Added metadata files

croissant_metadata/aliasing_dataset_c_croissant.json

@@ -0,0 +1,84 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-aliasing_dataset_c",
+  "schema:description": "Dataset for Benchmarking LLMs on pointer aliasing.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/aliasing_dataset_c.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/aliasing_dataset_c.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "aliasing_dataset_c",
+      "schema:sha256": "172f7e1859ad4b63e027850b524458aca0c1ad13c546671ce08eddb81246e11a"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "aliasing_dataset_c",
+      "description": "Contains C sample codes with proper annotation for pointer aliasing reasoning.",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset"
+        },
+        {
+          "name": "Programming Language",
+          "dataType": "sc:Text",
+          "description": "Programming language of the Source Code"
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking"
+        },
+        {
+          "name": "Selected Statement",
+          "dataType": "sc:Text",
+          "description": "Statement selected for aliasing reasoning."
+        },
+        {
+          "name": "Selected Pointer",
+          "dataType": "sc:Text",
+          "description": "One pointer used for aliasing reasoning."
+        },
+        {
+          "name": "Compared Statement",
+          "dataType": "sc:Text",
+          "description": "Statement associated with compared pointer"
+        },
+        {
+          "name": "Compared Pointer",
+          "dataType": "sc:Text",
+          "description": "Compared with Selected Pointer for aliasing reasoning."
+        },
+        {
+          "name": "Aliasing",
+          "dataType": "sc:Text",
+          "description": "Ground-Truth Label, Yes for aliasing, No for non-aliasing."
+        },
+        {
+          "name": "Function Input",
+          "dataType": "sc:StructuredValue",
+          "description": "Input of Source Code"
+        },
+        {
+          "name": "Program Information",
+          "dataType": "sc:StructuredValue",
+          "description": "Source Codes project information."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/block_analysis_c_croissant.json

@@ -0,0 +1,74 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-block_analysis_c",
+  "schema:description": "Dataset for block based code reasoning for C Language",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/block_analysis_c.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/block_analysis_c.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "block_analysis_c",
+      "schema:sha256": "891b8450d5b2e7bae0fb514f8ad1f2b17e34f314ef013e09c7a81d7a2c5fcb57"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "block_analysis_c",
+      "description": "Annotated samples for block size analysis for C codes.",
+      "field": [
+        {
+          "name": "Programming Language",
+          "dataType": "sc:Text",
+          "description": "Programming language of the Source Code."
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "Selected Statement",
+          "dataType": "sc:Text",
+          "description": "Statement selected for block reasoning."
+        },
+        {
+          "name": "Function Input",
+          "dataType": "sc:StructuredValue",
+          "description": "Input of Source Code."
+        },
+        {
+          "name": "Variable Values Before Statement",
+          "dataType": "sc:StructuredValue",
+          "description": "Variable values before the execution of the statement."
+        },
+        {
+          "name": "Value After Statement Execution",
+          "dataType": "sc:Text",
+          "description": "Ground Truth. Contains the value of the statement after execution."
+        },
+        {
+          "name": "Block_Size",
+          "dataType": "sc:Integer",
+          "description": "Block Size of the Selected Statement."
+        },
+        {
+          "name": "Project Information",
+          "dataType": "sc:Text",
+          "description": "Source Codes project information."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/block_analysis_python_croissant.json

@@ -0,0 +1,74 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-block_analysis_python",
+  "schema:description": "Dataset for block based code reasoning for Python Language",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/block_analysis_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/block_analysis_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "block_analysis_python",
+      "schema:sha256": "2f620e1cf8ff6cf70fa2fc4ef67e130c5134a560cfd6aa71aaadc7919d598bd3"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "block_analysis_python",
+      "description": "Annotated samples for block size analysis for Python codes.",
+      "field": [
+        {
+          "name": "Programming Language",
+          "dataType": "sc:Text",
+          "description": "Programming language of the Source Code."
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "Selected Statement",
+          "dataType": "sc:Text",
+          "description": "Statement selected for block reasoning."
+        },
+        {
+          "name": "Function Input",
+          "dataType": "sc:StructuredValue",
+          "description": "Input of Source Code."
+        },
+        {
+          "name": "Variable Values Before Statement",
+          "dataType": "sc:StructuredValue",
+          "description": "Variable values before the execution of the statement."
+        },
+        {
+          "name": "Value After Statement Execution",
+          "dataType": "sc:Text",
+          "description": "Ground Truth. Contains the value of the statement after execution."
+        },
+        {
+          "name": "Block_Size",
+          "dataType": "sc:Integer",
+          "description": "Block Size of the Selected Statement."
+        },
+        {
+          "name": "Project Information",
+          "dataType": "sc:Text",
+          "description": "Source Codes project information."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/conditional_dataset_python_croissant.json

