Convert samples to messages schema

README.md

@@ -4,17 +4,17 @@ language:
 license: other
 task_categories:
 - text-generation
-pretty_name: Luau Stack HQ Pretraining
+pretty_name: Luau Stack HQ ChatML
 size_categories:
 - 10K<n<100K
 ---
 
-# Dataset Card for YCWTG/luau-stack-hq-pretraining
+# Dataset Card for YCWTG/luau-stack-hq-ChatML
 
 ## Summary
 
 This dataset is a pretraining-oriented repackaging of `khtsly/luau-stack-hq`.
-It keeps only the source code content and stores it in a single field named `text`.
+It keeps only the source code content and stores it in a ChatML-style `messages` field.
 
 - Rows: 21,704
 - File format: JSONL
@@ -26,7 +26,11 @@ It keeps only the source code content and stores it in a single field named `tex
 Each row has one field:
 
 ```json
-{"text": "<luau source code>"}
+{
+  "messages": [
+    {"role": "assistant", "content": "<luau source code>"}
+  ]
+}
 ```
 
 ## Processing
@@ -34,23 +38,24 @@ Each row has one field:
 The source file `/luau-stack.jsonl` was transformed with the following rule:
 
 1. Read each JSON line.
-2. Extract the original `code` field.
-3. Write a new JSON line with `{"text": code}`.
+2. Extract the original `code` field as `text`.
+3. Wrap `text` into `messages` as:
+   `{"messages": [{"role": "assistant", "content": text}]}`.
 4. Drop all other fields.
 
-No filtering, deduplication, or normalization was applied beyond field selection/renaming.
+No filtering, deduplication, or normalization was applied beyond field selection/renaming/wrapping.
 
 ## Intended Use
 
-This dataset is intended for code language model pretraining or continued pretraining where a unified `text` field is required.
+This dataset is intended for code language model pretraining or continued pretraining where a ChatML-style assistant message format is required.
 
 ## Load with `datasets`
 
 ```python
 from datasets import load_dataset
 
-ds = load_dataset("YCWTG/luau-stack-hq-pretraining", split="train")
-print(ds[0]["text"][:200])
+ds = load_dataset("YCWTG/luau-stack-hq-ChatML", split="train")
+print(ds[0]["messages"][0]["content"][:200])
 ```
 
 ## Source and Attribution

train.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae74729e38d6ff0d9969437c08ea24b0e1f3309ff9d77bc6b04642ba3a833418
-size 65541252
+oid sha256:e041b5e166e7e58bc3f094433e441a40bd11d548c50d1b4eacb9a204b58d79ee
+size 66322596