gpt-5.5_v4_mini_medium v1 Evaluation

Generated: 2026-09-20 11:22:24 UTC

Pairs100
Distinct lemmas100
Slope0.898
R^20.985
BLEU-445.6%
chrF++69.3%
METEOR75.9%
ROUGE-L77.5%
BERTScore95.8%
COMET78.6%
BLEURT73.3%

Profile version id: 1765

First translation using this version: 2026-06-27

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=medium.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.898
Intercept2.100
R^20.985
P-value1.12e-90
Pearson r0.992
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words41.740
Mean absolute length residual2.841
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 = -0.360, R^2 = 0.129, p = 0.0002.
chrF++ vs passage length
chrF++: significant negative; r = -0.396, R^2 = 0.157, p = 4.52e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.508, R^2 = 0.258, p = 6.75e-08.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.503, R^2 = 0.253, p = 9.72e-08.
BERTScore vs passage length
BERTScore: significant negative; r = -0.576, R^2 = 0.332, p = 3.66e-10.
COMET vs passage length
COMET: significant negative; r = -0.436, R^2 = 0.190, p = 5.70e-06.
BLEURT vs passage length
BLEURT: significant negative; r = -0.709, R^2 = 0.503, p = 1.47e-16.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.383, R^2 = 0.147, p = 8.44e-05.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.393, R^2 = 0.154, p = 5.34e-05.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.389, R^2 = 0.151, p = 6.42e-05.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.386, R^2 = 0.149, p = 7.46e-05.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.360 0.129 0.0002 -0.00221 ok
chrF++ 100 significant negative -0.396 0.157 4.52e-05 -0.00153 ok
METEOR 100 significant negative -0.508 0.258 6.75e-08 -0.00230 ok
ROUGE-L 100 significant negative -0.503 0.253 9.72e-08 -0.00210 ok
BERTScore 100 significant negative -0.576 0.332 3.66e-10 -0.00044 ok
COMET 100 significant negative -0.436 0.190 5.70e-06 -0.00084 ok
BLEURT 100 significant negative -0.709 0.503 1.47e-16 -0.00241 ok
Trigram precision 100 significant negative -0.383 0.147 8.44e-05 -0.00276 ok
Trigram recall 100 significant negative -0.393 0.154 5.34e-05 -0.00269 ok
Trigram F1 100 significant negative -0.389 0.151 6.42e-05 -0.00272 ok
Trigram Jaccard 100 significant negative -0.386 0.149 7.46e-05 -0.00252 ok

Translation Similarity Metrics

Mean BLEU-445.6%
Mean chrF++69.3%
Mean METEOR75.9%
Mean ROUGE-L77.5%
Mean BERTScore95.8%
Mean COMET78.6%
Mean BLEURT73.3%
Legacy smoothed corpus BLEU45.7%
Unigram F177.3%
Bigram F153.2%
Trigram precision40.8%
Trigram recall38.5%
Trigram F139.6%
Trigram Jaccard24.7%
4-gram F129.9%

