Addyk24 commited on
Commit
642da08
·
verified ·
1 Parent(s): 162778f

Upload fine-tuned Intent Classifier model -- 23:43

Browse files
checkpoint-2000/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "hidden_dim": 3072,
10
+ "id2label": {
11
+ "0": "debug",
12
+ "1": "threat"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "label2id": {
16
+ "debug": 0,
17
+ "threat": 1
18
+ },
19
+ "max_position_embeddings": 512,
20
+ "model_type": "distilbert",
21
+ "n_heads": 12,
22
+ "n_layers": 6,
23
+ "pad_token_id": 0,
24
+ "problem_type": "single_label_classification",
25
+ "qa_dropout": 0.1,
26
+ "seq_classif_dropout": 0.2,
27
+ "sinusoidal_pos_embds": false,
28
+ "tie_weights_": true,
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.52.4",
31
+ "vocab_size": 30522
32
+ }
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e163f30ed9bbf0b4856c7f2453a2d5ffc2f1ec7e7f7fbe7fa34f5fc945ce1be
3
+ size 267832560
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b3631a04dca350ebb8847947b43f9bb55dabe4b7bb48fb04a8529897d26504c
3
+ size 535727290
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20df78ff05dba2dc3340668315cac48cb861b521987a7588f6045438a4ffb8cc
3
+ size 14244
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc447d7e1b85b9ccc116a56989dc4b3d9341fce2e743148a1cc5064d444188a7
3
+ size 1064
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,212 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2000,
3
+ "best_metric": 0.9999500948197181,
4
+ "best_model_checkpoint": "trained_models/intent_classifier/checkpoint-2000",
5
+ "epoch": 1.596169193934557,
6
+ "eval_steps": 1000,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0007980845969672786,
14
+ "grad_norm": 51713.75390625,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.695,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.07980845969672785,
21
+ "grad_norm": 91372.3984375,
22
+ "learning_rate": 1.98e-06,
23
+ "loss": 0.6689,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.1596169193934557,
28
+ "grad_norm": 47582.55078125,
29
+ "learning_rate": 3.980000000000001e-06,
30
+ "loss": 0.2889,
31
+ "step": 200
32
+ },
33
+ {
34
+ "epoch": 0.23942537909018355,
35
+ "grad_norm": 2023.180908203125,
36
+ "learning_rate": 5.98e-06,
37
+ "loss": 0.0206,
38
+ "step": 300
39
+ },
40
+ {
41
+ "epoch": 0.3192338387869114,
42
+ "grad_norm": 105.5784912109375,
43
+ "learning_rate": 7.980000000000002e-06,
44
+ "loss": 0.0006,
45
+ "step": 400
46
+ },
47
+ {
48
+ "epoch": 0.39904229848363926,
49
+ "grad_norm": 2.8826353549957275,
50
+ "learning_rate": 9.980000000000001e-06,
51
+ "loss": 0.0016,
52
+ "step": 500
53
+ },
54
+ {
55
+ "epoch": 0.4788507581803671,
56
+ "grad_norm": 0.18906153738498688,
57
+ "learning_rate": 1.198e-05,
58
+ "loss": 0.0,
59
+ "step": 600
60
+ },
61
+ {
62
+ "epoch": 0.5586592178770949,
63
+ "grad_norm": 0.06490982323884964,
64
+ "learning_rate": 1.398e-05,
65
+ "loss": 0.001,
66
+ "step": 700
67
+ },
68
+ {
69
+ "epoch": 0.6384676775738228,
70
+ "grad_norm": 0.07483379542827606,
71
+ "learning_rate": 1.5980000000000003e-05,
72
+ "loss": 0.0026,
73
+ "step": 800
74
+ },
75
+ {
76
+ "epoch": 0.7182761372705507,
77
+ "grad_norm": 0.029934018850326538,
78
+ "learning_rate": 1.798e-05,
79
+ "loss": 0.0027,
80
+ "step": 900
81
+ },
82
+ {
83
+ "epoch": 0.7980845969672785,
84
+ "grad_norm": 0.023010307922959328,
