Add files using upload-large-folder tool

.gitattributes

@@ -382,3 +382,94 @@ productivity_original_Swedish/Qwen3-4B-Base_productivity_splits_original_feature
 productivity_original_Swedish/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-4160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 productivity_original_Swedish/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-832/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-1266/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-1688/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-2110/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-2532/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-2954/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-3376/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-3798/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-422/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-4220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test2/checkpoint-844/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-1266/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-1688/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-2110/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-2532/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-2954/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-3376/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-3798/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-422/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-4220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_original_Estonian/Qwen3-4B-Base_productivity_splits_original_features_train_productivity_splits_original_features_test1/checkpoint-844/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-1248/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-1664/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2080/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2496/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2912/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-3328/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-3744/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-416/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-4160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-1248/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-832/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-1664/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2080/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2496/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2912/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-3328/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-3744/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-416/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-4160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Swedish/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-832/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-1266/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-1688/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2110/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2532/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-2954/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-3376/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-3798/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-422/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-4220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test2/checkpoint-844/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-1266/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-1688/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2110/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2532/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-2954/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-3376/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-3798/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-422/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-4220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+productivity_code_Estonian/Qwen3-4B-Base_productivity_splits_code_features_train_productivity_splits_code_features_test1/checkpoint-844/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1040/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1020/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1060/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1080/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1140/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1180/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1240/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1260/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1280/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1320/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1340/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1360/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1380/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-140/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1420/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1440/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1460/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1480/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1520/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1540/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+overgeneralisation_original_Swedish/Qwen3-4B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test2/checkpoint-1560/tokenizer.json filter=lfs diff=lfs merge=lfs -text

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3465/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3465/trainer_state.json

@@ -0,0 +1,823 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 3465,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2479030179977415,
+      "epoch": 0.1299545159194282,
+      "grad_norm": 1.519571304321289,
+      "learning_rate": 3.522207847653314e-05,
+      "loss": 2.093206329345703,
+      "mean_token_accuracy": 0.6068353663384914,
+      "num_tokens": 154518.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.932415626347065,
+      "epoch": 0.2599090318388564,
+      "grad_norm": 1.180830955505371,
+      "learning_rate": 7.11629748811588e-05,
+      "loss": 0.8930854797363281,
+      "mean_token_accuracy": 0.7708445385098457,
+      "num_tokens": 306733.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.7730373838543891,
+      "epoch": 0.3898635477582846,
+      "grad_norm": 0.7839977145195007,
+      "learning_rate": 0.00010710387128578447,
+      "loss": 0.7302116394042969,
+      "mean_token_accuracy": 0.8012136635184288,
+      "num_tokens": 446267.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6934178560972214,
+      "epoch": 0.5198180636777128,
+      "grad_norm": 0.666778564453125,
+      "learning_rate": 0.0001430447676904101,
+      "loss": 0.6505754852294922,
+      "mean_token_accuracy": 0.8195212116837501,
+      "num_tokens": 600256.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.6900296103954315,
+      "epoch": 0.649772579597141,
+      "grad_norm": 0.6762415766716003,
+      "learning_rate": 0.00017898566409503577,
+      "loss": 0.6378536987304687,
+      "mean_token_accuracy": 0.8223087686300278,
+      "num_tokens": 738649.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.667421719878912,
+      "epoch": 0.7797270955165692,
+      "grad_norm": 0.5047685503959656,
+      "learning_rate": 0.00021492656049966144,
+      "loss": 0.6148524856567383,
+      "mean_token_accuracy": 0.8280292323231697,
+      "num_tokens": 883494.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.6388977643847465,
+      "epoch": 0.9096816114359974,
+      "grad_norm": 0.4360353350639343,
+      "learning_rate": 0.0002508674569042871,
+      "loss": 0.5933729553222656,
+      "mean_token_accuracy": 0.8329134130477905,
+      "num_tokens": 1032111.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.6612381511009656,
+      "eval_loss": 0.6559221744537354,
+      "eval_mean_token_accuracy": 0.8195324820967821,
+      "eval_num_tokens": 1132140.0,
+      "eval_runtime": 53.4007,
+      "eval_samples_per_second": 31.03,
+      "eval_steps_per_second": 3.895,
+      "step": 385
+    },
+    {
+      "entropy": 0.6224366770916848,
+      "epoch": 1.0389863547758285,
+      "grad_norm": 0.5294668078422546,
+      "learning_rate": 0.00027673375518355765,
+      "loss": 0.5677951431274414,
+      "mean_token_accuracy": 0.8380465067211708,
+      "num_tokens": 1177556.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5827466724812984,
+      "epoch": 1.1689408706952567,
+      "grad_norm": 0.5172416567802429,
+      "learning_rate": 0.0002765120122346144,
+      "loss": 0.5423126983642578,
+      "mean_token_accuracy": 0.8467991036176682,
+      "num_tokens": 1325434.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5831517253816128,
+      "epoch": 1.2988953866146848,
+      "grad_norm": 0.41916292905807495,
+      "learning_rate": 0.0002760064270819138,
+      "loss": 0.534448013305664,
+      "mean_token_accuracy": 0.8456632816791534,
+      "num_tokens": 1474116.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5869986982643605,
+      "epoch": 1.428849902534113,
+      "grad_norm": 0.4387759566307068,
+      "learning_rate": 0.00027521803857633113,
+      "loss": 0.5367491912841796,
+      "mean_token_accuracy": 0.8462416216731071,
+      "num_tokens": 1621193.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5771756853163242,
+      "epoch": 1.5588044184535412,
+      "grad_norm": 0.49079665541648865,
+      "learning_rate": 0.00027414846665880935,
+      "loss": 0.5238623809814453,
+      "mean_token_accuracy": 0.84760089635849,
+      "num_tokens": 1767789.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5549105909466744,
+      "epoch": 1.6887589343729694,
+      "grad_norm": 0.4000363051891327,
+      "learning_rate": 0.0002727999090317863,
+      "loss": 0.510434226989746,
+      "mean_token_accuracy": 0.8517858856916427,
+      "num_tokens": 1918138.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.583413660377264,
+      "epoch": 1.8187134502923976,
+      "grad_norm": 0.33592426776885986,
+      "learning_rate": 0.00027117513664346674,
+      "loss": 0.5297993850708008,
+      "mean_token_accuracy": 0.846615691781044,
+      "num_tokens": 2057575.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.5732646904885769,
+      "epoch": 1.9486679662118258,
+      "grad_norm": 0.5528839230537415,
+      "learning_rate": 0.00026927748799421714,
+      "loss": 0.5219194793701172,
+      "mean_token_accuracy": 0.8489033079147339,
+      "num_tokens": 2208320.0,
+      "step": 750
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.6027900550801021,
+      "eval_loss": 0.5946928858757019,
+      "eval_mean_token_accuracy": 0.8318195798649237,
+      "eval_num_tokens": 2264280.0,
+      "eval_runtime": 53.3837,
+      "eval_samples_per_second": 31.039,
+      "eval_steps_per_second": 3.896,
+      "step": 770
+    },
+    {
+      "entropy": 0.5329899657611272,
+      "epoch": 2.077972709551657,
+      "grad_norm": 0.45793575048446655,
+      "learning_rate": 0.0002671108622767842,
+      "loss": 0.48420516967773436,
+      "mean_token_accuracy": 0.8578200301333289,
+      "num_tokens": 2348248.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.5142687204480171,
+      "epoch": 2.207927225471085,
+      "grad_norm": 0.4690960645675659,
+      "learning_rate": 0.0002646797113644295,
+      "loss": 0.4593114471435547,
+      "mean_token_accuracy": 0.8622670090198516,
+      "num_tokens": 2501427.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.5135884112119675,
+      "epoch": 2.3378817413905133,
+      "grad_norm": 0.3752821683883667,
+      "learning_rate": 0.00026198903066344565,
+      "loss": 0.4626216125488281,
+      "mean_token_accuracy": 0.8612511262297631,
+      "num_tokens": 2650794.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.5137367483973503,
+      "epoch": 2.4678362573099415,
+      "grad_norm": 0.3726271390914917,
+      "learning_rate": 0.0002590443488488465,
+      "loss": 0.4601683807373047,
+      "mean_token_accuracy": 0.8620512077212333,
+      "num_tokens": 2798180.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.5105714881420136,
+      "epoch": 2.5977907732293697,
+      "grad_norm": 0.41296717524528503,
+      "learning_rate": 0.00025585171650432525,
+      "loss": 0.46279102325439453,
+      "mean_token_accuracy": 0.8611763519048691,
+      "num_tokens": 2950301.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5169161760807037,
+      "epoch": 2.727745289148798,
+      "grad_norm": 0.4614253044128418,
+      "learning_rate": 0.0002524176936898197,
+      "loss": 0.45492774963378907,
+      "mean_token_accuracy": 0.8627680170536042,
+      "num_tokens": 3091810.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4989277676492929,
+      "epoch": 2.857699805068226,
+      "grad_norm": 0.37512704730033875,
+      "learning_rate": 0.00024874933646223225,
+      "loss": 0.4531984329223633,
+      "mean_token_accuracy": 0.8637665447592735,
+      "num_tokens": 3242184.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.5177617704868317,
+      "epoch": 2.9876543209876543,
+      "grad_norm": 0.3700532019138336,
+      "learning_rate": 0.00024485418237699976,
+      "loss": 0.45844474792480466,
+      "mean_token_accuracy": 0.8626988258957863,
+      "num_tokens": 3382605.0,
+      "step": 1150
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.5253989950108987,
+      "eval_loss": 0.5857328176498413,
+      "eval_mean_token_accuracy": 0.8360884573597175,
+      "eval_num_tokens": 3396420.0,
+      "eval_runtime": 53.341,
+      "eval_samples_per_second": 31.064,
+      "eval_steps_per_second": 3.899,
+      "step": 1155
+    },
+    {
+      "entropy": 0.4535361140517134,
+      "epoch": 3.116959064327485,
+      "grad_norm": 0.3412795662879944,
+      "learning_rate": 0.00024074023500030492,
+      "loss": 0.3942829132080078,
+      "mean_token_accuracy": 0.8781378038564519,
