KSvendsen commited on
Commit
8bc9fac
·
verified ·
1 Parent(s): 5d81e3d

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,7 @@ 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-11595/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-3865/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ checkpoint-7730/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoint-11595/config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Normal",
16
+ "1": "Abusive",
17
+ "2": "Hate"
18
+ },
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "is_decoder": false,
22
+ "label2id": {
23
+ "Abusive": 1,
24
+ "Hate": 2,
25
+ "Normal": 0
26
+ },
27
+ "layer_norm_eps": 1e-05,
28
+ "max_position_embeddings": 514,
29
+ "model_type": "xlm-roberta",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "output_past": true,
33
+ "pad_token_id": 1,
34
+ "position_embedding_type": "absolute",
35
+ "problem_type": "single_label_classification",
36
+ "tie_word_embeddings": true,
37
+ "transformers_version": "5.0.0",
38
+ "type_vocab_size": 1,
39
+ "use_cache": false,
40
+ "vocab_size": 250002
41
+ }
checkpoint-11595/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1acbfdddadff0de9e27b9ba47d3487bf31dc11f779255a61a6b3accf8c0bfe2d
3
+ size 1112208060
checkpoint-11595/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbd94d2672ab9faa52d349016ab8e527fa5e42a46360f68a6a1fd2bce20e7bc3
3
+ size 2224539019
checkpoint-11595/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11857e656f912d990c3d505022fb5fb7a76b4755862eacc4b5c07defb8183528
3
+ size 14645
checkpoint-11595/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c2f7beefaa5612f64139f052d1d5d9c10c178c2764102ac72dc544d655e69fc
3
+ size 1465
checkpoint-11595/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42d3da0b556fcc5dfcd7158b90066b5069bd23df2df684a17c4bcfd8575016ae
3
+ size 16781739
checkpoint-11595/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-11595/trainer_state.json ADDED
@@ -0,0 +1,869 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 11595,
3
+ "best_metric": 0.7562243780359653,
4
+ "best_model_checkpoint": "./EA-HS/checkpoint-11595",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 11595,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0258732212160414,
14
+ "grad_norm": 5.261500358581543,
15
+ "learning_rate": 1.7068965517241382e-06,
16
+ "loss": 1.100949478149414,
17
+ "step": 100
18
+ },
19
+ {
20
+ "epoch": 0.0517464424320828,
21
+ "grad_norm": 5.354814529418945,
22
+ "learning_rate": 3.431034482758621e-06,
23
+ "loss": 1.1074339294433593,
24
+ "step": 200
25
+ },
26
+ {
27
+ "epoch": 0.07761966364812418,
28
+ "grad_norm": 10.960789680480957,
29
+ "learning_rate": 5.155172413793104e-06,
30
+ "loss": 1.0751912689208984,
31
+ "step": 300
32
+ },
33
+ {
34
+ "epoch": 0.1034928848641656,
35
+ "grad_norm": 7.228652477264404,
36
+ "learning_rate": 6.8793103448275865e-06,
37
+ "loss": 1.0659678649902344,
38
+ "step": 400
39
+ },
40
+ {
41
+ "epoch": 0.129366106080207,
42
+ "grad_norm": 13.746113777160645,
43
+ "learning_rate": 8.603448275862069e-06,
44
+ "loss": 1.017400894165039,
45
+ "step": 500
46
+ },
47
+ {
48
+ "epoch": 0.15523932729624837,
49
+ "grad_norm": 9.63514232635498,
50
+ "learning_rate": 1.0327586206896552e-05,
51
+ "loss": 0.9742707824707031,
52
+ "step": 600
53
+ },
54
+ {
55
+ "epoch": 0.18111254851228978,
56
+ "grad_norm": 12.705816268920898,
57
+ "learning_rate": 1.2051724137931036e-05,
58
+ "loss": 0.9275926971435546,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 0.2069857697283312,
63
+ "grad_norm": 26.888172149658203,
64
+ "learning_rate": 1.3775862068965518e-05,
65
+ "loss": 0.9081744384765625,
66
+ "step": 800
67
+ },
68
+ {
69
+ "epoch": 0.23285899094437257,
70
+ "grad_norm": 21.919212341308594,
71
+ "learning_rate": 1.55e-05,
72
+ "loss": 0.8668842315673828,
73
+ "step": 900
74
+ },
75
+ {
76
+ "epoch": 0.258732212160414,
77
+ "grad_norm": 18.860923767089844,
78
+ "learning_rate": 1.7224137931034483e-05,
79
+ "loss": 0.8243231964111328,
80
+ "step": 1000
81
+ },
82
+ {
83
+ "epoch": 0.2846054333764554,
84
+ "grad_norm": 16.422523498535156,
85
+ "learning_rate": 1.894827586206897e-05,
86
+ "loss": 0.8356568145751954,
87
+ "step": 1100
88
+ },
89
+ {
90
+ "epoch": 0.31047865459249674,
91
+ "grad_norm": 6.305636882781982,
92
+ "learning_rate": 1.9925251557259225e-05,
93
+ "loss": 0.8148372650146485,
94
+ "step": 1200
95
+ },
96
+ {
97
+ "epoch": 0.33635187580853815,
98
+ "grad_norm": 55.4698600769043,
99
+ "learning_rate": 1.9733588883564927e-05,
100
+ "loss": 0.7789091491699218,
101
+ "step": 1300
102
+ },
103
+ {
104
+ "epoch": 0.36222509702457956,
105
+ "grad_norm": 7.873316287994385,
106
+ "learning_rate": 1.954192620987063e-05,
107
+ "loss": 0.7773941040039063,
108
+ "step": 1400
109
+ },
110
+ {
111
+ "epoch": 0.38809831824062097,
112
+ "grad_norm": 8.828251838684082,
113
+ "learning_rate": 1.935026353617633e-05,
114
+ "loss": 0.7520027160644531,
115
+ "step": 1500
116
+ },
117
+ {
118
+ "epoch": 0.4139715394566624,
119
+ "grad_norm": 10.744029998779297,
120
+ "learning_rate": 1.9158600862482032e-05,
121
+ "loss": 0.7416606903076172,
122
+ "step": 1600
123
+ },
124
+ {
125
+ "epoch": 0.4398447606727037,
126
+ "grad_norm": 13.313233375549316,
127
+ "learning_rate": 1.8966938188787737e-05,
128
+ "loss": 0.7262284851074219,
129
+ "step": 1700
130
+ },
131
+ {
132
+ "epoch": 0.46571798188874514,
133
+ "grad_norm": 16.973875045776367,
134
+ "learning_rate": 1.8775275515093435e-05,
135
+ "loss": 0.7252954864501953,
136
+ "step": 1800
137
+ },
138
+ {
139
+ "epoch": 0.49159120310478654,
140
+ "grad_norm": 11.87275218963623,
141
+ "learning_rate": 1.858361284139914e-05,
142
+ "loss": 0.7501118469238282,
143
+ "step": 1900
144
+ },
145
+ {
146
+ "epoch": 0.517464424320828,
147
+ "grad_norm": 18.290884017944336,
148
+ "learning_rate": 1.8391950167704842e-05,
149
+ "loss": 0.741920394897461,
150
+ "step": 2000
151
+ },
152
+ {
153
+ "epoch": 0.5433376455368694,
154
+ "grad_norm": 5.779555797576904,
155
+ "learning_rate": 1.8200287494010544e-05,
156
+ "loss": 0.7136674499511719,
157
+ "step": 2100
158
+ },
159
+ {
160
+ "epoch": 0.5692108667529108,
161
+ "grad_norm": 29.206987380981445,
162
+ "learning_rate": 1.8008624820316245e-05,
163
+ "loss": 0.7118588256835937,
164
+ "step": 2200
165
+ },
166
+ {
167
+ "epoch": 0.5950840879689522,
168
+ "grad_norm": 28.444412231445312,
169
+ "learning_rate": 1.7816962146621947e-05,
170
+ "loss": 0.7275136566162109,
171
+ "step": 2300
172
+ },
173
+ {
174
+ "epoch": 0.6209573091849935,
175
+ "grad_norm": 16.02486228942871,
176
+ "learning_rate": 1.762529947292765e-05,
177
+ "loss": 0.6934980773925781,
178
+ "step": 2400
179
+ },
180
+ {
181
+ "epoch": 0.6468305304010349,
182
+ "grad_norm": 16.08030891418457,
183