85
+ "learning_rate": 1.9980000000000002e-05,
86
+ "loss": 0.0049,
87
+ "step": 1000
88
+ },
89
+ {
90
+ "epoch": 0.7980845969672785,
91
+ "eval_accuracy": 0.9999001896396846,
92
+ "eval_f1": 0.9999001896386902,
93
+ "eval_f1_macro": 0.9999001896386903,
94
+ "eval_loss": 0.0016707783797755837,
95
+ "eval_runtime": 88.2175,
96
+ "eval_samples_per_second": 227.143,
97
+ "eval_steps_per_second": 1.78,
98
+ "step": 1000
99
+ },
100
+ {
101
+ "epoch": 0.8778930566640064,
102
+ "grad_norm": 0.030565178021788597,
103
+ "learning_rate": 1.9282348677056906e-05,
104
+ "loss": 0.0091,
105
+ "step": 1100
106
+ },
107
+ {
108
+ "epoch": 0.9577015163607342,
109
+ "grad_norm": 0.025367770344018936,
110
+ "learning_rate": 1.8557448350851758e-05,
111
+ "loss": 0.0029,
112
+ "step": 1200
113
+ },
114
+ {
115
+ "epoch": 1.037509976057462,
116
+ "grad_norm": 0.01835496723651886,
117
+ "learning_rate": 1.7832548024646613e-05,
118
+ "loss": 0.0025,
119
+ "step": 1300
120
+ },
121
+ {
122
+ "epoch": 1.1173184357541899,
123
+ "grad_norm": 0.013528961688280106,
124
+ "learning_rate": 1.7107647698441465e-05,
125
+ "loss": 0.0004,
126
+ "step": 1400
127
+ },
128
+ {
129
+ "epoch": 1.1971268954509178,
130
+ "grad_norm": 0.013094124384224415,
131
+ "learning_rate": 1.638274737223632e-05,
132
+ "loss": 0.0043,
133
+ "step": 1500
134
+ },
135
+ {
136
+ "epoch": 1.2769353551476457,
137
+ "grad_norm": 0.010992957279086113,
138
+ "learning_rate": 1.565784704603117e-05,
139
+ "loss": 0.0001,
140
+ "step": 1600
141
+ },
142
+ {
143
+ "epoch": 1.3567438148443736,
144
+ "grad_norm": 0.009696166031062603,
145
+ "learning_rate": 1.4932946719826025e-05,
146
+ "loss": 0.0001,
147
+ "step": 1700
148
+ },
149
+ {
150
+ "epoch": 1.4365522745411012,
151
+ "grad_norm": 0.008244643919169903,
152
+ "learning_rate": 1.4208046393620877e-05,
153
+ "loss": 0.0,
154
+ "step": 1800
155
+ },
156
+ {
157
+ "epoch": 1.5163607342378294,
158
+ "grad_norm": 0.00833104643970728,
159
+ "learning_rate": 1.348314606741573e-05,
160
+ "loss": 0.0,
161
+ "step": 1900
162
+ },
163
+ {
164
+ "epoch": 1.596169193934557,
165
+ "grad_norm": 0.008447665721178055,
166
+ "learning_rate": 1.2758245741210584e-05,
167
+ "loss": 0.0031,
168
+ "step": 2000
169
+ },
170
+ {
171
+ "epoch": 1.596169193934557,
172
+ "eval_accuracy": 0.9999500948198423,
173
+ "eval_f1": 0.9999500948197181,
174
+ "eval_f1_macro": 0.999950094819718,
175
+ "eval_loss": 0.000961420766543597,
176
+ "eval_runtime": 88.1534,
177
+ "eval_samples_per_second": 227.308,
178
+ "eval_steps_per_second": 1.781,
179
+ "step": 2000
180
+ }
181
+ ],
182
+ "logging_steps": 100,
183
+ "max_steps": 3759,
184
+ "num_input_tokens_seen": 0,
185
+ "num_train_epochs": 3,
186
+ "save_steps": 1000,
187
+ "stateful_callbacks": {
188
+ "EarlyStoppingCallback": {
189
+ "args": {
190
+ "early_stopping_patience": 3,
191
+ "early_stopping_threshold": 0.0
192
+ },
193
+ "attributes": {
194
+ "early_stopping_patience_counter": 0
195
+ }
196
+ },
197
+ "TrainerControl": {
198
+ "args": {
199
+ "should_epoch_stop": false,
200
+ "should_evaluate": false,
201
+ "should_log": false,
202
+ "should_save": true,
203
+ "should_training_stop": false
204
+ },
205
+ "attributes": {}
206
+ }
207
+ },
208
+ "total_flos": 8475131698612224.0,
209
+ "train_batch_size": 64,
210
+ "trial_name": null,
211
+ "trial_params": null
212
