gpt-5.5 v3 Evaluation

Generated: 2026-07-25 12:53:51 UTC

Pairs100
Distinct lemmas100
Slope0.929
R^20.986
BLEU-457.6%
chrF++77.2%
METEOR80.8%
ROUGE-L81.2%
BERTScoreN/A
COMETN/A
BLEURTN/A

Profile version id: 1101

First translation using this version: 2026-05-07

Total usable AI runs for this version: 2,015 runs across 380 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
Intercept2.247
R^20.986
P-value2.62e-93
Pearson r0.993
Mean source words30.400
Length fallback rows0
Mean human words44.160
Mean AI words43.290
Mean absolute length residual2.986
Mean absolute percent error6.5%

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 = -0.336, R^2 = 0.113, p = 0.0006.
chrF++ vs passage length
chrF++: significant negative; r = -0.417, R^2 = 0.174, p = 1.57e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.369, R^2 = 0.136, p = 0.0002.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.373, R^2 = 0.139, p = 0.0001.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.251, R^2 = 0.063, p = 0.0119.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.259, R^2 = 0.067, p = 0.0091.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.255, R^2 = 0.065, p = 0.0103.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.285, R^2 = 0.081, p = 0.0041.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.336 0.113 0.0006 -0.00252 ok
chrF++ 100 significant negative -0.417 0.174 1.57e-05 -0.00177 ok
METEOR 100 significant negative -0.369 0.136 0.0002 -0.00163 ok
ROUGE-L 100 significant negative -0.373 0.139 0.0001 -0.00155 ok
BERTScore 0 not enough data N/A N/A N/A N/A needs at least two rows with metric scores
COMET 0 not enough data N/A N/A N/A N/A needs at least two rows with metric scores
BLEURT 0 not enough data N/A N/A N/A N/A needs at least two rows with metric scores
Trigram precision 100 significant negative -0.251 0.063 0.0119 -0.00218 ok
Trigram recall 100 significant negative -0.259 0.067 0.0091 -0.00221 ok
Trigram F1 100 significant negative -0.255 0.065 0.0103 -0.00219 ok
Trigram Jaccard 100 significant negative -0.285 0.081 0.0041 -0.00263 ok

Translation Similarity Metrics

Mean BLEU-457.6%
Mean chrF++77.2%
Mean METEOR80.8%
Mean ROUGE-L81.2%
Mean BERTScoreN/A
Mean COMETN/A
Mean BLEURTN/A
Legacy smoothed corpus BLEU56.0%
Unigram F181.6%
Bigram F161.7%
Trigram precision50.0%
Trigram recall49.0%
Trigram F149.5%
Trigram Jaccard32.9%
4-gram F140.9%

