Hugging Face's logo

Tokyosaurus
/
taglish-xlm-binary

Safetensors
xlm-roberta
Model card Files
xet
 Community
Tokyosaurus commited on
Commit
9b1fede
·
verified ·
1 Parent(s): 8ac9e3f

Upload folder using huggingface_hub

Browse files
Files changed (27) hide show
  1. .gitattributes +3 -0
  2. checkpoint-340/config.json +39 -0
  3. checkpoint-340/model.safetensors +3 -0
  4. checkpoint-340/optimizer.pt +3 -0
  5. checkpoint-340/rng_state.pth +3 -0
  6. checkpoint-340/scaler.pt +3 -0
  7. checkpoint-340/scheduler.pt +3 -0
  8. checkpoint-340/tokenizer.json +3 -0
  9. checkpoint-340/tokenizer_config.json +14 -0
  10. checkpoint-340/trainer_state.json +139 -0
  11. checkpoint-340/training_args.bin +3 -0
  12. checkpoint-425/config.json +39 -0
  13. checkpoint-425/model.safetensors +3 -0
  14. checkpoint-425/optimizer.pt +3 -0
  15. checkpoint-425/rng_state.pth +3 -0
  16. checkpoint-425/scaler.pt +3 -0
  17. checkpoint-425/scheduler.pt +3 -0
  18. checkpoint-425/tokenizer.json +3 -0
  19. checkpoint-425/tokenizer_config.json +14 -0
  20. checkpoint-425/trainer_state.json +163 -0
  21. checkpoint-425/training_args.bin +3 -0
  22. comprehensive_results_binary.json +79 -0
  23. config.json +39 -0
  24. model.safetensors +3 -0
  25. tokenizer.json +3 -0
  26. tokenizer_config.json +14 -0
  27. training_args.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ checkpoint-340/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-425/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoint-340/config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
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": "Non-Gaslighting",
16
+ "1": "Gaslighting"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "Gaslighting": 1,
23
+ "Non-Gaslighting": 0
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "xlm-roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "position_embedding_type": "absolute",
33
+ "problem_type": "single_label_classification",
34
+ "tie_word_embeddings": true,
35
+ "transformers_version": "5.0.0",
36
+ "type_vocab_size": 1,
37
+ "use_cache": false,
38
+ "vocab_size": 250002
39
+ }
checkpoint-340/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d97a7ba9fe730fb08f6ed03685d40c26815b890d7769ff9d841ac2feeeea5e12
3
+ size 1112204984
checkpoint-340/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01b2cc2415077491cf41e9014296a793750913402eed3d28ce4339524ad13a5e
3
+ size 2224530059
checkpoint-340/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5c010487ae9edf9b823436ab5fb2c452efb3fc6d40777937dbeb8cfbfde9527
3
+ size 14645
checkpoint-340/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c38abe1eff32667ab62564b69fac6d423a608612a979824610a5092c87721827
3
+ size 1383
checkpoint-340/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:587757a448baa056a3ae0ede6e7f8e67137cc2957044c39d0eb0884a77eeeef8
3
+ size 1465
checkpoint-340/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5b633524ba90477daaba16ec27580a08a2856ae0ee8c33d9f5f9358378d3b35
3
+ size 16781751
checkpoint-340/tokenizer_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "is_local": false,
8
+ "mask_token": "<mask>",
9
+ "model_max_length": 512,
10
+ "pad_token": "<pad>",
11
+ "sep_token": "</s>",
12
+ "tokenizer_class": "XLMRobertaTokenizer",
13
+ "unk_token": "<unk>"
14
+ }
checkpoint-340/trainer_state.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 255,
3
+ "best_metric": 0.9158249158249159,
4
