EvalData commited on
Commit
1e5e73c
·
verified ·
1 Parent(s): 86b8b6e

Add real sha256; remove non-data FileObjects; add rai:personalSensitiveInformation + rai:hasSyntheticData

Browse files
Files changed (1) hide show
  1. croissant.json +33 -25
croissant.json CHANGED
@@ -60,29 +60,14 @@
60
  "inLanguage": "en",
61
  "citeAs": "@inproceedings{anonymous2026tabmibench, title={TabMI-Bench: Evaluating Mechanistic Interpretability Methods Across Tabular Foundation Model Architectures}, author={Anonymous}, booktitle={Advances in Neural Information Processing Systems (NeurIPS) Evaluations \\& Datasets Track}, year={2026}}",
62
  "distribution": [
63
- {
64
- "@type": "cr:FileObject",
65
- "@id": "github-repo",
66
- "name": "github-repo",
67
- "description": "Public GitHub repository containing all source code, hooks, evaluation scripts, tests, and frozen artifacts.",
68
- "contentUrl": "https://github.com/evaldataset/TabMI-Bench",
69
- "encodingFormat": "application/zip"
70
- },
71
- {
72
- "@type": "cr:FileObject",
73
- "@id": "huggingface-dataset",
74
- "name": "huggingface-dataset",
75
- "description": "HuggingFace dataset repository with frozen JSON aggregated artifacts driving paper-facing tables and figures.",
76
- "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench",
77
- "encodingFormat": "application/json"
78
- },
79
  {
80
  "@type": "cr:FileObject",
81
  "@id": "rd5-fullscale-aggregated",
82
  "name": "rd5-fullscale-aggregated",
83
  "description": "Aggregated 5-seed multi-model intermediary probing results (Table 1 source).",
84
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/rd5_fullscale_aggregated.json",
85
- "encodingFormat": "application/json"
 
86
  },
87
  {
88
  "@type": "cr:FileObject",
@@ -90,7 +75,8 @@
90
  "name": "tabdpt-probing-3seed",
91
  "description": "TabDPT in-family holdout probing results aggregated over 3 seeds.",
92
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabdpt_probing_3seed.json",
93
- "encodingFormat": "application/json"
 
94
  },
95
  {
96
  "@type": "cr:FileObject",
@@ -98,7 +84,8 @@
98
  "name": "tabdpt-causal-3seed",
99
  "description": "TabDPT noising-based causal tracing results aggregated over 3 seeds.",
100
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabdpt_causal_3seed.json",
101
- "encodingFormat": "application/json"
 
102
  },
103
  {
104
  "@type": "cr:FileObject",
@@ -106,7 +93,8 @@
106
  "name": "nam-holdout",
107
  "description": "NAM out-of-family holdout per-seed layer-wise R^2 profiles.",
108
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/nam_holdout.json",
109
- "encodingFormat": "application/json"
 
110
  },
111
  {
112
  "@type": "cr:FileObject",
@@ -114,7 +102,8 @@
114
  "name": "lofo-primary-endpoint",
115
  "description": "Leave-one-function-out robustness for the primary endpoint sigma2_profile.",
116
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/lofo_primary_endpoint.json",
117
- "encodingFormat": "application/json"
 
118
  },
119
  {
120
  "@type": "cr:FileObject",
@@ -122,7 +111,26 @@
122
  "name": "tabpfn25-fullscale-aggregated",
123
  "description": "TabPFN v2 vs v2.5 logit-lens, coefficient probing, intermediary probing, and CKA aggregated over 3 seeds.",
124
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabpfn25_fullscale_aggregated.json",
125
- "encodingFormat": "application/json"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
126
  }
127
  ],
128
  "recordSet": [
@@ -254,7 +262,7 @@
254
  "description": "MI technique name (e.g. linear-probing, logit-lens, activation-patching, sae, cka).",
255
  "dataType": "sc:Text",
256
  "source": {
257
- "fileObject": {"@id": "github-repo"},
258
  "extract": {"jsonPath": "$.technique"}
259
  }
260
  },
@@ -265,7 +273,7 @@
265
  "description": "TFM model name.",
266
  "dataType": "sc:Text",
267
  "source": {
268
- "fileObject": {"@id": "github-repo"},
269
  "extract": {"jsonPath": "$.model"}
270
  }
271
  },
@@ -276,7 +284,7 @@
276
  "description": "Applicability label: Supported, Limited, or Not established.",
277
  "dataType": "sc:Text",
278
  "source": {
279
- "fileObject": {"@id": "github-repo"},
280
  "extract": {"jsonPath": "$.label"}
281
  }
282
  }
 