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 270 236 -17.190 42.0% 65.8% 67.7% 70.8% N/A N/A N/A reviewed/approved 3612 gpt-5.5
2604 Καρχηδών 117 128 17.011 35.0% 63.9% 74.6% 69.6% N/A N/A N/A reviewed/approved 3349 gpt-5.5
2468 Καπετώλιον 138 118 -12.507 30.1% 50.3% 53.5% 63.3% N/A N/A N/A reviewed/approved 3326 gpt-5.5
2603 Κάρυστος 175 176 11.105 46.5% 69.5% 77.2% 74.1% N/A N/A N/A reviewed/approved 3346 gpt-5.5
2465 Κάνωπος 83 89 9.611 57.7% 77.4% 87.8% 83.7% N/A N/A N/A reviewed/approved 2418 gpt-5.5
2328 Καλλίπολις 43 50 7.788 38.3% 65.3% 71.2% 66.7% N/A N/A N/A reviewed/approved 2360 gpt-5.5
7254 Κύτα 84 88 7.682 51.9% 73.6% 79.1% 74.4% N/A N/A N/A reviewed/approved 3437 gpt-5.5
2056 Καβειρία 112 99 -7.342 43.9% 71.9% 59.8% 73.9% N/A N/A N/A reviewed/approved 3614 gpt-5.5
2119 Κάλλατις 65 70 7.341 31.8% 63.9% 75.5% 68.1% N/A N/A N/A reviewed/approved 3287 gpt-5.5
7260 Κυχρεῖος πάγος 61 66 7.059 49.6% 74.0% 79.1% 67.7% N/A N/A N/A reviewed/approved 3452 gpt-5.5
2470 Καππαδοκία 68 72 6.553 18.4% 55.5% 76.2% 75.7% N/A N/A N/A reviewed/approved 4057 gpt-5.5
3288 Κορώνεια 110 111 6.517 37.9% 67.9% 65.6% 70.3% N/A N/A N/A reviewed/approved 3615 gpt-5.5
2605 Κάσιον 56 60 5.706 53.0% 75.6% 85.4% 77.6% N/A N/A N/A reviewed/approved 3352 gpt-5.5
2342 Κάμιρος 54 58 5.565 65.5% 85.1% 85.5% 83.9% N/A N/A N/A reviewed/approved 2374 gpt-5.5
7255 Κυτέριον 22 28 5.306 53.0% 76.1% 73.5% 72.0% N/A N/A N/A reviewed/approved 3440 gpt-5.5
2326 Καλλίαρος 47 51 5.070 76.4% 88.2% 96.9% 91.8% N/A N/A N/A reviewed/approved 2358 gpt-5.5
7266 Κώμη 79 71 -4.671 13.8% 47.6% 53.7% 50.7% N/A N/A N/A reviewed/approved 3470 gpt-5.5
2116 Κάληρος 40 35 -4.424 29.3% 61.2% 73.5% 72.0% N/A N/A N/A reviewed/approved 3827 gpt-5.5
2609 Κάσπειρος 135 132 4.282 46.9% 68.4% 70.2% 68.9% N/A N/A N/A reviewed/approved 3358 gpt-5.5
2114 Κάλβιος 18 15 -3.976 36.3% 64.3% 59.2% 66.7% N/A N/A N/A reviewed/approved 3822 gpt-5.5
7258 Κύτωρος 30 34 3.871 69.5% 88.5% 96.3% 87.5% N/A N/A N/A reviewed/approved 3543 gpt-5.5
7267 Κωνώπη 64 58 -3.730 57.6% 75.7% 75.1% 82.0% N/A N/A N/A reviewed/approved 3473 gpt-5.5
2330 Καλύβη 24 21 -3.553 78.4% 89.3% 88.0% 93.3% N/A N/A N/A reviewed/approved 3302 gpt-5.5
2621 Κάστνιον 25 29 3.518 52.1% 79.3% 80.3% 77.8% N/A N/A N/A reviewed/approved 3364 gpt-5.5
2079 Καισάρεια 39 35 -3.494 64.2% 78.3% 85.4% 89.2% N/A N/A N/A reviewed/approved 3264 gpt-5.5
7244 Κυρήνη 54 49 -3.435 48.8% 70.4% 75.7% 75.7% N/A N/A N/A reviewed/approved 3408 gpt-5.5
3530 Κριώα 26 23 -3.412 38.0% 71.1% 65.7% 73.5% N/A N/A N/A reviewed/approved 3400 gpt-5.5
3496 Κοτιάειον 71 65 -3.236 41.0% 62.3% 64.4% 69.1% N/A N/A N/A reviewed/approved 4099 gpt-5.5
2602 Καρύανδα 34 37 3.153 70.0% 81.0% 93.2% 84.5% N/A N/A N/A reviewed/approved 3343 gpt-5.5
7249 Κύρρος 45 41 -3.071 54.8% 71.0% 81.2% 79.1% N/A N/A N/A reviewed/approved 3423 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: -0.208.