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 17.776 43.2% 66.5% 67.5% 68.8% 94.7% 77.6% 57.9% reviewed/approved 7307 gpt-5.4-mini
2484 Καρία 270 230 -14.520 27.0% 55.9% 56.2% 63.6% 90.5% 79.2% 56.4% reviewed/approved 7295 gpt-5.4-mini
2604 Καρχηδών 117 121 13.851 39.9% 66.4% 71.0% 66.7% 94.0% 78.4% 61.6% reviewed/approved 7310 gpt-5.4-mini
2056 Καβειρία 112 89 -13.659 26.6% 61.4% 53.6% 61.7% 94.5% 76.5% 55.7% reviewed/approved 7154 gpt-5.4-mini
2468 Καπετώλιον 138 113 -13.004 23.2% 48.8% 49.8% 52.6% 88.1% 67.9% 50.2% reviewed/approved 7286 gpt-5.4-mini
7254 Κύτα 84 89 11.481 29.6% 58.1% 59.5% 63.6% 92.1% 74.1% 56.5% reviewed/approved 7406 gpt-5.4-mini
7260 Κυχρεῖος πάγος 61 66 9.131 45.6% 69.9% 79.1% 67.7% 95.8% 77.6% 66.3% reviewed/approved 7424 gpt-5.4-mini
7255 Κυτέριον 22 30 8.147 34.9% 64.8% 67.3% 65.4% 95.7% 76.0% 74.5% reviewed/approved 7409 gpt-5.4-mini
2605 Κάσιον 56 60 7.620 42.6% 68.2% 78.5% 75.9% 95.6% 74.5% 64.1% reviewed/approved 7313 gpt-5.4-mini
7251 Κύρτος 66 54 -7.358 27.8% 54.2% 64.9% 66.1% 93.1% 69.9% 63.2% reviewed/approved 7397 gpt-5.4-mini
2342 Κάμιρος 54 57 6.416 43.5% 70.6% 69.4% 72.1% 95.0% 79.1% 64.3% reviewed/approved 7271 gpt-5.4-mini
7248 Κύρου πόλις 32 37 6.169 19.2% 63.2% 66.5% 66.7% 94.7% 78.0% 76.7% reviewed/approved 7388 gpt-5.4-mini
2054 Καβαλίς 65 66 5.540 51.9% 66.1% 67.7% 74.8% 94.5% 74.0% 69.1% reviewed/approved 5665 gpt-5.4-mini
2333 Κάλυμνα 38 41 4.782 54.7% 78.0% 84.1% 81.0% 96.0% 77.9% 74.6% reviewed/approved 7262 gpt-5.4-mini
2328 Καλλίπολις 43 45 4.293 25.4% 54.5% 57.7% 59.1% 94.1% 74.6% 68.4% reviewed/approved 7247 gpt-5.4-mini
2331 Κάλυδνα 43 45 4.293 64.5% 85.0% 92.2% 90.9% 96.7% 82.9% 81.4% reviewed/approved 7256 gpt-5.4-mini
2081 Καλάθη 49 42 -4.095 42.1% 69.4% 72.5% 83.5% 95.7% 73.6% 70.8% reviewed/approved 7199 gpt-5.4-mini
7266 Κώμη 79 69 -4.030 9.5% 39.9% 38.2% 43.2% 91.9% 70.8% 51.0% reviewed/approved 7442 gpt-5.4-mini
7264 Κωλιάς 61 53 -3.869 35.4% 59.7% 52.4% 64.3% 94.5% 71.8% 58.7% reviewed/approved 7436 gpt-5.4-mini
2346 Κάναι 77 75 3.766 54.7% 69.8% 80.5% 82.4% 95.9% 77.2% 66.6% reviewed/approved 7277 gpt-5.4-mini
7253 Κυρτώνιος 18 22 3.739 41.9% 69.0% 77.4% 70.0% 96.8% 77.1% 80.2% reviewed/approved 7403 gpt-5.4-mini
2465 Κάνωπος 83 80 3.378 48.8% 71.4% 75.6% 78.5% 95.8% 77.1% 71.6% reviewed/approved 7283 gpt-5.4-mini
2080 Καλαβρία 18 15 -3.261 30.7% 58.1% 65.3% 72.7% 92.6% 79.2% 76.0% reviewed/approved 7196 gpt-5.4-mini
2599 Καρπήσιοι 10 8 -3.078 23.1% 65.5% 70.6% 77.8% 97.1% 87.3% 78.2% reviewed/approved 7301 gpt-5.4-mini
2623 Κασώριον 20 17 -3.057 10.2% 54.0% 64.4% 70.3% 92.9% 72.7% 71.0% reviewed/approved 7331 gpt-5.4-mini
2116 Κάληρος 40 35 -3.014 28.9% 56.8% 68.7% 69.3% 93.1% 76.3% 71.0% reviewed/approved 7226 gpt-5.4-mini
7250 Κυρταία 40 35 -3.014 30.4% 63.9% 66.3% 66.7% 96.3% 72.9% 70.7% reviewed/approved 7394 gpt-5.4-mini
2078 Καινύς 22 19 -2.853 59.1% 68.7% 73.2% 78.0% 95.6% 82.6% 78.5% reviewed/approved 7190 gpt-5.4-mini
2470 Καππαδοκία 68 66 2.846 46.4% 67.8% 85.3% 83.6% 96.5% 77.1% 71.1% reviewed/approved 7289 gpt-5.4-mini
2326 Καλλίαρος 47 47 2.701 41.6% 67.4% 82.4% 78.7% 95.6% 70.2% 69.6% reviewed/approved 7241 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: 5.878; cross-validated R^2: -0.163.