+      "num_tokens": 3522582.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.44747897461056707,
+      "epoch": 3.246913580246914,
+      "grad_norm": 0.46647050976753235,
+      "learning_rate": 0.0002364159474637521,
+      "loss": 0.38986759185791015,
+      "mean_token_accuracy": 0.8777281475067139,
+      "num_tokens": 3670864.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.4480265176296234,
+      "epoch": 3.3768680961663415,
+      "grad_norm": 0.4068582355976105,
+      "learning_rate": 0.00023189020509529866,
+      "loss": 0.39444759368896487,
+      "mean_token_accuracy": 0.8774515727162361,
+      "num_tokens": 3822021.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.45180007234215736,
+      "epoch": 3.50682261208577,
+      "grad_norm": 0.4249928593635559,
+      "learning_rate": 0.00022717230716213122,
+      "loss": 0.3977077102661133,
+      "mean_token_accuracy": 0.8762744688987731,
+      "num_tokens": 3968736.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4614932192862034,
+      "epoch": 3.636777128005198,
+      "grad_norm": 0.561008095741272,
+      "learning_rate": 0.00022227194776300045,
+      "loss": 0.4022808456420898,
+      "mean_token_accuracy": 0.8760285252332687,
+      "num_tokens": 4113509.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.4414680179953575,
+      "epoch": 3.7667316439246266,
+      "grad_norm": 0.38943538069725037,
+      "learning_rate": 0.00021719919590927584,
+      "loss": 0.38586376190185545,
+      "mean_token_accuracy": 0.8783121705055237,
+      "num_tokens": 4267958.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.45685607343912127,
+      "epoch": 3.8966861598440543,
+      "grad_norm": 0.5362406969070435,
+      "learning_rate": 0.00021196447483564875,
+      "loss": 0.3983576583862305,
+      "mean_token_accuracy": 0.8764419692754746,
+      "num_tokens": 4415398.0,
+      "step": 1500
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.49981127894268584,
+      "eval_loss": 0.5997208952903748,
+      "eval_mean_token_accuracy": 0.8368159819107789,
+      "eval_num_tokens": 4528560.0,
+      "eval_runtime": 53.4304,
+      "eval_samples_per_second": 31.012,
+      "eval_steps_per_second": 3.893,
+      "step": 1540
+    },
+    {
+      "entropy": 0.4439909208060509,
+      "epoch": 4.025990903183885,
+      "grad_norm": 0.5490113496780396,
+      "learning_rate": 0.00020657854058299564,
+      "loss": 0.38307292938232423,
+      "mean_token_accuracy": 0.8795150506436525,
+      "num_tokens": 4559534.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3837250977009535,
+      "epoch": 4.155945419103314,
+      "grad_norm": 0.5567234754562378,
+      "learning_rate": 0.0002010524598974076,
+      "loss": 0.3182963752746582,
+      "mean_token_accuracy": 0.8964017608761787,
+      "num_tokens": 4707075.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.377872094810009,
+      "epoch": 4.2858999350227425,
+      "grad_norm": 0.4315710961818695,
+      "learning_rate": 0.00019539758749079845,
+      "loss": 0.318333683013916,
+      "mean_token_accuracy": 0.8963816618919372,
+      "num_tokens": 4851683.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.38739304527640345,
+      "epoch": 4.41585445094217,
+      "grad_norm": 0.49140632152557373,
+      "learning_rate": 0.00018962554270981555,
+      "loss": 0.32688804626464846,
+      "mean_token_accuracy": 0.8937860554456711,
+      "num_tokens": 4994086.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.39157475270330905,
+      "epoch": 4.545808966861598,
+      "grad_norm": 0.40667369961738586,
+      "learning_rate": 0.00018374818566099208,
+      "loss": 0.3305763626098633,
+      "mean_token_accuracy": 0.8916732975840569,
+      "num_tokens": 5137171.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.3838599680364132,
+      "epoch": 4.675763482781027,
+      "grad_norm": 0.4632417857646942,
+      "learning_rate": 0.0001777775928411983,
+      "loss": 0.3267818450927734,
+      "mean_token_accuracy": 0.8946500706672669,
+      "num_tokens": 5287076.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.38270787581801413,
+      "epoch": 4.805717998700455,
+      "grad_norm": 0.5529720187187195,
+      "learning_rate": 0.0001717260323234649,
+      "loss": 0.3264235305786133,
+      "mean_token_accuracy": 0.8948800846934318,
+      "num_tokens": 5436923.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.38736109718680384,
+      "epoch": 4.935672514619883,
+      "grad_norm": 0.5604785680770874,
+      "learning_rate": 0.00016560593854916497,
+      "loss": 0.3280513381958008,
+      "mean_token_accuracy": 0.8931388029456139,
+      "num_tokens": 5589195.0,
+      "step": 1900
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.4370518812479881,
+      "eval_loss": 0.6210553050041199,
+      "eval_mean_token_accuracy": 0.8379779781859654,
+      "eval_num_tokens": 5660700.0,
+      "eval_runtime": 53.4039,
+      "eval_samples_per_second": 31.028,
+      "eval_steps_per_second": 3.895,
+      "step": 1925
+    },
+    {
+      "entropy": 0.3330705868988181,
+      "epoch": 5.064977257959714,
+      "grad_norm": 0.5386723875999451,
+      "learning_rate": 0.0001594298867783512,
+      "loss": 0.2754818344116211,
+      "mean_token_accuracy": 0.9101201346771202,
+      "num_tokens": 5739445.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.2980620255321264,
+      "epoch": 5.1949317738791425,
+      "grad_norm": 0.5633581876754761,
+      "learning_rate": 0.00015321056725074549,
+      "loss": 0.23754241943359375,
+      "mean_token_accuracy": 0.9203532826900482,
+      "num_tokens": 5888043.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.3111592583358288,
+      "epoch": 5.32488628979857,
+      "grad_norm": 0.5031015872955322,
+      "learning_rate": 0.0001469607591104745,
+      "loss": 0.24428102493286133,
+      "mean_token_accuracy": 0.917181601524353,
+      "num_tokens": 6031284.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.31522042460739613,
+      "epoch": 5.454840805717999,
+      "grad_norm": 0.6432453393936157,
+      "learning_rate": 0.0001406933041481286,
+      "loss": 0.25112478256225584,
+      "mean_token_accuracy": 0.9152472382783889,
+      "num_tokens": 6179927.0,
+      "step": 2100
+    },
+    {
+      "entropy": 0.3046229027956724,
+      "epoch": 5.584795321637427,
+      "grad_norm": 0.5104537606239319,
+      "learning_rate": 0.00013442108041409814,
+      "loss": 0.2431495475769043,
+      "mean_token_accuracy": 0.917829519212246,
+      "num_tokens": 6322630.0,
+      "step": 2150
+    },
+    {
+      "entropy": 0.30440517760813235,
+      "epoch": 5.714749837556855,
+      "grad_norm": 0.5307765603065491,
+      "learning_rate": 0.0001281569757574053,
+      "loss": 0.24610313415527343,
+      "mean_token_accuracy": 0.9166415151953697,
+      "num_tokens": 6469843.0,
+      "step": 2200
+    },
+    {
+      "entropy": 0.304359400421381,
+      "epoch": 5.844704353476283,
+      "grad_norm": 0.5014523267745972,
+      "learning_rate": 0.00012191386134440133,
+      "loss": 0.24548973083496095,
+      "mean_token_accuracy": 0.9165477308630944,
+      "num_tokens": 6617768.0,
+      "step": 2250
+    },
+    {
+      "entropy": 0.3141418205201626,
+      "epoch": 5.974658869395712,
+      "grad_norm": 0.567398726940155,
+      "learning_rate": 0.00011570456521174339,
+      "loss": 0.24975168228149414,
+      "mean_token_accuracy": 0.9139353120326996,
+      "num_tokens": 6761187.0,
+      "step": 2300
+    },
+    {
+      "epoch": 6.0,
+      "eval_entropy": 0.3758909202252443,
+      "eval_loss": 0.6843022108078003,
+      "eval_mean_token_accuracy": 0.8348075449466705,
+      "eval_num_tokens": 6792840.0,
+      "eval_runtime": 53.3825,
+      "eval_samples_per_second": 31.04,
+      "eval_steps_per_second": 3.896,
+      "step": 2310
+    },
+    {
+      "entropy": 0.24346149432000203,
+      "epoch": 6.1039636127355426,
+      "grad_norm": 0.7140825986862183,
+      "learning_rate": 0.00010954184590799172,
+      "loss": 0.17231273651123047,
+      "mean_token_accuracy": 0.9407721275660261,
+      "num_tokens": 6909578.0,
+      "step": 2350
+    },
+    {
+      "entropy": 0.2160973483324051,
+      "epoch": 6.23391812865497,
+      "grad_norm": 0.49014952778816223,
+      "learning_rate": 0.00010343836627798716,
+      "loss": 0.15455107688903807,
+      "mean_token_accuracy": 0.9467655989527702,
+      "num_tokens": 7056244.0,
+      "step": 2400
+    },
+    {
+      "entropy": 0.21491924367845058,
+      "epoch": 6.363872644574399,
+      "grad_norm": 0.5529471635818481,
+      "learning_rate": 9.740666744387656e-05,
+      "loss": 0.1584029197692871,
+      "mean_token_accuracy": 0.9460993978381157,
+      "num_tokens": 7206950.0,
+      "step": 2450
+    },
+    {
+      "entropy": 0.22037068914622068,
+      "epoch": 6.493827160493828,
+      "grad_norm": 0.6232843995094299,
+      "learning_rate": 9.145914303624717e-05,
+      "loss": 0.15544342041015624,
+      "mean_token_accuracy": 0.9450622496008872,
+      "num_tokens": 7359429.0,
+      "step": 2500
+    },
+    {
+      "entropy": 0.2320463878661394,
+      "epoch": 6.623781676413255,
+      "grad_norm": 0.7459681630134583,
+      "learning_rate": 8.560801372831975e-05,
+      "loss": 0.16350215911865235,
+      "mean_token_accuracy": 0.9416968420147895,
+      "num_tokens": 7499281.0,
+      "step": 2550
+    },
+    {
+      "entropy": 0.22943626195192338,
+      "epoch": 6.753736192332683,
+      "grad_norm": 0.7482302784919739,
+      "learning_rate": 7.986530212552506e-05,
+      "loss": 0.16422538757324218,
+      "mean_token_accuracy": 0.9434959614276885,
+      "num_tokens": 7640758.0,
+      "step": 2600
+    },
+    {
+      "entropy": 0.21795938543975354,
+      "epoch": 6.883690708252112,
+      "grad_norm": 0.5210486054420471,
+      "learning_rate": 7.424280806206118e-05,
+      "loss": 0.15540474891662598,
+      "mean_token_accuracy": 0.9459306105971337,
+      "num_tokens": 7791986.0,
+      "step": 2650
+    },
+    {
+      "epoch": 7.0,
+      "eval_entropy": 0.3157534837149657,
+      "eval_loss": 0.7744874954223633,
+      "eval_mean_token_accuracy": 0.8346395036922052,
+      "eval_num_tokens": 7924980.0,
+      "eval_runtime": 53.3771,
+      "eval_samples_per_second": 31.043,
+      "eval_steps_per_second": 3.897,
+      "step": 2695
+    },
+    {
+      "entropy": 0.2124774060656677,
+      "epoch": 7.012995451591943,
+      "grad_norm": 0.42238789796829224,
+      "learning_rate": 6.875208435518865e-05,
+      "loss": 0.14792531967163086,
+      "mean_token_accuracy": 0.9490461115861059,
+      "num_tokens": 7940521.0,
+      "step": 2700
+    },
+    {
+      "entropy": 0.15692965138703585,
+      "epoch": 7.142949967511371,
+      "grad_norm": 0.4711572229862213,
+      "learning_rate": 6.340441306708468e-05,
+      "loss": 0.09051708221435546,
+      "mean_token_accuracy": 0.9700166273117066,
+      "num_tokens": 8084193.0,
+      "step": 2750
+    },
+    {
+      "entropy": 0.15241683423519134,
+      "epoch": 7.272904483430799,
+      "grad_norm": 0.4312196671962738,
+      "learning_rate": 5.821078232303016e-05,
+      "loss": 0.08812363624572754,
+      "mean_token_accuracy": 0.9699361199140548,
+      "num_tokens": 8230159.0,
+      "step": 2800
+    },
+    {
+      "entropy": 0.1459079357981682,
+      "epoch": 7.402858999350228,
+      "grad_norm": 0.4804084002971649,
+      "learning_rate": 5.3181863733564636e-05,
+      "loss": 0.08675944328308105,
+      "mean_token_accuracy": 0.9703072866797448,
+      "num_tokens": 8380556.0,
+      "step": 2850
+    },
+    {
+      "entropy": 0.15621621005237102,
+      "epoch": 7.532813515269655,
+      "grad_norm": 0.5435478091239929,
+      "learning_rate": 4.83279904669986e-05,
+      "loss": 0.09016354560852051,
+      "mean_token_accuracy": 0.9674961140751839,
+      "num_tokens": 8523248.0,
+      "step": 2900
+    },
+    {
+      "entropy": 0.1545175113901496,
+      "epoch": 7.662768031189084,
+      "grad_norm": 0.524286687374115,
+      "learning_rate": 4.365913601734056e-05,
+      "loss": 0.09049141883850098,
+      "mean_token_accuracy": 0.9679373624920845,
+      "num_tokens": 8672002.0,
+      "step": 2950
+    },
+    {
+      "entropy": 0.1553485019877553,
+      "epoch": 7.792722547108512,
+      "grad_norm": 0.5006484389305115,
+      "learning_rate": 3.9184893711264495e-05,
+      "loss": 0.08913107872009278,
+      "mean_token_accuracy": 0.9684559822082519,
+      "num_tokens": 8816090.0,
+      "step": 3000
+    },
+    {
+      "entropy": 0.15444331549108029,
+      "epoch": 7.92267706302794,
+      "grad_norm": 0.5613893866539001,
+      "learning_rate": 3.491445699622611e-05,
+      "loss": 0.08711207389831543,
+      "mean_token_accuracy": 0.9684525722265244,
+      "num_tokens": 8966004.0,
+      "step": 3050
+    },
+    {
+      "epoch": 8.0,
+      "eval_entropy": 0.26580205430778175,
+      "eval_loss": 0.8996392488479614,
+      "eval_mean_token_accuracy": 0.8319417791297803,
+      "eval_num_tokens": 9057120.0,
+      "eval_runtime": 53.4042,
+      "eval_samples_per_second": 31.028,
+      "eval_steps_per_second": 3.895,
+      "step": 3080
+    },
+    {
+      "entropy": 0.14060429509860187,
+      "epoch": 8.05198180636777,
+      "grad_norm": 0.25378674268722534,
+      "learning_rate": 3.085660055023035e-05,
+      "loss": 0.07468742847442628,
+      "mean_token_accuracy": 0.9737296260181983,
+      "num_tokens": 9115929.0,
+      "step": 3100
+    },
+    {
+      "entropy": 0.12152639016509056,
+      "epoch": 8.1819363222872,
+      "grad_norm": 0.3634810447692871,
+      "learning_rate": 2.7019662252065798e-05,
+      "loss": 0.05918361663818359,
+      "mean_token_accuracy": 0.9785374769568443,
+      "num_tokens": 9260646.0,
+      "step": 3150
+    },
+    {
+      "entropy": 0.1228457903675735,
+      "epoch": 8.311890838206628,
+      "grad_norm": 0.32873299717903137,
+      "learning_rate": 2.3411526049051643e-05,
+      "loss": 0.060924801826477054,
+      "mean_token_accuracy": 0.9778030979633331,
+      "num_tokens": 9404737.0,
+      "step": 3200
+    },
+    {