60
  "inLanguage": "en",
61
  "citeAs": "@inproceedings{anonymous2026tabmibench, title={TabMI-Bench: Evaluating Mechanistic Interpretability Methods Across Tabular Foundation Model Architectures}, author={Anonymous}, booktitle={Advances in Neural Information Processing Systems (NeurIPS) Evaluations \\& Datasets Track}, year={2026}}",
62
  "distribution": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
63
  {
64
  "@type": "cr:FileObject",
65
  "@id": "rd5-fullscale-aggregated",
66
  "name": "rd5-fullscale-aggregated",
67
  "description": "Aggregated 5-seed multi-model intermediary probing results (Table 1 source).",
68
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/rd5_fullscale_aggregated.json",
69
+ "encodingFormat": "application/json",
70
+ "sha256": "8bae47fa6f777ff1b8e4d8950470d99f3a024478d43997b09f4b24642881a1c3"
71
  },
72
  {
73
  "@type": "cr:FileObject",
 
75
  "name": "tabdpt-probing-3seed",
76
  "description": "TabDPT in-family holdout probing results aggregated over 3 seeds.",
77
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabdpt_probing_3seed.json",
78
+ "encodingFormat": "application/json",
79
+ "sha256": "09e6fe1cfcdec668ee6bbc0f8927aef63af4f10039c7db28145b4646ee53f055"
80
  },
81
  {
82
  "@type": "cr:FileObject",
 
84
  "name": "tabdpt-causal-3seed",
85
  "description": "TabDPT noising-based causal tracing results aggregated over 3 seeds.",
86
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabdpt_causal_3seed.json",
87
+ "encodingFormat": "application/json",
88
+ "sha256": "a19828d2f5ca3c3812651d5179532ff3fa882a01d92cf74735cab195859cf54b"
89
  },
90
  {
91
  "@type": "cr:FileObject",
 
93
  "name": "nam-holdout",
94
  "description": "NAM out-of-family holdout per-seed layer-wise R^2 profiles.",
95
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/nam_holdout.json",
96
+ "encodingFormat": "application/json",
97
+ "sha256": "ebf795a470ea11e5fa5b6d2a85c6287add70494cecba81e94ba038ed2c3653af"
98
  },
99
  {
100
  "@type": "cr:FileObject",
 
102
  "name": "lofo-primary-endpoint",
103
  "description": "Leave-one-function-out robustness for the primary endpoint sigma2_profile.",
104
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/lofo_primary_endpoint.json",
105
+ "encodingFormat": "application/json",
106
+ "sha256": "66fe13294502cd9c0a32616861b0ccb4aa11e7544da54f60b98943ad55173419"
107
  },
108
  {
109
  "@type": "cr:FileObject",
 
111
  "name": "tabpfn25-fullscale-aggregated",
112
  "description": "TabPFN v2 vs v2.5 logit-lens, coefficient probing, intermediary probing, and CKA aggregated over 3 seeds.",
113
  "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/tabpfn25_fullscale_aggregated.json",
114
+ "encodingFormat": "application/json",
115
+ "sha256": "3f81492cd2183a5e8bc82216162c75579bffccf5c19a0cc07b7533a85db94773"
116
+ },
117
+ {
118
+ "@type": "cr:FileObject",
119
+ "@id": "c1c2-baselines",
120
+ "name": "c1c2-baselines",
121
+ "description": "Negative-control baselines: shuffled-label probing and random-target probing per model and layer.",
122
+ "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/c1c2_baselines.json",
123
+ "encodingFormat": "application/json",
124
+ "sha256": "c19cd33ef8e1e7b03ad26956cff8f932d985ce4f1d234a030d619bd7a7aadf5f"
125
+ },
126
+ {
127
+ "@type": "cr:FileObject",
128
+ "@id": "scale-10k-multiseed",
129
+ "name": "scale-10k-multiseed",
130
+ "description": "N=10K probing scale validation results aggregated over 3 seeds.",
131
+ "contentUrl": "https://huggingface.co/datasets/EvalData/TabMI-Bench/resolve/main/scale_10k_multiseed.json",
132
+ "encodingFormat": "application/json",
133
+ "sha256": "ecf8cb8572dcd9f4ea1c6cd569b52ab9b38a2d406ab18bb1b2da766f8e853eca"
134
  }
135
  ],
136
  "recordSet": [
 
262
  "description": "MI technique name (e.g. linear-probing, logit-lens, activation-patching, sae, cka).",
263
  "dataType": "sc:Text",
264
  "source": {
265
+ "fileObject": {"@id": "rd5-fullscale-aggregated"},
266
  "extract": {"jsonPath": "$.technique"}
267
  }
268
  },
 
273
  "description": "TFM model name.",
274
  "dataType": "sc:Text",
275
  "source": {
276
+ "fileObject": {"@id": "rd5-fullscale-aggregated"},
277
  "extract": {"jsonPath": "$.model"}
278
  }
279
  },
 
284
  "description": "Applicability label: Supported, Limited, or Not established.",
285
  "dataType": "sc:Text",
286
  "source": {
287
+ "fileObject": {"@id": "rd5-fullscale-aggregated"},
288
  "extract": {"jsonPath": "$.label"}
289
  }
290
  }