+ "best_model_checkpoint": "D:/Thesis/Taglish_Gaslighting_V3\\model_outputs\\xlm-roberta_binary\\checkpoint-255",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 340,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 13.04288101196289,
15
+ "learning_rate": 1.8010471204188483e-05,
16
+ "loss": 0.6918394200942095,
17
+ "step": 85
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7093425605536332,
22
+ "eval_f1": 0.676923076923077,
23
+ "eval_loss": 0.5885154604911804,
24
+ "eval_macro_f1": 0.7063860667634252,
25
+ "eval_non_gas_f1": 0.7358490566037735,
26
+ "eval_non_gas_precision": 0.6842105263157895,
27
+ "eval_non_gas_recall": 0.7959183673469388,
28
+ "eval_precision": 0.7457627118644068,
29
+ "eval_recall": 0.6197183098591549,
30
+ "eval_roc_auc": 0.8050445530324807,
31
+ "eval_runtime": 13.2767,
32
+ "eval_samples_per_second": 21.768,
33
+ "eval_steps_per_second": 1.431,
34
+ "step": 85
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "grad_norm": 11.096158981323242,
39
+ "learning_rate": 1.356020942408377e-05,
40
+ "loss": 0.5428493724149817,
41
+ "step": 170
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.8027681660899654,
46
+ "eval_f1": 0.7764705882352941,
47
+ "eval_loss": 0.48855453729629517,
48
+ "eval_macro_f1": 0.8,
49
+ "eval_non_gas_f1": 0.8235294117647058,
50
+ "eval_non_gas_precision": 0.7556818181818182,
51
+ "eval_non_gas_recall": 0.9047619047619048,
52
+ "eval_precision": 0.8761061946902655,
53
+ "eval_recall": 0.6971830985915493,
54
+ "eval_roc_auc": 0.898438248538852,
55
+ "eval_runtime": 13.3951,
56
+ "eval_samples_per_second": 21.575,
57
+ "eval_steps_per_second": 1.418,
58
+ "step": 170
59
+ },
60
+ {
61
+ "epoch": 3.0,
62
+ "grad_norm": 5.319683074951172,
63
+ "learning_rate": 9.109947643979057e-06,
64
+ "loss": 0.4088932710535386,
65
+ "step": 255
66
+ },
67
+ {
68
+ "epoch": 3.0,
69
+ "eval_accuracy": 0.9134948096885813,
70
+ "eval_f1": 0.9158249158249159,
71
+ "eval_loss": 0.2529641091823578,
72
+ "eval_macro_f1": 0.9134284721473334,
73
+ "eval_non_gas_f1": 0.9110320284697508,
74
+ "eval_non_gas_precision": 0.9552238805970149,
75
+ "eval_non_gas_recall": 0.8707482993197279,
76
+ "eval_precision": 0.8774193548387097,
77
+ "eval_recall": 0.9577464788732394,
78
+ "eval_roc_auc": 0.9593752994155408,
79
+ "eval_runtime": 13.2242,
80
+ "eval_samples_per_second": 21.854,
81
+ "eval_steps_per_second": 1.437,
82
+ "step": 255
83
+ },
84
+ {
85
+ "epoch": 4.0,
86
+ "grad_norm": 4.501065254211426,
87
+ "learning_rate": 4.712041884816754e-06,
88
+ "loss": 0.2737856921027688,
89
+ "step": 340
90
+ },
91
+ {
92
+ "epoch": 4.0,
93
+ "eval_accuracy": 0.9100346020761245,
94
+ "eval_f1": 0.9084507042253521,
95
+ "eval_loss": 0.2273072451353073,
96
+ "eval_macro_f1": 0.9100076650378461,
97
+ "eval_non_gas_f1": 0.9115646258503401,
98
+ "eval_non_gas_precision": 0.9115646258503401,
99
+ "eval_non_gas_recall": 0.9115646258503401,
100
+ "eval_precision": 0.9084507042253521,
101
+ "eval_recall": 0.9084507042253521,
102
+ "eval_roc_auc": 0.9730765545654881,
103
+ "eval_runtime": 13.0718,
104
+ "eval_samples_per_second": 22.109,
105
+ "eval_steps_per_second": 1.454,
106
+ "step": 340
107
+ }
108
+ ],
109
+ "logging_steps": 500,
110
+ "max_steps": 425,
111
+ "num_input_tokens_seen": 0,
112
+ "num_train_epochs": 5,
113
+ "save_steps": 500,
114
+ "stateful_callbacks": {
115
+ "EarlyStoppingCallback": {
116
+ "args": {
117
+ "early_stopping_patience": 2,
118
+ "early_stopping_threshold": 0.0
119
+ },
120
+ "attributes": {
121
+ "early_stopping_patience_counter": 1
122
+ }
123
+ },
124
+ "TrainerControl": {
125
+ "args": {
126
+ "should_epoch_stop": false,