+      "entropy": 0.11790491977706552,
+      "epoch": 8.441845354126055,
+      "grad_norm": 0.27871423959732056,
+      "learning_rate": 2.0039605757500512e-05,
+      "loss": 0.05871880531311035,
+      "mean_token_accuracy": 0.9786837643384934,
+      "num_tokens": 9552241.0,
+      "step": 3250
+    },
+    {
+      "entropy": 0.12911307733505964,
+      "epoch": 8.571799870045485,
+      "grad_norm": 0.4569152593612671,
+      "learning_rate": 1.691082982918235e-05,
+      "loss": 0.06450970649719238,
+      "mean_token_accuracy": 0.9761316785216332,
+      "num_tokens": 9689407.0,
+      "step": 3300
+    },
+    {
+      "entropy": 0.11636774389073253,
+      "epoch": 8.701754385964913,
+      "grad_norm": 0.27477338910102844,
+      "learning_rate": 1.403162711509129e-05,
+      "loss": 0.05784036159515381,
+      "mean_token_accuracy": 0.9791285961866378,
+      "num_tokens": 9842204.0,
+      "step": 3350
+    },
+    {
+      "entropy": 0.11710284009575844,
+      "epoch": 8.83170890188434,
+      "grad_norm": 0.26900890469551086,
+      "learning_rate": 1.1407913655766755e-05,
+      "loss": 0.05737146377563476,
+      "mean_token_accuracy": 0.9788037702441216,
+      "num_tokens": 9994524.0,
+      "step": 3400
+    },
+    {
+      "entropy": 0.11474769543856382,
+      "epoch": 8.961663417803768,
+      "grad_norm": 0.3137633204460144,
+      "learning_rate": 9.045080525311815e-06,
+      "loss": 0.057830405235290525,
+      "mean_token_accuracy": 0.9789029136300087,
+      "num_tokens": 10147597.0,
+      "step": 3450
+    },
+    {
+      "epoch": 9.0,
+      "eval_entropy": 0.23872991701444754,
+      "eval_loss": 1.007972002029419,
+      "eval_mean_token_accuracy": 0.8318104110658169,
+      "eval_num_tokens": 10189260.0,
+      "eval_runtime": 53.4047,
+      "eval_samples_per_second": 31.027,
+      "eval_steps_per_second": 3.895,
+      "step": 3465
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3850,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.43404829763072e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-385/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-385/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.009078376988692594,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-385/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-385/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-385/trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 385,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2479030179977415,
+      "epoch": 0.1299545159194282,
+      "grad_norm": 1.519571304321289,
+      "learning_rate": 3.522207847653314e-05,
+      "loss": 2.093206329345703,
+      "mean_token_accuracy": 0.6068353663384914,
+      "num_tokens": 154518.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.932415626347065,
+      "epoch": 0.2599090318388564,
+      "grad_norm": 1.180830955505371,
+      "learning_rate": 7.11629748811588e-05,
+      "loss": 0.8930854797363281,
+      "mean_token_accuracy": 0.7708445385098457,
+      "num_tokens": 306733.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.7730373838543891,
+      "epoch": 0.3898635477582846,
+      "grad_norm": 0.7839977145195007,
+      "learning_rate": 0.00010710387128578447,
+      "loss": 0.7302116394042969,
+      "mean_token_accuracy": 0.8012136635184288,
+      "num_tokens": 446267.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6934178560972214,
+      "epoch": 0.5198180636777128,
+      "grad_norm": 0.666778564453125,
+      "learning_rate": 0.0001430447676904101,
+      "loss": 0.6505754852294922,
+      "mean_token_accuracy": 0.8195212116837501,
+      "num_tokens": 600256.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.6900296103954315,
+      "epoch": 0.649772579597141,
+      "grad_norm": 0.6762415766716003,
+      "learning_rate": 0.00017898566409503577,
+      "loss": 0.6378536987304687,
+      "mean_token_accuracy": 0.8223087686300278,
+      "num_tokens": 738649.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.667421719878912,
+      "epoch": 0.7797270955165692,
+      "grad_norm": 0.5047685503959656,
+      "learning_rate": 0.00021492656049966144,
+      "loss": 0.6148524856567383,
+      "mean_token_accuracy": 0.8280292323231697,
+      "num_tokens": 883494.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.6388977643847465,
+      "epoch": 0.9096816114359974,
+      "grad_norm": 0.4360353350639343,
+      "learning_rate": 0.0002508674569042871,
+      "loss": 0.5933729553222656,
+      "mean_token_accuracy": 0.8329134130477905,
+      "num_tokens": 1032111.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.6612381511009656,
+      "eval_loss": 0.6559221744537354,
+      "eval_mean_token_accuracy": 0.8195324820967821,
+      "eval_num_tokens": 1132140.0,
+      "eval_runtime": 53.4007,
+      "eval_samples_per_second": 31.03,
+      "eval_steps_per_second": 3.895,
+      "step": 385
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3850,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.91184805036032e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3850/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3850/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.009078376988692594,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3850/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3850/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Estonian/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-3850/trainer_state.json

@@ -0,0 +1,914 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 3850,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2479030179977415,
+      "epoch": 0.1299545159194282,
+      "grad_norm": 1.519571304321289,
+      "learning_rate": 3.522207847653314e-05,
+      "loss": 2.093206329345703,
+      "mean_token_accuracy": 0.6068353663384914,
+      "num_tokens": 154518.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.932415626347065,
+      "epoch": 0.2599090318388564,
+      "grad_norm": 1.180830955505371,
+      "learning_rate": 7.11629748811588e-05,
+      "loss": 0.8930854797363281,
+      "mean_token_accuracy": 0.7708445385098457,
+      "num_tokens": 306733.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.7730373838543891,
+      "epoch": 0.3898635477582846,
+      "grad_norm": 0.7839977145195007,
+      "learning_rate": 0.00010710387128578447,
+      "loss": 0.7302116394042969,
+      "mean_token_accuracy": 0.8012136635184288,
+      "num_tokens": 446267.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6934178560972214,
+      "epoch": 0.5198180636777128,
+      "grad_norm": 0.666778564453125,
+      "learning_rate": 0.0001430447676904101,
+      "loss": 0.6505754852294922,
+      "mean_token_accuracy": 0.8195212116837501,
+      "num_tokens": 600256.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.6900296103954315,
+      "epoch": 0.649772579597141,
+      "grad_norm": 0.6762415766716003,
+      "learning_rate": 0.00017898566409503577,
+      "loss": 0.6378536987304687,
+      "mean_token_accuracy": 0.8223087686300278,
+      "num_tokens": 738649.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.667421719878912,
+      "epoch": 0.7797270955165692,
+      "grad_norm": 0.5047685503959656,
+      "learning_rate": 0.00021492656049966144,
+      "loss": 0.6148524856567383,
+      "mean_token_accuracy": 0.8280292323231697,
+      "num_tokens": 883494.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.6388977643847465,
+      "epoch": 0.9096816114359974,
+      "grad_norm": 0.4360353350639343,
+      "learning_rate": 0.0002508674569042871,
+      "loss": 0.5933729553222656,
+      "mean_token_accuracy": 0.8329134130477905,
+      "num_tokens": 1032111.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.6612381511009656,
+      "eval_loss": 0.6559221744537354,
+      "eval_mean_token_accuracy": 0.8195324820967821,
+      "eval_num_tokens": 1132140.0,
+      "eval_runtime": 53.4007,
+      "eval_samples_per_second": 31.03,
+      "eval_steps_per_second": 3.895,
+      "step": 385
+    },
+    {
+      "entropy": 0.6224366770916848,
+      "epoch": 1.0389863547758285,
+      "grad_norm": 0.5294668078422546,
+      "learning_rate": 0.00027673375518355765,
+      "loss": 0.5677951431274414,
+      "mean_token_accuracy": 0.8380465067211708,
+      "num_tokens": 1177556.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5827466724812984,
+      "epoch": 1.1689408706952567,
+      "grad_norm": 0.5172416567802429,
+      "learning_rate": 0.0002765120122346144,
+      "loss": 0.5423126983642578,
+      "mean_token_accuracy": 0.8467991036176682,
+      "num_tokens": 1325434.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5831517253816128,
+      "epoch": 1.2988953866146848,
+      "grad_norm": 0.41916292905807495,
+      "learning_rate": 0.0002760064270819138,
+      "loss": 0.534448013305664,
+      "mean_token_accuracy": 0.8456632816791534,
+      "num_tokens": 1474116.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5869986982643605,
+      "epoch": 1.428849902534113,
+      "grad_norm": 0.4387759566307068,
+      "learning_rate": 0.00027521803857633113,
+      "loss": 0.5367491912841796,
+      "mean_token_accuracy": 0.8462416216731071,
+      "num_tokens": 1621193.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5771756853163242,
+      "epoch": 1.5588044184535412,
+      "grad_norm": 0.49079665541648865,
+      "learning_rate": 0.00027414846665880935,
+      "loss": 0.5238623809814453,
+      "mean_token_accuracy": 0.84760089635849,
+      "num_tokens": 1767789.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5549105909466744,
+      "epoch": 1.6887589343729694,
+      "grad_norm": 0.4000363051891327,
+      "learning_rate": 0.0002727999090317863,
+      "loss": 0.510434226989746,
+      "mean_token_accuracy": 0.8517858856916427,
+      "num_tokens": 1918138.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.583413660377264,
+      "epoch": 1.8187134502923976,
+      "grad_norm": 0.33592426776885986,
+      "learning_rate": 0.00027117513664346674,
+      "loss": 0.5297993850708008,
+      "mean_token_accuracy": 0.846615691781044,
+      "num_tokens": 2057575.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.5732646904885769,
+      "epoch": 1.9486679662118258,
+      "grad_norm": 0.5528839230537415,
+      "learning_rate": 0.00026927748799421714,
+      "loss": 0.5219194793701172,
+      "mean_token_accuracy": 0.8489033079147339,
+      "num_tokens": 2208320.0,
+      "step": 750
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.6027900550801021,
+      "eval_loss": 0.5946928858757019,
+      "eval_mean_token_accuracy": 0.8318195798649237,
+      "eval_num_tokens": 2264280.0,
+      "eval_runtime": 53.3837,
+      "eval_samples_per_second": 31.039,
+      "eval_steps_per_second": 3.896,
+      "step": 770
+    },
+    {
+      "entropy": 0.5329899657611272,
+      "epoch": 2.077972709551657,
+      "grad_norm": 0.45793575048446655,
+      "learning_rate": 0.0002671108622767842,
+      "loss": 0.48420516967773436,
+      "mean_token_accuracy": 0.8578200301333289,
+      "num_tokens": 2348248.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.5142687204480171,
+      "epoch": 2.207927225471085,
+      "grad_norm": 0.4690960645675659,
+      "learning_rate": 0.0002646797113644295,
+      "loss": 0.4593114471435547,
+      "mean_token_accuracy": 0.8622670090198516,
+      "num_tokens": 2501427.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.5135884112119675,
+      "epoch": 2.3378817413905133,
+      "grad_norm": 0.3752821683883667,
+      "learning_rate": 0.00026198903066344565,
+      "loss": 0.4626216125488281,
+      "mean_token_accuracy": 0.8612511262297631,
+      "num_tokens": 2650794.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.5137367483973503,
+      "epoch": 2.4678362573099415,
+      "grad_norm": 0.3726271390914917,
+      "learning_rate": 0.0002590443488488465,
+      "loss": 0.4601683807373047,
+      "mean_token_accuracy": 0.8620512077212333,
+      "num_tokens": 2798180.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.5105714881420136,
+      "epoch": 2.5977907732293697,
+      "grad_norm": 0.41296717524528503,
+      "learning_rate": 0.00025585171650432525,
+      "loss": 0.46279102325439453,
+      "mean_token_accuracy": 0.8611763519048691,
+      "num_tokens": 2950301.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5169161760807037,
+      "epoch": 2.727745289148798,
+      "grad_norm": 0.4614253044128418,
+      "learning_rate": 0.0002524176936898197,
+      "loss": 0.45492774963378907,
+      "mean_token_accuracy": 0.8627680170536042,
+      "num_tokens": 3091810.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4989277676492929,
+      "epoch": 2.857699805068226,
+      "grad_norm": 0.37512704730033875,
+      "learning_rate": 0.00024874933646223225,
+      "loss": 0.4531984329223633,
+      "mean_token_accuracy": 0.8637665447592735,
+      "num_tokens": 3242184.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.5177617704868317,
+      "epoch": 2.9876543209876543,
+      "grad_norm": 0.3700532019138336,
+      "learning_rate": 0.00024485418237699976,
+      "loss": 0.45844474792480466,
+      "mean_token_accuracy": 0.8626988258957863,
+      "num_tokens": 3382605.0,
+      "step": 1150
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.5253989950108987,
+      "eval_loss": 0.5857328176498413,
+      "eval_mean_token_accuracy": 0.8360884573597175,
+      "eval_num_tokens": 3396420.0,
+      "eval_runtime": 53.341,
+      "eval_samples_per_second": 31.064,
+      "eval_steps_per_second": 3.899,
+      "step": 1155
+    },
+    {
+      "entropy": 0.4535361140517134,
+      "epoch": 3.116959064327485,
+      "grad_norm": 0.3412795662879944,
+      "learning_rate": 0.00024074023500030492,
+      "loss": 0.3942829132080078,
+      "mean_token_accuracy": 0.8781378038564519,
+      "num_tokens": 3522582.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.44747897461056707,
+      "epoch": 3.246913580246914,
+      "grad_norm": 0.46647050976753235,
+      "learning_rate": 0.0002364159474637521,
+      "loss": 0.38986759185791015,
+      "mean_token_accuracy": 0.8777281475067139,
