gpt-4.1-2025-04-14 v3 Evaluation

Generated: 2026-09-20 11:20:05 UTC

Pairs100
Distinct lemmas100
Slope0.976
R^20.989
BLEU-443.7%
chrF++68.4%
METEOR75.8%
ROUGE-L75.9%
BERTScore95.6%
COMET78.7%
BLEURT73.0%

Profile version id: 2415

First translation using this version: 2026-07-16

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

Created: 2026-07-16 06:54:03.500756+10:00

Active: True

Notes: Model timeline evaluation: GPT-4.1 (2025-04-14) 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
Slope0.976
Intercept0.998
R^20.989
P-value3.23e-98
Pearson r0.995
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words44.110
Mean absolute length residual2.965
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 = -0.327, R^2 = 0.107, p = 0.0009.
chrF++ vs passage length
chrF++: significant negative; r = -0.373, R^2 = 0.139, p = 0.0001.
METEOR vs passage length
METEOR: significant negative; r = -0.481, R^2 = 0.232, p = 4.03e-07.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.488, R^2 = 0.238, p = 2.62e-07.
BERTScore vs passage length
BERTScore: significant negative; r = -0.633, R^2 = 0.401, p = 1.54e-12.
COMET vs passage length
COMET: significant negative; r = -0.424, R^2 = 0.180, p = 1.09e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.754, R^2 = 0.569, p = 1.35e-19.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.366, R^2 = 0.134, p = 0.0002.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.362, R^2 = 0.131, p = 0.0002.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.364, R^2 = 0.132, p = 0.0002.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.352, R^2 = 0.124, p = 0.0003.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.327 0.107 0.0009 -0.00206 ok
chrF++ 100 significant negative -0.373 0.139 0.0001 -0.00143 ok
METEOR 100 significant negative -0.481 0.232 4.03e-07 -0.00215 ok
ROUGE-L 100 significant negative -0.488 0.238 2.62e-07 -0.00206 ok
BERTScore 100 significant negative -0.633 0.401 1.54e-12 -0.00049 ok
COMET 100 significant negative -0.424 0.180 1.09e-05 -0.00082 ok
BLEURT 100 significant negative -0.754 0.569 1.35e-19 -0.00258 ok
Trigram precision 100 significant negative -0.366 0.134 0.0002 -0.00290 ok
Trigram recall 100 significant negative -0.362 0.131 0.0002 -0.00285 ok
Trigram F1 100 significant negative -0.364 0.132 0.0002 -0.00286 ok
Trigram Jaccard 100 significant negative -0.352 0.124 0.0003 -0.00290 ok