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e28c46ef7bbe258c7c76e4214a74b22862afd61a1216391d20f64e53f25ddc3
3
+ size 5240
checkpoint-3000/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "hidden_dim": 3072,
10
+ "id2label": {
11
+ "0": "debug",
12
+ "1": "threat"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "label2id": {
16
+ "debug": 0,
17
+ "threat": 1
18
+ },
19
+ "max_position_embeddings": 512,
20
+ "model_type": "distilbert",
21
+ "n_heads": 12,
22
+ "n_layers": 6,
23
+ "pad_token_id": 0,
24
+ "problem_type": "single_label_classification",
25
+ "qa_dropout": 0.1,
26
+ "seq_classif_dropout": 0.2,
27
+ "sinusoidal_pos_embds": false,
28
+ "tie_weights_": true,
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.52.4",
31
+ "vocab_size": 30522
32
+ }
checkpoint-3000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1241e02e2df501cd1a21d8ce4900a0927ffdf3666fc2962849d92390aea69554
3
+ size 267832560
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b8254741c62c345cf306ed7e121832c33f71f8e29656a7e92b7b36ffc6d892e
3
+ size 535727290
checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88b5f1a5dfce4c46f6e46274af5806ee8126fb1e19258e1096134ca597bfa748
3
+ size 14244
checkpoint-3000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d6fef9910ec885ed4f1cf24c1af154e089de0b99f10b6d0f631010271c3574a
3
+ size 1064
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,293 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2000,
3
+ "best_metric": 0.9999500948197181,
4
+ "best_model_checkpoint": "trained_models/intent_classifier/checkpoint-2000",
5
+ "epoch": 2.3942537909018355,
6
+ "eval_steps": 1000,
7
+ "global_step": 3000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0007980845969672786,
14
+ "grad_norm": 51713.75390625,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.695,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.07980845969672785,
21
+ "grad_norm": 91372.3984375,
22
+ "learning_rate": 1.98e-06,
23
+ "loss": 0.6689,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.1596169193934557,
28
+ "grad_norm": 47582.55078125,
29
+ "learning_rate": 3.980000000000001e-06,
30
+ "loss": 0.2889,
31
+ "step": 200
32
+ },
33
+ {
34
+ "epoch": 0.23942537909018355,
35
+ "grad_norm": 2023.180908203125,
36
+ "learning_rate": 5.98e-06,
37
+ "loss": 0.0206,
38
+ "step": 300
39
+ },
40
+ {
41
+ "epoch": 0.3192338387869114,
42
+ "grad_norm": 105.5784912109375,
43
+ "learning_rate": 7.980000000000002e-06,
44
+ "loss": 0.0006,
45
+ "step": 400
46
+ },
47
+ {
48
+ "epoch": 0.39904229848363926,
49
+ "grad_norm": 2.8826353549957275,
50
+ "learning_rate": 9.980000000000001e-06,
51
+ "loss": 0.0016,
52
+ "step": 500
53
+ },
54
+ {
55
+ "epoch": 0.4788507581803671,
56
+ "grad_norm": 0.18906153738498688,
57
+ "learning_rate": 1.198e-05,
58
+ "loss": 0.0,
59
+ "step": 600
60
+ },
61
+ {
62
+ "epoch": 0.5586592178770949,
63
+ "grad_norm": 0.06490982323884964,
64
+ "learning_rate": 1.398e-05,
65
+ "loss": 0.001,
66
+ "step": 700
67
+ },
68
+ {
69
+ "epoch": 0.6384676775738228,
70
+ "grad_norm": 0.07483379542827606,
71
+ "learning_rate": 1.5980000000000003e-05,
72
+ "loss": 0.0026,
73
+ "step": 800
74
+ },
75
+ {
76
+ "epoch": 0.7182761372705507,
77
+ "grad_norm": 0.029934018850326538,
78
+ "learning_rate": 1.798e-05,
79
+ "loss": 0.0027,
80
+ "step": 900
81
+ },
82
+ {
83