+      "num_tokens": 3670864.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.4480265176296234,
+      "epoch": 3.3768680961663415,
+      "grad_norm": 0.4068582355976105,
+      "learning_rate": 0.00023189020509529866,
+      "loss": 0.39444759368896487,
+      "mean_token_accuracy": 0.8774515727162361,
+      "num_tokens": 3822021.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.45180007234215736,
+      "epoch": 3.50682261208577,
+      "grad_norm": 0.4249928593635559,
+      "learning_rate": 0.00022717230716213122,
+      "loss": 0.3977077102661133,
+      "mean_token_accuracy": 0.8762744688987731,
+      "num_tokens": 3968736.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4614932192862034,
+      "epoch": 3.636777128005198,
+      "grad_norm": 0.561008095741272,
+      "learning_rate": 0.00022227194776300045,
+      "loss": 0.4022808456420898,
+      "mean_token_accuracy": 0.8760285252332687,
+      "num_tokens": 4113509.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.4414680179953575,
+      "epoch": 3.7667316439246266,
+      "grad_norm": 0.38943538069725037,
+      "learning_rate": 0.00021719919590927584,
+      "loss": 0.38586376190185545,
+      "mean_token_accuracy": 0.8783121705055237,
+      "num_tokens": 4267958.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.45685607343912127,
+      "epoch": 3.8966861598440543,
+      "grad_norm": 0.5362406969070435,
+      "learning_rate": 0.00021196447483564875,
+      "loss": 0.3983576583862305,
+      "mean_token_accuracy": 0.8764419692754746,
+      "num_tokens": 4415398.0,
+      "step": 1500
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.49981127894268584,
+      "eval_loss": 0.5997208952903748,
+      "eval_mean_token_accuracy": 0.8368159819107789,
+      "eval_num_tokens": 4528560.0,
+      "eval_runtime": 53.4304,
+      "eval_samples_per_second": 31.012,
+      "eval_steps_per_second": 3.893,
+      "step": 1540
+    },
+    {
+      "entropy": 0.4439909208060509,
+      "epoch": 4.025990903183885,
+      "grad_norm": 0.5490113496780396,
+      "learning_rate": 0.00020657854058299564,
+      "loss": 0.38307292938232423,
+      "mean_token_accuracy": 0.8795150506436525,
+      "num_tokens": 4559534.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3837250977009535,
+      "epoch": 4.155945419103314,
+      "grad_norm": 0.5567234754562378,
+      "learning_rate": 0.0002010524598974076,
+      "loss": 0.3182963752746582,
+      "mean_token_accuracy": 0.8964017608761787,
+      "num_tokens": 4707075.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.377872094810009,
+      "epoch": 4.2858999350227425,
+      "grad_norm": 0.4315710961818695,
+      "learning_rate": 0.00019539758749079845,
+      "loss": 0.318333683013916,
+      "mean_token_accuracy": 0.8963816618919372,
+      "num_tokens": 4851683.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.38739304527640345,
+      "epoch": 4.41585445094217,
+      "grad_norm": 0.49140632152557373,
+      "learning_rate": 0.00018962554270981555,
+      "loss": 0.32688804626464846,
+      "mean_token_accuracy": 0.8937860554456711,
+      "num_tokens": 4994086.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.39157475270330905,
+      "epoch": 4.545808966861598,
+      "grad_norm": 0.40667369961738586,
+      "learning_rate": 0.00018374818566099208,
+      "loss": 0.3305763626098633,
+      "mean_token_accuracy": 0.8916732975840569,
+      "num_tokens": 5137171.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.3838599680364132,
+      "epoch": 4.675763482781027,
+      "grad_norm": 0.4632417857646942,
+      "learning_rate": 0.0001777775928411983,
+      "loss": 0.3267818450927734,
+      "mean_token_accuracy": 0.8946500706672669,
+      "num_tokens": 5287076.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.38270787581801413,
+      "epoch": 4.805717998700455,
+      "grad_norm": 0.5529720187187195,
+      "learning_rate": 0.0001717260323234649,
+      "loss": 0.3264235305786133,
+      "mean_token_accuracy": 0.8948800846934318,
+      "num_tokens": 5436923.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.38736109718680384,
+      "epoch": 4.935672514619883,
+      "grad_norm": 0.5604785680770874,
+      "learning_rate": 0.00016560593854916497,
+      "loss": 0.3280513381958008,
+      "mean_token_accuracy": 0.8931388029456139,
+      "num_tokens": 5589195.0,
+      "step": 1900
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.4370518812479881,
+      "eval_loss": 0.6210553050041199,
+      "eval_mean_token_accuracy": 0.8379779781859654,
+      "eval_num_tokens": 5660700.0,
+      "eval_runtime": 53.4039,
+      "eval_samples_per_second": 31.028,
+      "eval_steps_per_second": 3.895,
+      "step": 1925
+    },
+    {
+      "entropy": 0.3330705868988181,
+      "epoch": 5.064977257959714,
+      "grad_norm": 0.5386723875999451,
+      "learning_rate": 0.0001594298867783512,
+      "loss": 0.2754818344116211,
+      "mean_token_accuracy": 0.9101201346771202,
+      "num_tokens": 5739445.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.2980620255321264,
+      "epoch": 5.1949317738791425,
+      "grad_norm": 0.5633581876754761,
+      "learning_rate": 0.00015321056725074549,
+      "loss": 0.23754241943359375,
+      "mean_token_accuracy": 0.9203532826900482,
+      "num_tokens": 5888043.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.3111592583358288,
+      "epoch": 5.32488628979857,
+      "grad_norm": 0.5031015872955322,
+      "learning_rate": 0.0001469607591104745,
+      "loss": 0.24428102493286133,
+      "mean_token_accuracy": 0.917181601524353,
+      "num_tokens": 6031284.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.31522042460739613,
+      "epoch": 5.454840805717999,
+      "grad_norm": 0.6432453393936157,
+      "learning_rate": 0.0001406933041481286,
+      "loss": 0.25112478256225584,
+      "mean_token_accuracy": 0.9152472382783889,
+      "num_tokens": 6179927.0,
+      "step": 2100
+    },
+    {
+      "entropy": 0.3046229027956724,
+      "epoch": 5.584795321637427,
+      "grad_norm": 0.5104537606239319,
+      "learning_rate": 0.00013442108041409814,
+      "loss": 0.2431495475769043,
+      "mean_token_accuracy": 0.917829519212246,
+      "num_tokens": 6322630.0,
+      "step": 2150
+    },
+    {
+      "entropy": 0.30440517760813235,
+      "epoch": 5.714749837556855,
+      "grad_norm": 0.5307765603065491,
+      "learning_rate": 0.0001281569757574053,
+      "loss": 0.24610313415527343,
+      "mean_token_accuracy": 0.9166415151953697,
+      "num_tokens": 6469843.0,
+      "step": 2200
+    },
+    {
+      "entropy": 0.304359400421381,
+      "epoch": 5.844704353476283,
+      "grad_norm": 0.5014523267745972,
+      "learning_rate": 0.00012191386134440133,
+      "loss": 0.24548973083496095,
+      "mean_token_accuracy": 0.9165477308630944,
+      "num_tokens": 6617768.0,
+      "step": 2250
+    },
+    {
+      "entropy": 0.3141418205201626,
+      "epoch": 5.974658869395712,
+      "grad_norm": 0.567398726940155,
+      "learning_rate": 0.00011570456521174339,
+      "loss": 0.24975168228149414,
+      "mean_token_accuracy": 0.9139353120326996,
+      "num_tokens": 6761187.0,
+      "step": 2300
+    },
+    {
+      "epoch": 6.0,
+      "eval_entropy": 0.3758909202252443,
+      "eval_loss": 0.6843022108078003,
+      "eval_mean_token_accuracy": 0.8348075449466705,
+      "eval_num_tokens": 6792840.0,
+      "eval_runtime": 53.3825,
+      "eval_samples_per_second": 31.04,
+      "eval_steps_per_second": 3.896,
+      "step": 2310
+    },
+    {
+      "entropy": 0.24346149432000203,
+      "epoch": 6.1039636127355426,
+      "grad_norm": 0.7140825986862183,
+      "learning_rate": 0.00010954184590799172,
+      "loss": 0.17231273651123047,
+      "mean_token_accuracy": 0.9407721275660261,
+      "num_tokens": 6909578.0,
+      "step": 2350
+    },
+    {
+      "entropy": 0.2160973483324051,
+      "epoch": 6.23391812865497,
+      "grad_norm": 0.49014952778816223,
+      "learning_rate": 0.00010343836627798716,
+      "loss": 0.15455107688903807,
+      "mean_token_accuracy": 0.9467655989527702,
+      "num_tokens": 7056244.0,
+      "step": 2400
+    },
+    {
+      "entropy": 0.21491924367845058,
+      "epoch": 6.363872644574399,
+      "grad_norm": 0.5529471635818481,
+      "learning_rate": 9.740666744387656e-05,
+      "loss": 0.1584029197692871,
+      "mean_token_accuracy": 0.9460993978381157,
+      "num_tokens": 7206950.0,
+      "step": 2450
+    },
+    {
+      "entropy": 0.22037068914622068,
+      "epoch": 6.493827160493828,
+      "grad_norm": 0.6232843995094299,
+      "learning_rate": 9.145914303624717e-05,
+      "loss": 0.15544342041015624,
+      "mean_token_accuracy": 0.9450622496008872,
+      "num_tokens": 7359429.0,
+      "step": 2500
+    },
+    {
+      "entropy": 0.2320463878661394,
+      "epoch": 6.623781676413255,
+      "grad_norm": 0.7459681630134583,
+      "learning_rate": 8.560801372831975e-05,
+      "loss": 0.16350215911865235,
+      "mean_token_accuracy": 0.9416968420147895,
+      "num_tokens": 7499281.0,
+      "step": 2550
+    },
+    {
+      "entropy": 0.22943626195192338,
+      "epoch": 6.753736192332683,
+      "grad_norm": 0.7482302784919739,
+      "learning_rate": 7.986530212552506e-05,
+      "loss": 0.16422538757324218,
+      "mean_token_accuracy": 0.9434959614276885,
+      "num_tokens": 7640758.0,
+      "step": 2600
+    },
+    {
+      "entropy": 0.21795938543975354,
+      "epoch": 6.883690708252112,
+      "grad_norm": 0.5210486054420471,
+      "learning_rate": 7.424280806206118e-05,
+      "loss": 0.15540474891662598,
+      "mean_token_accuracy": 0.9459306105971337,
+      "num_tokens": 7791986.0,
+      "step": 2650
+    },
+    {
+      "epoch": 7.0,
+      "eval_entropy": 0.3157534837149657,
+      "eval_loss": 0.7744874954223633,
+      "eval_mean_token_accuracy": 0.8346395036922052,
+      "eval_num_tokens": 7924980.0,
+      "eval_runtime": 53.3771,
+      "eval_samples_per_second": 31.043,
+      "eval_steps_per_second": 3.897,
+      "step": 2695
+    },
+    {
+      "entropy": 0.2124774060656677,
+      "epoch": 7.012995451591943,
+      "grad_norm": 0.42238789796829224,
+      "learning_rate": 6.875208435518865e-05,
+      "loss": 0.14792531967163086,
+      "mean_token_accuracy": 0.9490461115861059,
+      "num_tokens": 7940521.0,
+      "step": 2700
+    },
+    {
+      "entropy": 0.15692965138703585,
+      "epoch": 7.142949967511371,
+      "grad_norm": 0.4711572229862213,
+      "learning_rate": 6.340441306708468e-05,
+      "loss": 0.09051708221435546,
+      "mean_token_accuracy": 0.9700166273117066,
+      "num_tokens": 8084193.0,
+      "step": 2750
+    },
+    {
+      "entropy": 0.15241683423519134,
+      "epoch": 7.272904483430799,
+      "grad_norm": 0.4312196671962738,
+      "learning_rate": 5.821078232303016e-05,
+      "loss": 0.08812363624572754,
+      "mean_token_accuracy": 0.9699361199140548,
+      "num_tokens": 8230159.0,
+      "step": 2800
+    },
+    {
+      "entropy": 0.1459079357981682,
+      "epoch": 7.402858999350228,
+      "grad_norm": 0.4804084002971649,
+      "learning_rate": 5.3181863733564636e-05,
+      "loss": 0.08675944328308105,
+      "mean_token_accuracy": 0.9703072866797448,
+      "num_tokens": 8380556.0,
+      "step": 2850
+    },
+    {
+      "entropy": 0.15621621005237102,
+      "epoch": 7.532813515269655,
+      "grad_norm": 0.5435478091239929,
+      "learning_rate": 4.83279904669986e-05,
+      "loss": 0.09016354560852051,
+      "mean_token_accuracy": 0.9674961140751839,
+      "num_tokens": 8523248.0,
+      "step": 2900
+    },
+    {
+      "entropy": 0.1545175113901496,
+      "epoch": 7.662768031189084,
+      "grad_norm": 0.524286687374115,
+      "learning_rate": 4.365913601734056e-05,
+      "loss": 0.09049141883850098,
+      "mean_token_accuracy": 0.9679373624920845,
+      "num_tokens": 8672002.0,
+      "step": 2950
+    },
+    {
+      "entropy": 0.1553485019877553,
+      "epoch": 7.792722547108512,
+      "grad_norm": 0.5006484389305115,
+      "learning_rate": 3.9184893711264495e-05,
+      "loss": 0.08913107872009278,
+      "mean_token_accuracy": 0.9684559822082519,
+      "num_tokens": 8816090.0,
+      "step": 3000
+    },
+    {
+      "entropy": 0.15444331549108029,
+      "epoch": 7.92267706302794,
+      "grad_norm": 0.5613893866539001,
+      "learning_rate": 3.491445699622611e-05,
+      "loss": 0.08711207389831543,
+      "mean_token_accuracy": 0.9684525722265244,
+      "num_tokens": 8966004.0,
+      "step": 3050
+    },
+    {
+      "epoch": 8.0,
+      "eval_entropy": 0.26580205430778175,
+      "eval_loss": 0.8996392488479614,
+      "eval_mean_token_accuracy": 0.8319417791297803,
+      "eval_num_tokens": 9057120.0,
+      "eval_runtime": 53.4042,
+      "eval_samples_per_second": 31.028,
+      "eval_steps_per_second": 3.895,
+      "step": 3080
+    },
+    {
+      "entropy": 0.14060429509860187,
+      "epoch": 8.05198180636777,
+      "grad_norm": 0.25378674268722534,
+      "learning_rate": 3.085660055023035e-05,
+      "loss": 0.07468742847442628,
+      "mean_token_accuracy": 0.9737296260181983,
+      "num_tokens": 9115929.0,
+      "step": 3100
+    },
+    {
+      "entropy": 0.12152639016509056,
+      "epoch": 8.1819363222872,
+      "grad_norm": 0.3634810447692871,
+      "learning_rate": 2.7019662252065798e-05,
+      "loss": 0.05918361663818359,
+      "mean_token_accuracy": 0.9785374769568443,
+      "num_tokens": 9260646.0,
+      "step": 3150
+    },
+    {
+      "entropy": 0.1228457903675735,
+      "epoch": 8.311890838206628,
+      "grad_norm": 0.32873299717903137,
+      "learning_rate": 2.3411526049051643e-05,
+      "loss": 0.060924801826477054,
+      "mean_token_accuracy": 0.9778030979633331,
+      "num_tokens": 9404737.0,
+      "step": 3200
+    },
+    {
+      "entropy": 0.11790491977706552,
+      "epoch": 8.441845354126055,
+      "grad_norm": 0.27871423959732056,
+      "learning_rate": 2.0039605757500512e-05,
+      "loss": 0.05871880531311035,
+      "mean_token_accuracy": 0.9786837643384934,
+      "num_tokens": 9552241.0,