+ "learning_rate": 1.743363679923335e-05,
184
+ "loss": 0.6814935302734375,
185
+ "step": 2500
186
+ },
187
+ {
188
+ "epoch": 0.6727037516170763,
189
+ "grad_norm": 15.731101036071777,
190
+ "learning_rate": 1.7241974125539052e-05,
191
+ "loss": 0.6848414611816406,
192
+ "step": 2600
193
+ },
194
+ {
195
+ "epoch": 0.6985769728331177,
196
+ "grad_norm": 14.289274215698242,
197
+ "learning_rate": 1.7050311451844754e-05,
198
+ "loss": 0.6762744903564453,
199
+ "step": 2700
200
+ },
201
+ {
202
+ "epoch": 0.7244501940491591,
203
+ "grad_norm": 14.445661544799805,
204
+ "learning_rate": 1.6858648778150455e-05,
205
+ "loss": 0.6762972259521485,
206
+ "step": 2800
207
+ },
208
+ {
209
+ "epoch": 0.7503234152652005,
210
+ "grad_norm": 18.009662628173828,
211
+ "learning_rate": 1.6666986104456157e-05,
212
+ "loss": 0.6624829864501953,
213
+ "step": 2900
214
+ },
215
+ {
216
+ "epoch": 0.7761966364812419,
217
+ "grad_norm": 19.931297302246094,
218
+ "learning_rate": 1.647532343076186e-05,
219
+ "loss": 0.7022187805175781,
220
+ "step": 3000
221
+ },
222
+ {
223
+ "epoch": 0.8020698576972833,
224
+ "grad_norm": 6.394979476928711,
225
+ "learning_rate": 1.6283660757067564e-05,
226
+ "loss": 0.6697378540039063,
227
+ "step": 3100
228
+ },
229
+ {
230
+ "epoch": 0.8279430789133247,
231
+ "grad_norm": 12.148775100708008,
232
+ "learning_rate": 1.6091998083373262e-05,
233
+ "loss": 0.6663771057128907,
234
+ "step": 3200
235
+ },
236
+ {
237
+ "epoch": 0.8538163001293662,
238
+ "grad_norm": 15.607839584350586,
239
+ "learning_rate": 1.5900335409678967e-05,
240
+ "loss": 0.6489146423339843,
241
+ "step": 3300
242
+ },
243
+ {
244
+ "epoch": 0.8796895213454075,
245
+ "grad_norm": 9.942991256713867,
246
+ "learning_rate": 1.570867273598467e-05,
247
+ "loss": 0.6574240112304688,
248
+ "step": 3400
249
+ },
250
+ {
251
+ "epoch": 0.9055627425614489,
252
+ "grad_norm": 32.7762336730957,
253
+ "learning_rate": 1.551701006229037e-05,
254
+ "loss": 0.6326899719238281,
255
+ "step": 3500
256
+ },
257
+ {
258
+ "epoch": 0.9314359637774903,
259
+ "grad_norm": 10.126937866210938,
260
+ "learning_rate": 1.5325347388596072e-05,
261
+ "loss": 0.6766317749023437,
262
+ "step": 3600
263
+ },
264
+ {
265
+ "epoch": 0.9573091849935317,
266
+ "grad_norm": 22.445276260375977,
267
+ "learning_rate": 1.5133684714901776e-05,
268
+ "loss": 0.6632305145263672,
269
+ "step": 3700
270
+ },
271
+ {
272
+ "epoch": 0.9831824062095731,
273
+ "grad_norm": 14.928740501403809,
274
+ "learning_rate": 1.4942022041207476e-05,
275
+ "loss": 0.6465906524658203,
276
+ "step": 3800
277
+ },
278
+ {
279
+ "epoch": 1.0,
280
+ "eval_accuracy": 0.7215118332744613,
281
+ "eval_f1": 0.7178718846283108,
282
+ "eval_loss": 0.6563947200775146,
283
+ "eval_runtime": 86.9654,
284
+ "eval_samples_per_second": 162.766,
285
+ "eval_steps_per_second": 10.176,
286
+ "step": 3865
287
+ },
288
+ {
289
+ "epoch": 1.0090556274256144,
290
+ "grad_norm": 12.22783374786377,
291
+ "learning_rate": 1.4750359367513179e-05,
292
+ "loss": 0.6397914123535157,
293
+ "step": 3900
294
+ },
295
+ {
296
+ "epoch": 1.034928848641656,
297
+ "grad_norm": 25.15692901611328,
298
+ "learning_rate": 1.4558696693818879e-05,
299
+ "loss": 0.5944495010375976,
300
+ "step": 4000
301
+ },
302
+ {
303
+ "epoch": 1.0608020698576972,
304
+ "grad_norm": 6.247220993041992,
305
+ "learning_rate": 1.4367034020124582e-05,
306
+ "loss": 0.5855656051635743,
307
+ "step": 4100
308
+ },
309
+ {
310
+ "epoch": 1.0866752910737387,
311
+ "grad_norm": 8.999340057373047,
312
+ "learning_rate": 1.4175371346430284e-05,
313
+ "loss": 0.6091019821166992,
314
+ "step": 4200
315
+ },
316
+ {
317
+ "epoch": 1.11254851228978,
318
+ "grad_norm": 10.196805953979492,
319
+ "learning_rate": 1.3983708672735986e-05,
320
+ "loss": 0.5973617553710937,
321
+ "step": 4300
322
+ },
323
+ {
324
+ "epoch": 1.1384217335058215,
325
+ "grad_norm": 9.179534912109375,
326
+ "learning_rate": 1.3792045999041688e-05,
327
+ "loss": 0.5922854232788086,
328
+ "step": 4400
329
+ },
330
+ {
331
+ "epoch": 1.1642949547218628,
332
+ "grad_norm": 11.18813419342041,
333
+ "learning_rate": 1.3600383325347391e-05,
334
+ "loss": 0.575289421081543,
335
+ "step": 4500
336
+ },
337
+ {
338
+ "epoch": 1.1901681759379044,
339
+ "grad_norm": 12.966670036315918,
340
+ "learning_rate": 1.3408720651653091e-05,
341
+ "loss": 0.5534341049194336,
342
+ "step": 4600
343
+ },
344
+ {
345
+ "epoch": 1.2160413971539457,
346
+ "grad_norm": 15.65532112121582,
347
+ "learning_rate": 1.3217057977958794e-05,
348
+ "loss": 0.5662339401245117,
349
+ "step": 4700
350
+ },
351
+ {
352
+ "epoch": 1.2419146183699872,
353
+ "grad_norm": 12.402419090270996,
354
+ "learning_rate": 1.3025395304264496e-05,
355
+ "loss": 0.6043228912353515,
356
+ "step": 4800
357
+ },
358
+ {
359
+ "epoch": 1.2677878395860285,
360
+ "grad_norm": 21.828113555908203,
361
+ "learning_rate": 1.2833732630570198e-05,
362
+ "loss": 0.5857475662231445,
363
+ "step": 4900
364
+ },
365
+ {
366
+ "epoch": 1.2936610608020698,
367
+ "grad_norm": 20.76739501953125,
368
+ "learning_rate": 1.26420699568759e-05,
369
+ "loss": 0.5899971389770508,
370
+ "step": 5000
371
+ },
372
+ {
373
+ "epoch": 1.3195342820181113,
374
+ "grad_norm": 13.431411743164062,
375
+ "learning_rate": 1.2450407283181601e-05,
376
+ "loss": 0.5802528381347656,
377
+ "step": 5100
378
+ },
379
+ {
380
+ "epoch": 1.3454075032341526,
381
+ "grad_norm": 17.105060577392578,
382
+ "learning_rate": 1.2258744609487303e-05,
383
+ "loss": 0.5847416305541993,
384
+ "step": 5200
385
+ },
386
+ {
387
+ "epoch": 1.371280724450194,
388
+ "grad_norm": 51.55683135986328,
389
+ "learning_rate": 1.2067081935793006e-05,
390
+ "loss": 0.5702993392944335,
391
+ "step": 5300
392
+ },
393
+ {
394
+ "epoch": 1.3971539456662354,
395
+ "grad_norm": 19.022991180419922,
396
+ "learning_rate": 1.1875419262098706e-05,
397
+ "loss": 0.5837166595458985,
398
+ "step": 5400
399
+ },
400
+ {
401
+ "epoch": 1.4230271668822767,
402
+ "grad_norm": 12.546523094177246,
403
+ "learning_rate": 1.168375658840441e-05,
404
+ "loss": 0.5941635131835937,
405
+ "step": 5500
406
+ },
407
+ {
408
+ "epoch": 1.4489003880983182,
409
+ "grad_norm": 16.194042205810547,
410
+ "learning_rate": 1.1492093914710111e-05,
411
+ "loss": 0.6047253799438477,
412
+ "step": 5600
413
+ },
414
+ {
415
+ "epoch": 1.4747736093143597,
416
+ "grad_norm": 14.654067993164062,
417
+ "learning_rate": 1.1300431241015813e-05,
418
+ "loss": 0.5888873291015625,
419
+ "step": 5700
420
+ },
421
+ {
422
+ "epoch": 1.500646830530401,
423
+ "grad_norm": 15.194084167480469,
424
+ "learning_rate": 1.1108768567321514e-05,
425
+ "loss": 0.5698460769653321,
426
+ "step": 5800
427
+ },
428
+ {
429
+ "epoch": 1.5265200517464423,
430
+ "grad_norm": 13.003668785095215,
431
+ "learning_rate": 1.0917105893627218e-05,