127
+ "should_evaluate": false,
128
+ "should_log": false,
129
+ "should_save": true,
130
+ "should_training_stop": false
131
+ },
132
+ "attributes": {}
133
+ }
134
+ },
135
+ "total_flos": 354673702625280.0,
136
+ "train_batch_size": 16,
137
+ "trial_name": null,
138
+ "trial_params": null
139
+ }
checkpoint-340/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdb9630f2ec73d46c11a0a22c41425e17dd472a7f7557576b8eff38cf51da9b9
3
+ size 5265
checkpoint-425/config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
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": "Non-Gaslighting",
16
+ "1": "Gaslighting"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "Gaslighting": 1,
23
+ "Non-Gaslighting": 0
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "xlm-roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "position_embedding_type": "absolute",
33
+ "problem_type": "single_label_classification",
34
+ "tie_word_embeddings": true,
35
+ "transformers_version": "5.0.0",
36
+ "type_vocab_size": 1,
37
+ "use_cache": false,
38
+ "vocab_size": 250002
39
+ }
checkpoint-425/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7ac699d23a6df393b55ade26aa532b95acf638fd46911ba61f86fff5d26ea88
3
+ size 1112204984
checkpoint-425/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02d8430f09364201de53bb264bbc88219edd9dc9fb35969e11b930834bbfe2fc
3
+ size 2224530059
checkpoint-425/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ab4de3002df3f8a9f907f3b84af4619db19c85d48785e2f3a1d83c32547a1e4
3
+ size 14645
checkpoint-425/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b22192c62e378b4a155269efe220c2cacf478062f193bb1de8ca6293c63b3ff
3
+ size 1383
checkpoint-425/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1843d69d836283fb093fd990cc35bcfd014eccedd2c630705f4fec60bd7ebb92
3
+ size 1465
checkpoint-425/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5b633524ba90477daaba16ec27580a08a2856ae0ee8c33d9f5f9358378d3b35
3
+ size 16781751
checkpoint-425/tokenizer_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "is_local": false,
8
+ "mask_token": "<mask>",
9
+ "model_max_length": 512,
10
+ "pad_token": "<pad>",
11
+ "sep_token": "</s>",
12
+ "tokenizer_class": "XLMRobertaTokenizer",
13
+ "unk_token": "<unk>"
14
+ }
checkpoint-425/trainer_state.json ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 425,
3
+ "best_metric": 0.924187725631769,
4
+ "best_model_checkpoint": "D:/Thesis/Taglish_Gaslighting_V3\\model_outputs\\xlm-roberta_binary\\checkpoint-425",
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 425,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 13.04288101196289,
15
+ "learning_rate": 1.8010471204188483e-05,
16
+ "loss": 0.6918394200942095,
17
+ "step": 85
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.7093425605536332,
22
+ "eval_f1": 0.676923076923077,
23
+ "eval_loss": 0.5885154604911804,
24
+ "eval_macro_f1": 0.7063860667634252,
25
+ "eval_non_gas_f1": 0.7358490566037735,
26
+ "eval_non_gas_precision": 0.6842105263157895,
27
+ "eval_non_gas_recall": 0.7959183673469388,
28
+ "eval_precision": 0.7457627118644068,
29
+ "eval_recall": 0.6197183098591549,
30
+ "eval_roc_auc": 0.8050445530324807,
31
+ "eval_runtime": 13.2767,
32
+ "eval_samples_per_second": 21.768,
33
+ "eval_steps_per_second": 1.431,
34
+ "step": 85
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "grad_norm": 11.096158981323242,
39
+ "learning_rate": 1.356020942408377e-05,
40
+ "loss": 0.5428493724149817,
41
+ "step": 170
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.8027681660899654,