+      "step": 3250
+    },
+    {
+      "entropy": 0.12911307733505964,
+      "epoch": 8.571799870045485,
+      "grad_norm": 0.4569152593612671,
+      "learning_rate": 1.691082982918235e-05,
+      "loss": 0.06450970649719238,
+      "mean_token_accuracy": 0.9761316785216332,
+      "num_tokens": 9689407.0,
+      "step": 3300
+    },
+    {
+      "entropy": 0.11636774389073253,
+      "epoch": 8.701754385964913,
+      "grad_norm": 0.27477338910102844,
+      "learning_rate": 1.403162711509129e-05,
+      "loss": 0.05784036159515381,
+      "mean_token_accuracy": 0.9791285961866378,
+      "num_tokens": 9842204.0,
+      "step": 3350
+    },
+    {
+      "entropy": 0.11710284009575844,
+      "epoch": 8.83170890188434,
+      "grad_norm": 0.26900890469551086,
+      "learning_rate": 1.1407913655766755e-05,
+      "loss": 0.05737146377563476,
+      "mean_token_accuracy": 0.9788037702441216,
+      "num_tokens": 9994524.0,
+      "step": 3400
+    },
+    {
+      "entropy": 0.11474769543856382,
+      "epoch": 8.961663417803768,
+      "grad_norm": 0.3137633204460144,
+      "learning_rate": 9.045080525311815e-06,
+      "loss": 0.057830405235290525,
+      "mean_token_accuracy": 0.9789029136300087,
+      "num_tokens": 10147597.0,
+      "step": 3450
+    },
+    {
+      "epoch": 9.0,
+      "eval_entropy": 0.23872991701444754,
+      "eval_loss": 1.007972002029419,
+      "eval_mean_token_accuracy": 0.8318104110658169,
+      "eval_num_tokens": 10189260.0,
+      "eval_runtime": 53.4047,
+      "eval_samples_per_second": 31.027,
+      "eval_steps_per_second": 3.895,
+      "step": 3465
+    },
+    {
+      "entropy": 0.10747776249769944,
+      "epoch": 9.0909681611436,
+      "grad_norm": 0.21027140319347382,
+      "learning_rate": 6.9479827540858e-06,
+      "loss": 0.05341584682464599,
+      "mean_token_accuracy": 0.9819158309668152,
+      "num_tokens": 10294278.0,
+      "step": 3500
+    },
+    {
+      "entropy": 0.10646019088104368,
+      "epoch": 9.220922677063028,
+      "grad_norm": 0.1820213794708252,
+      "learning_rate": 5.120929352832946e-06,
+      "loss": 0.04993132591247559,
+      "mean_token_accuracy": 0.9813734939694405,
+      "num_tokens": 10446433.0,
+      "step": 3550
+    },
+    {
+      "entropy": 0.11276215925812721,
+      "epoch": 9.350877192982455,
+      "grad_norm": 0.28350868821144104,
+      "learning_rate": 3.5676744587442527e-06,
+      "loss": 0.05132147789001465,
+      "mean_token_accuracy": 0.9798016020655632,
+      "num_tokens": 10596480.0,
+      "step": 3600
+    },
+    {
+      "entropy": 0.11684846783056856,
+      "epoch": 9.480831708901885,
+      "grad_norm": 0.24154022336006165,
+      "learning_rate": 2.2914096216458985e-06,
+      "loss": 0.05330245018005371,
+      "mean_token_accuracy": 0.9795207896828652,
+      "num_tokens": 10740528.0,
+      "step": 3650
+    },
+    {
+      "entropy": 0.11312318585813046,
+      "epoch": 9.610786224821313,
+      "grad_norm": 0.28576743602752686,
+      "learning_rate": 1.2947572461634096e-06,
+      "loss": 0.05499160289764404,
+      "mean_token_accuracy": 0.9797447052598,
+      "num_tokens": 10881750.0,
+      "step": 3700
+    },
+    {
+      "entropy": 0.11110325066372752,
+      "epoch": 9.74074074074074,
+      "grad_norm": 0.22744601964950562,
+      "learning_rate": 5.79765203336998e-07,
+      "loss": 0.05335733413696289,
+      "mean_token_accuracy": 0.9804390069842338,
+      "num_tokens": 11027552.0,
+      "step": 3750
+    },
+    {
+      "entropy": 0.11025484301149845,
+      "epoch": 9.870695256660168,
+      "grad_norm": 0.2376026064157486,
+      "learning_rate": 1.4790262275940392e-07,
+      "loss": 0.053631534576416014,
+      "mean_token_accuracy": 0.9807139033079147,
+      "num_tokens": 11172587.0,
+      "step": 3800
+    },
+    {
+      "entropy": 0.11222033412312743,
+      "epoch": 10.0,
+      "grad_norm": 0.22163568437099457,
+      "learning_rate": 5.6873882486966634e-11,
+      "loss": 0.05134555339813232,
+      "mean_token_accuracy": 0.9805273500519183,
+      "num_tokens": 11321400.0,
+      "step": 3850
+    },
+    {
+      "epoch": 10.0,
+      "eval_entropy": 0.23073764632527644,
+      "eval_loss": 1.059328556060791,
+      "eval_mean_token_accuracy": 0.8315239631785796,
+      "eval_num_tokens": 11321400.0,
+      "eval_runtime": 53.4589,
+      "eval_samples_per_second": 30.996,
+      "eval_steps_per_second": 3.891,
+      "step": 3850
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3850,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.925137653703373e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: transformers
+model_name: Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1
+tags:
+- generated_from_trainer
+- sft
+- trl
+licence: license
+---
+
+# Model Card for Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1
+
+This model is a fine-tuned version of [Qwen/Qwen3-4B-Base](https://huggingface.co/Qwen/Qwen3-4B-Base).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/katriin-kukk/Cross_lingual_morphological_generalization/runs/ranbbj8x) 
+
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.29.0
+- Transformers: 5.5.4
+- Pytorch: 2.10.0
+- Datasets: 4.6.1
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-1122/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test1/checkpoint-1122/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: transformers
+model_name: Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2
+tags:
+- generated_from_trainer
+- sft
+- trl
+licence: license
+---
+
+# Model Card for Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2
+
+This model is a fine-tuned version of [Qwen/Qwen3-4B-Base](https://huggingface.co/Qwen/Qwen3-4B-Base).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/katriin-kukk/Cross_lingual_morphological_generalization/runs/y6cm94yy) 
+
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.29.0
+- Transformers: 5.5.4
+- Pytorch: 2.10.0
+- Datasets: 4.6.1
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1122/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1122/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1122/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1122/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1122/trainer_state.json

@@ -0,0 +1,287 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 1122,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.23531258243029e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1496/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1496/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1496/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1496/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1496/trainer_state.json

@@ -0,0 +1,368 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 1496,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    },
+    {
+      "entropy": 0.41342759042075183,
+      "epoch": 3.074966532797858,
+      "grad_norm": 0.34829744696617126,
+      "learning_rate": 0.0002045690972655427,
+      "loss": 0.3644887542724609,
+      "mean_token_accuracy": 0.8879408977850519,
+      "num_tokens": 2955424.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.3948360003530979,
+      "epoch": 3.208835341365462,
+      "grad_norm": 0.312427282333374,
+      "learning_rate": 0.00020086347837665854,
+      "loss": 0.34146129608154296,
+      "mean_token_accuracy": 0.8914449456334114,
+      "num_tokens": 3078799.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.38602679744362833,
+      "epoch": 3.3427041499330654,
+      "grad_norm": 0.2715625762939453,
+      "learning_rate": 0.0001969750898863629,
+      "loss": 0.34105979919433593,
+      "mean_token_accuracy": 0.8928995525836945,
+      "num_tokens": 3211945.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.3923524462431669,
+      "epoch": 3.4765729585006695,
+      "grad_norm": 0.27295640110969543,
+      "learning_rate": 0.00019291239826442992,
+      "loss": 0.3473458099365234,
+      "mean_token_accuracy": 0.8913829082250595,
+      "num_tokens": 3343933.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.40172914519906044,
+      "epoch": 3.610441767068273,
+      "grad_norm": 0.32684192061424255,
+      "learning_rate": 0.0001886842495034615,
+      "loss": 0.35543827056884764,
+      "mean_token_accuracy": 0.8904094022512435,
+      "num_tokens": 3470087.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4010512103140354,
+      "epoch": 3.7443105756358768,
+      "grad_norm": 0.23922984302043915,
+      "learning_rate": 0.00018429984985786734,
+      "loss": 0.3535212326049805,
+      "mean_token_accuracy": 0.8891122484207153,
+      "num_tokens": 3590858.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.39226082623004915,
+      "epoch": 3.878179384203481,
+      "grad_norm": 0.25130993127822876,
+      "learning_rate": 0.00017976874579842046,
+      "loss": 0.3484851837158203,
+      "mean_token_accuracy": 0.8916165816783905,
+      "num_tokens": 3725806.0,
+      "step": 1450
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.45332007586956025,
+      "eval_loss": 0.5243425965309143,
+      "eval_mean_token_accuracy": 0.8532214590907097,
+      "eval_num_tokens": 3838972.0,
+      "eval_runtime": 51.3837,
+      "eval_samples_per_second": 31.099,
+      "eval_steps_per_second": 3.892,
+      "step": 1496
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.647043150972508e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1870/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1870/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1870/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1870/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-1870/trainer_state.json

@@ -0,0 +1,459 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 1870,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    },
+    {
+      "entropy": 0.41342759042075183,
+      "epoch": 3.074966532797858,
+      "grad_norm": 0.34829744696617126,
+      "learning_rate": 0.0002045690972655427,
+      "loss": 0.3644887542724609,
+      "mean_token_accuracy": 0.8879408977850519,
+      "num_tokens": 2955424.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.3948360003530979,
+      "epoch": 3.208835341365462,
+      "grad_norm": 0.312427282333374,
+      "learning_rate": 0.00020086347837665854,
+      "loss": 0.34146129608154296,
+      "mean_token_accuracy": 0.8914449456334114,
+      "num_tokens": 3078799.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.38602679744362833,
+      "epoch": 3.3427041499330654,
+      "grad_norm": 0.2715625762939453,
+      "learning_rate": 0.0001969750898863629,
+      "loss": 0.34105979919433593,
+      "mean_token_accuracy": 0.8928995525836945,
+      "num_tokens": 3211945.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.3923524462431669,
+      "epoch": 3.4765729585006695,
+      "grad_norm": 0.27295640110969543,
+      "learning_rate": 0.00019291239826442992,
+      "loss": 0.3473458099365234,
+      "mean_token_accuracy": 0.8913829082250595,
+      "num_tokens": 3343933.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.40172914519906044,
+      "epoch": 3.610441767068273,
+      "grad_norm": 0.32684192061424255,
+      "learning_rate": 0.0001886842495034615,
+      "loss": 0.35543827056884764,
+      "mean_token_accuracy": 0.8904094022512435,
+      "num_tokens": 3470087.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4010512103140354,
+      "epoch": 3.7443105756358768,
+      "grad_norm": 0.23922984302043915,
+      "learning_rate": 0.00018429984985786734,
+      "loss": 0.3535212326049805,
+      "mean_token_accuracy": 0.8891122484207153,
+      "num_tokens": 3590858.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.39226082623004915,
+      "epoch": 3.878179384203481,
+      "grad_norm": 0.25130993127822876,
+      "learning_rate": 0.00017976874579842046,
+      "loss": 0.3484851837158203,
+      "mean_token_accuracy": 0.8916165816783905,
+      "num_tokens": 3725806.0,
+      "step": 1450
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.45332007586956025,
+      "eval_loss": 0.5243425965309143,
+      "eval_mean_token_accuracy": 0.8532214590907097,
+      "eval_num_tokens": 3838972.0,
+      "eval_runtime": 51.3837,
+      "eval_samples_per_second": 31.099,
+      "eval_steps_per_second": 3.892,
+      "step": 1496
+    },
+    {
+      "entropy": 0.39358587043754983,
+      "epoch": 4.010709504685408,
+      "grad_norm": 0.27960067987442017,
+      "learning_rate": 0.0001751008032260355,
+      "loss": 0.34616813659667967,
+      "mean_token_accuracy": 0.8923709010234987,
+      "num_tokens": 3849380.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.3206369188427925,
+      "epoch": 4.144578313253012,
+      "grad_norm": 0.37261858582496643,
+      "learning_rate": 0.00017030618599002818,
+      "loss": 0.2684581565856934,
+      "mean_token_accuracy": 0.9131761506199837,
+      "num_tokens": 3976694.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3254615054279566,
+      "epoch": 4.278447121820616,
+      "grad_norm": 0.39803799986839294,
+      "learning_rate": 0.00016539533375763032,
+      "loss": 0.2769618606567383,
+      "mean_token_accuracy": 0.9102409112453461,
+      "num_tokens": 4110204.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.32003962114453316,
+      "epoch": 4.412315930388219,
+      "grad_norm": 0.35707366466522217,
+      "learning_rate": 0.0001603789392829468,
+      "loss": 0.2749842834472656,
+      "mean_token_accuracy": 0.910626070201397,
+      "num_tokens": 4240883.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.32672152675688265,
+      "epoch": 4.546184738955823,
+      "grad_norm": 0.47052621841430664,
+      "learning_rate": 0.00015526792512484774,
+      "loss": 0.27983531951904295,
+      "mean_token_accuracy": 0.9093958771228791,
+      "num_tokens": 4365381.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.33449163861572745,
+      "epoch": 4.680053547523427,
+      "grad_norm": 0.330709844827652,
+      "learning_rate": 0.00015007341986449012,
+      "loss": 0.28533639907836916,
+      "mean_token_accuracy": 0.9082232251763344,
