gpt-5.1 v3 Evaluation

Generated: 2026-09-20 11:21:21 UTC

Pairs100
Distinct lemmas100
Slope1.005
R^20.989
BLEU-446.3%
chrF++70.7%
METEOR76.9%
ROUGE-L76.9%
BERTScore95.7%
COMET79.0%
BLEURT73.5%

Profile version id: 2345

First translation using this version: 2026-07-15

Total usable AI runs for this version: 100 runs across 100 lemmas

Created: 2026-07-15 13:49:25.227272+10:00

Active: True

Notes: Model timeline evaluation: GPT-5.1 using prompt text from gpt-5.5 v3; approved-human-only for the 100-row Kappa corpus.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope1.005
Intercept0.825
R^20.989
P-value3.80e-97
Pearson r0.994
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words45.180
Mean absolute length residual3.052
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 = -0.336, R^2 = 0.113, p = 0.0006.
chrF++ vs passage length
chrF++: significant negative; r = -0.375, R^2 = 0.141, p = 0.0001.
METEOR vs passage length
METEOR: significant negative; r = -0.466, R^2 = 0.217, p = 1.01e-06.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.479, R^2 = 0.229, p = 4.75e-07.
BERTScore vs passage length
BERTScore: significant negative; r = -0.595, R^2 = 0.354, p = 6.92e-11.
COMET vs passage length
COMET: significant negative; r = -0.411, R^2 = 0.169, p = 2.21e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.716, R^2 = 0.512, p = 5.80e-17.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.358, R^2 = 0.128, p = 0.0003.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.360, R^2 = 0.129, p = 0.0002.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.359, R^2 = 0.129, p = 0.0002.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.345, R^2 = 0.119, p = 0.0004.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.336 0.113 0.0006 -0.00197 ok
chrF++ 100 significant negative -0.375 0.141 0.0001 -0.00134 ok
METEOR 100 significant negative -0.466 0.217 1.01e-06 -0.00210 ok
ROUGE-L 100 significant negative -0.479 0.229 4.75e-07 -0.00191 ok
BERTScore 100 significant negative -0.595 0.354 6.92e-11 -0.00043 ok
COMET 100 significant negative -0.411 0.169 2.21e-05 -0.00078 ok
BLEURT 100 significant negative -0.716 0.512 5.80e-17 -0.00237 ok
Trigram precision 100 significant negative -0.358 0.128 0.0003 -0.00270 ok
Trigram recall 100 significant negative -0.360 0.129 0.0002 -0.00266 ok
Trigram F1 100 significant negative -0.359 0.129 0.0002 -0.00266 ok
Trigram Jaccard 100 significant negative -0.345 0.119 0.0004 -0.00268 ok

Translation Similarity Metrics

Mean BLEU-446.3%
Mean chrF++70.7%
Mean METEOR76.9%
Mean ROUGE-L76.9%
Mean BERTScore95.7%
Mean COMET79.0%
Mean BLEURT73.5%
Legacy smoothed corpus BLEU47.3%
Unigram F177.6%
Bigram F153.8%
Trigram precision39.9%
Trigram recall40.8%
Trigram F140.3%
Trigram Jaccard25.3%
4-gram F131.1%

