Update README.md

README.md

@@ -5,4 +5,8 @@ language:
 base_model:
 - mistralai/Mistral-Small-3.2-24B-Instruct-2506
 ---
-Trained on https://huggingface.co/datasets/Nitral-AI/antirep_sharegpt for 2 epochs at 3e-5
+# Trained on https://huggingface.co/datasets/Nitral-AI/antirep_sharegpt for 2 epochs at 3e-5, 4bit qlora. at rank/alpha 32.
+
+## Original set was created by: "Preference dataset meant to decrease repetition, measured as either copying n-grams from input or infinite / semi-infite repetition of tokens; the chosen split is V3 03/24 instructed to avoid n-gram repetition, while the rejected split consists of either V3 03/24 instructed to copy from the input or Qwen 3 8B with a rep pen of 0.7."
+
+## just took the preference rows, removed the rejected, converted to sharegpt jsonl to create the sft version of the set used above.