+      "num_tokens": 4490711.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.33353066638112067,
+      "epoch": 4.813922356091031,
+      "grad_norm": 0.3990134298801422,
+      "learning_rate": 0.00014480673387425272,
+      "loss": 0.28489078521728517,
+      "mean_token_accuracy": 0.908001911342144,
+      "num_tokens": 4618532.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.3272412090748549,
+      "epoch": 4.947791164658635,
+      "grad_norm": 0.3183020353317261,
+      "learning_rate": 0.00013947933469084315,
+      "loss": 0.2772365379333496,
+      "mean_token_accuracy": 0.908946952521801,
+      "num_tokens": 4752261.0,
+      "step": 1850
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.3932068006694317,
+      "eval_loss": 0.5647156834602356,
+      "eval_mean_token_accuracy": 0.85078628718853,
+      "eval_num_tokens": 4798715.0,
+      "eval_runtime": 51.3578,
+      "eval_samples_per_second": 31.115,
+      "eval_steps_per_second": 3.894,
+      "step": 1870
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.0568507616161485e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2244/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2244/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2244/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2244/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2244/trainer_state.json

@@ -0,0 +1,540 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 500,
+  "global_step": 2244,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    },
+    {
+      "entropy": 0.41342759042075183,
+      "epoch": 3.074966532797858,
+      "grad_norm": 0.34829744696617126,
+      "learning_rate": 0.0002045690972655427,
+      "loss": 0.3644887542724609,
+      "mean_token_accuracy": 0.8879408977850519,
+      "num_tokens": 2955424.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.3948360003530979,
+      "epoch": 3.208835341365462,
+      "grad_norm": 0.312427282333374,
+      "learning_rate": 0.00020086347837665854,
+      "loss": 0.34146129608154296,
+      "mean_token_accuracy": 0.8914449456334114,
+      "num_tokens": 3078799.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.38602679744362833,
+      "epoch": 3.3427041499330654,
+      "grad_norm": 0.2715625762939453,
+      "learning_rate": 0.0001969750898863629,
+      "loss": 0.34105979919433593,
+      "mean_token_accuracy": 0.8928995525836945,
+      "num_tokens": 3211945.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.3923524462431669,
+      "epoch": 3.4765729585006695,
+      "grad_norm": 0.27295640110969543,
+      "learning_rate": 0.00019291239826442992,
+      "loss": 0.3473458099365234,
+      "mean_token_accuracy": 0.8913829082250595,
+      "num_tokens": 3343933.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.40172914519906044,
+      "epoch": 3.610441767068273,
+      "grad_norm": 0.32684192061424255,
+      "learning_rate": 0.0001886842495034615,
+      "loss": 0.35543827056884764,
+      "mean_token_accuracy": 0.8904094022512435,
+      "num_tokens": 3470087.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4010512103140354,
+      "epoch": 3.7443105756358768,
+      "grad_norm": 0.23922984302043915,
+      "learning_rate": 0.00018429984985786734,
+      "loss": 0.3535212326049805,
+      "mean_token_accuracy": 0.8891122484207153,
+      "num_tokens": 3590858.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.39226082623004915,
+      "epoch": 3.878179384203481,
+      "grad_norm": 0.25130993127822876,
+      "learning_rate": 0.00017976874579842046,
+      "loss": 0.3484851837158203,
+      "mean_token_accuracy": 0.8916165816783905,
+      "num_tokens": 3725806.0,
+      "step": 1450
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.45332007586956025,
+      "eval_loss": 0.5243425965309143,
+      "eval_mean_token_accuracy": 0.8532214590907097,
+      "eval_num_tokens": 3838972.0,
+      "eval_runtime": 51.3837,
+      "eval_samples_per_second": 31.099,
+      "eval_steps_per_second": 3.892,
+      "step": 1496
+    },
+    {
+      "entropy": 0.39358587043754983,
+      "epoch": 4.010709504685408,
+      "grad_norm": 0.27960067987442017,
+      "learning_rate": 0.0001751008032260355,
+      "loss": 0.34616813659667967,
+      "mean_token_accuracy": 0.8923709010234987,
+      "num_tokens": 3849380.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.3206369188427925,
+      "epoch": 4.144578313253012,
+      "grad_norm": 0.37261858582496643,
+      "learning_rate": 0.00017030618599002818,
+      "loss": 0.2684581565856934,
+      "mean_token_accuracy": 0.9131761506199837,
+      "num_tokens": 3976694.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3254615054279566,
+      "epoch": 4.278447121820616,
+      "grad_norm": 0.39803799986839294,
+      "learning_rate": 0.00016539533375763032,
+      "loss": 0.2769618606567383,
+      "mean_token_accuracy": 0.9102409112453461,
+      "num_tokens": 4110204.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.32003962114453316,
+      "epoch": 4.412315930388219,
+      "grad_norm": 0.35707366466522217,
+      "learning_rate": 0.0001603789392829468,
+      "loss": 0.2749842834472656,
+      "mean_token_accuracy": 0.910626070201397,
+      "num_tokens": 4240883.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.32672152675688265,
+      "epoch": 4.546184738955823,
+      "grad_norm": 0.47052621841430664,
+      "learning_rate": 0.00015526792512484774,
+      "loss": 0.27983531951904295,
+      "mean_token_accuracy": 0.9093958771228791,
+      "num_tokens": 4365381.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.33449163861572745,
+      "epoch": 4.680053547523427,
+      "grad_norm": 0.330709844827652,
+      "learning_rate": 0.00015007341986449012,
+      "loss": 0.28533639907836916,
+      "mean_token_accuracy": 0.9082232251763344,
+      "num_tokens": 4490711.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.33353066638112067,
+      "epoch": 4.813922356091031,
+      "grad_norm": 0.3990134298801422,
+      "learning_rate": 0.00014480673387425272,
+      "loss": 0.28489078521728517,
+      "mean_token_accuracy": 0.908001911342144,
+      "num_tokens": 4618532.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.3272412090748549,
+      "epoch": 4.947791164658635,
+      "grad_norm": 0.3183020353317261,
+      "learning_rate": 0.00013947933469084315,
+      "loss": 0.2772365379333496,
+      "mean_token_accuracy": 0.908946952521801,
+      "num_tokens": 4752261.0,
+      "step": 1850
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.3932068006694317,
+      "eval_loss": 0.5647156834602356,
+      "eval_mean_token_accuracy": 0.85078628718853,
+      "eval_num_tokens": 4798715.0,
+      "eval_runtime": 51.3578,
+      "eval_samples_per_second": 31.115,
+      "eval_steps_per_second": 3.894,
+      "step": 1870
+    },
+    {
+      "entropy": 0.2832252390005372,
+      "epoch": 5.080321285140562,
+      "grad_norm": 0.3697633147239685,
+      "learning_rate": 0.00013410282204620014,
+      "loss": 0.2279021453857422,
+      "mean_token_accuracy": 0.9252248072262966,
+      "num_tokens": 4879271.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.250804705247283,
+      "epoch": 5.214190093708166,
+      "grad_norm": 0.3890162706375122,
+      "learning_rate": 0.00012868890261055722,
+      "loss": 0.1980854606628418,
+      "mean_token_accuracy": 0.9338876655697823,
+      "num_tokens": 5005076.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.2531572911888361,
+      "epoch": 5.34805890227577,
+      "grad_norm": 0.43466225266456604,
+      "learning_rate": 0.0001232493645026623,
+      "loss": 0.20114482879638673,
+      "mean_token_accuracy": 0.9317018255591393,
+      "num_tokens": 5133591.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.25918263107538225,
+      "epoch": 5.481927710843373,
+      "grad_norm": 0.38253673911094666,
+      "learning_rate": 0.00011779605162265297,
+      "loss": 0.2056061363220215,
+      "mean_token_accuracy": 0.9302830925583839,
+      "num_tokens": 5257252.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.2553627458959818,
+      "epoch": 5.615796519410977,
+      "grad_norm": 0.4536231458187103,
+      "learning_rate": 0.00011234083786347563,
+      "loss": 0.20531394958496094,
+      "mean_token_accuracy": 0.9302299374341965,
+      "num_tokens": 5388652.0,
+      "step": 2100
+    },
+    {
+      "entropy": 0.2575570110231638,
+      "epoch": 5.749665327978581,
+      "grad_norm": 0.36399731040000916,
+      "learning_rate": 0.00010689560125699833,
+      "loss": 0.2048162841796875,
+      "mean_token_accuracy": 0.9306997761130333,
+      "num_tokens": 5515488.0,
+      "step": 2150
+    },
+    {
+      "entropy": 0.24660897620022296,
+      "epoch": 5.883534136546185,
+      "grad_norm": 0.43602702021598816,
+      "learning_rate": 0.00010147219811111233,
+      "loss": 0.1986431884765625,
+      "mean_token_accuracy": 0.9335029146075249,
+      "num_tokens": 5644323.0,
+      "step": 2200
+    },
+    {
+      "epoch": 6.0,
+      "eval_entropy": 0.32280903935432437,
+      "eval_loss": 0.6383674144744873,
+      "eval_mean_token_accuracy": 0.849581449329853,
+      "eval_num_tokens": 5758458.0,
+      "eval_runtime": 51.3251,
+      "eval_samples_per_second": 31.135,
+      "eval_steps_per_second": 3.897,
+      "step": 2244
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.465057375940987e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2618/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2618/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2618/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2618/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2618/trainer_state.json

@@ -0,0 +1,631 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 2618,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    },
+    {
+      "entropy": 0.41342759042075183,
+      "epoch": 3.074966532797858,
+      "grad_norm": 0.34829744696617126,
+      "learning_rate": 0.0002045690972655427,
+      "loss": 0.3644887542724609,
+      "mean_token_accuracy": 0.8879408977850519,
+      "num_tokens": 2955424.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.3948360003530979,
+      "epoch": 3.208835341365462,
+      "grad_norm": 0.312427282333374,
+      "learning_rate": 0.00020086347837665854,
+      "loss": 0.34146129608154296,
+      "mean_token_accuracy": 0.8914449456334114,
+      "num_tokens": 3078799.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.38602679744362833,
+      "epoch": 3.3427041499330654,
+      "grad_norm": 0.2715625762939453,
+      "learning_rate": 0.0001969750898863629,
+      "loss": 0.34105979919433593,
+      "mean_token_accuracy": 0.8928995525836945,
+      "num_tokens": 3211945.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.3923524462431669,
+      "epoch": 3.4765729585006695,
+      "grad_norm": 0.27295640110969543,
+      "learning_rate": 0.00019291239826442992,
+      "loss": 0.3473458099365234,
+      "mean_token_accuracy": 0.8913829082250595,
+      "num_tokens": 3343933.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.40172914519906044,
+      "epoch": 3.610441767068273,
+      "grad_norm": 0.32684192061424255,
+      "learning_rate": 0.0001886842495034615,
+      "loss": 0.35543827056884764,
+      "mean_token_accuracy": 0.8904094022512435,
+      "num_tokens": 3470087.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4010512103140354,
+      "epoch": 3.7443105756358768,
+      "grad_norm": 0.23922984302043915,
+      "learning_rate": 0.00018429984985786734,
+      "loss": 0.3535212326049805,
+      "mean_token_accuracy": 0.8891122484207153,
+      "num_tokens": 3590858.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.39226082623004915,
+      "epoch": 3.878179384203481,
+      "grad_norm": 0.25130993127822876,
+      "learning_rate": 0.00017976874579842046,
+      "loss": 0.3484851837158203,
+      "mean_token_accuracy": 0.8916165816783905,
+      "num_tokens": 3725806.0,
+      "step": 1450
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.45332007586956025,
+      "eval_loss": 0.5243425965309143,
+      "eval_mean_token_accuracy": 0.8532214590907097,
+      "eval_num_tokens": 3838972.0,
+      "eval_runtime": 51.3837,
+      "eval_samples_per_second": 31.099,
+      "eval_steps_per_second": 3.892,
+      "step": 1496
+    },
+    {
+      "entropy": 0.39358587043754983,
+      "epoch": 4.010709504685408,
+      "grad_norm": 0.27960067987442017,
+      "learning_rate": 0.0001751008032260355,
+      "loss": 0.34616813659667967,
+      "mean_token_accuracy": 0.8923709010234987,
+      "num_tokens": 3849380.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.3206369188427925,
+      "epoch": 4.144578313253012,
+      "grad_norm": 0.37261858582496643,
+      "learning_rate": 0.00017030618599002818,
+      "loss": 0.2684581565856934,
+      "mean_token_accuracy": 0.9131761506199837,
+      "num_tokens": 3976694.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3254615054279566,
+      "epoch": 4.278447121820616,
+      "grad_norm": 0.39803799986839294,
+      "learning_rate": 0.00016539533375763032,
+      "loss": 0.2769618606567383,
+      "mean_token_accuracy": 0.9102409112453461,
+      "num_tokens": 4110204.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.32003962114453316,
+      "epoch": 4.412315930388219,
+      "grad_norm": 0.35707366466522217,
+      "learning_rate": 0.0001603789392829468,
+      "loss": 0.2749842834472656,
+      "mean_token_accuracy": 0.910626070201397,
+      "num_tokens": 4240883.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.32672152675688265,
+      "epoch": 4.546184738955823,
+      "grad_norm": 0.47052621841430664,
+      "learning_rate": 0.00015526792512484774,
+      "loss": 0.27983531951904295,
+      "mean_token_accuracy": 0.9093958771228791,
+      "num_tokens": 4365381.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.33449163861572745,
+      "epoch": 4.680053547523427,
+      "grad_norm": 0.330709844827652,
+      "learning_rate": 0.00015007341986449012,
+      "loss": 0.28533639907836916,
+      "mean_token_accuracy": 0.9082232251763344,