+ "epoch": 0.7980845969672785,
84
+ "grad_norm": 0.023010307922959328,
85
+ "learning_rate": 1.9980000000000002e-05,
86
+ "loss": 0.0049,
87
+ "step": 1000
88
+ },
89
+ {
90
+ "epoch": 0.7980845969672785,
91
+ "eval_accuracy": 0.9999001896396846,
92
+ "eval_f1": 0.9999001896386902,
93
+ "eval_f1_macro": 0.9999001896386903,
94
+ "eval_loss": 0.0016707783797755837,
95
+ "eval_runtime": 88.2175,
96
+ "eval_samples_per_second": 227.143,
97
+ "eval_steps_per_second": 1.78,
98
+ "step": 1000
99
+ },
100
+ {
101
+ "epoch": 0.8778930566640064,
102
+ "grad_norm": 0.030565178021788597,
103
+ "learning_rate": 1.9282348677056906e-05,
104
+ "loss": 0.0091,
105
+ "step": 1100
106
+ },
107
+ {
108
+ "epoch": 0.9577015163607342,
109
+ "grad_norm": 0.025367770344018936,
110
+ "learning_rate": 1.8557448350851758e-05,
111
+ "loss": 0.0029,
112
+ "step": 1200
113
+ },
114
+ {
115
+ "epoch": 1.037509976057462,
116
+ "grad_norm": 0.01835496723651886,
117
+ "learning_rate": 1.7832548024646613e-05,
118
+ "loss": 0.0025,
119
+ "step": 1300
120
+ },
121
+ {
122
+ "epoch": 1.1173184357541899,
123
+ "grad_norm": 0.013528961688280106,
124
+ "learning_rate": 1.7107647698441465e-05,
125
+ "loss": 0.0004,
126
+ "step": 1400
127
+ },
128
+ {
129
+ "epoch": 1.1971268954509178,
130
+ "grad_norm": 0.013094124384224415,
131
+ "learning_rate": 1.638274737223632e-05,
132
+ "loss": 0.0043,
133
+ "step": 1500
134
+ },
135
+ {
136
+ "epoch": 1.2769353551476457,
137
+ "grad_norm": 0.010992957279086113,
138
+ "learning_rate": 1.565784704603117e-05,
139
+ "loss": 0.0001,
140
+ "step": 1600
141
+ },
142
+ {
143
+ "epoch": 1.3567438148443736,
144
+ "grad_norm": 0.009696166031062603,
145
+ "learning_rate": 1.4932946719826025e-05,
146
+ "loss": 0.0001,
147
+ "step": 1700
148
+ },
149
+ {
150
+ "epoch": 1.4365522745411012,
151
+ "grad_norm": 0.008244643919169903,
152
+ "learning_rate": 1.4208046393620877e-05,
153
+ "loss": 0.0,
154
+ "step": 1800
155
+ },
156
+ {
157
+ "epoch": 1.5163607342378294,
158
+ "grad_norm": 0.00833104643970728,
159
+ "learning_rate": 1.348314606741573e-05,
160
+ "loss": 0.0,
161
+ "step": 1900
162
+ },
163
+ {
164
+ "epoch": 1.596169193934557,
165
+ "grad_norm": 0.008447665721178055,
166
+ "learning_rate": 1.2758245741210584e-05,
167
+ "loss": 0.0031,
168
+ "step": 2000
169
+ },
170
+ {
171
+ "epoch": 1.596169193934557,
172
+ "eval_accuracy": 0.9999500948198423,
173
+ "eval_f1": 0.9999500948197181,
174
+ "eval_f1_macro": 0.999950094819718,
175
+ "eval_loss": 0.000961420766543597,
176
+ "eval_runtime": 88.1534,
177
+ "eval_samples_per_second": 227.308,
178
+ "eval_steps_per_second": 1.781,
179
+ "step": 2000
180
+ },
181
+ {
182
+ "epoch": 1.675977653631285,
183
+ "grad_norm": 0.008628464303910732,
184
+ "learning_rate": 1.2033345415005438e-05,
185
+ "loss": 0.0036,
186
+ "step": 2100
187
+ },
188
+ {
189
+ "epoch": 1.7557861133280128,
190
+ "grad_norm": 0.006202944088727236,
191
+ "learning_rate": 1.1308445088800291e-05,
192
+ "loss": 0.0,
193
+ "step": 2200
194
+ },
195
+ {
196
+ "epoch": 1.8355945730247405,
197
+ "grad_norm": 70615.15625,
198
+ "learning_rate": 1.0583544762595143e-05,
199
+ "loss": 0.0061,
200
+ "step": 2300
201
+ },
202
+ {
203
+ "epoch": 1.9154030327214686,
204
+ "grad_norm": 0.007761036045849323,
205
+ "learning_rate": 9.858644436389997e-06,
206
+ "loss": 0.0,
207
+ "step": 2400
208
+ },
209
+ {
210