46
+ "eval_f1": 0.7764705882352941,
47
+ "eval_loss": 0.48855453729629517,
48
+ "eval_macro_f1": 0.8,
49
+ "eval_non_gas_f1": 0.8235294117647058,
50
+ "eval_non_gas_precision": 0.7556818181818182,
51
+ "eval_non_gas_recall": 0.9047619047619048,
52
+ "eval_precision": 0.8761061946902655,
53
+ "eval_recall": 0.6971830985915493,
54
+ "eval_roc_auc": 0.898438248538852,
55
+ "eval_runtime": 13.3951,
56
+ "eval_samples_per_second": 21.575,
57
+ "eval_steps_per_second": 1.418,
58
+ "step": 170
59
+ },
60
+ {
61
+ "epoch": 3.0,
62
+ "grad_norm": 5.319683074951172,
63
+ "learning_rate": 9.109947643979057e-06,
64
+ "loss": 0.4088932710535386,
65
+ "step": 255
66
+ },
67
+ {
68
+ "epoch": 3.0,
69
+ "eval_accuracy": 0.9134948096885813,
70
+ "eval_f1": 0.9158249158249159,
71
+ "eval_loss": 0.2529641091823578,
72
+ "eval_macro_f1": 0.9134284721473334,
73
+ "eval_non_gas_f1": 0.9110320284697508,
74
+ "eval_non_gas_precision": 0.9552238805970149,
75
+ "eval_non_gas_recall": 0.8707482993197279,
76
+ "eval_precision": 0.8774193548387097,
77
+ "eval_recall": 0.9577464788732394,
78
+ "eval_roc_auc": 0.9593752994155408,
79
+ "eval_runtime": 13.2242,
80
+ "eval_samples_per_second": 21.854,
81
+ "eval_steps_per_second": 1.437,
82
+ "step": 255
83
+ },
84
+ {
85
+ "epoch": 4.0,
86
+ "grad_norm": 4.501065254211426,
87
+ "learning_rate": 4.712041884816754e-06,
88
+ "loss": 0.2737856921027688,
89
+ "step": 340
90
+ },
91
+ {
92
+ "epoch": 4.0,
93
+ "eval_accuracy": 0.9100346020761245,
94
+ "eval_f1": 0.9084507042253521,
95
+ "eval_loss": 0.2273072451353073,
96
+ "eval_macro_f1": 0.9100076650378461,
97
+ "eval_non_gas_f1": 0.9115646258503401,
98
+ "eval_non_gas_precision": 0.9115646258503401,
99
+ "eval_non_gas_recall": 0.9115646258503401,
100
+ "eval_precision": 0.9084507042253521,
101
+ "eval_recall": 0.9084507042253521,
102
+ "eval_roc_auc": 0.9730765545654881,
103
+ "eval_runtime": 13.0718,
104
+ "eval_samples_per_second": 22.109,
105
+ "eval_steps_per_second": 1.454,
106
+ "step": 340
107
+ },
108
+ {
109
+ "epoch": 5.0,
110
+ "grad_norm": 4.042298316955566,
111
+ "learning_rate": 2.617801047120419e-07,
112
+ "loss": 0.2015650805305032,
113
+ "step": 425
114
+ },
115
+ {
116
+ "epoch": 5.0,
117
+ "eval_accuracy": 0.9273356401384083,
118
+ "eval_f1": 0.924187725631769,
119
+ "eval_loss": 0.2408331334590912,
120
+ "eval_macro_f1": 0.927210141885652,
121
+ "eval_non_gas_f1": 0.9302325581395349,
122
+ "eval_non_gas_precision": 0.9090909090909091,
123
+ "eval_non_gas_recall": 0.9523809523809523,
124
+ "eval_precision": 0.9481481481481482,
125
+ "eval_recall": 0.9014084507042254,
126
+ "eval_roc_auc": 0.9746095621347131,
127
+ "eval_runtime": 13.1098,
128
+ "eval_samples_per_second": 22.045,
129
+ "eval_steps_per_second": 1.449,
130
+ "step": 425
131
+ }
132
+ ],
133
+ "logging_steps": 500,
134
+ "max_steps": 425,
135
+ "num_input_tokens_seen": 0,
136
+ "num_train_epochs": 5,
137
+ "save_steps": 500,
138
+ "stateful_callbacks": {
139
+ "EarlyStoppingCallback": {
140
+ "args": {
141
+ "early_stopping_patience": 2,
142
+ "early_stopping_threshold": 0.0
143
+ },
144
+ "attributes": {
145
+ "early_stopping_patience_counter": 0
146
+ }
147
+ },
148
+ "TrainerControl": {
149
+ "args": {
150
+ "should_epoch_stop": false,
151
+ "should_evaluate": false,
152
+ "should_log": false,
153
+ "should_save": true,
154
+ "should_training_stop": true
155
+ },
156
+ "attributes": {}
157
+ }
158
+ },
159
+ "total_flos": 443342128281600.0,
160
+ "train_batch_size": 16,
161
+ "trial_name": null,