432
+ "loss": 0.5704814147949219,
433
+ "step": 5900
434
+ },
435
+ {
436
+ "epoch": 1.5523932729624839,
437
+ "grad_norm": 9.508420944213867,
438
+ "learning_rate": 1.0725443219932918e-05,
439
+ "loss": 0.5655577087402344,
440
+ "step": 6000
441
+ },
442
+ {
443
+ "epoch": 1.5782664941785254,
444
+ "grad_norm": 14.1275634765625,
445
+ "learning_rate": 1.0533780546238621e-05,
446
+ "loss": 0.5951072311401367,
447
+ "step": 6100
448
+ },
449
+ {
450
+ "epoch": 1.6041397153945667,
451
+ "grad_norm": 12.208888053894043,
452
+ "learning_rate": 1.0342117872544325e-05,
453
+ "loss": 0.6105654144287109,
454
+ "step": 6200
455
+ },
456
+ {
457
+ "epoch": 1.630012936610608,
458
+ "grad_norm": 14.291622161865234,
459
+ "learning_rate": 1.0150455198850025e-05,
460
+ "loss": 0.5704663467407226,
461
+ "step": 6300
462
+ },
463
+ {
464
+ "epoch": 1.6558861578266493,
465
+ "grad_norm": 11.807310104370117,
466
+ "learning_rate": 9.958792525155726e-06,
467
+ "loss": 0.5667978286743164,
468
+ "step": 6400
469
+ },
470
+ {
471
+ "epoch": 1.6817593790426908,
472
+ "grad_norm": 7.61679744720459,
473
+ "learning_rate": 9.767129851461428e-06,
474
+ "loss": 0.5553795623779297,
475
+ "step": 6500
476
+ },
477
+ {
478
+ "epoch": 1.7076326002587323,
479
+ "grad_norm": 9.418617248535156,
480
+ "learning_rate": 9.57546717776713e-06,
481
+ "loss": 0.5939513778686524,
482
+ "step": 6600
483
+ },
484
+ {
485
+ "epoch": 1.7335058214747736,
486
+ "grad_norm": 11.112833976745605,
487
+ "learning_rate": 9.383804504072831e-06,
488
+ "loss": 0.5403312683105469,
489
+ "step": 6700
490
+ },
491
+ {
492
+ "epoch": 1.759379042690815,
493
+ "grad_norm": 15.817110061645508,
494
+ "learning_rate": 9.192141830378535e-06,
495
+ "loss": 0.5703708267211914,
496
+ "step": 6800
497
+ },
498
+ {
499
+ "epoch": 1.7852522639068564,
500
+ "grad_norm": 8.80798053741455,
501
+ "learning_rate": 9.000479156684236e-06,
502
+ "loss": 0.5544892120361328,
503
+ "step": 6900
504
+ },
505
+ {
506
+ "epoch": 1.811125485122898,
507
+ "grad_norm": 16.89642333984375,
508
+ "learning_rate": 8.808816482989938e-06,
509
+ "loss": 0.573361701965332,
510
+ "step": 7000
511
+ },
512
+ {
513
+ "epoch": 1.8369987063389392,
514
+ "grad_norm": 16.808788299560547,
515
+ "learning_rate": 8.617153809295641e-06,
516
+ "loss": 0.5603259658813476,
517
+ "step": 7100
518
+ },
519
+ {
520
+ "epoch": 1.8628719275549805,
521
+ "grad_norm": 13.838523864746094,
522
+ "learning_rate": 8.425491135601343e-06,
523
+ "loss": 0.5797721481323242,
524
+ "step": 7200
525
+ },
526
+ {
527
+ "epoch": 1.8887451487710218,
528
+ "grad_norm": 8.298638343811035,
529
+ "learning_rate": 8.233828461907045e-06,
530
+ "loss": 0.5604195022583007,
531
+ "step": 7300
532
+ },
533
+ {
534
+ "epoch": 1.9146183699870634,
535
+ "grad_norm": 25.5234432220459,
536
+ "learning_rate": 8.042165788212747e-06,
537
+ "loss": 0.5867827606201171,
538
+ "step": 7400
539
+ },
540
+ {
541
+ "epoch": 1.9404915912031049,
542
+ "grad_norm": 15.7322416305542,
543
+ "learning_rate": 7.850503114518448e-06,
544
+ "loss": 0.5443745803833008,
545
+ "step": 7500
546
+ },
547
+ {
548
+ "epoch": 1.9663648124191462,
549
+ "grad_norm": 9.020604133605957,
550
+ "learning_rate": 7.65884044082415e-06,
551
+ "loss": 0.5062921905517578,
552
+ "step": 7600
553
+ },
554
+ {
555
+ "epoch": 1.9922380336351875,
556
+ "grad_norm": 11.164002418518066,
557
+ "learning_rate": 7.467177767129852e-06,
558
+ "loss": 0.5686885070800781,
559
+ "step": 7700
560
+ },
561
+ {
562
+ "epoch": 2.0,
563
+ "eval_accuracy": 0.7476510067114094,
564
+ "eval_f1": 0.7474800395521614,
565
+ "eval_loss": 0.5974893569946289,
566
+ "eval_runtime": 86.7366,
567
+ "eval_samples_per_second": 163.195,
568
+ "eval_steps_per_second": 10.203,
569
+ "step": 7730
570
+ },
571
+ {
572
+ "epoch": 2.0181112548512288,
573
+ "grad_norm": 7.4816789627075195,
574
+ "learning_rate": 7.275515093435554e-06,
575
+ "loss": 0.5046030426025391,
576
+ "step": 7800
577
+ },
578
+ {
579
+ "epoch": 2.0439844760672705,
580
+ "grad_norm": 21.1260986328125,
581
+ "learning_rate": 7.083852419741256e-06,
582
+ "loss": 0.49911834716796877,
583
+ "step": 7900
584
+ },
585
+ {
586
+ "epoch": 2.069857697283312,
587
+ "grad_norm": 30.835697174072266,
588
+ "learning_rate": 6.8921897460469575e-06,
589
+ "loss": 0.48409915924072267,
590
+ "step": 8000
591
+ },
592
+ {
593
+ "epoch": 2.095730918499353,
594
+ "grad_norm": 15.94166374206543,
595
+ "learning_rate": 6.700527072352659e-06,
596
+ "loss": 0.5172428131103516,
597
+ "step": 8100
598
+ },
599
+ {
600
+ "epoch": 2.1216041397153944,
601
+ "grad_norm": 17.08721160888672,
602
+ "learning_rate": 6.508864398658363e-06,
603
+ "loss": 0.4927161407470703,
604
+ "step": 8200
605
+ },
606
+ {
607
+ "epoch": 2.147477360931436,
608
+ "grad_norm": 12.646923065185547,
609
+ "learning_rate": 6.317201724964063e-06,
610
+ "loss": 0.4802434539794922,
611
+ "step": 8300
612
+ },
613
+ {
614
+ "epoch": 2.1733505821474774,
615
+ "grad_norm": 5.114115238189697,
616
+ "learning_rate": 6.125539051269765e-06,
617
+ "loss": 0.4971760177612305,
618
+ "step": 8400
619
+ },
620
+ {
621
+ "epoch": 2.1992238033635187,
622
+ "grad_norm": 18.957698822021484,
623
+ "learning_rate": 5.9338763775754685e-06,
624
+ "loss": 0.5098771667480468,
625
+ "step": 8500
626
+ },
627
+ {
628
+ "epoch": 2.22509702457956,
629
+ "grad_norm": 13.89101505279541,
630
+ "learning_rate": 5.74221370388117e-06,
631
+ "loss": 0.47926040649414064,
632
+ "step": 8600
633
+ },
634
+ {
635
+ "epoch": 2.2509702457956013,
636
+ "grad_norm": 10.93714427947998,
637
+ "learning_rate": 5.550551030186872e-06,
638
+ "loss": 0.46799674987792966,
639
+ "step": 8700
640
+ },
641
+ {
642
+ "epoch": 2.276843467011643,
643
+ "grad_norm": 23.81261444091797,
644
+ "learning_rate": 5.358888356492573e-06,
645
+ "loss": 0.5347587203979492,
646
+ "step": 8800
647
+ },
648
+ {
649
+ "epoch": 2.3027166882276844,
650
+ "grad_norm": 9.246687889099121,
651
+ "learning_rate": 5.167225682798276e-06,
652
+ "loss": 0.4957209396362305,
653
+ "step": 8900
654
+ },
655
+ {
656
+ "epoch": 2.3285899094437257,
657
+ "grad_norm": 15.216452598571777,
658
+ "learning_rate": 4.975563009103978e-06,
659
+ "loss": 0.4704705047607422,
660
+ "step": 9000
661
+ },
662
+ {
663
+ "epoch": 2.354463130659767,
664
+ "grad_norm": 10.991480827331543,
665
+ "learning_rate": 4.7839003354096794e-06,
666
+ "loss": 0.4816410064697266,
667
+ "step": 9100
668
+ },
669
+ {
670
+ "epoch": 2.3803363518758087,
671
+ "grad_norm": 33.290924072265625,
672
+ "learning_rate": 4.592237661715381e-06,
673
+ "loss": 0.4527182388305664,
674
+ "step": 9200
675
+ },
676
+ {
677
+ "epoch": 2.40620957309185,
678
+ "grad_norm": 28.04972267150879,
679
+ "learning_rate": 4.400574988021084e-06,
680
+ "loss": 0.5087837600708007,
681
+ "step": 9300