Translation Similarity Metrics

Mean BLEU-443.7%
Mean chrF++68.4%
Mean METEOR75.8%
Mean ROUGE-L75.9%
Mean BERTScore95.6%
Mean COMET78.7%
Mean BLEURT73.0%
Legacy smoothed corpus BLEU45.9%
Unigram F176.6%
Bigram F151.9%
Trigram precision38.2%
Trigram recall38.1%
Trigram F138.2%
Trigram Jaccard23.6%
4-gram F129.2%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2468 Καπετώλιον 138 124 -11.753 19.8% 47.7% 50.8% 55.0% 87.3% 64.6% 50.0% reviewed/approved 13274 gpt-4.1-2025-04-14
2604 Καρχηδών 117 127 11.753 37.8% 65.1% 68.0% 63.7% 93.9% 78.0% 59.9% reviewed/approved 13298 gpt-4.1-2025-04-14
2056 Καβειρία 112 100 -10.365 35.0% 67.8% 63.2% 69.8% 95.2% 78.7% 62.5% reviewed/approved 13142 gpt-4.1-2025-04-14
2626 Κατάνη 95 84 -9.764 35.1% 60.7% 64.1% 69.3% 93.9% 74.8% 63.7% reviewed/approved 13331 gpt-4.1-2025-04-14
3288 Κορώνεια 110 117 8.588 30.7% 63.1% 73.0% 70.2% 94.6% 76.1% 64.5% reviewed/approved 13346 gpt-4.1-2025-04-14
2455 Κάναστρον 60 51 -8.587 17.4% 49.1% 51.7% 57.7% 93.0% 71.4% 61.3% reviewed/approved 13268 gpt-4.1-2025-04-14
2342 Κάμιρος 54 62 8.272 41.8% 69.0% 70.0% 67.2% 93.9% 77.9% 64.5% reviewed/approved 13259 gpt-4.1-2025-04-14
7266 Κώμη 79 70 -8.141 18.6% 49.8% 48.1% 51.0% 93.3% 71.4% 57.4% reviewed/approved 13430 gpt-4.1-2025-04-14
2328 Καλλίπολις 43 51 8.013 29.6% 58.9% 54.4% 57.4% 93.2% 75.5% 70.2% reviewed/approved 13235 gpt-4.1-2025-04-14
2609 Κάσπειρος 135 125 -7.824 41.5% 67.3% 69.0% 71.5% 90.8% 74.2% 66.3% reviewed/approved 13307 gpt-4.1-2025-04-14
2470 Καππαδοκία 68 75 7.601 27.0% 62.9% 76.1% 74.1% 95.0% 76.7% 68.6% reviewed/approved 13277 gpt-4.1-2025-04-14
7260 Κυχρεῖος πάγος 61 68 7.436 44.4% 68.6% 73.5% 68.2% 95.6% 76.2% 59.1% reviewed/approved 13412 gpt-4.1-2025-04-14
2329 Κάλπη 57 50 -6.658 40.8% 64.8% 64.6% 70.4% 94.5% 72.7% 67.2% reviewed/approved 13238 gpt-4.1-2025-04-14
7259 Κύφος 54 60 6.272 46.2% 73.0% 82.8% 80.7% 96.3% 75.5% 75.6% reviewed/approved 13409 gpt-4.1-2025-04-14
7247 Κύρνος 50 44 -5.822 40.0% 65.6% 74.6% 66.0% 96.2% 77.5% 79.1% reviewed/approved 13373 gpt-4.1-2025-04-14
7248 Κύρου πόλις 32 38 5.754 23.6% 59.8% 71.0% 62.9% 93.1% 77.4% 79.6% reviewed/approved 13376 gpt-4.1-2025-04-14
2621 Κάστνιον 25 31 5.590 49.9% 75.7% 72.3% 75.0% 97.4% 81.3% 77.2% reviewed/approved 13313 gpt-4.1-2025-04-14
2602 Καρύανδα 34 29 -5.199 36.6% 65.9% 74.7% 69.8% 96.3% 78.5% 74.2% reviewed/approved 13292 gpt-4.1-2025-04-14
7244 Κυρήνη 54 49 -4.728 36.1% 64.5% 69.2% 69.9% 95.4% 77.0% 72.5% reviewed/approved 13364 gpt-4.1-2025-04-14
7252 Κύρτωνες 27 32 4.637 44.0% 71.0% 80.3% 74.6% 96.5% 74.6% 74.3% reviewed/approved 13388 gpt-4.1-2025-04-14
2055 Καβασσός 109 103 -4.435 27.9% 58.9% 54.0% 60.4% 93.5% 77.4% 51.2% reviewed/approved 13139 gpt-4.1-2025-04-14
2603 Κάρυστος 175 176 4.116 32.0% 59.4% 58.5% 56.8% 93.2% 75.2% 53.8% reviewed/approved 13295 gpt-4.1-2025-04-14
7254 Κύτα 84 87 3.977 38.4% 66.3% 70.6% 73.7% 93.1% 76.9% 64.1% reviewed/approved 13394 gpt-4.1-2025-04-14
2077 Καιρή 17 14 -3.598 47.8% 67.1% 67.4% 71.0% 96.9% 84.6% 76.9% reviewed/approved 13178 gpt-4.1-2025-04-14
2628 Καταονία 24 28 3.566 18.1% 53.4% 47.9% 61.5% 94.2% 77.0% 74.2% reviewed/approved 13337 gpt-4.1-2025-04-14
7251 Κύρτος 66 69 3.554 38.0% 63.2% 79.2% 71.1% 93.5% 70.6% 66.9% reviewed/approved 13385 gpt-4.1-2025-04-14
2115 Καλὴ ἀκτή 64 67 3.507 20.3% 46.8% 62.4% 58.0% 92.0% 66.9% 52.1% reviewed/approved 13211 gpt-4.1-2025-04-14
2059 Καδμεία 20 24 3.472 17.8% 57.8% 68.9% 72.7% 95.7% 80.7% 74.4% reviewed/approved 13151 gpt-4.1-2025-04-14
2802 Κάδρεμα 24 21 -3.434 48.8% 67.1% 65.7% 71.1% 96.6% 75.5% 67.1% reviewed/approved 13160 gpt-4.1-2025-04-14
2114 Κάλβιος 18 22 3.425 24.4% 53.6% 75.3% 65.0% 93.5% 72.1% 70.9% reviewed/approved 13208 gpt-4.1-2025-04-14

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: -0.150.