162
+ "trial_params": null
163
+ }
checkpoint-425/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdb9630f2ec73d46c11a0a22c41425e17dd472a7f7557576b8eff38cf51da9b9
3
+ size 5265
comprehensive_results_binary.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "val": {
3
+ "loss": 0.2408331334590912,
4
+ "accuracy": 0.9273356401384083,
5
+ "precision": 0.9481481481481482,
6
+ "recall": 0.9014084507042254,
7
+ "f1": 0.924187725631769,
8
+ "macro_f1": 0.927210141885652,
9
+ "roc_auc": 0.9746095621347131,
10
+ "non_gas_precision": 0.9090909090909091,
11
+ "non_gas_recall": 0.9523809523809523,
12
+ "non_gas_f1": 0.9302325581395349,
13
+ "runtime": 13.2248,
14
+ "samples_per_second": 21.853,
15
+ "steps_per_second": 1.437,
16
+ "confusion_matrix": [
17
+ [
18
+ 140,
19
+ 7
20
+ ],
21
+ [
22
+ 14,
23
+ 128
24
+ ]
25
+ ]
26
+ },
27
+ "test_id": {
28
+ "loss": 0.33420857787132263,
29
+ "accuracy": 0.8927335640138409,
30
+ "precision": 0.8992805755395683,
31
+ "recall": 0.8802816901408451,
32
+ "f1": 0.8896797153024911,
33
+ "macro_f1": 0.8926513054626934,
34
+ "roc_auc": 0.951806074542493,
35
+ "non_gas_precision": 0.8866666666666667,
36
+ "non_gas_recall": 0.9047619047619048,
37
+ "non_gas_f1": 0.8956228956228957,
38
+ "runtime": 13.0723,
39
+ "samples_per_second": 22.108,
40
+ "steps_per_second": 1.453,
41
+ "confusion_matrix": [
42
+ [
43
+ 133,
44
+ 14
45
+ ],
46
+ [
47
+ 17,
48
+ 125
49
+ ]
50
+ ]
51
+ },
52
+ "test_ood": {
53
+ "loss": 0.5717462301254272,
54
+ "accuracy": 0.8170289855072463,
55
+ "precision": 0.9970414201183432,
56
+ "recall": 0.7711670480549199,
57
+ "f1": 0.8696774193548387,
58
+ "macro_f1": 0.781343268947936,
59
+ "roc_auc": 0.9866480947169436,
60
+ "non_gas_precision": 0.5327102803738317,
61
+ "non_gas_recall": 0.991304347826087,
62
+ "non_gas_f1": 0.6930091185410334,
63
+ "runtime": 13.3686,
64
+ "samples_per_second": 41.291,
65
+ "steps_per_second": 2.618,
66
+ "confusion_matrix": [
67
+ [
68
+ 114,
69
+ 1
70
+ ],
71
+ [
72
+ 100,
73
+ 337
74
+ ]
75
+ ]
76
+ },
77
+ "delta_f1": 0.02000229594765246,
78
+ "delta_f1_interpretation": "\u2705 Robust generalization (minimal domain sensitivity)"
79
+ }
config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
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": "Non-Gaslighting",
16
+ "1": "Gaslighting"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "Gaslighting": 1,
23
+ "Non-Gaslighting": 0
24
+ },
25
+ "layer_norm_eps": 1e-05,
26
+ "max_position_embeddings": 514,
27
+ "model_type": "xlm-roberta",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "output_past": true,
31
+ "pad_token_id": 1,
32
+ "position_embedding_type": "absolute",
33
+ "problem_type": "single_label_classification",
34
+ "tie_word_embeddings": true,
35
+ "transformers_version": "5.0.0",
36
+ "type_vocab_size": 1,
37
+ "use_cache": false,
38
+ "vocab_size": 250002
39
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7ac699d23a6df393b55ade26aa532b95acf638fd46911ba61f86fff5d26ea88
3
+ size 1112204984
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5b633524ba90477daaba16ec27580a08a2856ae0ee8c33d9f5f9358378d3b35
3
+ size 16781751
tokenizer_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "is_local": false,
8
+ "mask_token": "<mask>",
9
+ "model_max_length": 512,
10
+ "pad_token": "<pad>",
11
+ "sep_token": "</s>",
12
+ "tokenizer_class": "XLMRobertaTokenizer",
13
+ "unk_token": "<unk>"
14
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdb9630f2ec73d46c11a0a22c41425e17dd472a7f7557576b8eff38cf51da9b9
3
+ size 5265