+      "num_tokens": 4490711.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.33353066638112067,
+      "epoch": 4.813922356091031,
+      "grad_norm": 0.3990134298801422,
+      "learning_rate": 0.00014480673387425272,
+      "loss": 0.28489078521728517,
+      "mean_token_accuracy": 0.908001911342144,
+      "num_tokens": 4618532.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.3272412090748549,
+      "epoch": 4.947791164658635,
+      "grad_norm": 0.3183020353317261,
+      "learning_rate": 0.00013947933469084315,
+      "loss": 0.2772365379333496,
+      "mean_token_accuracy": 0.908946952521801,
+      "num_tokens": 4752261.0,
+      "step": 1850
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.3932068006694317,
+      "eval_loss": 0.5647156834602356,
+      "eval_mean_token_accuracy": 0.85078628718853,
+      "eval_num_tokens": 4798715.0,
+      "eval_runtime": 51.3578,
+      "eval_samples_per_second": 31.115,
+      "eval_steps_per_second": 3.894,
+      "step": 1870
+    },
+    {
+      "entropy": 0.2832252390005372,
+      "epoch": 5.080321285140562,
+      "grad_norm": 0.3697633147239685,
+      "learning_rate": 0.00013410282204620014,
+      "loss": 0.2279021453857422,
+      "mean_token_accuracy": 0.9252248072262966,
+      "num_tokens": 4879271.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.250804705247283,
+      "epoch": 5.214190093708166,
+      "grad_norm": 0.3890162706375122,
+      "learning_rate": 0.00012868890261055722,
+      "loss": 0.1980854606628418,
+      "mean_token_accuracy": 0.9338876655697823,
+      "num_tokens": 5005076.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.2531572911888361,
+      "epoch": 5.34805890227577,
+      "grad_norm": 0.43466225266456604,
+      "learning_rate": 0.0001232493645026623,
+      "loss": 0.20114482879638673,
+      "mean_token_accuracy": 0.9317018255591393,
+      "num_tokens": 5133591.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.25918263107538225,
+      "epoch": 5.481927710843373,
+      "grad_norm": 0.38253673911094666,
+      "learning_rate": 0.00011779605162265297,
+      "loss": 0.2056061363220215,
+      "mean_token_accuracy": 0.9302830925583839,
+      "num_tokens": 5257252.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.2553627458959818,
+      "epoch": 5.615796519410977,
+      "grad_norm": 0.4536231458187103,
+      "learning_rate": 0.00011234083786347563,
+      "loss": 0.20531394958496094,
+      "mean_token_accuracy": 0.9302299374341965,
+      "num_tokens": 5388652.0,
+      "step": 2100
+    },
+    {
+      "entropy": 0.2575570110231638,
+      "epoch": 5.749665327978581,
+      "grad_norm": 0.36399731040000916,
+      "learning_rate": 0.00010689560125699833,
+      "loss": 0.2048162841796875,
+      "mean_token_accuracy": 0.9306997761130333,
+      "num_tokens": 5515488.0,
+      "step": 2150
+    },
+    {
+      "entropy": 0.24660897620022296,
+      "epoch": 5.883534136546185,
+      "grad_norm": 0.43602702021598816,
+      "learning_rate": 0.00010147219811111233,
+      "loss": 0.1986431884765625,
+      "mean_token_accuracy": 0.9335029146075249,
+      "num_tokens": 5644323.0,
+      "step": 2200
+    },
+    {
+      "epoch": 6.0,
+      "eval_entropy": 0.32280903935432437,
+      "eval_loss": 0.6383674144744873,
+      "eval_mean_token_accuracy": 0.849581449329853,
+      "eval_num_tokens": 5758458.0,
+      "eval_runtime": 51.3251,
+      "eval_samples_per_second": 31.135,
+      "eval_steps_per_second": 3.897,
+      "step": 2244
+    },
+    {
+      "entropy": 0.24638524448329752,
+      "epoch": 6.016064257028113,
+      "grad_norm": 0.3538878262042999,
+      "learning_rate": 9.608243719413435e-05,
+      "loss": 0.19203664779663085,
+      "mean_token_accuracy": 0.9353304363862432,
+      "num_tokens": 5773027.0,
+      "step": 2250
+    },
+    {
+      "entropy": 0.17814311504364014,
+      "epoch": 6.149933065595716,
+      "grad_norm": 0.3886161148548126,
+      "learning_rate": 9.07380540227205e-05,
+      "loss": 0.12442682266235351,
+      "mean_token_accuracy": 0.9582101872563362,
+      "num_tokens": 5904840.0,
+      "step": 2300
+    },
+    {
+      "entropy": 0.17237136442214251,
+      "epoch": 6.28380187416332,
+      "grad_norm": 0.38807374238967896,
+      "learning_rate": 8.545068530927622e-05,
+      "loss": 0.12445520401000977,
+      "mean_token_accuracy": 0.9580146077275277,
+      "num_tokens": 6037457.0,
+      "step": 2350
+    },
+    {
+      "entropy": 0.18334724467247723,
+      "epoch": 6.417670682730924,
+      "grad_norm": 0.48334068059921265,
+      "learning_rate": 8.023184362449975e-05,
+      "loss": 0.12853397369384767,
+      "mean_token_accuracy": 0.956232733130455,
+      "num_tokens": 6161042.0,
+      "step": 2400
+    },
+    {
+      "entropy": 0.17894859783351422,
+      "epoch": 6.551539491298527,
+      "grad_norm": 0.3343403935432434,
+      "learning_rate": 7.509289233022861e-05,
+      "loss": 0.12605968475341797,
+      "mean_token_accuracy": 0.9566894540190697,
+      "num_tokens": 6291748.0,
+      "step": 2450
+    },
+    {
+      "entropy": 0.17900108266621828,
+      "epoch": 6.685408299866131,
+      "grad_norm": 0.41615480184555054,
+      "learning_rate": 7.00450208371691e-05,
+      "loss": 0.12843725204467774,
+      "mean_token_accuracy": 0.956638223528862,
+      "num_tokens": 6419265.0,
+      "step": 2500
+    },
+    {
+      "entropy": 0.18341445792466401,
+      "epoch": 6.8192771084337345,
+      "grad_norm": 0.3722545802593231,
+      "learning_rate": 6.509922024138231e-05,
+      "loss": 0.13251185417175293,
+      "mean_token_accuracy": 0.9549383011460304,
+      "num_tokens": 6544758.0,
+      "step": 2550
+    },
+    {
+      "entropy": 0.17789534136652946,
+      "epoch": 6.953145917001339,
+      "grad_norm": 0.33068087697029114,
+      "learning_rate": 6.02662593925748e-05,
+      "loss": 0.126302547454834,
+      "mean_token_accuracy": 0.9568320420384407,
+      "num_tokens": 6674626.0,
+      "step": 2600
+    },
+    {
+      "epoch": 7.0,
+      "eval_entropy": 0.2639452085644007,
+      "eval_loss": 0.7510635852813721,
+      "eval_mean_token_accuracy": 0.8456276795268058,
+      "eval_num_tokens": 6718201.0,
+      "eval_runtime": 51.3435,
+      "eval_samples_per_second": 31.124,
+      "eval_steps_per_second": 3.895,
+      "step": 2618
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.8789404553397146e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2992/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2992/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2992/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2992/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-2992/trainer_state.json

@@ -0,0 +1,712 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 2992,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.2357245934009553,
+      "epoch": 0.13386880856760375,
+      "grad_norm": 1.1940524578094482,
+      "learning_rate": 3.0637587087867373e-05,
+      "loss": 2.0509376525878906,
+      "mean_token_accuracy": 0.6194104523956776,
+      "num_tokens": 127704.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.8588631230592728,
+      "epoch": 0.2677376171352075,
+      "grad_norm": 0.6621095538139343,
+      "learning_rate": 6.190043105507899e-05,
+      "loss": 0.8043325805664062,
+      "mean_token_accuracy": 0.7855384379625321,
+      "num_tokens": 256077.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.6726470375061036,
+      "epoch": 0.40160642570281124,
+      "grad_norm": 0.5255736112594604,
+      "learning_rate": 9.316327502229059e-05,
+      "loss": 0.6415129089355469,
+      "mean_token_accuracy": 0.8183896672725678,
+      "num_tokens": 387735.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6119109469652176,
+      "epoch": 0.535475234270415,
+      "grad_norm": 0.45975860953330994,
+      "learning_rate": 0.0001244261189895022,
+      "loss": 0.576114501953125,
+      "mean_token_accuracy": 0.8376823288202285,
+      "num_tokens": 522202.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.5972725109755993,
+      "epoch": 0.6693440428380187,
+      "grad_norm": 0.40055611729621887,
+      "learning_rate": 0.0001556889629567138,
+      "loss": 0.5613529205322265,
+      "mean_token_accuracy": 0.8418879929184914,
+      "num_tokens": 648663.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.5673307004570961,
+      "epoch": 0.8032128514056225,
+      "grad_norm": 0.7427454590797424,
+      "learning_rate": 0.0001869518069239254,
+      "loss": 0.5325925827026368,
+      "mean_token_accuracy": 0.8488189685344696,
+      "num_tokens": 778245.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.5615521620213986,
+      "epoch": 0.9370816599732262,
+      "grad_norm": 0.4017025828361511,
+      "learning_rate": 0.000218214650891137,
+      "loss": 0.522266616821289,
+      "mean_token_accuracy": 0.8500416606664658,
+      "num_tokens": 905328.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.5758187392354012,
+      "eval_loss": 0.5664522051811218,
+      "eval_mean_token_accuracy": 0.8383084440231323,
+      "eval_num_tokens": 959743.0,
+      "eval_runtime": 51.7964,
+      "eval_samples_per_second": 30.852,
+      "eval_steps_per_second": 3.861,
+      "step": 374
+    },
+    {
+      "entropy": 0.5440343178883947,
+      "epoch": 1.069611780455154,
+      "grad_norm": 0.3084093928337097,
+      "learning_rate": 0.00023381424541885068,
+      "loss": 0.5051319122314453,
+      "mean_token_accuracy": 0.8554374280602041,
+      "num_tokens": 1022581.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.5245272906124592,
+      "epoch": 1.2034805890227578,
+      "grad_norm": 0.3342743515968323,
+      "learning_rate": 0.00023355972972676628,
+      "loss": 0.4855318450927734,
+      "mean_token_accuracy": 0.8584430786967278,
+      "num_tokens": 1151430.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5234513898193837,
+      "epoch": 1.3373493975903614,
+      "grad_norm": 0.26953577995300293,
+      "learning_rate": 0.00023305125251804043,
+      "loss": 0.4811768341064453,
+      "mean_token_accuracy": 0.8606387570500373,
+      "num_tokens": 1282481.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.5147616830468178,
+      "epoch": 1.4712182061579653,
+      "grad_norm": 0.28791534900665283,
+      "learning_rate": 0.0002322899209369128,
+      "loss": 0.4762062835693359,
+      "mean_token_accuracy": 0.8612849581241607,
+      "num_tokens": 1414166.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5060296922922134,
+      "epoch": 1.605087014725569,
+      "grad_norm": 0.2496163249015808,
+      "learning_rate": 0.0002312773926857543,
+      "loss": 0.4695176315307617,
+      "mean_token_accuracy": 0.8636157616972924,
+      "num_tokens": 1547902.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5008939932286739,
+      "epoch": 1.7389558232931726,
+      "grad_norm": 0.2329542636871338,
+      "learning_rate": 0.00023001587241563198,
+      "loss": 0.46317913055419924,
+      "mean_token_accuracy": 0.8652430367469788,
+      "num_tokens": 1678635.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.4986082436144352,
+      "epoch": 1.8728246318607764,
+      "grad_norm": 0.2459402084350586,
+      "learning_rate": 0.00022850810692596235,
+      "loss": 0.4617066192626953,
+      "mean_token_accuracy": 0.8672981086373329,
+      "num_tokens": 1803328.0,
+      "step": 700
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.546445109397173,
+      "eval_loss": 0.5307288765907288,
+      "eval_mean_token_accuracy": 0.8434162598848343,
+      "eval_num_tokens": 1919486.0,
+      "eval_runtime": 51.3627,
+      "eval_samples_per_second": 31.112,
+      "eval_steps_per_second": 3.894,
+      "step": 748
+    },
+    {
+      "entropy": 0.5069879525237613,
+      "epoch": 2.005354752342704,
+      "grad_norm": 0.21533548831939697,
+      "learning_rate": 0.00022675737918370628,
+      "loss": 0.4585062026977539,
+      "mean_token_accuracy": 0.865652882390552,
+      "num_tokens": 1925307.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.4457989126443863,
+      "epoch": 2.139223560910308,
+      "grad_norm": 0.24705350399017334,
+      "learning_rate": 0.00022476750117512737,
+      "loss": 0.4026710891723633,
+      "mean_token_accuracy": 0.8774338760972022,
+      "num_tokens": 2055627.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.45084999009966853,
+      "epoch": 2.2730923694779115,
+      "grad_norm": 0.26643863320350647,
+      "learning_rate": 0.00022254280560567822,
+      "loss": 0.40950340270996094,
+      "mean_token_accuracy": 0.8752825647592545,
+      "num_tokens": 2177850.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.4516983331739903,
+      "epoch": 2.4069611780455156,
+      "grad_norm": 0.26972198486328125,
+      "learning_rate": 0.00022008813646608725,
+      "loss": 0.4115512466430664,
+      "mean_token_accuracy": 0.8761435833573341,
+      "num_tokens": 2304612.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.45280722543597224,
+      "epoch": 2.540829986613119,
+      "grad_norm": 0.2643600106239319,
+      "learning_rate": 0.00021740883848518684,
+      "loss": 0.41181053161621095,
+      "mean_token_accuracy": 0.8756420350074768,
+      "num_tokens": 2430946.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.4487862553447485,
+      "epoch": 2.674698795180723,
+      "grad_norm": 0.2849285304546356,
+      "learning_rate": 0.00021451074549244846,
+      "loss": 0.4094270706176758,
+      "mean_token_accuracy": 0.8771369129419326,
+      "num_tokens": 2557241.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.45000545382499696,
+      "epoch": 2.8085676037483265,
+      "grad_norm": 0.24081671237945557,
+      "learning_rate": 0.0002114001677155633,
+      "loss": 0.4073855972290039,
+      "mean_token_accuracy": 0.8779223081469536,
+      "num_tokens": 2692775.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.4576124830543995,
+      "epoch": 2.9424364123159306,
+      "grad_norm": 0.2341010719537735,
+      "learning_rate": 0.00020808387804072673,
+      "loss": 0.4154107666015625,
+      "mean_token_accuracy": 0.8756425747275353,
