Upload 11 files

.gitattributes

@@ -59,3 +59,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
 HyperScholar-OmniPython-50K-CodeOnly.jsonl filter=lfs diff=lfs merge=lfs -text
 HyperScholar-OmniPython-50K-HyperReason.jsonl filter=lfs diff=lfs merge=lfs -text
+HyperScholar-OmniPython-50K-AutoTrain-Chat.jsonl filter=lfs diff=lfs merge=lfs -text

HyperScholar-OmniPython-50K-AutoTrain-Chat.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:520d6412a9e18acdd059c6deaef7ac9ca62145c23687f01aceec92551b735019
+size 67691405

README.md

@@ -24,6 +24,8 @@ configs:
   data_files: HyperScholar-OmniPython-50K-CodeOnly.jsonl
 - config_name: hyperreason
   data_files: HyperScholar-OmniPython-50K-HyperReason.jsonl
+- config_name: autotrain_chat
+  data_files: HyperScholar-OmniPython-50K-AutoTrain-Chat.jsonl
 ---
 
 # HyperScholar-OmniPython (50K) — WithIn Us AI
@@ -107,3 +109,8 @@ To train the structured “engineering discipline” behavior:
 
 Train a model on this dataset and publish it on Hugging Face, then add your result in `LEADERBOARD.md`
 using `MODEL_SUBMISSION_TEMPLATE.md`.
+
+
+## AutoTrain (no-code) ready
+
+Use config `autotrain_chat` in AutoTrain Advanced (LLM → SFT). Column mapping: `{ "text": "text" }` and set `chat_template=tokenizer`.