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 175 192 15.364 31.4% 62.6% 61.4% 63.8% 93.1% 78.1% 54.2% reviewed/approved 12094 gpt-5.1-2025-11-13
2604 Καρχηδών 117 132 13.632 35.8% 68.3% 75.9% 70.9% 93.6% 79.6% 63.2% reviewed/approved 12097 gpt-5.1-2025-11-13
2056 Καβειρία 112 101 -12.344 26.8% 64.7% 51.3% 63.8% 94.6% 77.5% 59.0% reviewed/approved 11941 gpt-5.1-2025-11-13
2484 Καρία 270 261 -11.076 37.6% 63.6% 63.6% 66.3% 91.7% 80.0% 53.8% reviewed/approved 12082 gpt-5.1-2025-11-13
2626 Κατάνη 95 86 -10.266 39.6% 61.1% 61.4% 65.2% 93.6% 72.6% 61.8% reviewed/approved 12130 gpt-5.1-2025-11-13
7254 Κύτα 84 93 7.785 35.2% 64.7% 72.3% 67.8% 93.5% 72.9% 60.6% reviewed/approved 12193 gpt-5.1-2025-11-13
2055 Καβασσός 109 103 -7.330 39.2% 64.6% 65.4% 63.2% 92.6% 77.3% 54.6% reviewed/approved 11938 gpt-5.1-2025-11-13
3496 Κοτιάειον 71 65 -7.154 19.8% 52.9% 58.9% 58.8% 93.6% 74.3% 67.2% reviewed/approved 12148 gpt-5.1-2025-11-13
7255 Κυτέριον 22 30 7.073 29.0% 61.7% 55.4% 61.5% 93.7% 71.5% 63.9% reviewed/approved 12196 gpt-5.1-2025-11-13
2328 Καλλίπολις 43 51 6.975 36.5% 64.8% 70.2% 63.8% 92.1% 74.8% 68.1% reviewed/approved 12034 gpt-5.1-2025-11-13
7259 Κύφος 54 62 6.924 42.7% 72.0% 78.0% 75.9% 95.5% 74.7% 70.2% reviewed/approved 12208 gpt-5.1-2025-11-13
2054 Καβαλίς 65 73 6.873 52.4% 72.5% 61.0% 73.9% 95.5% 77.2% 73.5% reviewed/approved 11935 gpt-5.1-2025-11-13
3288 Κορώνεια 110 118 6.665 36.3% 67.5% 69.7% 70.7% 94.2% 76.1% 66.0% reviewed/approved 12145 gpt-5.1-2025-11-13
2477 Καρδαμύλη 43 50 5.975 27.7% 62.1% 74.2% 68.8% 95.2% 78.0% 73.9% reviewed/approved 12079 gpt-5.1-2025-11-13
2081 Καλάθη 49 56 5.948 42.4% 70.7% 70.6% 80.0% 95.4% 73.1% 70.1% reviewed/approved 11986 gpt-5.1-2025-11-13
2119 Κάλλατις 65 72 5.873 27.2% 61.4% 72.4% 62.8% 93.0% 70.7% 64.6% reviewed/approved 12022 gpt-5.1-2025-11-13
2470 Καππαδοκία 68 75 5.859 27.1% 63.9% 74.4% 75.5% 95.5% 75.2% 67.5% reviewed/approved 12076 gpt-5.1-2025-11-13
2465 Κάνωπος 83 90 5.790 42.2% 69.3% 73.3% 75.1% 95.1% 75.9% 68.0% reviewed/approved 12070 gpt-5.1-2025-11-13
2455 Κάναστρον 60 56 -5.103 21.7% 57.2% 47.4% 60.3% 93.1% 70.7% 61.8% reviewed/approved 12067 gpt-5.1-2025-11-13
7244 Κυρήνη 54 50 -5.076 41.2% 68.1% 75.4% 69.2% 95.4% 75.6% 70.5% reviewed/approved 12163 gpt-5.1-2025-11-13
7260 Κυχρεῖος πάγος 61 67 4.892 43.8% 69.2% 76.1% 65.6% 94.9% 76.7% 60.0% reviewed/approved 12211 gpt-5.1-2025-11-13
2115 Καλὴ ἀκτή 64 61 -4.122 32.8% 59.6% 65.2% 62.4% 93.6% 72.7% 58.8% reviewed/approved 12010 gpt-5.1-2025-11-13
7247 Κύρνος 50 47 -4.057 41.8% 63.1% 72.3% 74.2% 96.3% 77.5% 78.4% reviewed/approved 12172 gpt-5.1-2025-11-13
7261 Κύψελα 34 31 -3.983 47.3% 70.9% 69.9% 70.8% 95.5% 80.7% 75.2% reviewed/approved 12214 gpt-5.1-2025-11-13
2331 Κάλυδνα 43 48 3.975 41.4% 71.3% 89.6% 76.9% 95.2% 78.7% 74.6% reviewed/approved 12043 gpt-5.1-2025-11-13
2335 Κάλυτις 26 23 -3.946 38.1% 61.2% 62.0% 73.5% 96.3% 79.4% 68.4% reviewed/approved 12055 gpt-5.1-2025-11-13
2330 Καλύβη 24 21 -3.937 56.4% 76.9% 79.4% 84.4% 96.2% 80.9% 75.8% reviewed/approved 12040 gpt-5.1-2025-11-13
2342 Κάμιρος 54 59 3.924 45.2% 73.4% 77.1% 69.0% 94.6% 80.0% 68.6% reviewed/approved 12058 gpt-5.1-2025-11-13
2334 Κάλυνδα 15 19 3.105 48.6% 80.8% 82.4% 82.4% 95.0% 79.7% 76.5% reviewed/approved 12052 gpt-5.1-2025-11-13
2329 Κάλπη 57 55 -3.090 49.2% 70.9% 70.4% 76.1% 94.9% 75.3% 72.3% reviewed/approved 12037 gpt-5.1-2025-11-13

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: 4.125; cross-validated R^2: -0.094.