@@ -0,0 +1,54 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-conditional_dataset_python",
+  "schema:description": "Dataset for Branch Prediciton capabilities of LLMs",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/conditional_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/conditional_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "conditional_dataset_python",
+      "schema:sha256": "9050a7dab251f41f260efaa4080e97a2bc862f59510ba481f923bd246e080e57"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "conditional_dataset_python",
+      "description": "Annoated samples for branch prediciton task",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "question",
+          "dataType": "sc:Text",
+          "description": "Contains specific branch line and question whether that line will be executed based on a function call with input value"
+        },
+        {
+          "name": "answer",
+          "dataType": "sc:Text",
+          "description": "Ground-Truth, 'Yes' if the branch was taken, 'No' for not taken."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/input_output_dataset_c_croissant.json

@@ -0,0 +1,69 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-input_output_dataset_c",
+  "schema:description": "Input Output Prediciton dataset for C Language.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/input_output_dataset_c.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/input_output_dataset_c.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "input_output_dataset_c",
+      "schema:sha256": "85dd58229d95d8de00e91c31a1bf6b2186f0e3351f76257416dfa753aa8eb48a"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "input_output_dataset_c",
+      "description": "Samples containing required fields for input output prediciton task.",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "input",
+          "dataType": "sc:Text",
+          "description": "Function Input"
+        },
+        {
+          "name": "basic_input",
+          "dataType": "sc:Text",
+          "description": "Primitive input types."
+        },
+        {
+          "name": "output",
+          "dataType": "sc:Text",
+          "description": "Output of the function based on the given input."
+        },
+        {
+          "name": "cyclomatic_complexity",
+          "dataType": "sc:Integer",
+          "description": "cyclomatic_complexity of the code"
+        },
+        {
+          "name": "code_length",
+          "dataType": "sc:Integer",
+          "description": "No. of code lines."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/input_output_dataset_java_croissant.json

@@ -0,0 +1,69 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-input_output_dataset_java",
+  "schema:description": "Input Output Prediciton dataset for Java Language.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/input_output_dataset_java.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/input_output_dataset_java.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "input_output_dataset_java",
+      "schema:sha256": "006c404e50822aa6d0b083654915afe7a10c9ec66a415098dc3a4cd9e65fcd37"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "input_output_dataset_java",
+      "description": "Samples containing required fields for input output prediciton task",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "basic_input",
+          "dataType": "sc:Text",
+          "description": "Primitive input types."
+        },
+        {
+          "name": "output",
+          "dataType": "sc:Text",
+          "description": "Output of the function based on the given input."
+        },
+        {
+          "name": "cyclomatic_complexity",
+          "dataType": "sc:Integer",
+          "description": "cyclomatic_complexity of the code"
+        },
+        {
+          "name": "code_length",
+          "dataType": "sc:Integer",
+          "description": "No. of code lines."
+        },
+        {
+          "name": "category",
+          "dataType": "sc:Text",
+          "description": "Complexity category based on cyclomatic complexity and code_length"
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/input_output_dataset_python_croissant.json

@@ -0,0 +1,74 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-input_output_dataset_python",
+  "schema:description": "Input Output Prediciton dataset for Python Language.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/input_output_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/input_output_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "input_output_dataset_python",
+      "schema:sha256": "41ab975435a86db443c9ec81628621a1590f52d776d39b4175318633e982a7bc"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "input_output_dataset_python",
+      "description": "Samples containing required fields for input output prediciton task",
+      "field": [
+        {
+          "name": "index",
+          "dataType": "sc:Integer",
+          "description": "Random number associated with each sample."
+        },
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking."
+        },
+        {
+          "name": "input",
+          "dataType": "sc:Text",
+          "description": "Function Input"
+        },
+        {
+          "name": "output",
+          "dataType": "sc:Text",
+          "description": "Output of the function based on the given input."
+        },
+        {
+          "name": "cyclomatic_complexity",
+          "dataType": "sc:Integer",
+          "description": "cyclomatic_complexity of the code"
+        },
+        {
+          "name": "code_length",
+          "dataType": "sc:Integer",
+          "description": "No. of code lines."
+        },
+        {
+          "name": "category",
+          "dataType": "sc:Text",
+          "description": "Complexity category based on cyclomatic complexity and code_length"
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/loop_after_dataset_python_croissant.json

