Upload dataset

README.md

@@ -65,12 +65,12 @@ and neural-network applications in Bashkir, Tatar, the Mari languages, the Osset
 dialects, and the other languages for which the data was collected.
 
 This dataset is the result of the project: native-speaker translations, validated with
-the **Claude Fable 5** model, which also produced the machine translations (`mt` rows)
-for the prompts no volunteer had covered.
+the **Claude Fable 5** model, which also produced the translations (`fable` rows) for
+the prompts no volunteer had covered.
 
 ## Languages
 
-| code | language | human | mt |
+| code | language | human | fable |
 |---|---|---|---|
 | ru | Russian (pivot) | — | — |
 | alt | Southern Altai | 124 | 0 |
@@ -109,9 +109,9 @@ and carries two metadata columns prefixed with its code:
 
 - `<code>` — the prompt in that language;
 - `<code>_source` — `human` (translated by a native-speaker volunteer on the SuperMinor
-  portal) or `mt` (no volunteer translation existed; rendered by the validation model);
+  portal) or `fable` (no volunteer translation existed; translated by Claude Fable 5);
 - `<code>_edited` — `true` if the text was touched during the validation pass (typo /
-  consistency / mistranslation fixes for human rows; always `true` for `mt` rows).
+  consistency / mistranslation fixes for human rows; always `true` for `fable` rows).
 
 When pristine native-speaker text is required, filter on `<code>_source == "human"`.
 

data/prompts.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4fa78343769cc9f80e1be5f6d43c75c57d479ab82dbd5d873f4e207cbe6b916
-size 101731
+oid sha256:d54a6b9efd0348ea454901a0e4fd5cbbb7c71c976a78caaec7442c964dc567aa
+size 101813