+ "epoch": 1.9952114924181963,
211
+ "grad_norm": 0.006005772389471531,
212
+ "learning_rate": 9.13374411018485e-06,
213
+ "loss": 0.0,
214
+ "step": 2500
215
+ },
216
+ {
217
+ "epoch": 2.075019952114924,
218
+ "grad_norm": 0.005284695420414209,
219
+ "learning_rate": 8.408843783979704e-06,
220
+ "loss": 0.0,
221
+ "step": 2600
222
+ },
223
+ {
224
+ "epoch": 2.154828411811652,
225
+ "grad_norm": 0.004640200641006231,
226
+ "learning_rate": 7.683943457774557e-06,
227
+ "loss": 0.0,
228
+ "step": 2700
229
+ },
230
+ {
231
+ "epoch": 2.2346368715083798,
232
+ "grad_norm": 0.004441382829099894,
233
+ "learning_rate": 6.95904313156941e-06,
234
+ "loss": 0.0,
235
+ "step": 2800
236
+ },
237
+ {
238
+ "epoch": 2.314445331205108,
239
+ "grad_norm": 0.007967362180352211,
240
+ "learning_rate": 6.234142805364263e-06,
241
+ "loss": 0.0029,
242
+ "step": 2900
243
+ },
244
+ {
245
+ "epoch": 2.3942537909018355,
246
+ "grad_norm": 0.0051362947560846806,
247
+ "learning_rate": 5.509242479159116e-06,
248
+ "loss": 0.0,
249
+ "step": 3000
250
+ },
251
+ {
252
+ "epoch": 2.3942537909018355,
253
+ "eval_accuracy": 0.9999500948198423,
254
+ "eval_f1": 0.9999500948197181,
255
+ "eval_f1_macro": 0.999950094819718,
256
+ "eval_loss": 0.0008828217978589237,
257
+ "eval_runtime": 87.5417,
258
+ "eval_samples_per_second": 228.897,
259
+ "eval_steps_per_second": 1.793,
260
+ "step": 3000
261
+ }
262
+ ],
263
+ "logging_steps": 100,
264
+ "max_steps": 3759,
265
+ "num_input_tokens_seen": 0,
266
+ "num_train_epochs": 3,
267
+ "save_steps": 1000,
268
+ "stateful_callbacks": {
269
+ "EarlyStoppingCallback": {
270
+ "args": {
271
+ "early_stopping_patience": 3,
272
+ "early_stopping_threshold": 0.0
273
+ },
274
+ "attributes": {
275
+ "early_stopping_patience_counter": 1
276
+ }
277
+ },
278
+ "TrainerControl": {
279
+ "args": {
280
+ "should_epoch_stop": false,
281
+ "should_evaluate": false,
282
+ "should_log": false,
283
+ "should_save": true,
284
+ "should_training_stop": false
285
+ },
286
+ "attributes": {}
287
+ }
288
+ },
289
+ "total_flos": 1.2711306640232448e+16,
290
+ "train_batch_size": 64,
291
+ "trial_name": null,
292
+ "trial_params": null
293
+ }
checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e28c46ef7bbe258c7c76e4214a74b22862afd61a1216391d20f64e53f25ddc3
3
+ size 5240
checkpoint-3759/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "hidden_dim": 3072,
10
+ "id2label": {
11
+ "0": "debug",
12
+ "1": "threat"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "label2id": {
16
+ "debug": 0,
17
+ "threat": 1
18
+ },
19
+ "max_position_embeddings": 512,
20
+ "model_type": "distilbert",
21
+ "n_heads": 12,
22
+ "n_layers": 6,
23
+ "pad_token_id": 0,
24
+ "problem_type": "single_label_classification",
25
+ "qa_dropout": 0.1,
26
+ "seq_classif_dropout": 0.2,
27
+ "sinusoidal_pos_embds": false,
28
+ "tie_weights_": true,
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.52.4",
31
+ "vocab_size": 30522
32
+ }
checkpoint-3759/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:479909e91032977bf85d53beadb66b5b4efb39bfa7b9d714449acfee0f99b09e
3
+ size 267832560
checkpoint-3759/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd9e3b740a4e9ff2bf3ef2ce730d1ec74d895bf7391e7b35d3b2b95bad95576f
3
+ size 535727290
checkpoint-3759/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:858670a83d02ea0df01a01f5161b4421423098c663625a3c07bf27da5c02de2b
