Hugging Face's logo

CIRCL
/
vulnerability-attack-technique-classification-pilot

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
text-embeddings-inference
Model card Files
xet
 Community
cedricbonhomme commited on
Commit
a6f297f
·
verified ·
1 Parent(s): 4366b55

End of training

Browse files
Files changed (8) hide show
  1. README.md +103 -0
  2. config.json +153 -0
  3. emissions.csv +2 -0
  4. metrics.json +13 -0
  5. model.safetensors +3 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +17 -0
  8. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: mit
4
+ base_model: roberta-base
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: vulnerability-attack-technique-classification-pilot
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # vulnerability-attack-technique-classification-pilot
16
+
17
+ This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.6123
20
+ - F1 Micro: 0.3952
21
+ - F1 Macro: 0.1641
22
+ - Precision Micro: 0.2887
23
+ - Recall Micro: 0.6264
24
+ - Recall At 3: 0.4912
25
+ - Recall At 5: 0.6328
26
+
27
+ ## Model description
28
+
29
+ More information needed
30
+
31
+ ## Intended uses & limitations
32
+
33
+ More information needed
34
+
35
+ ## Training and evaluation data
36
+
37
+ More information needed
38
+
39
+ ## Training procedure
40
+
41
+ ### Training hyperparameters
42
+
43
+ The following hyperparameters were used during training:
44
+ - learning_rate: 1e-05
45
+ - train_batch_size: 32
46
+ - eval_batch_size: 32
47
+ - seed: 42
48
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
+ - lr_scheduler_type: linear
50
+ - num_epochs: 40
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | Precision Micro | Recall Micro | Recall At 3 | Recall At 5 |
55
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------------:|:------------:|:-----------:|:-----------:|
56
+ | 0.8293 | 1.0 | 44 | 0.7935 | 0.2010 | 0.0348 | 0.1365 | 0.3811 | 0.2169 | 0.2724 |
57
+ | 0.7495 | 2.0 | 88 | 0.7544 | 0.2326 | 0.0326 | 0.1605 | 0.4226 | 0.2708 | 0.3669 |
58
+ | 0.7045 | 3.0 | 132 | 0.7379 | 0.2970 | 0.0539 | 0.2481 | 0.3698 | 0.3581 | 0.4528 |
59
+ | 0.7120 | 4.0 | 176 | 0.7184 | 0.2972 | 0.0682 | 0.2139 | 0.4868 | 0.3732 | 0.4926 |
60
+ | 0.6766 | 5.0 | 220 | 0.7017 | 0.2996 | 0.0870 | 0.2097 | 0.5245 | 0.3405 | 0.4634 |
61
+ | 0.6569 | 6.0 | 264 | 0.6817 | 0.3559 | 0.1129 | 0.2664 | 0.5358 | 0.4208 | 0.5801 |
62
+ | 0.6366 | 7.0 | 308 | 0.6658 | 0.3408 | 0.1129 | 0.2380 | 0.6 | 0.4301 | 0.5406 |
63
+ | 0.6025 | 8.0 | 352 | 0.6517 | 0.3713 | 0.1286 | 0.2719 | 0.5849 | 0.4378 | 0.5888 |
64
+ | 0.5755 | 9.0 | 396 | 0.6468 | 0.3695 | 0.1210 | 0.2700 | 0.5849 | 0.4205 | 0.5651 |