682
+ },
683
+ {
684
+ "epoch": 2.4320827943078913,
685
+ "grad_norm": 13.985801696777344,
686
+ "learning_rate": 4.208912314326785e-06,
687
+ "loss": 0.46610614776611325,
688
+ "step": 9400
689
+ },
690
+ {
691
+ "epoch": 2.4579560155239326,
692
+ "grad_norm": 14.219202995300293,
693
+ "learning_rate": 4.017249640632487e-06,
694
+ "loss": 0.43952152252197263,
695
+ "step": 9500
696
+ },
697
+ {
698
+ "epoch": 2.4838292367399744,
699
+ "grad_norm": 14.186090469360352,
700
+ "learning_rate": 3.825586966938189e-06,
701
+ "loss": 0.4782567977905273,
702
+ "step": 9600
703
+ },
704
+ {
705
+ "epoch": 2.5097024579560157,
706
+ "grad_norm": 13.511507987976074,
707
+ "learning_rate": 3.6339242932438913e-06,
708
+ "loss": 0.46348812103271486,
709
+ "step": 9700
710
+ },
711
+ {
712
+ "epoch": 2.535575679172057,
713
+ "grad_norm": 18.06315803527832,
714
+ "learning_rate": 3.442261619549593e-06,
715
+ "loss": 0.49416404724121094,
716
+ "step": 9800
717
+ },
718
+ {
719
+ "epoch": 2.5614489003880982,
720
+ "grad_norm": 20.560895919799805,
721
+ "learning_rate": 3.250598945855295e-06,
722
+ "loss": 0.48670326232910155,
723
+ "step": 9900
724
+ },
725
+ {
726
+ "epoch": 2.5873221216041395,
727
+ "grad_norm": 7.430022716522217,
728
+ "learning_rate": 3.0589362721609967e-06,
729
+ "loss": 0.4578776550292969,
730
+ "step": 10000
731
+ },
732
+ {
733
+ "epoch": 2.6131953428201813,
734
+ "grad_norm": 21.980758666992188,
735
+ "learning_rate": 2.867273598466699e-06,
736
+ "loss": 0.47359500885009764,
737
+ "step": 10100
738
+ },
739
+ {
740
+ "epoch": 2.6390685640362226,
741
+ "grad_norm": 18.348045349121094,
742
+ "learning_rate": 2.675610924772401e-06,
743
+ "loss": 0.509437484741211,
744
+ "step": 10200
745
+ },
746
+ {
747
+ "epoch": 2.664941785252264,
748
+ "grad_norm": 15.071292877197266,
749
+ "learning_rate": 2.4839482510781026e-06,
750
+ "loss": 0.4916699981689453,
751
+ "step": 10300
752
+ },
753
+ {
754
+ "epoch": 2.690815006468305,
755
+ "grad_norm": 9.668122291564941,
756
+ "learning_rate": 2.2922855773838048e-06,
757
+ "loss": 0.4785163497924805,
758
+ "step": 10400
759
+ },
760
+ {
761
+ "epoch": 2.7166882276843465,
762
+ "grad_norm": 19.475624084472656,
763
+ "learning_rate": 2.100622903689507e-06,
764
+ "loss": 0.4679045867919922,
765
+ "step": 10500
766
+ },
767
+ {
768
+ "epoch": 2.742561448900388,
769
+ "grad_norm": 9.452678680419922,
770
+ "learning_rate": 1.9089602299952085e-06,
771
+ "loss": 0.49099754333496093,
772
+ "step": 10600
773
+ },
774
+ {
775
+ "epoch": 2.7684346701164295,
776
+ "grad_norm": 18.706396102905273,
777
+ "learning_rate": 1.7172975563009107e-06,
778
+ "loss": 0.4827518081665039,
779
+ "step": 10700
780
+ },
781
+ {
782
+ "epoch": 2.794307891332471,
783
+ "grad_norm": 22.398160934448242,
784
+ "learning_rate": 1.5256348826066126e-06,
785
+ "loss": 0.47274169921875,
786
+ "step": 10800
787
+ },
788
+ {
789
+ "epoch": 2.8201811125485126,
790
+ "grad_norm": 9.312556266784668,
791
+ "learning_rate": 1.3339722089123145e-06,
792
+ "loss": 0.4818600082397461,
793
+ "step": 10900
794
+ },
795
+ {
796
+ "epoch": 2.8460543337645534,
797
+ "grad_norm": 13.301252365112305,
798
+ "learning_rate": 1.1423095352180164e-06,
799
+ "loss": 0.4749205780029297,
800
+ "step": 11000
801
+ },
802
+ {
803
+ "epoch": 2.871927554980595,
804
+ "grad_norm": 20.057945251464844,
805
+ "learning_rate": 9.506468615237182e-07,
806
+ "loss": 0.48763519287109375,
807
+ "step": 11100
808
+ },
809
+ {
810
+ "epoch": 2.8978007761966365,
811
+ "grad_norm": 10.004188537597656,
812
+ "learning_rate": 7.589841878294204e-07,
813
+ "loss": 0.4786335372924805,
814
+ "step": 11200
815
+ },
816
+ {
817
+ "epoch": 2.9236739974126777,
818
+ "grad_norm": 13.496678352355957,
819
+ "learning_rate": 5.673215141351223e-07,
820
+ "loss": 0.45396602630615235,
821
+ "step": 11300
822
+ },
823
+ {
824
+ "epoch": 2.9495472186287195,
825
+ "grad_norm": 16.282756805419922,
826
+ "learning_rate": 3.7565884044082415e-07,
827
+ "loss": 0.4245071029663086,
828
+ "step": 11400
829
+ },
830
+ {
831
+ "epoch": 2.975420439844761,
832
+ "grad_norm": 11.617134094238281,
833
+ "learning_rate": 1.8399616674652613e-07,
834
+ "loss": 0.5004544067382812,
835
+ "step": 11500
836
+ },
837
+ {
838
+ "epoch": 3.0,
839
+ "eval_accuracy": 0.7586012009890498,
840
+ "eval_f1": 0.7562243780359653,
841
+ "eval_loss": 0.6050583720207214,
842
+ "eval_runtime": 87.1393,
843
+ "eval_samples_per_second": 162.441,
844
+ "eval_steps_per_second": 10.156,
845
+ "step": 11595
846
+ }
847
+ ],
848
+ "logging_steps": 100,
849
+ "max_steps": 11595,
850
+ "num_input_tokens_seen": 0,
851
+ "num_train_epochs": 3,
852
+ "save_steps": 500,
853
+ "stateful_callbacks": {
854
+ "TrainerControl": {
855
+ "args": {
856
+ "should_epoch_stop": false,
857
+ "should_evaluate": false,
858
+ "should_log": false,
859
+ "should_save": true,
860
+ "should_training_stop": true
861
+ },
862
+ "attributes": {}
863
+ }
864
+ },
865
+ "total_flos": 1.2123143945733744e+16,
866
+ "train_batch_size": 16,
867
+ "trial_name": null,
868
+ "trial_params": null
869
+ }
checkpoint-11595/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fabf78786d2ae6dcda35b22fbd2da223968e53ae585fc7647c873f62e2caaa5
3
+ size 5201
checkpoint-3865/config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Normal",
16
+ "1": "Abusive",
17
+ "2": "Hate"
18
+ },
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "is_decoder": false,
22
+ "label2id": {
23
+ "Abusive": 1,
24
+ "Hate": 2,
25
+ "Normal": 0
26
+ },
27
+ "layer_norm_eps": 1e-05,
28
+ "max_position_embeddings": 514,
29
+ "model_type": "xlm-roberta",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "output_past": true,
33
+ "pad_token_id": 1,
34
+ "position_embedding_type": "absolute",
35
+ "problem_type": "single_label_classification",
36
+ "tie_word_embeddings": true,
37
+ "transformers_version": "5.0.0",
38
+ "type_vocab_size": 1,
39
+ "use_cache": false,
40
+ "vocab_size": 250002
41
+ }
checkpoint-3865/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67b81c2ad0d33efc0a53aa302e54d5658fad23d39a5f01e2ec348803b1cc248c
3
+ size 1112208060
checkpoint-3865/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e5e0aa2b1c7234dbd1c5a541985d6f430c24c3f3c08c43cd1528fb81f7ef595
3
+ size 2224539019
checkpoint-3865/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:788d3bdda0c8ed4f88f8e05180274be9d103d89d5775d9797b9d5108a6c6761f
3
+ size 14645
checkpoint-3865/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2552fa0e348fb2f379264bcdce3f4b389b2474de73f6628fe78d1c1f93a0f224
3
+ size 1465
checkpoint-3865/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42d3da0b556fcc5dfcd7158b90066b5069bd23df2df684a17c4bcfd8575016ae
3
+ size 16781739
checkpoint-3865/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-3865/trainer_state.json ADDED