Positive termCoefficientDocsMean presentMean absent
δε και 1.1967 8 3.898 -0.339
οι 1.1229 11 4.123 -0.510
τη 1.0820 15 2.335 -0.412
πολιτης 0.9687 18 3.026 -0.664
απο 0.9343 27 1.477 -0.546
ος 0.9092 6 3.910 -0.250
εκληθη 0.8645 4 8.679 -0.362
στραβων 0.8529 15 3.011 -0.531
λεγεται 0.8463 9 3.482 -0.344
πολις 0.8179 68 0.448 -0.951
λεγεται και 0.8108 8 3.630 -0.316
εν τη 0.7905 7 3.675 -0.277
πολις προς 0.7666 3 8.527 -0.264
ου και 0.7492 4 9.083 -0.378
και 0.7325 66 0.345 -0.670
φησι 0.6888 12 2.061 -0.281
εκαλειτο δε 0.6786 3 5.702 -0.176
εστι δε και 0.6670 3 6.093 -0.188
των 0.6655 17 1.792 -0.367
αφ ου 0.6546 5 2.523 -0.133
Negative termCoefficientDocsMean presentMean absent
γαρ -0.9675 9 -0.223 0.022
του -0.8886 37 0.579 -0.340
μεν -0.7200 4 -9.030 0.376
το -0.6906 70 0.039 -0.092
οικητωρ -0.6830 6 -4.395 0.281
δια του -0.6440 10 -1.466 0.163
εθνος -0.5765 8 -3.243 0.282
παυσανιας -0.5400 4 -4.707 0.196
ης -0.5295 3 -4.073 0.126
ηρωδιανος -0.4935 3 -9.425 0.291
τω περι -0.4754 4 -4.987 0.208
εν τω περι -0.4754 4 -4.987 0.208
δια -0.4739 16 0.807 -0.154
παρα το -0.4706 3 -3.840 0.119
ως το -0.4644 3 -6.328 0.196
εν τω -0.4338 6 -0.671 0.043
παρα -0.4211 7 -3.025 0.228
του δε -0.4012 3 -3.992 0.123
περι -0.3982 11 -2.120 0.262
ζευς -0.3945 3 -7.919 0.245

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 5.878; cross-validated R^2: -1.385.

Positive termCoefficientDocsMean presentMean absent
γαρ 1.2306 9 7.774 2.354
την 1.1919 13 8.190 2.042
και 1.1679 66 3.507 1.550
τω 1.1261 16 5.739 2.290
δε 0.9740 46 3.958 1.890
ος 0.8672 6 8.527 2.479
αφ 0.8433 6 7.533 2.542
εκαλειτο δε 0.7928 3 15.382 2.454
απο 0.7832 27 5.162 1.983
δε και 0.7708 8 7.808 2.410
αφ ου 0.7551 5 8.651 2.536
λεγεται και 0.6926 8 8.268 2.370
δια 0.6832 16 5.385 2.357
της 0.6644 35 4.485 1.956
λεγεται 0.6460 9 7.604 2.370
εν 0.6294 37 3.890 2.226
το θηλυκον 0.6022 6 8.323 2.492
απο της 0.5977 6 8.930 2.453
τη 0.5943 15 4.952 2.469
εκαλειτο 0.5912 7 7.555 2.487
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.7511 56 2.774 2.927
εθνικον -0.7511 56 2.774 2.927
ως -0.4838 52 3.091 2.571
προς τη -0.4501 4 1.173 2.911
ηροδοτος -0.4232 9 2.851 2.841
τους -0.4179 4 1.019 2.917
τεταρτω -0.3771 3 0.906 2.901
φασι -0.3725 8 3.161 2.814
πολυβιος -0.3702 5 1.118 2.932
τινες -0.3653 6 1.259 2.942
δευτερω -0.3648 3 1.200 2.892
καππαδοκιας -0.3494 3 1.266 2.890
λυκιας -0.3077 3 0.981 2.899
τινες δε -0.2966 3 0.555 2.912
μια -0.2849 3 1.180 2.893
τον -0.2796 10 3.303 2.790
τις -0.2582 3 1.114 2.895
πολις -0.2538 68 3.071 2.354
εκ -0.2418 6 2.647 2.854
νησος -0.2268 9 2.362 2.889

AI English terms predicting large absolute residuals

Features: 317; ridge alpha: 8.377; cross-validated R^2: -1.465.