Positive termCoefficientDocsMean presentMean absent
απο 0.4951 27 2.025 -0.749
και 0.4769 66 0.270 -0.525
οι 0.4575 11 3.954 -0.489
δε και 0.4440 8 3.094 -0.269
ου 0.3588 14 2.354 -0.383
εστι δε και 0.3517 3 8.772 -0.271
και το 0.3343 9 3.487 -0.345
ος 0.3238 6 3.760 -0.240
τη 0.3059 15 1.711 -0.302
εν τη 0.3053 7 3.580 -0.269
εστι δε 0.2964 5 4.249 -0.224
υπο 0.2785 4 4.656 -0.194
ου και 0.2729 4 7.003 -0.292
πολιτης 0.2612 18 2.606 -0.572
εκληθη 0.2589 4 6.397 -0.267
αφ ου 0.2506 5 3.094 -0.163
εκαλειτο δε 0.2457 3 3.642 -0.113
πολιχνιον 0.2352 3 5.439 -0.168
πολις 0.2329 68 0.480 -1.019
εδει 0.2296 3 5.804 -0.179
Negative termCoefficientDocsMean presentMean absent
γαρ -0.3714 9 -0.947 0.094
το -0.3382 70 -0.013 0.031
οικητωρ -0.3077 6 -4.987 0.318
δια -0.2617 16 0.235 -0.045
δια του -0.2453 10 -1.312 0.146
εθνος -0.2270 8 -3.125 0.272
το εθνικον -0.2166 56 -0.231 0.294
εθνικον -0.2166 56 -0.231 0.294
ηρωδιανος -0.1995 3 -10.107 0.313
θρακης -0.1960 5 -1.746 0.092
εκαταιος ασια -0.1949 5 -1.511 0.080
ασια -0.1949 5 -1.511 0.080
πολις θρακης -0.1890 4 -1.929 0.080
τινες -0.1867 6 -2.045 0.131
του -0.1808 37 0.940 -0.552
ζευς -0.1793 3 -9.503 0.294
ης -0.1621 3 -3.083 0.095
εστι και -0.1594 20 0.056 -0.014
δευτερω -0.1553 3 -1.165 0.036
τινες δε -0.1484 3 -2.875 0.089

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 2.894; cross-validated R^2: -0.200.

Positive termCoefficientDocsMean presentMean absent
την 1.6962 13 7.213 2.355
γαρ 1.6790 9 6.813 2.608
δια 1.6421 16 5.988 2.414
και 1.6231 66 3.528 1.935
ος 1.5738 6 9.008 2.602
εκαλειτο δε 1.4794 3 15.102 2.612
δε και 1.4694 8 8.265 2.527
δια του 1.4244 10 7.281 2.509
απο 1.3221 27 5.413 2.089
εις 1.2586 7 9.140 2.523
δε 1.2003 46 3.967 2.151
εκαλειτο 1.1712 7 7.456 2.650
εκαλειτο δε και 0.9971 3 15.102 2.612
εστι δε και 0.9933 3 8.772 2.807
εν 0.9928 37 3.889 2.456
της 0.9148 35 4.484 2.180
εστι 0.8724 24 4.508 2.506
εστι δε 0.8479 5 7.186 2.765
αφ ου 0.8136 5 6.567 2.798
τη 0.8018 15 5.005 2.630
Negative termCoefficientDocsMean presentMean absent
προς τη -0.8332 4 1.512 3.048
το εθνικον -0.7562 56 2.985 2.988
εθνικον -0.7562 56 2.985 2.988
τον -0.7507 10 2.721 3.016
προς -0.7345 15 2.976 2.988
εκ -0.7297 6 1.839 3.059
του δε -0.6827 3 0.824 3.053
νησος -0.6635 9 2.220 3.062
καππαδοκιας -0.6541 3 1.498 3.032
και πολις -0.5984 5 2.430 3.015
ακρα -0.5383 4 1.250 3.059
πλησιον -0.4690 8 1.943 3.077
μια -0.4618 3 1.447 3.034
τεταρτω -0.4489 3 1.784 3.023
θεοπομπος -0.4246 5 1.395 3.070
στραβων -0.3685 15 3.709 2.859
αιγυπτου -0.3594 3 2.655 2.996
εκαταιος -0.3534 14 2.361 3.088
τινες -0.3423 6 2.379 3.025
τοπικα -0.3234 3 1.675 3.027

AI English terms predicting large absolute residuals

Features: 326; ridge alpha: 4.125; cross-validated R^2: -0.213.

