4996 4 8.230 2.608
to 0.4905 23 4.453 2.349
cape 0.4854 4 6.960 2.661
called 0.4840 29 4.052 2.335
that 0.4688 21 4.236 2.460
also 0.4674 46 3.597 2.182
after 0.4646 21 4.747 2.324
among 0.4577 6 5.837 2.641
was called 0.4409 4 7.959 2.619
most 0.4399 4 9.637 2.549
because 0.4358 11 5.854 2.459
be 0.4313 10 5.039 2.588
Negative termCoefficientDocsMean presentMean absent
city in -0.5826 39 2.598 2.983
a city in -0.5826 39 2.598 2.983
the ethnonym is -0.4433 58 2.820 2.851
the ethnonym -0.4237 59 2.865 2.787
ethnonym is -0.3770 59 2.963 2.645
ethnonym -0.3583 60 3.005 2.575
a city -0.3111 64 3.018 2.504
inhabitant is -0.2752 8 1.825 2.920
herodotos book -0.2433 3 0.864 2.894
inhabitant -0.2328 9 2.542 2.862
in his -0.2217 13 2.736 2.847
a region -0.2151 4 1.519 2.888
written -0.1987 4 1.528 2.887
city -0.1981 67 3.042 2.409
second -0.1971 3 0.727 2.898
city in thrace -0.1891 3 0.655 2.900
in libya -0.1851 3 1.088 2.887
strabo book -0.1849 8 2.215 2.887
in -0.1846 80 2.905 2.543
polybios book -0.1826 5 1.446 2.906

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 2.894; cross-validated R^2: -0.434.

Positive termCoefficientDocsMean presentMean absent
the 1.4690 90 3.010 1.235
and 1.2288 52 3.807 1.777
to 1.1685 31 4.509 2.080
to be 1.0570 9 7.481 2.373
who 1.0091 13 5.870 2.379
says 0.9534 17 5.250 2.338
there 0.9403 34 4.263 2.096
also 0.9330 43 3.863 2.056
this 0.9257 15 5.759 2.317
among 0.8778 4 7.118 2.654
used 0.8736 8 7.672 2.412
used to 0.8736 8 7.672 2.412
used to be 0.8636 6 8.095 2.497
as 0.8480 60 3.280 2.162
be 0.8469 15 6.002 2.274
them 0.8167 5 8.506 2.534
called 0.8160 21 4.956 2.269
after 0.7863 22 4.817 2.273
for 0.7583 11 5.591 2.492
it 0.7139 37 3.989 2.154
Negative termCoefficientDocsMean presentMean absent
a city in -0.8869 40 2.665 2.945
city in -0.8356 41 2.726 2.907
the ethnonym is -0.7695 52 2.534 3.156
ethnonym is -0.7695 52 2.534 3.156
an -0.5010 22 2.373 2.963
the ethnonym -0.4994 56 2.911 2.734
they -0.4912 12 3.149 2.790
ethnonym -0.4759 57 2.956 2.670
a city -0.4720 64 3.018 2.503
and the -0.3976 15 3.278 2.754
just as -0.3924 5 1.178 2.920
or after -0.3727 3 2.111 2.855
polybios book the -0.3695 3 0.313 2.911
one in -0.3596 3 1.516 2.874
territory -0.3581 5 1.288 2.914
that they -0.3358 3 1.218 2.883
town -0.3319 3 1.936 2.861
name -0.3037 7 2.662 2.846
just -0.2956 6 1.773 2.900
one -0.2562 12 3.079 2.799

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.