Positive termCoefficientDocsMean presentMean absent
the 1.2225 92 2.995 1.080
when 0.7646 6 10.930 2.325
after 0.7112 22 5.239 2.165
and 0.7086 58 3.648 1.728
of 0.6806 68 3.295 1.877
was 0.6196 16 5.672 2.302
called 0.6018 29 5.017 1.953
whom 0.5527 3 13.019 2.527
to 0.5436 24 5.216 2.091
while 0.4825 4 14.952 2.337
because of 0.4736 4 9.313 2.572
is 0.4732 89 2.956 1.915
usage 0.4324 4 8.507 2.605
was called 0.4324 4 13.820 2.384
also 0.4249 46 3.890 1.948
for 0.4230 12 6.352 2.363
who 0.4048 11 5.147 2.556
there 0.3976 32 4.089 2.254
so 0.3741 7 8.278 2.432
the feminine 0.3728 13 6.961 2.226
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.4239 57 2.651 3.094
the ethnonym is -0.4039 57 2.651 3.094
just as -0.3723 14 1.339 3.086
just -0.3723 14 1.339 3.086
ethnonym -0.3555 58 2.840 2.843
ethnonym is -0.3357 58 2.840 2.843
book -0.3320 39 2.825 2.852
they -0.3070 9 3.045 2.821
city in -0.2640 35 2.563 2.991
a city in -0.2640 35 2.563 2.991
some -0.2465 8 1.543 2.954
herodotos book -0.2299 5 1.069 2.935
book of his -0.2107 13 2.700 2.863
book of -0.2107 13 2.700 2.863
herodotos -0.2083 9 2.851 2.841
a city -0.2047 64 3.151 2.292
book the ethnonym -0.1947 7 1.321 2.956
strabo book -0.1912 9 2.123 2.913
say -0.1881 9 2.213 2.904
about -0.1862 3 0.771 2.905

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 11.938; cross-validated R^2: -1.522.

Positive termCoefficientDocsMean presentMean absent
the 0.8096 90 3.029 1.156
also 0.5521 43 4.092 1.898
with 0.4977 16 4.429 2.539
was 0.4346 20 4.886 2.330
is also 0.4240 31 4.379 2.151
for 0.4205 11 6.327 2.411
called 0.3964 21 5.753 2.068
and 0.3885 52 3.673 1.940
there 0.3815 34 3.996 2.247
is 0.3534 87 2.944 2.154
after 0.3447 22 4.810 2.286
named 0.3423 5 7.856 2.578
who 0.3192 13 5.588 2.431
to 0.3061 31 4.317 2.179
is with 0.2862 4 10.409 2.526
the form 0.2818 9 6.902 2.440
from 0.2798 37 3.631 2.378
from the 0.2779 13 6.188 2.341
from the form 0.2748 5 6.423 2.653
usage 0.2727 4 7.355 2.653
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.3717 52 2.335 3.390
ethnonym is -0.3717 52 2.335 3.390
a city in -0.2936 40 2.802 2.868
book -0.2642 39 2.825 2.852
city in -0.2639 41 2.920 2.787
ethnonym -0.2618 57 2.872 2.801
the ethnonym -0.2600 56 2.887 2.784
a city -0.2142 64 3.056 2.461
as in -0.2062 29 2.751 2.878
herodotos book -0.1993 5 1.069 2.935
herodotos -0.1914 9 2.851 2.841
in book of -0.1794 14 1.848 3.003
book of his -0.1712 16 2.560 2.895
book of -0.1712 16 2.560 2.895
an -0.1566 22 2.703 2.880
about -0.1547 5 1.169 2.929
kappadokia -0.1534 4 1.661 2.891
of his -0.1506 18 2.731 2.866
one -0.1322 12 2.881 2.836
just as -0.1306 5 1.258 2.925

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.