Update README.md

README.md

@@ -19,8 +19,12 @@ tags:
 
 Qwen2.5-Coder-7B-Instruct trained on a merged dataset of Unity3d q&a from these two datasets:
 [ibranze/codellama_unity3d_v2](https://huggingface.co/datasets/ibranze/codellama_unity3d_v2) (Full)
-[Hypersniper/unity_api_2022_3](https://huggingface.co/datasets/Hypersniper/unity_api_2022_3) (5%)
+[Hypersniper/unity_api_2022_3](https://huggingface.co/datasets/Hypersniper/unity_api_2022_3) (10%)
 
+preview 2:
+26210 rows, of which ca 1000 are from my own multi response dataset
+
+preview 1:
 15062 rows in total with a 10% validation split.
 
 Trained with native chat template (minus tools usage, see this issue: https://github.com/unslothai/unsloth/issues/1053). With a little superficial testing done, it seems to respond well to the mistral template.
@@ -42,7 +46,7 @@ This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unsloth
 
 # Training details
 
-About 1 epoch.
+About 1.5 epochs. It's probably a bit overfitting and I should introduce some general coding questions to my validation set to ensure it doesn't lose too much general performance.
 
 Rank: 128
 
@@ -74,15 +78,23 @@ TrainingArguments(
 
 
 Step 	Training Loss 	Validation Loss
-10 	2.097300 	1.165832
-20 	1.058100 	1.013441
-30 	0.898500 	0.969640
-40 	0.866600 	0.943687
-50 	0.847300 	0.926879
-60 	0.838200 	0.903914
-70 	0.797600 	0.888580
-80 	0.777700 	0.873389
-90 	0.793900 	0.859501
-100 	0.725500 	0.846339
-110 	0.739400 	0.843786
-120 	0.675200 	0.833775
+
+20 	2.043000 	1.197104
+
+40 	1.087300 	0.933553
+
+60 	0.942200 	0.890801
+
+80 	0.865600 	0.866198
+
+100 	0.851400 	0.849733
+
+120 	0.812900 	0.837039
+
+140 	0.812400 	0.827064
+
+160 	0.817300 	0.818410
+
+180 	0.802600 	0.810163
+
+200 	0.788600 	0.803399