@@ -0,0 +1,310 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 3865,
3
+ "best_metric": 0.7178718846283108,
4
+ "best_model_checkpoint": "./EA-HS/checkpoint-3865",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3865,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0258732212160414,
14
+ "grad_norm": 5.261500358581543,
15
+ "learning_rate": 1.7068965517241382e-06,
16
+ "loss": 1.100949478149414,
17
+ "step": 100
18
+ },
19
+ {
20
+ "epoch": 0.0517464424320828,
21
+ "grad_norm": 5.354814529418945,
22
+ "learning_rate": 3.431034482758621e-06,
23
+ "loss": 1.1074339294433593,
24
+ "step": 200
25
+ },
26
+ {
27
+ "epoch": 0.07761966364812418,
28
+ "grad_norm": 10.960789680480957,
29
+ "learning_rate": 5.155172413793104e-06,
30
+ "loss": 1.0751912689208984,
31
+ "step": 300
32
+ },
33
+ {
34
+ "epoch": 0.1034928848641656,
35
+ "grad_norm": 7.228652477264404,
36
+ "learning_rate": 6.8793103448275865e-06,
37
+ "loss": 1.0659678649902344,
38
+ "step": 400
39
+ },
40
+ {
41
+ "epoch": 0.129366106080207,
42
+ "grad_norm": 13.746113777160645,
43
+ "learning_rate": 8.603448275862069e-06,
44
+ "loss": 1.017400894165039,
45
+ "step": 500
46
+ },
47
+ {
48
+ "epoch": 0.15523932729624837,
49
+ "grad_norm": 9.63514232635498,
50
+ "learning_rate": 1.0327586206896552e-05,
51
+ "loss": 0.9742707824707031,
52
+ "step": 600
53
+ },
54
+ {
55
+ "epoch": 0.18111254851228978,
56
+ "grad_norm": 12.705816268920898,
57
+ "learning_rate": 1.2051724137931036e-05,
58
+ "loss": 0.9275926971435546,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 0.2069857697283312,
63
+ "grad_norm": 26.888172149658203,
64
+ "learning_rate": 1.3775862068965518e-05,
65
+ "loss": 0.9081744384765625,
66
+ "step": 800
67
+ },
68
+ {
69
+ "epoch": 0.23285899094437257,
70
+ "grad_norm": 21.919212341308594,
71
+ "learning_rate": 1.55e-05,
72
+ "loss": 0.8668842315673828,
73
+ "step": 900
74
+ },
75
+ {
76
+ "epoch": 0.258732212160414,
77
+ "grad_norm": 18.860923767089844,
78
+ "learning_rate": 1.7224137931034483e-05,
79
+ "loss": 0.8243231964111328,
80
+ "step": 1000
81
+ },
82
+ {
83
+ "epoch": 0.2846054333764554,
84
+ "grad_norm": 16.422523498535156,
85
+ "learning_rate": 1.894827586206897e-05,
86
+ "loss": 0.8356568145751954,
87
+ "step": 1100
88
+ },
89
+ {
90
+ "epoch": 0.31047865459249674,
91
+ "grad_norm": 6.305636882781982,
92
+ "learning_rate": 1.9925251557259225e-05,
93
+ "loss": 0.8148372650146485,
94
+ "step": 1200
95
+ },
96
+ {
97
+ "epoch": 0.33635187580853815,
98
+ "grad_norm": 55.4698600769043,
99
+ "learning_rate": 1.9733588883564927e-05,
100
+ "loss": 0.7789091491699218,
101
+ "step": 1300
102
+ },
103
+ {
104
+ "epoch": 0.36222509702457956,
105
+ "grad_norm": 7.873316287994385,
106
+ "learning_rate": 1.954192620987063e-05,
107
+ "loss": 0.7773941040039063,
108
+ "step": 1400
109
+ },
110
+ {
111
+ "epoch": 0.38809831824062097,
112
+ "grad_norm": 8.828251838684082,
113
+ "learning_rate": 1.935026353617633e-05,
114
+ "loss": 0.7520027160644531,
115
+ "step": 1500
116
+ },
117
+ {
118
+ "epoch": 0.4139715394566624,
119
+ "grad_norm": 10.744029998779297,
120
+ "learning_rate": 1.9158600862482032e-05,
121
+ "loss": 0.7416606903076172,
122
+ "step": 1600
123
+ },
124
+ {
125
+ "epoch": 0.4398447606727037,
126
+ "grad_norm": 13.313233375549316,
127
+ "learning_rate": 1.8966938188787737e-05,
128
+ "loss": 0.7262284851074219,
129
+ "step": 1700
130
+ },
131
+ {
132
+ "epoch": 0.46571798188874514,
133
+ "grad_norm": 16.973875045776367,
134
+ "learning_rate": 1.8775275515093435e-05,
135
+ "loss": 0.7252954864501953,
136
+ "step": 1800
137
+ },
138
+ {
139
+ "epoch": 0.49159120310478654,
140
+ "grad_norm": 11.87275218963623,
141
+ "learning_rate": 1.858361284139914e-05,
142
+ "loss": 0.7501118469238282,
143
+ "step": 1900
144
+ },
145
+ {
146
+ "epoch": 0.517464424320828,
147
+ "grad_norm": 18.290884017944336,
148
+ "learning_rate": 1.8391950167704842e-05,
149
+ "loss": 0.741920394897461,
150
+ "step": 2000
151
+ },
152
+ {
153
+ "epoch": 0.5433376455368694,
154
+ "grad_norm": 5.779555797576904,
155
+ "learning_rate": 1.8200287494010544e-05,
156
+ "loss": 0.7136674499511719,
157
+ "step": 2100
158
+ },
159
+ {
160
+ "epoch": 0.5692108667529108,
161
+ "grad_norm": 29.206987380981445,
162
+ "learning_rate": 1.8008624820316245e-05,
163
+ "loss": 0.7118588256835937,
164
+ "step": 2200
165
+ },
166
+ {
167
+ "epoch": 0.5950840879689522,
168
+ "grad_norm": 28.444412231445312,
169
+ "learning_rate": 1.7816962146621947e-05,
170
+ "loss": 0.7275136566162109,
171
+ "step": 2300
172
+ },
173
+ {
174
+ "epoch": 0.6209573091849935,
175
+ "grad_norm": 16.02486228942871,
176
+ "learning_rate": 1.762529947292765e-05,
177
+ "loss": 0.6934980773925781,
178
+ "step": 2400
179
+ },
180
+ {
181
+ "epoch": 0.6468305304010349,
182
+ "grad_norm": 16.08030891418457,
183
+ "learning_rate": 1.743363679923335e-05,
184
+ "loss": 0.6814935302734375,
185
+ "step": 2500
186
+ },
187
+ {
188
+ "epoch": 0.6727037516170763,
189
+ "grad_norm": 15.731101036071777,
190
+ "learning_rate": 1.7241974125539052e-05,
191
+ "loss": 0.6848414611816406,
192
+ "step": 2600
193
+ },
194
+ {
195
+ "epoch": 0.6985769728331177,
196
+ "grad_norm": 14.289274215698242,
197
+ "learning_rate": 1.7050311451844754e-05,
198
+ "loss": 0.6762744903564453,
199
+ "step": 2700
200
+ },
201
+ {
202
+ "epoch": 0.7244501940491591,
203
+ "grad_norm": 14.445661544799805,
204
+ "learning_rate": 1.6858648778150455e-05,
205
+ "loss": 0.6762972259521485,
206
+ "step": 2800
207
+ },
208
+ {
209
+ "epoch": 0.7503234152652005,
210
+ "grad_norm": 18.009662628173828,
211
+ "learning_rate": 1.6666986104456157e-05,
212
+ "loss": 0.6624829864501953,
213
+ "step": 2900
214
+ },
215
+ {
216
+ "epoch": 0.7761966364812419,
217
+ "grad_norm": 19.931297302246094,
218
+ "learning_rate": 1.647532343076186e-05,
219
+ "loss": 0.7022187805175781,
220
+ "step": 3000
221
+ },
222
+ {
223
+ "epoch": 0.8020698576972833,
224
+ "grad_norm": 6.394979476928711,
225
+ "learning_rate": 1.6283660757067564e-05,
226
+ "loss": 0.6697378540039063,
227
+ "step": 3100
228
+ },
229
+ {
230
+ "epoch": 0.8279430789133247,
231
+ "grad_norm": 12.148775100708008,
232
+ "learning_rate": 1.6091998083373262e-05,
233
+ "loss": 0.6663771057128907,
234
+ "step": 3200
235
+ },
236
+ {
237
+ "epoch": 0.8538163001293662,
238
+ "grad_norm": 15.607839584350586,
239
+ "learning_rate": 1.5900335409678967e-05,
240
+ "loss": 0.6489146423339843,
241
+ "step": 3300
242
+ },
243
+ {
244
+ "epoch": 0.8796895213454075,
245
+ "grad_norm": 9.942991256713867,
246
+ "learning_rate": 1.570867273598467e-05,
247
+ "loss": 0.6574240112304688,
248
+ "step": 3400
249
+ },
250
+ {
251
+ "epoch": 0.9055627425614489,
252
+ "grad_norm": 32.7762336730957,
253
+ "learning_rate": 1.551701006229037e-05,