Positive termCoefficientDocsMean presentMean absent
δε και 1.6739 8 4.568 -0.397
οι 1.5082 11 4.124 -0.510
τη 1.2825 15 2.307 -0.407
εστι δε και 1.2709 3 8.758 -0.271
απο 1.2699 27 1.476 -0.546
πολιτης 1.1845 18 3.192 -0.701
ου 1.0622 14 2.093 -0.341
λεγεται 0.9961 9 3.290 -0.325
εκαλειτο 0.9741 7 1.932 -0.145
πολις 0.9590 68 0.213 -0.453
εν τη 0.9511 7 3.380 -0.254
παιδος 0.9359 4 3.965 -0.165
ος 0.9338 6 3.925 -0.251
εκαλειτο δε 0.9148 3 5.973 -0.185
εις 0.9051 7 2.875 -0.216
στραβων 0.8985 15 2.106 -0.372
ου και 0.8028 4 6.781 -0.283
λεγεται και 0.7736 8 2.955 -0.257
αυτην 0.7504 9 3.104 -0.307
πολιχνιον 0.6231 3 4.222 -0.131
Negative termCoefficientDocsMean presentMean absent
το -1.2345 70 -0.375 0.876
ην -1.1168 13 -1.406 0.210
θρακης -1.0287 5 -3.376 0.178
παρα το -0.8615 3 -4.837 0.150
πολις της -0.8296 3 -6.560 0.203
πολις θρακης -0.8054 4 -2.945 0.123
μετα -0.8018 3 -7.598 0.235
τα -0.7603 12 -1.814 0.247
θεοπομπος -0.7348 5 -2.445 0.129
εστι και -0.7068 20 -0.083 0.021
παυσανιας -0.6910 4 -4.275 0.178
το κτητικον -0.6842 7 -3.428 0.258
κτητικον -0.6842 7 -3.428 0.258
ης -0.6682 3 -3.765 0.116
εθνος -0.6461 8 -2.636 0.229
τινες -0.6319 6 -2.191 0.140
ποταμου -0.6209 3 -2.346 0.073
ως το -0.5998 3 -5.935 0.184
τινες δε -0.5841 3 -3.670 0.113
δε της -0.5828 6 -1.421 0.091

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 1.425; cross-validated R^2: 0.156.

Positive termCoefficientDocsMean presentMean absent
δε 3.5437 46 4.270 2.014
της 2.2782 35 4.609 2.213
εκαλειτο δε 2.0362 3 13.358 2.733
την 1.8133 13 6.978 2.465
το θηλυκον 1.6860 6 7.667 2.757
ος 1.6826 6 6.798 2.813
εστι δε 1.6716 5 8.548 2.762
και 1.6504 66 3.617 1.955
απο 1.5901 27 5.227 2.247
δε και 1.5231 8 7.338 2.679
εκαλειτο δε και 1.4820 3 13.358 2.733
εστι 1.4800 24 5.193 2.376
αλλη 1.4771 4 9.190 2.796
αφ 1.4594 6 6.654 2.822
αλεξανδρος 1.4236 3 7.001 2.930
και αλλη 1.4124 3 10.561 2.819
τη 1.3359 15 4.651 2.769
εστι δε και 1.3258 3 8.758 2.875
μεν 1.3194 4 8.053 2.843
εκαλειτο 1.2939 7 7.217 2.738
Negative termCoefficientDocsMean presentMean absent
προς τη -1.5801 4 0.565 3.155
και πολις -1.4706 5 1.618 3.127
τους -1.2437 4 2.314 3.082
το εθνικον -1.1109 56 3.084 3.011
εθνικον -1.1109 56 3.084 3.011
τε -1.1022 5 2.526 3.079
οικητωρ -0.9045 6 3.051 3.052
του δε -0.8269 3 1.970 3.085
πολυβιος -0.8155 5 1.219 3.148
πολις και -0.8117 6 3.001 3.055
μια -0.7491 3 1.333 3.105
αιγυπτου -0.6604 3 1.966 3.085
χωρα -0.6511 6 3.673 3.012
δια του -0.6094 10 4.125 2.932
ακρα -0.6003 4 1.841 3.102
διονυσιος -0.5984 4 1.644 3.110
τεταρτω -0.5677 3 1.955 3.086
ιταλιας -0.4585 3 1.370 3.104
προς -0.4571 15 3.511 2.971
ως στραβων -0.4528 4 3.193 3.046

AI English terms predicting large absolute residuals

Features: 355; ridge alpha: 0.492; cross-validated R^2: 0.193.