@@ -0,0 +1,74 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-loop_after_dataset_python",
+  "schema:description": "Dataset for Post-Loop value reasoning of LLMs",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/loop_after_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/loop_after_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "loop_after_dataset_python",
+      "schema:sha256": "b37b21ca8896fb2085a2482ce1fa9945b38dfc3890074d747f375cb832663b34"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "loop_after_dataset_python",
+      "description": "A collection of Python code examples designed to evaluate a model’s ability to reason about variable values after loop execution. ",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "scratchpad_format",
+          "dataType": "sc:Text",
+          "description": "code in scratchpad format, the variable states defined in comments."
+        },
+        {
+          "name": "loop_code",
+          "dataType": "sc:Text",
+          "description": "code used for LLM prompting."
+        },
+        {
+          "name": "question",
+          "dataType": "sc:Text",
+          "description": "Query regarding a specific line post-loop in the code."
+        },
+        {
+          "name": "answer",
+          "dataType": "sc:Text",
+          "description": "Ground-Truth, contains the exact value of the variable after loop execution."
+        },
+        {
+          "name": "variable_assignment",
+          "dataType": "sc:Text",
+          "description": "Line that was selected for prompting."
+        },
+        {
+          "name": "loop_range",
+          "dataType": "sc:ItemList",
+          "description": "Range of loop lines that was selected."
+        },
+        {
+          "name": "post_loop_line",
+          "dataType": "sc:Integer",
+          "description": "Line no. of selected line/variable_assignment."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/loop_body_dataset_python_croissant.json

@@ -0,0 +1,64 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-loop_body_dataset_python",
+  "schema:description": "Dataset for In-Loop value reasoning of LLMs",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/loop_body_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/loop_body_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "loop_body_dataset_python",
+      "schema:sha256": "4eca6a5382cecc7b4a54fc269093de0ed7f06c1ba89226f3aff3353c8ee84b29"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "loop_body_dataset_python",
+      "description": "A collection of Python code examples designed to evaluate a model’s ability to reason about variable values inside loop execution. ",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "scratchpad_format",
+          "dataType": "sc:Text",
+          "description": "code in scratchpad format, the variable states defined in comments."
+        },
+        {
+          "name": "loop_code",
+          "dataType": "sc:Text",
+          "description": "code used for LLM prompting."
+        },
+        {
+          "name": "question",
+          "dataType": "sc:Text",
+          "description": "Query regarding a specific line in-loop in the code."
+        },
+        {
+          "name": "answer",
+          "dataType": "sc:Text",
+          "description": "Ground-Truth, contains the exact value of the variable in-loop execution."
+        },
+        {
+          "name": "variable_assignment",
+          "dataType": "sc:Text",
+          "description": "selected variables value as assignment."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/loop_iteration_dataset_python_croissant.json

@@ -0,0 +1,59 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-loop_iteration_dataset_python",
+  "schema:description": "Dataset for loop iteration values of LLMs",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/loop_iteration_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/loop_iteration_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "loop_iteration_dataset_python",
+      "schema:sha256": "12961156c536e2e21c7e2d42c06814bafe5e2fa652a971d0700f308c2f830bd6"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "loop_iteration_dataset_python",
+      "description": "A collection of Python code examples designed to evaluate a model’s ability to reason about loop iteration.",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset."
+        },
+        {
+          "name": "scratchpad_format",
+          "dataType": "sc:Text",
+          "description": "code in scratchpad format, the variable states defined in comments."
+        },
+        {
+          "name": "loop_code",
+          "dataType": "sc:Text",
+          "description": "code used for LLM prompting."
+        },
+        {
+          "name": "question",
+          "dataType": "sc:Text",
+          "description": "Query regarding a specific loop line for number of iterations."
+        },
+        {
+          "name": "answer",
+          "dataType": "sc:Text",
+          "description": "Ground-Truth, contains the exact value of the number of loop iterations."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/statement_prediction_dataset_C_croissant.json