254
+ "loss": 0.6326899719238281,
255
+ "step": 3500
256
+ },
257
+ {
258
+ "epoch": 0.9314359637774903,
259
+ "grad_norm": 10.126937866210938,
260
+ "learning_rate": 1.5325347388596072e-05,
261
+ "loss": 0.6766317749023437,
262
+ "step": 3600
263
+ },
264
+ {
265
+ "epoch": 0.9573091849935317,
266
+ "grad_norm": 22.445276260375977,
267
+ "learning_rate": 1.5133684714901776e-05,
268
+ "loss": 0.6632305145263672,
269
+ "step": 3700
270
+ },
271
+ {
272
+ "epoch": 0.9831824062095731,
273
+ "grad_norm": 14.928740501403809,
274
+ "learning_rate": 1.4942022041207476e-05,
275
+ "loss": 0.6465906524658203,
276
+ "step": 3800
277
+ },
278
+ {
279
+ "epoch": 1.0,
280
+ "eval_accuracy": 0.7215118332744613,
281
+ "eval_f1": 0.7178718846283108,
282
+ "eval_loss": 0.6563947200775146,
283
+ "eval_runtime": 86.9654,
284
+ "eval_samples_per_second": 162.766,
285
+ "eval_steps_per_second": 10.176,
286
+ "step": 3865
287
+ }
288
+ ],
289
+ "logging_steps": 100,
290
+ "max_steps": 11595,
291
+ "num_input_tokens_seen": 0,
292
+ "num_train_epochs": 3,
293
+ "save_steps": 500,
294
+ "stateful_callbacks": {
295
+ "TrainerControl": {
296
+ "args": {
297
+ "should_epoch_stop": false,
298
+ "should_evaluate": false,
299
+ "should_log": false,
300
+ "should_save": true,
301
+ "should_training_stop": false
302
+ },
303
+ "attributes": {}
304
+ }
305
+ },
306
+ "total_flos": 4041183649767264.0,
307
+ "train_batch_size": 16,
308
+ "trial_name": null,
309
+ "trial_params": null
310
+ }
checkpoint-3865/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fabf78786d2ae6dcda35b22fbd2da223968e53ae585fc7647c873f62e2caaa5
3
+ size 5201
checkpoint-7730/config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Normal",
16
+ "1": "Abusive",
17
+ "2": "Hate"
18
+ },
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "is_decoder": false,
22
+ "label2id": {
23
+ "Abusive": 1,
24
+ "Hate": 2,
25
+ "Normal": 0
26
+ },
27
+ "layer_norm_eps": 1e-05,
28
+ "max_position_embeddings": 514,
29
+ "model_type": "xlm-roberta",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "output_past": true,
33
+ "pad_token_id": 1,
34
+ "position_embedding_type": "absolute",
35
+ "problem_type": "single_label_classification",
36
+ "tie_word_embeddings": true,
37
+ "transformers_version": "5.0.0",
38
+ "type_vocab_size": 1,
39
+ "use_cache": false,
40
+ "vocab_size": 250002
41
+ }
checkpoint-7730/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3bce1a233387c2c190c60873cb44537cebeee337425b40c36eb40b539a1eba3
3
+ size 1112208060
checkpoint-7730/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:728a06f0fb25bfdb37e57ab9a4e0d07a70db6341a33f070c69a52d683ff727e7
3
+ size 2224539019
checkpoint-7730/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:091a79b606ac3f4895c82f5fd2754ace04bf4bb156f7f8671565fb1f5697b350
3
+ size 14645
checkpoint-7730/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd9aeab2b453b2a0e60e5068a958cff66324b419715e0d34fdef5a65c47c1cf5
3
+ size 1465
checkpoint-7730/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42d3da0b556fcc5dfcd7158b90066b5069bd23df2df684a17c4bcfd8575016ae
3
+ size 16781739
checkpoint-7730/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-7730/trainer_state.json ADDED
@@ -0,0 +1,593 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 7730,
3
+ "best_metric": 0.7474800395521614,
4
+ "best_model_checkpoint": "./EA-HS/checkpoint-7730",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 7730,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0258732212160414,
14
+ "grad_norm": 5.261500358581543,
15
+ "learning_rate": 1.7068965517241382e-06,
16
+ "loss": 1.100949478149414,
17
+ "step": 100
18
+ },
19
+ {
20
+ "epoch": 0.0517464424320828,
21
+ "grad_norm": 5.354814529418945,
22
+ "learning_rate": 3.431034482758621e-06,
23
+ "loss": 1.1074339294433593,
24
+ "step": 200
25
+ },
26
+ {
27
+ "epoch": 0.07761966364812418,
28
+ "grad_norm": 10.960789680480957,
29
+ "learning_rate": 5.155172413793104e-06,
30
+ "loss": 1.0751912689208984,
31
+ "step": 300
32
+ },
33
+ {
34
+ "epoch": 0.1034928848641656,
35
+ "grad_norm": 7.228652477264404,
36
+ "learning_rate": 6.8793103448275865e-06,
37
+ "loss": 1.0659678649902344,
38
+ "step": 400
39
+ },
40
+ {
41
+ "epoch": 0.129366106080207,
42
+ "grad_norm": 13.746113777160645,
43
+ "learning_rate": 8.603448275862069e-06,
44
+ "loss": 1.017400894165039,
45
+ "step": 500
46
+ },
47
+ {
48
+ "epoch": 0.15523932729624837,
49
+ "grad_norm": 9.63514232635498,
50
+ "learning_rate": 1.0327586206896552e-05,
51
+ "loss": 0.9742707824707031,
52
+ "step": 600
53
+ },
54
+ {
55
+ "epoch": 0.18111254851228978,
56
+ "grad_norm": 12.705816268920898,
57
+ "learning_rate": 1.2051724137931036e-05,
58
+ "loss": 0.9275926971435546,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 0.2069857697283312,
63
+ "grad_norm": 26.888172149658203,
64
+ "learning_rate": 1.3775862068965518e-05,
65
+ "loss": 0.9081744384765625,
66
+ "step": 800
67
+ },
68
+ {
69
+ "epoch": 0.23285899094437257,
70
+ "grad_norm": 21.919212341308594,
71
+ "learning_rate": 1.55e-05,
72
+ "loss": 0.8668842315673828,
73
+ "step": 900
74
+ },
75
+ {
76
+ "epoch": 0.258732212160414,
77
+ "grad_norm": 18.860923767089844,
78
+ "learning_rate": 1.7224137931034483e-05,
79
+ "loss": 0.8243231964111328,
80
+ "step": 1000
81
+ },
82
+ {
83
+ "epoch": 0.2846054333764554,
84
+ "grad_norm": 16.422523498535156,
85
+ "learning_rate": 1.894827586206897e-05,
86
+ "loss": 0.8356568145751954,
87
+ "step": 1100
88
+ },
89
+ {
90
+ "epoch": 0.31047865459249674,
91
+ "grad_norm": 6.305636882781982,
92
+ "learning_rate": 1.9925251557259225e-05,
93
+ "loss": 0.8148372650146485,
94
+ "step": 1200
95
+ },
96
+ {
97
+ "epoch": 0.33635187580853815,
98
+ "grad_norm": 55.4698600769043,
99
+ "learning_rate": 1.9733588883564927e-05,
100
+ "loss": 0.7789091491699218,
101
+ "step": 1300
102
+ },
103
+ {
104
+ "epoch": 0.36222509702457956,
105
+ "grad_norm": 7.873316287994385,
106
+ "learning_rate": 1.954192620987063e-05,
107
+ "loss": 0.7773941040039063,
108
+ "step": 1400
109
+ },
110
+ {
111
+ "epoch": 0.38809831824062097,
112
+ "grad_norm": 8.828251838684082,
113
+ "learning_rate": 1.935026353617633e-05,
114
+ "loss": 0.7520027160644531,
115
+ "step": 1500
116
+ },
117
+ {
118
+ "epoch": 0.4139715394566624,
119
+ "grad_norm": 10.744029998779297,
120
+ "learning_rate": 1.9158600862482032e-05,
121
+ "loss": 0.7416606903076172,
122
+ "step": 1600
123
+ },
124
+ {
125
+ "epoch": 0.4398447606727037,
126
+ "grad_norm": 13.313233375549316,
127
+ "learning_rate": 1.8966938188787737e-05,
128
+ "loss": 0.7262284851074219,
129
+ "step": 1700
130
+ },
131
+ {
132
+ "epoch": 0.46571798188874514,
133
+ "grad_norm": 16.973875045776367,
134
+ "learning_rate": 1.8775275515093435e-05,
135
+ "loss": 0.7252954864501953,
136