Positive termCoefficientDocsMean presentMean absent
with 1.2196 17 5.332 2.506
there 1.0042 34 4.388 2.264
used to be 0.9390 4 10.100 2.690
used 0.8796 11 6.829 2.511
used to 0.8766 5 8.727 2.684
to 0.8666 28 5.112 2.160
it with 0.8652 3 11.224 2.731
on 0.8406 14 5.167 2.631
there is 0.8340 27 4.777 2.324
be called 0.7852 6 8.606 2.628
to be called 0.7852 6 8.606 2.628
be 0.7371 15 5.917 2.469
to be 0.7327 13 6.377 2.480
by 0.7241 12 6.361 2.526
it 0.6972 38 4.284 2.191
and 0.6811 56 3.808 1.941
for 0.6614 12 5.878 2.592
is the 0.6577 4 8.359 2.762
however 0.6467 6 9.004 2.602
who 0.6404 9 6.163 2.672
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.6511 56 2.775 3.255
the ethnonym is -0.6272 53 2.819 3.175
ethnonym -0.6113 57 2.855 3.160
ethnonym is -0.5864 54 2.902 3.085
a city -0.5295 64 3.051 2.871
promontory -0.4848 5 1.304 3.075
kappadokia -0.4330 3 1.498 3.032
about -0.4303 3 0.584 3.061
which the -0.4186 5 1.424 3.068
island -0.3985 14 2.576 3.053
been -0.3686 5 2.273 3.024
book -0.3583 39 2.868 3.062
in the -0.3560 21 2.866 3.018
as in -0.3470 25 2.825 3.040
a city in -0.3432 45 3.203 2.809
an island -0.3432 10 2.139 3.080
near -0.3341 19 3.035 2.975
also a city -0.3113 4 1.597 3.044
city in -0.3106 46 3.258 2.755
one of -0.3031 3 1.525 3.031

Human English terms predicting large absolute residuals

Features: 336; ridge alpha: 4.125; cross-validated R^2: -0.268.

Positive termCoefficientDocsMean presentMean absent
with 1.4637 16 5.118 2.580
and 1.1009 52 3.866 2.033
is with 1.0618 4 12.671 2.583
the 0.9734 90 3.115 1.828
also 0.8277 43 3.931 2.274
who 0.7924 13 5.620 2.593
on 0.7740 16 4.510 2.696
was 0.7699 20 5.012 2.480
there 0.7042 34 4.224 2.349
he 0.6491 3 10.643 2.749
forms 0.6378 4 10.898 2.657
for 0.6238 11 5.542 2.670
their 0.6116 3 15.102 2.612
sanctuary 0.6012 3 9.275 2.792
to 0.5843 31 4.258 2.415
be 0.5806 15 6.318 2.398
usage 0.5545 4 6.986 2.820
into 0.5498 3 8.058 2.829
is 0.5481 87 3.092 2.281
after 0.5409 22 5.135 2.380
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.6640 52 2.750 3.242
ethnonym is -0.6640 52 2.750 3.242
ethnonym -0.5540 57 2.955 3.027
the ethnonym -0.5271 56 2.998 2.971
a city -0.5049 64 3.067 2.842
as in -0.4416 29 3.037 2.965
ethnonyms -0.3984 3 0.800 3.054
a city in -0.3842 40 3.100 2.911
herodotos book -0.3569 5 1.626 3.058
island -0.3545 14 2.576 3.053
city in -0.3466 41 3.163 2.863
near -0.3440 16 2.916 3.000
it is -0.3426 24 3.858 2.711
city -0.3150 70 3.153 2.598
an -0.3124 22 2.863 3.021
one of the -0.3115 4 1.471 3.049
one of -0.3105 5 1.734 3.052
kappadokia -0.3051 3 2.518 3.001
an island -0.2984 9 2.227 3.061
in the -0.2855 25 3.406 2.846

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.