65
+ | 0.5695 | 10.0 | 440 | 0.6354 | 0.3807 | 0.1382 | 0.2707 | 0.6415 | 0.4596 | 0.5838 |
66
+ | 0.5580 | 11.0 | 484 | 0.6348 | 0.3709 | 0.1433 | 0.2603 | 0.6453 | 0.4295 | 0.5954 |
67
+ | 0.5485 | 12.0 | 528 | 0.6277 | 0.3636 | 0.1307 | 0.2562 | 0.6264 | 0.4272 | 0.5432 |
68
+ | 0.5319 | 13.0 | 572 | 0.6196 | 0.3865 | 0.1482 | 0.2752 | 0.6491 | 0.4596 | 0.6022 |
69
+ | 0.5063 | 14.0 | 616 | 0.6214 | 0.3850 | 0.1577 | 0.2717 | 0.6604 | 0.4495 | 0.6057 |
70
+ | 0.4967 | 15.0 | 660 | 0.6181 | 0.3709 | 0.1342 | 0.2655 | 0.6151 | 0.4433 | 0.5817 |
71
+ | 0.4838 | 16.0 | 704 | 0.6162 | 0.3866 | 0.1522 | 0.2788 | 0.6302 | 0.4558 | 0.6095 |
72
+ | 0.4641 | 17.0 | 748 | 0.6123 | 0.3952 | 0.1641 | 0.2887 | 0.6264 | 0.4912 | 0.6328 |
73
+ | 0.4619 | 18.0 | 792 | 0.6073 | 0.3902 | 0.1466 | 0.2826 | 0.6302 | 0.4836 | 0.6314 |
74
+ | 0.4555 | 19.0 | 836 | 0.6082 | 0.3753 | 0.1515 | 0.2672 | 0.6302 | 0.4717 | 0.5845 |
75
+ | 0.4339 | 20.0 | 880 | 0.6087 | 0.3810 | 0.1541 | 0.2696 | 0.6491 | 0.4714 | 0.5820 |
76
+ | 0.4439 | 21.0 | 924 | 0.6103 | 0.3942 | 0.1372 | 0.2908 | 0.6113 | 0.4842 | 0.5956 |
77
+ | 0.4251 | 22.0 | 968 | 0.6090 | 0.4034 | 0.1550 | 0.2984 | 0.6226 | 0.4856 | 0.6207 |
78
+ | 0.4196 | 23.0 | 1012 | 0.6000 | 0.3693 | 0.1596 | 0.2587 | 0.6453 | 0.4644 | 0.6045 |
79
+ | 0.4222 | 24.0 | 1056 | 0.6066 | 0.3985 | 0.1540 | 0.2939 | 0.6189 | 0.4801 | 0.6192 |
80
+ | 0.4026 | 25.0 | 1100 | 0.6083 | 0.4039 | 0.1541 | 0.2980 | 0.6264 | 0.4912 | 0.6189 |
81
+ | 0.4028 | 26.0 | 1144 | 0.6082 | 0.3975 | 0.1538 | 0.2945 | 0.6113 | 0.4801 | 0.6342 |
82
+ | 0.4056 | 27.0 | 1188 | 0.6093 | 0.3937 | 0.1522 | 0.2903 | 0.6113 | 0.4829 | 0.6196 |
83
+ | 0.4020 | 28.0 | 1232 | 0.6052 | 0.4050 | 0.1544 | 0.3038 | 0.6075 | 0.5037 | 0.6213 |
84
+ | 0.3867 | 29.0 | 1276 | 0.6090 | 0.3965 | 0.1504 | 0.2961 | 0.6 | 0.4912 | 0.6145 |
85
+ | 0.3840 | 30.0 | 1320 | 0.6033 | 0.3932 | 0.1551 | 0.2890 | 0.6151 | 0.4912 | 0.6233 |
86
+ | 0.3730 | 31.0 | 1364 | 0.6056 | 0.3995 | 0.1522 | 0.2985 | 0.6038 | 0.5023 | 0.6050 |
87
+ | 0.3661 | 32.0 | 1408 | 0.6063 | 0.4131 | 0.1578 | 0.3100 | 0.6189 | 0.5190 | 0.6414 |
88
+ | 0.3630 | 33.0 | 1452 | 0.6058 | 0.4090 | 0.1573 | 0.3054 | 0.6189 | 0.5044 | 0.6150 |
89
+ | 0.3707 | 34.0 | 1496 | 0.6058 | 0.4044 | 0.1560 | 0.3004 | 0.6189 | 0.4981 | 0.6233 |
90
+ | 0.3607 | 35.0 | 1540 | 0.6031 | 0.4160 | 0.1629 | 0.3114 | 0.6264 | 0.5190 | 0.6525 |
91
+ | 0.3588 | 36.0 | 1584 | 0.6069 | 0.4046 | 0.1548 | 0.3042 | 0.6038 | 0.5051 | 0.6200 |
92
+ | 0.3591 | 37.0 | 1628 | 0.6069 | 0.4106 | 0.1553 | 0.3092 | 0.6113 | 0.5127 | 0.6117 |
93