+      "num_tokens": 2823060.0,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "eval_entropy": 0.4812743577361107,
+      "eval_loss": 0.523208498954773,
+      "eval_mean_token_accuracy": 0.8494922530651092,
+      "eval_num_tokens": 2879229.0,
+      "eval_runtime": 51.3707,
+      "eval_samples_per_second": 31.107,
+      "eval_steps_per_second": 3.893,
+      "step": 1122
+    },
+    {
+      "entropy": 0.41342759042075183,
+      "epoch": 3.074966532797858,
+      "grad_norm": 0.34829744696617126,
+      "learning_rate": 0.0002045690972655427,
+      "loss": 0.3644887542724609,
+      "mean_token_accuracy": 0.8879408977850519,
+      "num_tokens": 2955424.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.3948360003530979,
+      "epoch": 3.208835341365462,
+      "grad_norm": 0.312427282333374,
+      "learning_rate": 0.00020086347837665854,
+      "loss": 0.34146129608154296,
+      "mean_token_accuracy": 0.8914449456334114,
+      "num_tokens": 3078799.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.38602679744362833,
+      "epoch": 3.3427041499330654,
+      "grad_norm": 0.2715625762939453,
+      "learning_rate": 0.0001969750898863629,
+      "loss": 0.34105979919433593,
+      "mean_token_accuracy": 0.8928995525836945,
+      "num_tokens": 3211945.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.3923524462431669,
+      "epoch": 3.4765729585006695,
+      "grad_norm": 0.27295640110969543,
+      "learning_rate": 0.00019291239826442992,
+      "loss": 0.3473458099365234,
+      "mean_token_accuracy": 0.8913829082250595,
+      "num_tokens": 3343933.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.40172914519906044,
+      "epoch": 3.610441767068273,
+      "grad_norm": 0.32684192061424255,
+      "learning_rate": 0.0001886842495034615,
+      "loss": 0.35543827056884764,
+      "mean_token_accuracy": 0.8904094022512435,
+      "num_tokens": 3470087.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.4010512103140354,
+      "epoch": 3.7443105756358768,
+      "grad_norm": 0.23922984302043915,
+      "learning_rate": 0.00018429984985786734,
+      "loss": 0.3535212326049805,
+      "mean_token_accuracy": 0.8891122484207153,
+      "num_tokens": 3590858.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.39226082623004915,
+      "epoch": 3.878179384203481,
+      "grad_norm": 0.25130993127822876,
+      "learning_rate": 0.00017976874579842046,
+      "loss": 0.3484851837158203,
+      "mean_token_accuracy": 0.8916165816783905,
+      "num_tokens": 3725806.0,
+      "step": 1450
+    },
+    {
+      "epoch": 4.0,
+      "eval_entropy": 0.45332007586956025,
+      "eval_loss": 0.5243425965309143,
+      "eval_mean_token_accuracy": 0.8532214590907097,
+      "eval_num_tokens": 3838972.0,
+      "eval_runtime": 51.3837,
+      "eval_samples_per_second": 31.099,
+      "eval_steps_per_second": 3.892,
+      "step": 1496
+    },
+    {
+      "entropy": 0.39358587043754983,
+      "epoch": 4.010709504685408,
+      "grad_norm": 0.27960067987442017,
+      "learning_rate": 0.0001751008032260355,
+      "loss": 0.34616813659667967,
+      "mean_token_accuracy": 0.8923709010234987,
+      "num_tokens": 3849380.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.3206369188427925,
+      "epoch": 4.144578313253012,
+      "grad_norm": 0.37261858582496643,
+      "learning_rate": 0.00017030618599002818,
+      "loss": 0.2684581565856934,
+      "mean_token_accuracy": 0.9131761506199837,
+      "num_tokens": 3976694.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.3254615054279566,
+      "epoch": 4.278447121820616,
+      "grad_norm": 0.39803799986839294,
+      "learning_rate": 0.00016539533375763032,
+      "loss": 0.2769618606567383,
+      "mean_token_accuracy": 0.9102409112453461,
+      "num_tokens": 4110204.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.32003962114453316,
+      "epoch": 4.412315930388219,
+      "grad_norm": 0.35707366466522217,
+      "learning_rate": 0.0001603789392829468,
+      "loss": 0.2749842834472656,
+      "mean_token_accuracy": 0.910626070201397,
+      "num_tokens": 4240883.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.32672152675688265,
+      "epoch": 4.546184738955823,
+      "grad_norm": 0.47052621841430664,
+      "learning_rate": 0.00015526792512484774,
+      "loss": 0.27983531951904295,
+      "mean_token_accuracy": 0.9093958771228791,
+      "num_tokens": 4365381.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.33449163861572745,
+      "epoch": 4.680053547523427,
+      "grad_norm": 0.330709844827652,
+      "learning_rate": 0.00015007341986449012,
+      "loss": 0.28533639907836916,
+      "mean_token_accuracy": 0.9082232251763344,
+      "num_tokens": 4490711.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.33353066638112067,
+      "epoch": 4.813922356091031,
+      "grad_norm": 0.3990134298801422,
+      "learning_rate": 0.00014480673387425272,
+      "loss": 0.28489078521728517,
+      "mean_token_accuracy": 0.908001911342144,
+      "num_tokens": 4618532.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.3272412090748549,
+      "epoch": 4.947791164658635,
+      "grad_norm": 0.3183020353317261,
+      "learning_rate": 0.00013947933469084315,
+      "loss": 0.2772365379333496,
+      "mean_token_accuracy": 0.908946952521801,
+      "num_tokens": 4752261.0,
+      "step": 1850
+    },
+    {
+      "epoch": 5.0,
+      "eval_entropy": 0.3932068006694317,
+      "eval_loss": 0.5647156834602356,
+      "eval_mean_token_accuracy": 0.85078628718853,
+      "eval_num_tokens": 4798715.0,
+      "eval_runtime": 51.3578,
+      "eval_samples_per_second": 31.115,
+      "eval_steps_per_second": 3.894,
+      "step": 1870
+    },
+    {
+      "entropy": 0.2832252390005372,
+      "epoch": 5.080321285140562,
+      "grad_norm": 0.3697633147239685,
+      "learning_rate": 0.00013410282204620014,
+      "loss": 0.2279021453857422,
+      "mean_token_accuracy": 0.9252248072262966,
+      "num_tokens": 4879271.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.250804705247283,
+      "epoch": 5.214190093708166,
+      "grad_norm": 0.3890162706375122,
+      "learning_rate": 0.00012868890261055722,
+      "loss": 0.1980854606628418,
+      "mean_token_accuracy": 0.9338876655697823,
+      "num_tokens": 5005076.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.2531572911888361,
+      "epoch": 5.34805890227577,
+      "grad_norm": 0.43466225266456604,
+      "learning_rate": 0.0001232493645026623,
+      "loss": 0.20114482879638673,
+      "mean_token_accuracy": 0.9317018255591393,
+      "num_tokens": 5133591.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.25918263107538225,
+      "epoch": 5.481927710843373,
+      "grad_norm": 0.38253673911094666,
+      "learning_rate": 0.00011779605162265297,
+      "loss": 0.2056061363220215,
+      "mean_token_accuracy": 0.9302830925583839,
+      "num_tokens": 5257252.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.2553627458959818,
+      "epoch": 5.615796519410977,
+      "grad_norm": 0.4536231458187103,
+      "learning_rate": 0.00011234083786347563,
+      "loss": 0.20531394958496094,
+      "mean_token_accuracy": 0.9302299374341965,
+      "num_tokens": 5388652.0,
+      "step": 2100
+    },
+    {
+      "entropy": 0.2575570110231638,
+      "epoch": 5.749665327978581,
+      "grad_norm": 0.36399731040000916,
+      "learning_rate": 0.00010689560125699833,
+      "loss": 0.2048162841796875,
+      "mean_token_accuracy": 0.9306997761130333,
+      "num_tokens": 5515488.0,
+      "step": 2150
+    },
+    {
+      "entropy": 0.24660897620022296,
+      "epoch": 5.883534136546185,
+      "grad_norm": 0.43602702021598816,
+      "learning_rate": 0.00010147219811111233,
+      "loss": 0.1986431884765625,
+      "mean_token_accuracy": 0.9335029146075249,
+      "num_tokens": 5644323.0,
+      "step": 2200
+    },
+    {
+      "epoch": 6.0,
+      "eval_entropy": 0.32280903935432437,
+      "eval_loss": 0.6383674144744873,
+      "eval_mean_token_accuracy": 0.849581449329853,
+      "eval_num_tokens": 5758458.0,
+      "eval_runtime": 51.3251,
+      "eval_samples_per_second": 31.135,
+      "eval_steps_per_second": 3.897,
+      "step": 2244
+    },
+    {
+      "entropy": 0.24638524448329752,
+      "epoch": 6.016064257028113,
+      "grad_norm": 0.3538878262042999,
+      "learning_rate": 9.608243719413435e-05,
+      "loss": 0.19203664779663085,
+      "mean_token_accuracy": 0.9353304363862432,
+      "num_tokens": 5773027.0,
+      "step": 2250
+    },
+    {
+      "entropy": 0.17814311504364014,
+      "epoch": 6.149933065595716,
+      "grad_norm": 0.3886161148548126,
+      "learning_rate": 9.07380540227205e-05,
+      "loss": 0.12442682266235351,
+      "mean_token_accuracy": 0.9582101872563362,
+      "num_tokens": 5904840.0,
+      "step": 2300
+    },
+    {
+      "entropy": 0.17237136442214251,
+      "epoch": 6.28380187416332,
+      "grad_norm": 0.38807374238967896,
+      "learning_rate": 8.545068530927622e-05,
+      "loss": 0.12445520401000977,
+      "mean_token_accuracy": 0.9580146077275277,
+      "num_tokens": 6037457.0,
+      "step": 2350
+    },
+    {
+      "entropy": 0.18334724467247723,
+      "epoch": 6.417670682730924,
+      "grad_norm": 0.48334068059921265,
+      "learning_rate": 8.023184362449975e-05,
+      "loss": 0.12853397369384767,
+      "mean_token_accuracy": 0.956232733130455,
+      "num_tokens": 6161042.0,
+      "step": 2400
+    },
+    {
+      "entropy": 0.17894859783351422,
+      "epoch": 6.551539491298527,
+      "grad_norm": 0.3343403935432434,
+      "learning_rate": 7.509289233022861e-05,
+      "loss": 0.12605968475341797,
+      "mean_token_accuracy": 0.9566894540190697,
+      "num_tokens": 6291748.0,
+      "step": 2450
+    },
+    {
+      "entropy": 0.17900108266621828,
+      "epoch": 6.685408299866131,
+      "grad_norm": 0.41615480184555054,
+      "learning_rate": 7.00450208371691e-05,
+      "loss": 0.12843725204467774,
+      "mean_token_accuracy": 0.956638223528862,
+      "num_tokens": 6419265.0,
+      "step": 2500
+    },
+    {
+      "entropy": 0.18341445792466401,
+      "epoch": 6.8192771084337345,
+      "grad_norm": 0.3722545802593231,
+      "learning_rate": 6.509922024138231e-05,
+      "loss": 0.13251185417175293,
+      "mean_token_accuracy": 0.9549383011460304,
+      "num_tokens": 6544758.0,
+      "step": 2550
+    },
+    {
+      "entropy": 0.17789534136652946,
+      "epoch": 6.953145917001339,
+      "grad_norm": 0.33068087697029114,
+      "learning_rate": 6.02662593925748e-05,
+      "loss": 0.126302547454834,
+      "mean_token_accuracy": 0.9568320420384407,
+      "num_tokens": 6674626.0,
+      "step": 2600
+    },
+    {
+      "epoch": 7.0,
+      "eval_entropy": 0.2639452085644007,
+      "eval_loss": 0.7510635852813721,
+      "eval_mean_token_accuracy": 0.8456276795268058,
+      "eval_num_tokens": 6718201.0,
+      "eval_runtime": 51.3435,
+      "eval_samples_per_second": 31.124,
+      "eval_steps_per_second": 3.895,
+      "step": 2618
+    },
+    {
+      "entropy": 0.15149398003187445,
+      "epoch": 7.085676037483267,
+      "grad_norm": 0.3362366855144501,
+      "learning_rate": 5.5556661446302733e-05,
+      "loss": 0.09618576049804688,
+      "mean_token_accuracy": 0.9674260706612559,
+      "num_tokens": 6803830.0,
+      "step": 2650
+    },
+    {
+      "entropy": 0.1279136904887855,
+      "epoch": 7.21954484605087,
+      "grad_norm": 0.2925446033477783,
+      "learning_rate": 5.0980680951143166e-05,
+      "loss": 0.07902004718780517,
+      "mean_token_accuracy": 0.9733691918849945,
+      "num_tokens": 6936289.0,
+      "step": 2700
+    },
+    {
+      "entropy": 0.13701909594237804,
+      "epoch": 7.353413654618474,
+      "grad_norm": 0.2659797668457031,
+      "learning_rate": 4.6548281520723104e-05,
+      "loss": 0.08250561714172364,
+      "mean_token_accuracy": 0.971816695034504,
+      "num_tokens": 7057823.0,
+      "step": 2750
+    },
+    {
+      "entropy": 0.12771729078143834,
+      "epoch": 7.4872824631860775,
+      "grad_norm": 0.34247028827667236,
+      "learning_rate": 4.2269114139222296e-05,
+      "loss": 0.08109721183776855,
+      "mean_token_accuracy": 0.9735026282072067,
+      "num_tokens": 7187020.0,
+      "step": 2800
+    },
+    {
+      "entropy": 0.12350119687616826,
+      "epoch": 7.621151271753681,
+      "grad_norm": 0.40673893690109253,
+      "learning_rate": 3.8152496147586614e-05,
+      "loss": 0.07707037448883057,
+      "mean_token_accuracy": 0.9743763041496277,
+      "num_tokens": 7323457.0,
+      "step": 2850
+    },
+    {
+      "entropy": 0.12642662361264229,
+      "epoch": 7.755020080321285,
+      "grad_norm": 0.3064998984336853,
+      "learning_rate": 3.4207390956206875e-05,
+      "loss": 0.07985133647918702,
+      "mean_token_accuracy": 0.9733496251702308,
+      "num_tokens": 7451765.0,
+      "step": 2900
+    },
+    {
+      "entropy": 0.1269074559956789,
+      "epoch": 7.888888888888889,
+      "grad_norm": 0.24833732843399048,
+      "learning_rate": 3.0442388528236647e-05,
+      "loss": 0.0821513843536377,
+      "mean_token_accuracy": 0.9733479696512223,
+      "num_tokens": 7578222.0,
+      "step": 2950
+    },
+    {
+      "epoch": 8.0,
+      "eval_entropy": 0.2316589906811714,
+      "eval_loss": 0.851987898349762,
+      "eval_mean_token_accuracy": 0.8467304027080536,
+      "eval_num_tokens": 7677944.0,
+      "eval_runtime": 51.3501,
+      "eval_samples_per_second": 31.12,
+      "eval_steps_per_second": 3.895,
+      "step": 2992
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.2928370910154854e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-3366/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-4B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-4B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-3366/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0094403300459725,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_code_Swedish/Qwen3-4B-Base_code_features_structural_train_code_features_structural_test2/checkpoint-3366/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}