3
+ size 14244
checkpoint-3759/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-3759/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d42694c09a3d1e4eeeb582d2e746c1f7a70932c533556e2a30401816641deb7
3
+ size 1064
checkpoint-3759/trainer_state.json ADDED
@@ -0,0 +1,342 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2000,
3
+ "best_metric": 0.9999500948197181,
4
+ "best_model_checkpoint": "trained_models/intent_classifier/checkpoint-2000",
5
+ "epoch": 3.0,
6
+ "eval_steps": 1000,
7
+ "global_step": 3759,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0007980845969672786,
14
+ "grad_norm": 51713.75390625,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.695,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.07980845969672785,
21
+ "grad_norm": 91372.3984375,
22
+ "learning_rate": 1.98e-06,
23
+ "loss": 0.6689,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.1596169193934557,
28
+ "grad_norm": 47582.55078125,
29
+ "learning_rate": 3.980000000000001e-06,
30
+ "loss": 0.2889,
31
+ "step": 200
32
+ },
33
+ {
34
+ "epoch": 0.23942537909018355,
35
+ "grad_norm": 2023.180908203125,
36
+ "learning_rate": 5.98e-06,
37
+ "loss": 0.0206,
38
+ "step": 300
39
+ },
40
+ {
41
+ "epoch": 0.3192338387869114,
42
+ "grad_norm": 105.5784912109375,
43
+ "learning_rate": 7.980000000000002e-06,
44
+ "loss": 0.0006,
45
+ "step": 400
46
+ },
47
+ {
48
+ "epoch": 0.39904229848363926,
49
+ "grad_norm": 2.8826353549957275,
50
+ "learning_rate": 9.980000000000001e-06,
51
+ "loss": 0.0016,
52
+ "step": 500
53
+ },
54
+ {
55
+ "epoch": 0.4788507581803671,
56
+ "grad_norm": 0.18906153738498688,
57
+ "learning_rate": 1.198e-05,
58
+ "loss": 0.0,
59
+ "step": 600
60
+ },
61
+ {
62
+ "epoch": 0.5586592178770949,
63
+ "grad_norm": 0.06490982323884964,
64
+ "learning_rate": 1.398e-05,
65
+ "loss": 0.001,
66
+ "step": 700
67
+ },
68
+ {
69
+ "epoch": 0.6384676775738228,
70
+ "grad_norm": 0.07483379542827606,
71
+ "learning_rate": 1.5980000000000003e-05,
72
+ "loss": 0.0026,
73
+ "step": 800
74
+ },
75
+ {
76
+ "epoch": 0.7182761372705507,
77
+ "grad_norm": 0.029934018850326538,
78
+ "learning_rate": 1.798e-05,
79
+ "loss": 0.0027,
80
+ "step": 900
81
+ },
82
+ {
83
+ "epoch": 0.7980845969672785,
84
+ "grad_norm": 0.023010307922959328,
85
+ "learning_rate": 1.9980000000000002e-05,
86
+ "loss": 0.0049,
87
+ "step": 1000
88
+ },
89
+ {
90
+ "epoch": 0.7980845969672785,
91
+ "eval_accuracy": 0.9999001896396846,
92
+ "eval_f1": 0.9999001896386902,
93
+ "eval_f1_macro": 0.9999001896386903,
94
+ "eval_loss": 0.0016707783797755837,
95
+ "eval_runtime": 88.2175,
96
+ "eval_samples_per_second": 227.143,
97
+ "eval_steps_per_second": 1.78,
98
+ "step": 1000
99
+ },
100
+ {
101
+ "epoch": 0.8778930566640064,
102
+ "grad_norm": 0.030565178021788597,
103
+ "learning_rate": 1.9282348677056906e-05,
104
+ "loss": 0.0091,
105
+ "step": 1100
106
+ },
107
+ {
108
+ "epoch": 0.9577015163607342,
109
+ "grad_norm": 0.025367770344018936,
110
+ "learning_rate": 1.8557448350851758e-05,
111
+ "loss": 0.0029,
112
+ "step": 1200
113
+ },
114
+ {
115
+ "epoch": 1.037509976057462,
116
+ "grad_norm": 0.01835496723651886,
117
+ "learning_rate": 1.7832548024646613e-05,
118
+ "loss": 0.0025,
119
+ "step": 1300
120
+ },
121
+ {
122
+ "epoch": 1.1173184357541899,
123
+ "grad_norm": 0.013528961688280106,