Positive termCoefficientDocsMean presentMean absent
δε και 1.1697 8 4.697 -0.408
πολιτης 1.0810 18 3.100 -0.681
απο 1.0621 27 1.672 -0.618
οι 0.8064 11 2.734 -0.338
εστι δε και 0.7628 3 7.785 -0.241
παιδος 0.7465 4 4.566 -0.190
ος 0.7368 6 3.281 -0.209
εις 0.7131 7 2.403 -0.181
εκ 0.7098 6 3.195 -0.204
τη 0.6533 15 1.831 -0.323
εν τη 0.6480 7 2.991 -0.225
πολις 0.6321 68 0.415 -0.882
εστι δε 0.6189 5 3.299 -0.174
πολις προς 0.6116 3 4.424 -0.137
λεγεται 0.5989 9 2.788 -0.276
την 0.5675 13 0.330 -0.049
δυο 0.5631 4 3.642 -0.152
εκαλειτο δε 0.5519 3 6.073 -0.188
λεγεται και 0.5298 8 2.760 -0.240
ου και 0.5289 4 3.750 -0.156
Negative termCoefficientDocsMean presentMean absent
τα -1.2987 12 -3.351 0.457
γαρ -1.0697 9 -2.085 0.206
το -0.9521 70 -0.416 0.971
εκαταιος -0.6830 14 -1.284 0.209
θεοπομπος -0.6569 5 -2.284 0.120
το εθνικον -0.5967 56 -0.809 1.030
εθνικον -0.5967 56 -0.809 1.030
μετα -0.5651 3 -5.079 0.157
τε -0.5642 5 -0.868 0.046
θρακης -0.5561 5 -3.032 0.160
ουδετερως -0.5229 3 -4.054 0.125
ει -0.5064 4 -2.962 0.123
ουτως -0.5031 5 -2.869 0.151
εθνος -0.4917 8 -1.420 0.123
οικητωρ -0.4349 6 -1.801 0.115
ειναι -0.4282 8 -1.742 0.151
πλησιον -0.4141 8 -0.816 0.071
ποταμου -0.4061 3 -1.716 0.053
δε της -0.3979 6 -1.986 0.127
αλλ -0.3819 3 -2.937 0.091

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 2.031; cross-validated R^2: 0.045.

Positive termCoefficientDocsMean presentMean absent
και 2.6705 66 3.512 1.903
δε 2.5051 46 4.169 1.939
την 1.8884 13 6.037 2.506
γαρ 1.3049 9 5.185 2.746
ος 1.2189 6 7.199 2.695
δε και 1.1200 8 5.467 2.748
αλλ 1.0669 3 8.004 2.809
βοιωτιας 1.0041 3 7.863 2.814
εστι δε 0.9806 5 7.445 2.729
τε 0.9484 5 5.518 2.831
τα 0.9481 12 4.784 2.717
εις 0.9153 7 5.761 2.755
αλλη 0.9119 4 7.556 2.774
της 0.8558 35 3.846 2.491
εστι δε και 0.8269 3 7.785 2.816
και αλλη 0.8146 3 8.498 2.794
δε της 0.7644 6 6.665 2.729
μετα 0.7608 3 6.646 2.851
τριτη 0.7358 3 6.561 2.854
αι 0.6998 3 7.145 2.836
Negative termCoefficientDocsMean presentMean absent
και το -1.4379 9 1.642 3.096
στραβων -1.1255 15 2.198 3.101
θηλυκον -1.0657 12 2.900 2.974
και πολις -0.9518 5 1.393 3.048
εστι και -0.9334 20 2.965 2.965
δε απο -0.8102 3 0.777 3.033
μια -0.7962 3 0.554 3.040
οικητωρ -0.7950 6 2.585 2.989
τεταρτω -0.7658 3 1.678 3.005
το εθνικον -0.6958 56 3.030 2.883
εθνικον -0.6958 56 3.030 2.883
τοπος -0.6825 4 1.790 3.014
πορρω -0.6659 3 0.364 3.046
ου πορρω -0.6659 3 0.364 3.046
ακρα -0.6165 4 1.350 3.033
προς τω -0.5971 4 1.873 3.011
προς τη -0.5767 4 1.982 3.006
τους -0.5666 4 2.724 2.975
αυτην φησιν -0.5657 3 1.601 3.007
εκαταιος ασια -0.5522 5 1.232 3.056

AI English terms predicting large absolute residuals

Features: 341; ridge alpha: 2.031; cross-validated R^2: -0.037.