Positive termCoefficientDocsMean presentMean absent
of 3.8706 66 3.738 1.719
the 3.4488 95 3.155 1.083
is 3.1822 92 3.192 1.435
is the 3.1225 7 8.202 2.664
to 2.9810 26 5.257 2.277
however 2.8122 10 6.865 2.628
when 2.7638 7 8.356 2.653
and there is 2.6781 4 10.415 2.745
from whom 2.6206 3 9.874 2.841
whose 2.5236 3 7.881 2.902
europe 2.4541 6 6.966 2.802
is a 2.2886 24 5.196 2.375
used 2.2635 10 7.557 2.551
he 2.2015 3 9.867 2.841
in 2.1302 82 3.228 2.251
his europe 2.0334 5 6.802 2.854
in his europe 2.0334 5 6.802 2.854
while 2.0113 3 10.641 2.817
there 1.9992 31 4.975 2.188
is a city 1.9918 3 6.699 2.939
Negative termCoefficientDocsMean presentMean absent
and a -1.9843 6 2.454 3.090
the ethnonym is -1.8920 55 2.912 3.222
the ethnonym -1.7545 57 2.974 3.154
and in -1.4707 3 4.208 3.016
ethnonym is -1.4702 56 3.081 3.015
one of the -1.3771 3 1.333 3.105
one of -1.3771 3 1.333 3.105
island -1.3726 14 2.491 3.143
ethnonym -1.3433 58 3.136 2.936
them -1.2160 4 2.646 3.069
and the -1.1680 19 3.503 2.946
out -1.1572 3 2.745 3.061
the island -1.1316 3 2.299 3.075
so -1.1066 8 4.041 2.966
but -1.0664 20 3.813 2.861
near -1.0576 19 3.101 3.040
called -1.0145 24 4.217 2.684
dionysios -1.0140 4 1.644 3.110
forms are -0.9678 3 2.035 3.083
forms -0.9678 3 2.035 3.083

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 1.425; cross-validated R^2: 0.139.

Positive termCoefficientDocsMean presentMean absent
the 2.6040 90 3.254 1.234
there 2.2969 34 4.881 2.109
of 2.1436 66 3.631 1.928
who 1.6646 13 6.357 2.558
and 1.4955 52 3.942 2.087
into 1.4631 3 9.296 2.859
to 1.4244 31 4.663 2.328
when 1.3690 7 6.079 2.824
also 1.3432 43 4.311 2.102
from the 1.3251 13 5.978 2.615
in 1.3018 87 3.213 1.974
was 1.2545 20 4.728 2.633
their 1.2272 3 13.358 2.733
for 1.1931 11 6.257 2.656
europe 1.1790 6 6.966 2.802
from 1.1514 37 4.207 2.373
is also 1.1473 31 4.954 2.197
this 1.1440 15 6.536 2.437
karia 1.1117 3 7.052 2.928
his europe 1.1030 5 6.802 2.854
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -1.2547 52 2.631 3.507
ethnonym is -1.2547 52 2.631 3.507
and a -1.1321 3 0.724 3.124
with -0.9567 16 3.404 2.985
with a -0.7930 4 1.259 3.126
but -0.7829 12 2.568 3.118
the ethnonym -0.7072 56 3.101 2.990
ethnonym -0.6420 57 3.175 2.889
island -0.6180 14 2.491 3.143
or after -0.6078 3 2.636 3.065
in the -0.6042 25 3.560 2.882
and the -0.5605 15 3.444 2.982
near -0.5494 16 3.131 3.037
they -0.5482 12 3.700 2.963
now -0.5284 4 2.532 3.073
polybios book -0.5118 5 1.219 3.148
polybios -0.5118 5 1.219 3.148
those who -0.5049 3 2.164 3.079
which -0.5011 20 3.727 2.883
also a city -0.4919 4 4.290 3.000

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.