124
+ "learning_rate": 1.7107647698441465e-05,
125
+ "loss": 0.0004,
126
+ "step": 1400
127
+ },
128
+ {
129
+ "epoch": 1.1971268954509178,
130
+ "grad_norm": 0.013094124384224415,
131
+ "learning_rate": 1.638274737223632e-05,
132
+ "loss": 0.0043,
133
+ "step": 1500
134
+ },
135
+ {
136
+ "epoch": 1.2769353551476457,
137
+ "grad_norm": 0.010992957279086113,
138
+ "learning_rate": 1.565784704603117e-05,
139
+ "loss": 0.0001,
140
+ "step": 1600
141
+ },
142
+ {
143
+ "epoch": 1.3567438148443736,
144
+ "grad_norm": 0.009696166031062603,
145
+ "learning_rate": 1.4932946719826025e-05,
146
+ "loss": 0.0001,
147
+ "step": 1700
148
+ },
149
+ {
150
+ "epoch": 1.4365522745411012,
151
+ "grad_norm": 0.008244643919169903,
152
+ "learning_rate": 1.4208046393620877e-05,
153
+ "loss": 0.0,
154
+ "step": 1800
155
+ },
156
+ {
157
+ "epoch": 1.5163607342378294,
158
+ "grad_norm": 0.00833104643970728,
159
+ "learning_rate": 1.348314606741573e-05,
160
+ "loss": 0.0,
161
+ "step": 1900
162
+ },
163
+ {
164
+ "epoch": 1.596169193934557,
165
+ "grad_norm": 0.008447665721178055,
166
+ "learning_rate": 1.2758245741210584e-05,
167
+ "loss": 0.0031,
168
+ "step": 2000
169
+ },
170
+ {
171
+ "epoch": 1.596169193934557,
172
+ "eval_accuracy": 0.9999500948198423,
173
+ "eval_f1": 0.9999500948197181,
174
+ "eval_f1_macro": 0.999950094819718,
175
+ "eval_loss": 0.000961420766543597,
176
+ "eval_runtime": 88.1534,
177
+ "eval_samples_per_second": 227.308,
178
+ "eval_steps_per_second": 1.781,
179
+ "step": 2000
180
+ },
181
+ {
182
+ "epoch": 1.675977653631285,
183
+ "grad_norm": 0.008628464303910732,
184
+ "learning_rate": 1.2033345415005438e-05,
185
+ "loss": 0.0036,
186
+ "step": 2100
187
+ },
188
+ {
189
+ "epoch": 1.7557861133280128,
190
+ "grad_norm": 0.006202944088727236,
191
+ "learning_rate": 1.1308445088800291e-05,
192
+ "loss": 0.0,
193
+ "step": 2200
194
+ },
195
+ {
196
+ "epoch": 1.8355945730247405,
197
+ "grad_norm": 70615.15625,
198
+ "learning_rate": 1.0583544762595143e-05,
199
+ "loss": 0.0061,
200
+ "step": 2300
201
+ },
202
+ {
203
+ "epoch": 1.9154030327214686,
204
+ "grad_norm": 0.007761036045849323,
205
+ "learning_rate": 9.858644436389997e-06,
206
+ "loss": 0.0,
207
+ "step": 2400
208
+ },
209
+ {
210
+ "epoch": 1.9952114924181963,
211
+ "grad_norm": 0.006005772389471531,
212
+ "learning_rate": 9.13374411018485e-06,
213
+ "loss": 0.0,
214
+ "step": 2500
215
+ },
216
+ {
217
+ "epoch": 2.075019952114924,
218
+ "grad_norm": 0.005284695420414209,
219
+ "learning_rate": 8.408843783979704e-06,
220
+ "loss": 0.0,
221
+ "step": 2600
222
+ },
223
+ {
224
+ "epoch": 2.154828411811652,
225
+ "grad_norm": 0.004640200641006231,
226
+ "learning_rate": 7.683943457774557e-06,
227
+ "loss": 0.0,
228
+ "step": 2700
229
+ },
230
+ {
231
+ "epoch": 2.2346368715083798,
232
+ "grad_norm": 0.004441382829099894,
233
+ "learning_rate": 6.95904313156941e-06,
234
+ "loss": 0.0,
235
+ "step": 2800
236
+ },
237
+ {
238
+ "epoch": 2.314445331205108,
239
+ "grad_norm": 0.007967362180352211,
240
+ "learning_rate": 6.234142805364263e-06,
241
+ "loss": 0.0029,
242
+ "step": 2900
243
+ },
244
+ {
245
+ "epoch": 2.3942537909018355,
246
+ "grad_norm": 0.0051362947560846806,
247
+ "learning_rate": 5.509242479159116e-06,
248
+ "loss": 0.0,
249
+ "step": 3000
250
+ },
251
+ {
252
+ "epoch": 2.3942537909018355,
253