+ "step": 1800
137
+ },
138
+ {
139
+ "epoch": 0.49159120310478654,
140
+ "grad_norm": 11.87275218963623,
141
+ "learning_rate": 1.858361284139914e-05,
142
+ "loss": 0.7501118469238282,
143
+ "step": 1900
144
+ },
145
+ {
146
+ "epoch": 0.517464424320828,
147
+ "grad_norm": 18.290884017944336,
148
+ "learning_rate": 1.8391950167704842e-05,
149
+ "loss": 0.741920394897461,
150
+ "step": 2000
151
+ },
152
+ {
153
+ "epoch": 0.5433376455368694,
154
+ "grad_norm": 5.779555797576904,
155
+ "learning_rate": 1.8200287494010544e-05,
156
+ "loss": 0.7136674499511719,
157
+ "step": 2100
158
+ },
159
+ {
160
+ "epoch": 0.5692108667529108,
161
+ "grad_norm": 29.206987380981445,
162
+ "learning_rate": 1.8008624820316245e-05,
163
+ "loss": 0.7118588256835937,
164
+ "step": 2200
165
+ },
166
+ {
167
+ "epoch": 0.5950840879689522,
168
+ "grad_norm": 28.444412231445312,
169
+ "learning_rate": 1.7816962146621947e-05,
170
+ "loss": 0.7275136566162109,
171
+ "step": 2300
172
+ },
173
+ {
174
+ "epoch": 0.6209573091849935,
175
+ "grad_norm": 16.02486228942871,
176
+ "learning_rate": 1.762529947292765e-05,
177
+ "loss": 0.6934980773925781,
178
+ "step": 2400
179
+ },
180
+ {
181
+ "epoch": 0.6468305304010349,
182
+ "grad_norm": 16.08030891418457,
183
+ "learning_rate": 1.743363679923335e-05,
184
+ "loss": 0.6814935302734375,
185
+ "step": 2500
186
+ },
187
+ {
188
+ "epoch": 0.6727037516170763,
189
+ "grad_norm": 15.731101036071777,
190
+ "learning_rate": 1.7241974125539052e-05,
191
+ "loss": 0.6848414611816406,
192
+ "step": 2600
193
+ },
194
+ {
195
+ "epoch": 0.6985769728331177,
196
+ "grad_norm": 14.289274215698242,
197
+ "learning_rate": 1.7050311451844754e-05,
198
+ "loss": 0.6762744903564453,
199
+ "step": 2700
200
+ },
201
+ {
202
+ "epoch": 0.7244501940491591,
203
+ "grad_norm": 14.445661544799805,
204
+ "learning_rate": 1.6858648778150455e-05,
205
+ "loss": 0.6762972259521485,
206
+ "step": 2800
207
+ },
208
+ {
209
+ "epoch": 0.7503234152652005,
210
+ "grad_norm": 18.009662628173828,
211
+ "learning_rate": 1.6666986104456157e-05,
212
+ "loss": 0.6624829864501953,
213
+ "step": 2900
214
+ },
215
+ {
216
+ "epoch": 0.7761966364812419,
217
+ "grad_norm": 19.931297302246094,
218
+ "learning_rate": 1.647532343076186e-05,
219
+ "loss": 0.7022187805175781,
220
+ "step": 3000
221
+ },
222
+ {
223
+ "epoch": 0.8020698576972833,
224
+ "grad_norm": 6.394979476928711,
225
+ "learning_rate": 1.6283660757067564e-05,
226
+ "loss": 0.6697378540039063,
227
+ "step": 3100
228
+ },
229
+ {
230
+ "epoch": 0.8279430789133247,
231
+ "grad_norm": 12.148775100708008,
232
+ "learning_rate": 1.6091998083373262e-05,
233
+ "loss": 0.6663771057128907,
234
+ "step": 3200
235
+ },
236
+ {
237
+ "epoch": 0.8538163001293662,
238
+ "grad_norm": 15.607839584350586,
239
+ "learning_rate": 1.5900335409678967e-05,
240
+ "loss": 0.6489146423339843,
241
+ "step": 3300
242
+ },
243
+ {
244
+ "epoch": 0.8796895213454075,
245
+ "grad_norm": 9.942991256713867,
246
+ "learning_rate": 1.570867273598467e-05,
247
+ "loss": 0.6574240112304688,
248
+ "step": 3400
249
+ },
250
+ {
251
+ "epoch": 0.9055627425614489,
252
+ "grad_norm": 32.7762336730957,
253
+ "learning_rate": 1.551701006229037e-05,
254
+ "loss": 0.6326899719238281,
255
+ "step": 3500
256
+ },
257
+ {
258
+ "epoch": 0.9314359637774903,
259
+ "grad_norm": 10.126937866210938,
260
+ "learning_rate": 1.5325347388596072e-05,
261
+ "loss": 0.6766317749023437,
262
+ "step": 3600
263
+ },
264
+ {
265
+ "epoch": 0.9573091849935317,
266
+ "grad_norm": 22.445276260375977,
267
+ "learning_rate": 1.5133684714901776e-05,
268
+ "loss": 0.6632305145263672,
269
+ "step": 3700
270
+ },
271
+ {
272
+ "epoch": 0.9831824062095731,
273
+ "grad_norm": 14.928740501403809,
274
+ "learning_rate": 1.4942022041207476e-05,
275
+ "loss": 0.6465906524658203,
276
+ "step": 3800
277
+ },
278
+ {
279
+ "epoch": 1.0,
280
+ "eval_accuracy": 0.7215118332744613,
281
+ "eval_f1": 0.7178718846283108,
282
+ "eval_loss": 0.6563947200775146,
283
+ "eval_runtime": 86.9654,
284
+ "eval_samples_per_second": 162.766,
285
+ "eval_steps_per_second": 10.176,
286
+ "step": 3865
287
+ },
288
+ {
289
+ "epoch": 1.0090556274256144,
290
+ "grad_norm": 12.22783374786377,
291
+ "learning_rate": 1.4750359367513179e-05,
292
+ "loss": 0.6397914123535157,
293
+ "step": 3900
294
+ },
295
+ {
296
+ "epoch": 1.034928848641656,
297
+ "grad_norm": 25.15692901611328,
298
+ "learning_rate": 1.4558696693818879e-05,
299
+ "loss": 0.5944495010375976,
300
+ "step": 4000
301
+ },
302
+ {
303
+ "epoch": 1.0608020698576972,
304
+ "grad_norm": 6.247220993041992,
305
+ "learning_rate": 1.4367034020124582e-05,
306
+ "loss": 0.5855656051635743,
307
+ "step": 4100
308
+ },
309
+ {
310
+ "epoch": 1.0866752910737387,
311
+ "grad_norm": 8.999340057373047,
312
+ "learning_rate": 1.4175371346430284e-05,
313
+ "loss": 0.6091019821166992,
314
+ "step": 4200
315
+ },
316
+ {
317
+ "epoch": 1.11254851228978,
318
+ "grad_norm": 10.196805953979492,
319
+ "learning_rate": 1.3983708672735986e-05,
320
+ "loss": 0.5973617553710937,
321
+ "step": 4300
322
+ },
323
+ {
324
+ "epoch": 1.1384217335058215,
325
+ "grad_norm": 9.179534912109375,
326
+ "learning_rate": 1.3792045999041688e-05,
327
+ "loss": 0.5922854232788086,
328
+ "step": 4400
329
+ },
330
+ {
331
+ "epoch": 1.1642949547218628,
332
+ "grad_norm": 11.18813419342041,
333
+ "learning_rate": 1.3600383325347391e-05,
334
+ "loss": 0.575289421081543,
335
+ "step": 4500
336
+ },
337
+ {
338
+ "epoch": 1.1901681759379044,
339
+ "grad_norm": 12.966670036315918,
340
+ "learning_rate": 1.3408720651653091e-05,
341
+ "loss": 0.5534341049194336,
342
+ "step": 4600
343
+ },
344
+ {
345
+ "epoch": 1.2160413971539457,
346
+ "grad_norm": 15.65532112121582,
347
+ "learning_rate": 1.3217057977958794e-05,
348
+ "loss": 0.5662339401245117,
349
+ "step": 4700
350
+ },
351
+ {
352
+ "epoch": 1.2419146183699872,
353
+ "grad_norm": 12.402419090270996,
354
+ "learning_rate": 1.3025395304264496e-05,
355
+ "loss": 0.6043228912353515,
356
+ "step": 4800
357
+ },
358
+ {
359
+ "epoch": 1.2677878395860285,
360
+ "grad_norm": 21.828113555908203,
361
+ "learning_rate": 1.2833732630570198e-05,
362
+ "loss": 0.5857475662231445,
363
+ "step": 4900
364
+ },
365
+ {
366
+ "epoch": 1.2936610608020698,
367
+ "grad_norm": 20.76739501953125,
368
+ "learning_rate": 1.26420699568759e-05,
369
+ "loss": 0.5899971389770508,
370
+ "step": 5000
371
+ },
372
+ {
373
+ "epoch": 1.3195342820181113,
374
+ "grad_norm": 13.431411743164062,
375
+ "learning_rate": 1.2450407283181601e-05,
376
+ "loss": 0.5802528381347656,
377
+ "step": 5100
378
+ },
379
+ {
380
+ "epoch": 1.3454075032341526,
381
+ "grad_norm": 17.105060577392578,
382
+ "learning_rate": 1.2258744609487303e-05,
383
+ "loss": 0.5847416305541993,
384
+ "step": 5200
385
+ },
386