Positive termCoefficientDocsMean presentMean absent
and 2.9172 54 3.848 1.929
for 1.6225 14 5.030 2.629
the 1.5229 94 3.049 1.658
on 1.1286 15 4.861 2.631
among 1.0882 8 5.385 2.755
iota 1.0799 5 5.463 2.834
of 1.0599 62 3.339 2.355
from 1.0421 45 3.824 2.263
was 1.0277 21 4.491 2.560
these 0.9996 5 7.245 2.740
renowned 0.9970 4 8.336 2.741
who was 0.9681 3 9.154 2.774
boiotia 0.9220 3 7.863 2.814
to 0.8874 21 4.198 2.637
bithynia 0.8620 3 8.808 2.785
it 0.8345 39 3.951 2.335
his europe 0.8341 3 8.058 2.808
in his europe 0.8341 3 8.058 2.808
also 0.8226 43 3.785 2.347
is a city 0.8181 4 5.497 2.860
Negative termCoefficientDocsMean presentMean absent
strabo book -0.7321 8 1.221 3.117
book -0.6916 40 2.499 3.276
the feminine -0.6838 12 2.652 3.008
strabo -0.6662 15 2.198 3.101
the son of -0.6131 6 1.595 3.053
its -0.6065 4 1.453 3.028
one of -0.5769 3 0.765 3.033
a city in -0.5560 43 2.786 3.100
in his asia -0.5530 4 1.087 3.043
his asia -0.5530 4 1.087 3.043
not far from -0.5408 3 0.364 3.046
not far -0.5408 3 0.364 3.046
far from -0.5408 3 0.364 3.046
far -0.5408 3 0.364 3.046
was also -0.5375 5 1.901 3.021
syria -0.5227 3 0.556 3.040
in syria -0.5227 3 0.556 3.040
sea -0.5180 7 2.860 2.973
just as -0.5035 11 2.494 3.023
just -0.5035 11 2.494 3.023

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 1.425; cross-validated R^2: 0.059.

Positive termCoefficientDocsMean presentMean absent
and 3.8293 52 3.950 1.898
the 2.2711 90 3.106 1.700
was 1.8816 20 4.694 2.533
with 1.7349 16 4.172 2.735
for 1.5381 11 5.584 2.642
at 1.3909 6 7.142 2.699
on 1.3454 16 4.003 2.768
from 1.2840 37 3.608 2.588
of 1.2612 66 3.467 1.992
this 1.2408 15 5.702 2.482
for the 1.1886 7 6.065 2.732
forms 1.1682 4 8.017 2.755
were 1.1406 4 7.050 2.795
has 1.0884 6 6.555 2.736
boiotia 1.0850 3 7.863 2.814
all 1.0231 3 8.142 2.805
also 0.9732 43 3.662 2.440
too 0.9659 3 7.679 2.819
are 0.9123 25 4.146 2.571
on the 0.9111 10 4.485 2.796
Negative termCoefficientDocsMean presentMean absent
strabo -1.0041 15 2.198 3.101
book -0.9796 39 2.349 3.359
strabo book -0.8640 10 1.788 3.096
the ethnonym is -0.7493 52 2.953 2.978
ethnonym is -0.7493 52 2.953 2.978
from the -0.7445 13 2.725 3.001
the possessive -0.7273 6 1.796 3.040
the ethnonym -0.6714 56 3.067 2.835
to be -0.6670 9 3.153 2.947
one of -0.6647 5 2.316 2.999
ethnonym -0.6530 57 3.091 2.798
place -0.6396 6 3.217 2.949
in his asia -0.6171 5 1.232 3.056
his asia -0.6171 5 1.232 3.056
is the -0.6073 3 2.078 2.993
not far from -0.6070 3 0.364 3.046
not far -0.6070 3 0.364 3.046
far from -0.6070 3 0.364 3.046
far -0.6070 3 0.364 3.046
also a city -0.5986 4 1.914 3.009

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.