+ | 0.3647 | 38.0 | 1672 | 0.6062 | 0.4050 | 0.1541 | 0.3038 | 0.6075 | 0.5023 | 0.6217 |
94
+ | 0.3483 | 39.0 | 1716 | 0.6058 | 0.4090 | 0.1565 | 0.3064 | 0.6151 | 0.4995 | 0.6133 |
95
+ | 0.3508 | 40.0 | 1760 | 0.6063 | 0.4111 | 0.1574 | 0.3087 | 0.6151 | 0.5044 | 0.6217 |
96
+
97
+
98
+ ### Framework versions
99
+
100
+ - Transformers 5.13.0
101
+ - Pytorch 2.12.1+cu130
102
+ - Datasets 4.8.5
103
+ - Tokenizers 0.22.2
config.json ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "T1003",
16
+ "1": "T1005",
17
+ "2": "T1021",
18
+ "3": "T1027",
19
+ "4": "T1036",
20
+ "5": "T1040",
21
+ "6": "T1041",
22
+ "7": "T1046",
23
+ "8": "T1055",
24
+ "9": "T1059",
25
+ "10": "T1068",
26
+ "11": "T1070",
27
+ "12": "T1071",
28
+ "13": "T1078",
29
+ "14": "T1082",
30
+ "15": "T1083",
31
+ "16": "T1087",
32
+ "17": "T1091",
33
+ "18": "T1098",
34
+ "19": "T1105",
35
+ "20": "T1106",
36
+ "21": "T1110",
37
+ "22": "T1114",
38
+ "23": "T1133",
39
+ "24": "T1136",
40
+ "25": "T1185",
41
+ "26": "T1189",
42
+ "27": "T1190",
43
+ "28": "T1202",
44
+ "29": "T1203",
45
+ "30": "T1204",
46
+ "31": "T1210",
47
+ "32": "T1211",
48
+ "33": "T1212",
49
+ "34": "T1213",
50
+ "35": "T1485",
51
+ "36": "T1486",
52
+ "37": "T1489",
53
+ "38": "T1496",
54
+ "39": "T1497",
55
+ "40": "T1498",
56
+ "41": "T1499",
57
+ "42": "T1505",
58
+ "43": "T1528",
59
+ "44": "T1539",
60
+ "45": "T1542",
61
+ "46": "T1543",
62
+ "47": "T1548",
63
+ "48": "T1550",
64
+ "49": "T1552",
65
+ "50": "T1553",
66
+ "51": "T1555",
67
+ "52": "T1557",
68
+ "53": "T1563",
69
+ "54": "T1565",
70
+ "55": "T1566",
71
+ "56": "T1574",
72
+ "57": "T1588",
73
+ "58": "T1608",
74
+ "59": "T1685"
75
+ },
76
+ "initializer_range": 0.02,
77
+ "intermediate_size": 3072,
78
+ "is_decoder": false,
79
+ "label2id": {
80
+ "T1003": 0,
81
+ "T1005": 1,
82
+ "T1021": 2,
83
+ "T1027": 3,
84
+ "T1036": 4,
85
+ "T1040": 5,
86
+ "T1041": 6,
87
+ "T1046": 7,
88
+ "T1055": 8,
89
+ "T1059": 9,
90
+ "T1068": 10,
91
+ "T1070": 11,
92
+ "T1071": 12,
93
+ "T1078": 13,
94
+ "T1082": 14,
95
+ "T1083": 15,
96
+ "T1087": 16,
97
+ "T1091": 17,
98
+ "T1098": 18,
99
+ "T1105": 19,
100
+ "T1106": 20,
101
+ "T1110": 21,
102
+ "T1114": 22,
103
+ "T1133": 23,
104
+ "T1136": 24,
105
+ "T1185": 25,
106
+ "T1189": 26,
107
+ "T1190": 27,
108
+ "T1202": 28,
109
+ "T1203": 29,
110
+ "T1204": 30,
111
+ "T1210": 31,
112
+ "T1211": 32,
113
+ "T1212": 33,
114
+ "T1213": 34,
115
+ "T1485": 35,
116
+ "T1486": 36,
117
+ "T1489": 37,
118
+ "T1496": 38,
119
+ "T1497": 39,
120
+ "T1498": 40,
121
+ "T1499": 41,
122
+ "T1505": 42,
123
+ "T1528": 43,
124
+ "T1539": 44,
125
+ "T1542": 45,
126
+ "T1543": 46,