+ {
387
+ "epoch": 1.371280724450194,
388
+ "grad_norm": 51.55683135986328,
389
+ "learning_rate": 1.2067081935793006e-05,
390
+ "loss": 0.5702993392944335,
391
+ "step": 5300
392
+ },
393
+ {
394
+ "epoch": 1.3971539456662354,
395
+ "grad_norm": 19.022991180419922,
396
+ "learning_rate": 1.1875419262098706e-05,
397
+ "loss": 0.5837166595458985,
398
+ "step": 5400
399
+ },
400
+ {
401
+ "epoch": 1.4230271668822767,
402
+ "grad_norm": 12.546523094177246,
403
+ "learning_rate": 1.168375658840441e-05,
404
+ "loss": 0.5941635131835937,
405
+ "step": 5500
406
+ },
407
+ {
408
+ "epoch": 1.4489003880983182,
409
+ "grad_norm": 16.194042205810547,
410
+ "learning_rate": 1.1492093914710111e-05,
411
+ "loss": 0.6047253799438477,
412
+ "step": 5600
413
+ },
414
+ {
415
+ "epoch": 1.4747736093143597,
416
+ "grad_norm": 14.654067993164062,
417
+ "learning_rate": 1.1300431241015813e-05,
418
+ "loss": 0.5888873291015625,
419
+ "step": 5700
420
+ },
421
+ {
422
+ "epoch": 1.500646830530401,
423
+ "grad_norm": 15.194084167480469,
424
+ "learning_rate": 1.1108768567321514e-05,
425
+ "loss": 0.5698460769653321,
426
+ "step": 5800
427
+ },
428
+ {
429
+ "epoch": 1.5265200517464423,
430
+ "grad_norm": 13.003668785095215,
431
+ "learning_rate": 1.0917105893627218e-05,
432
+ "loss": 0.5704814147949219,
433
+ "step": 5900
434
+ },
435
+ {
436
+ "epoch": 1.5523932729624839,
437
+ "grad_norm": 9.508420944213867,
438
+ "learning_rate": 1.0725443219932918e-05,
439
+ "loss": 0.5655577087402344,
440
+ "step": 6000
441
+ },
442
+ {
443
+ "epoch": 1.5782664941785254,
444
+ "grad_norm": 14.1275634765625,
445
+ "learning_rate": 1.0533780546238621e-05,
446
+ "loss": 0.5951072311401367,
447
+ "step": 6100
448
+ },
449
+ {
450
+ "epoch": 1.6041397153945667,
451
+ "grad_norm": 12.208888053894043,
452
+ "learning_rate": 1.0342117872544325e-05,
453
+ "loss": 0.6105654144287109,
454
+ "step": 6200
455
+ },
456
+ {
457
+ "epoch": 1.630012936610608,
458
+ "grad_norm": 14.291622161865234,
459
+ "learning_rate": 1.0150455198850025e-05,
460
+ "loss": 0.5704663467407226,
461
+ "step": 6300
462
+ },
463
+ {
464
+ "epoch": 1.6558861578266493,
465
+ "grad_norm": 11.807310104370117,
466
+ "learning_rate": 9.958792525155726e-06,
467
+ "loss": 0.5667978286743164,
468
+ "step": 6400
469
+ },
470
+ {
471
+ "epoch": 1.6817593790426908,
472
+ "grad_norm": 7.61679744720459,
473
+ "learning_rate": 9.767129851461428e-06,
474
+ "loss": 0.5553795623779297,
475
+ "step": 6500
476
+ },
477
+ {
478
+ "epoch": 1.7076326002587323,
479
+ "grad_norm": 9.418617248535156,
480
+ "learning_rate": 9.57546717776713e-06,
481
+ "loss": 0.5939513778686524,
482
+ "step": 6600
483
+ },
484
+ {
485
+ "epoch": 1.7335058214747736,
486
+ "grad_norm": 11.112833976745605,
487
+ "learning_rate": 9.383804504072831e-06,
488
+ "loss": 0.5403312683105469,
489
+ "step": 6700
490
+ },
491
+ {
492
+ "epoch": 1.759379042690815,
493
+ "grad_norm": 15.817110061645508,
494
+ "learning_rate": 9.192141830378535e-06,
495
+ "loss": 0.5703708267211914,
496
+ "step": 6800
497
+ },
498
+ {
499
+ "epoch": 1.7852522639068564,
500
+ "grad_norm": 8.80798053741455,
501
+ "learning_rate": 9.000479156684236e-06,
502
+ "loss": 0.5544892120361328,
503
+ "step": 6900
504
+ },
505
+ {
506
+ "epoch": 1.811125485122898,
507
+ "grad_norm": 16.89642333984375,
508
+ "learning_rate": 8.808816482989938e-06,
509
+ "loss": 0.573361701965332,
510
+ "step": 7000
511
+ },
512
+ {
513
+ "epoch": 1.8369987063389392,
514
+ "grad_norm": 16.808788299560547,
515
+ "learning_rate": 8.617153809295641e-06,
516
+ "loss": 0.5603259658813476,
517
+ "step": 7100
518
+ },
519
+ {
520
+ "epoch": 1.8628719275549805,
521
+ "grad_norm": 13.838523864746094,
522
+ "learning_rate": 8.425491135601343e-06,
523
+ "loss": 0.5797721481323242,
524
+ "step": 7200
525
+ },
526
+ {
527
+ "epoch": 1.8887451487710218,
528
+ "grad_norm": 8.298638343811035,
529
+ "learning_rate": 8.233828461907045e-06,
530
+ "loss": 0.5604195022583007,
531
+ "step": 7300
532
+ },
533
+ {
534
+ "epoch": 1.9146183699870634,
535
+ "grad_norm": 25.5234432220459,
536
+ "learning_rate": 8.042165788212747e-06,
537
+ "loss": 0.5867827606201171,
538
+ "step": 7400
539
+ },
540
+ {
541
+ "epoch": 1.9404915912031049,
542
+ "grad_norm": 15.7322416305542,
543
+ "learning_rate": 7.850503114518448e-06,
544
+ "loss": 0.5443745803833008,
545
+ "step": 7500
546
+ },
547
+ {
548
+ "epoch": 1.9663648124191462,
549
+ "grad_norm": 9.020604133605957,
550
+ "learning_rate": 7.65884044082415e-06,
551
+ "loss": 0.5062921905517578,
552
+ "step": 7600
553
+ },
554
+ {
555
+ "epoch": 1.9922380336351875,
556
+ "grad_norm": 11.164002418518066,
557
+ "learning_rate": 7.467177767129852e-06,
558
+ "loss": 0.5686885070800781,
559
+ "step": 7700
560
+ },
561
+ {
562
+ "epoch": 2.0,
563
+ "eval_accuracy": 0.7476510067114094,
564
+ "eval_f1": 0.7474800395521614,
565
+ "eval_loss": 0.5974893569946289,
566
+ "eval_runtime": 86.7366,
567
+ "eval_samples_per_second": 163.195,
568
+ "eval_steps_per_second": 10.203,
569
+ "step": 7730
570
+ }
571
+ ],
572
+ "logging_steps": 100,
573
+ "max_steps": 11595,
574
+ "num_input_tokens_seen": 0,
575
+ "num_train_epochs": 3,
576
+ "save_steps": 500,
577
+ "stateful_callbacks": {
578
+ "TrainerControl": {
579
+ "args": {
580
+ "should_epoch_stop": false,
581
+ "should_evaluate": false,
582
+ "should_log": false,
583
+ "should_save": true,
584
+ "should_training_stop": false
585
+ },
586
+ "attributes": {}
587
+ }
588
+ },
589
+ "total_flos": 8083999424953872.0,
590
+ "train_batch_size": 16,
591
+ "trial_name": null,
592
+ "trial_params": null
593
+ }
checkpoint-7730/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fabf78786d2ae6dcda35b22fbd2da223968e53ae585fc7647c873f62e2caaa5
3
+ size 5201
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "Normal",
16
+ "1": "Abusive",
17
+ "2": "Hate"
18
+ },
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "is_decoder": false,
22
+ "label2id": {
23
+ "Abusive": 1,
24
+ "Hate": 2,
25
+ "Normal": 0
26
+ },
27
+ "layer_norm_eps": 1e-05,
28
+ "max_position_embeddings": 514,
29
+ "model_type": "xlm-roberta",
30
+ "num_attention_heads": 12,
31
+ "num_hidden_layers": 12,
32
+ "output_past": true,
33
+ "pad_token_id": 1,
34
+ "position_embedding_type": "absolute",
35
+ "problem_type": "single_label_classification",
36
+ "tie_word_embeddings": true,
37
+ "transformers_version": "5.0.0",
38
+ "type_vocab_size": 1,
39
+ "use_cache": false,
40
+ "vocab_size": 250002
41
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1acbfdddadff0de9e27b9ba47d3487bf31dc11f779255a61a6b3accf8c0bfe2d
3
+ size 1112208060
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42d3da0b556fcc5dfcd7158b90066b5069bd23df2df684a17c4bcfd8575016ae
3
+ size 16781739
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
+ }