fix: reduce number of languages, add lora

adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "RobertaForSequenceClassification",
+    "parent_library": "transformers.models.roberta.modeling_roberta"
+  },
+  "base_model_name_or_path": "huggingface/CodeBERTa-small-v1",
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": [
+    "classifier"
+  ],
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "value",
+    "query"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2402db45b928e2129ec461c2b8c3cfe2814fc4cdc2c7fc1ef90ce0c52033df25
+size 3570544

config.json

@@ -11,114 +11,26 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "Assembly",
-    "1": "Batchfile",
-    "2": "C",
-    "3": "C#",
-    "4": "C++",
-    "5": "CMake",
-    "6": "COBOL",
-    "7": "CSS",
-    "8": "CSV",
-    "9": "Clojure",
-    "10": "CoffeeScript",
-    "11": "DM",
-    "12": "Dart",
-    "13": "Dockerfile",
-    "14": "Elixir",
-    "15": "Erlang",
-    "16": "Fortran",
-    "17": "Go",
-    "18": "Groovy",
-    "19": "HTML",
-    "20": "Haskell",
-    "21": "INI",
-    "22": "JSON",
-    "23": "Java",
-    "24": "JavaScript",
-    "25": "Julia",
-    "26": "Kotlin",
-    "27": "Lisp",
-    "28": "Lua",
-    "29": "Makefile",
-    "30": "Matlab",
-    "31": "OCaml",
-    "32": "Objective-C",
-    "33": "PHP",
-    "34": "Pascal",
-    "35": "Perl",
-    "36": "PowerShell",
-    "37": "Prolog",
-    "38": "Python",
-    "39": "R",
-    "40": "Ruby",
-    "41": "Rust",
-    "42": "SQL",
-    "43": "Scala",
-    "44": "Shell",
-    "45": "Swift",
-    "46": "TeX",
-    "47": "TypeScript",
-    "48": "Verilog",
-    "49": "Visual Basic",
-    "50": "XML",
-    "51": "YAML"
+    "0": "C#",
+    "1": "Go",
+    "2": "JSON",
+    "3": "Java",
+    "4": "TypeScript",
+    "5": "Python",
+    "6": "Rust",
+    "7": "Shell"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "Assembly": 0,
-    "Batchfile": 1,
-    "C": 2,
-    "C#": 3,
-    "C++": 4,
-    "CMake": 5,
-    "COBOL": 6,
-    "CSS": 7,
-    "CSV": 8,
-    "Clojure": 9,
-    "CoffeeScript": 10,
-    "DM": 11,
-    "Dart": 12,
-    "Dockerfile": 13,
-    "Elixir": 14,
-    "Erlang": 15,
-    "Fortran": 16,
-    "Go": 17,
-    "Groovy": 18,
-    "HTML": 19,
-    "Haskell": 20,
-    "INI": 21,
-    "JSON": 22,
-    "Java": 23,
-    "JavaScript": 24,
-    "Julia": 25,
-    "Kotlin": 26,
-    "Lisp": 27,
-    "Lua": 28,
-    "Makefile": 29,
-    "Matlab": 30,
-    "OCaml": 31,
-    "Objective-C": 32,
-    "PHP": 33,
-    "Pascal": 34,
-    "Perl": 35,
-    "PowerShell": 36,
-    "Prolog": 37,
-    "Python": 38,
-    "R": 39,
-    "Ruby": 40,
-    "Rust": 41,
-    "SQL": 42,
-    "Scala": 43,
-    "Shell": 44,
-    "Swift": 45,
-    "TeX": 46,
-    "TypeScript": 47,
-    "Verilog": 48,
-    "Visual Basic": 49,
-    "XML": 50,
-    "YAML": 51
+    "C#": 0,
+    "Go": 1,
+    "JSON": 2,
+    "Java": 3,
+    "Python": 5,
+    "Rust": 6,
+    "Shell": 7,
+    "TypeScript": 4
   },
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
@@ -127,7 +39,6 @@
   "num_hidden_layers": 6,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
-  "problem_type": "single_label_classification",
   "transformers_version": "4.57.0",
   "type_vocab_size": 1,
   "use_cache": true,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0bb8a3f7117d4dc8dc40f73b2b7620be6cd78430b171049938670e999bb8b84
-size 333976000
+oid sha256:15f29975ded74b6bf071753e05b25401f099419d06cf9ed1845bfba94a02746c
+size 333840656

tokenizer_config.json

@@ -54,6 +54,5 @@
   "pad_token": "<pad>",
   "sep_token": "</s>",
   "tokenizer_class": "RobertaTokenizer",
-  "trim_offsets": true,
   "unk_token": "<unk>"
 }