127
+ "T1548": 47,
128
+ "T1550": 48,
129
+ "T1552": 49,
130
+ "T1553": 50,
131
+ "T1555": 51,
132
+ "T1557": 52,
133
+ "T1563": 53,
134
+ "T1565": 54,
135
+ "T1566": 55,
136
+ "T1574": 56,
137
+ "T1588": 57,
138
+ "T1608": 58,
139
+ "T1685": 59
140
+ },
141
+ "layer_norm_eps": 1e-05,
142
+ "max_position_embeddings": 514,
143
+ "model_type": "roberta",
144
+ "num_attention_heads": 12,
145
+ "num_hidden_layers": 12,
146
+ "pad_token_id": 1,
147
+ "problem_type": "multi_label_classification",
148
+ "tie_word_embeddings": true,
149
+ "transformers_version": "5.13.0",
150
+ "type_vocab_size": 1,
151
+ "use_cache": false,
152
+ "vocab_size": 50265
153
+ }
emissions.csv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ timestamp,project_name,run_id,experiment_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,water_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,cpu_utilization_percent,gpu_utilization_percent,ram_utilization_percent,ram_used_gb,on_cloud,pue,wue
2
+ 2026-07-14T05:43:20,VulnTrain,ce58936d-aceb-404c-ac40-fbdeb5970871,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,287.4541791751981,0.005600634657486989,1.948357360312895e-05,70.00013690517649,516.5056235887321,70.0,0.005386460057389911,0.04243354450238712,0.0053861044955129415,0.05320610905528997,0.0,Luxembourg,LUX,luxembourg,,,Linux-6.8.0-106-generic-x86_64-with-glibc2.39,3.12.3,3.2.8,224,Intel(R) Xeon(R) Platinum 8480+,2,2 x NVIDIA H100 NVL,6.1327,49.6098,2015.336296081543,machine,0.9485915492957746,54.026408450704224,1.730281690140845,35.15862096867091,N,1.0,0.0
metrics.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eval_loss": 0.6123166680335999,
3
+ "eval_f1_micro": 0.3952380952380952,
4
+ "eval_f1_macro": 0.1641093244848836,
5
+ "eval_precision_micro": 0.288695652173913,
6
+ "eval_recall_micro": 0.6264150943396226,
7
+ "eval_recall_at_3": 0.4911904761904761,
8
+ "eval_recall_at_5": 0.6327876984126984,
9
+ "eval_runtime": 0.2477,
10
+ "eval_samples_per_second": 484.423,
11
+ "eval_steps_per_second": 16.147,
12
+ "epoch": 40.0
13
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2d72d7e93f612f723c15f9e85acd7a5972f6d17a5d1fe1ca9bad8acd697f501
3
+ size 498791232
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "is_local": false,
9
+ "local_files_only": false,
10
+ "mask_token": "<mask>",
11
+ "model_max_length": 512,
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "tokenizer_class": "RobertaTokenizer",
15
+ "trim_offsets": true,
16
+ "unk_token": "<unk>"
17
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0846c18592fb472562f25da50281e44bf5761ab7760a4ceb4bedff2407a5879
3
+ size 5265