@@ -0,0 +1,79 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-statement_prediction_dataset_C",
+  "schema:description": "Dataset containing five type of statements to benchmark LLMs reason about different type of statement executions.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/statement_prediction_dataset_C.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/statement_prediction_dataset_C.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "statement_prediction_dataset_C",
+      "schema:sha256": "dec9546ca7a8d948f2441816bea7486d6c97d6c7ddf9eb28dd2e84778772acc3"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "statement_prediction_dataset_C",
+      "description": "Annotated C programs containing information for different statement type reaasoning.",
+      "field": [
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset"
+        },
+        {
+          "name": "Programming Language",
+          "dataType": "sc:Text",
+          "description": "Programming language of the Source Code"
+        },
+        {
+          "name": "Statement Type",
+          "dataType": "sc:Text",
+          "description": "Type of the Selected Statement"
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking"
+        },
+        {
+          "name": "Selected Statement",
+          "dataType": "sc:Text",
+          "description": "The statment selected from Source Code for quering the LLMs."
+        },
+        {
+          "name": "Function Input",
+          "dataType": "sc:StructuredValue",
+          "description": "Input of Source Code"
+        },
+        {
+          "name": "Variable Values Before Statement",
+          "dataType": "sc:StructuredValue",
+          "description": "Variable values before the execution of the statement."
+        },
+        {
+          "name": "Value After Statement Execution",
+          "dataType": "sc:Text",
+          "description": "Ground Truth. Contains the value of the statement after execution."
+        },
+        {
+          "name": "Project Information",
+          "dataType": "sc:Text",
+          "description": "Source Codes project information."
+        }
+      ]
+    }
+  ]
+}

croissant_metadata/statement_prediction_dataset_python_croissant.json

@@ -0,0 +1,84 @@
+{
+  "@context": {
+    "ml": "http://mlcommons.org/croissant/",
+    "schema": "https://schema.org/",
+    "sc": "https://schema.org/",
+    "codesense": "https://codesense-bench.github.io/schema/"
+  },
+  "@type": "schema:Dataset",
+  "schema:name": "CodeSense-Bench-statement_prediction_dataset_python",
+  "schema:description": "Dataset containing five type of statements to benchmark LLMs reason about different type of statement executions.",
+  "schema:license": "CC-BY-4.0",
+  "schema:url": "https://huggingface.co/datasets/codesense-bench/codesense/blob/main/Task%20Specific%20Datasets/statement_prediction_dataset_python.jsonl",
+  "schema:creator": {
+    "@type": "schema:Organization",
+    "schema:name": "CodeSense Research Team"
+  },
+  "schema:distribution": [
+    {
+      "@type": "schema:FileObject",
+      "schema:contentUrl": "https://huggingface.co/datasets/codesense-bench/codesense/resolve/main/Task%20Specific%20Datasets/statement_prediction_dataset_python.jsonl",
+      "schema:encodingFormat": "application/jsonl",
+      "schema:name": "statement_prediction_dataset_python",
+      "schema:sha256": "1fe6432f2cf154d6d19eee63fc43a77db018d400968464187c8683c0e11e3e11"
+    }
+  ],
+  "recordSet": [
+    {
+      "name": "statement_prediction_dataset_python",
+      "description": "Annotated Python programs containing information for different statement type reaasoning.",
+      "field": [
+        {
+          "name": "Programming Language",
+          "dataType": "sc:Text",
+          "description": "Programming language of the Source Code"
+        },
+        {
+          "name": "Statement Type",
+          "dataType": "sc:Text",
+          "description": "Type of the Selected Statement"
+        },
+        {
+          "name": "Source Code",
+          "dataType": "sc:Text",
+          "description": "Code snippet used for benchmarking"
+        },
+        {
+          "name": "Selected Statement",
+          "dataType": "sc:Text",
+          "description": "The statment selected from Source Code for quering the LLMs."
+        },
+        {
+          "name": "Function Input",
+          "dataType": "sc:StructuredValue",
+          "description": "Input of Source Code"
+        },
+        {
+          "name": "Variable Values Before Statement",
+          "dataType": "sc:StructuredValue",
+          "description": "Variable values before the execution of the statement."
+        },
+        {
+          "name": "Value After Statement Execution",
+          "dataType": "sc:Text",
+          "description": "Ground Truth. Contains the value of the statement after execution."
+        },
+        {
+          "name": "Variable States During Runtime",
+          "dataType": "sc:StructuredValue",
+          "description": "All the variable states with line number during the runtime of the program."
+        },
+        {
+          "name": "Program Information",
+          "dataType": "sc:Text",
+          "description": "Source Codes project information."
+        },
+        {
+          "name": "idx",
+          "dataType": "sc:Integer",
+          "description": "Unique identifier for each record in the dataset"
+        }
+      ]
+    }
+  ]
+}