+ "eval_accuracy": 0.9999500948198423,
254
+ "eval_f1": 0.9999500948197181,
255
+ "eval_f1_macro": 0.999950094819718,
256
+ "eval_loss": 0.0008828217978589237,
257
+ "eval_runtime": 87.5417,
258
+ "eval_samples_per_second": 228.897,
259
+ "eval_steps_per_second": 1.793,
260
+ "step": 3000
261
+ },
262
+ {
263
+ "epoch": 2.474062250598563,
264
+ "grad_norm": 0.004741505719721317,
265
+ "learning_rate": 4.784342152953969e-06,
266
+ "loss": 0.0,
267
+ "step": 3100
268
+ },
269
+ {
270
+ "epoch": 2.5538707102952913,
271
+ "grad_norm": 0.004082200583070517,
272
+ "learning_rate": 4.059441826748822e-06,
273
+ "loss": 0.0,
274
+ "step": 3200
275
+ },
276
+ {
277
+ "epoch": 2.633679169992019,
278
+ "grad_norm": 0.004417167976498604,
279
+ "learning_rate": 3.3345415005436755e-06,
280
+ "loss": 0.0,
281
+ "step": 3300
282
+ },
283
+ {
284
+ "epoch": 2.713487629688747,
285
+ "grad_norm": 0.003513456555083394,
286
+ "learning_rate": 2.6096411743385286e-06,
287
+ "loss": 0.0,
288
+ "step": 3400
289
+ },
290
+ {
291
+ "epoch": 2.793296089385475,
292
+ "grad_norm": 0.0046548256650567055,
293
+ "learning_rate": 1.8847408481333817e-06,
294
+ "loss": 0.0031,
295
+ "step": 3500
296
+ },
297
+ {
298
+ "epoch": 2.8731045490822025,
299
+ "grad_norm": 0.004223688039928675,
300
+ "learning_rate": 1.1598405219282348e-06,
301
+ "loss": 0.0,
302
+ "step": 3600
303
+ },
304
+ {
305
+ "epoch": 2.9529130087789306,
306
+ "grad_norm": 0.00418103625997901,
307
+ "learning_rate": 4.349401957230881e-07,
308
+ "loss": 0.0006,
309
+ "step": 3700
310
+ }
311
+ ],
312
+ "logging_steps": 100,
313
+ "max_steps": 3759,
314
+ "num_input_tokens_seen": 0,
315
+ "num_train_epochs": 3,
316
+ "save_steps": 1000,
317
+ "stateful_callbacks": {
318
+ "EarlyStoppingCallback": {
319
+ "args": {
320
+ "early_stopping_patience": 3,
321
+ "early_stopping_threshold": 0.0
322
+ },
323
+ "attributes": {
324
+ "early_stopping_patience_counter": 1
325
+ }
326
+ },
327
+ "TrainerControl": {
328
+ "args": {
329
+ "should_epoch_stop": false,
330
+ "should_evaluate": false,
331
+ "should_log": false,
332
+ "should_save": true,
333
+ "should_training_stop": true
334
+ },
335
+ "attributes": {}
336
+ }
337
+ },
338
+ "total_flos": 1.59258930034176e+16,
339
+ "train_batch_size": 64,
340
+ "trial_name": null,
341
+ "trial_params": null
342
+ }
checkpoint-3759/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e28c46ef7bbe258c7c76e4214a74b22862afd61a1216391d20f64e53f25ddc3
3
+ size 5240
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "hidden_dim": 3072,
10
+ "id2label": {
11
+ "0": "debug",
12
+ "1": "threat"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "label2id": {
16
+ "debug": 0,
17
+ "threat": 1
18
+ },
19
+ "max_position_embeddings": 512,
20
+ "model_type": "distilbert",
21
+ "n_heads": 12,
22
+ "n_layers": 6,
23
+ "pad_token_id": 0,
24
+ "problem_type": "single_label_classification",
25
+ "qa_dropout": 0.1,
26
+ "seq_classif_dropout": 0.2,
27
+ "sinusoidal_pos_embds": false,
28
+ "tie_weights_": true,
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.52.4",
31
+ "vocab_size": 30522
32
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e163f30ed9bbf0b4856c7f2453a2d5ffc2f1ec7e7f7fbe7fa34f5fc945ce1be
3
+ size 267832560
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e28c46ef7bbe258c7c76e4214a74b22862afd61a1216391d20f64e53f25ddc3
3
+ size 5240