spacesedan commited on
Commit
ca3d9c0
·
verified ·
1 Parent(s): 147eec7

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ library_name: transformers
4
+ tags:
5
+ - autotrain
6
+ - text-classification
7
+ base_model: allenai/longformer-base-4096
8
+ widget:
9
+ - text: "I love AutoTrain"
10
+ datasets:
11
+ - spacesedan/Amazon_Reviews_Sentiment_10K
12
+ ---
13
+
14
+ # Model Trained Using AutoTrain
15
+
16
+ - Problem type: Text Classification
17
+
18
+ ## Validation Metrics
19
+ loss: 1.0463045835494995
20
+
21
+ f1_macro: 0.509019730829119
22
+
23
+ f1_micro: 0.5233333333333333
24
+
25
+ f1_weighted: 0.516464791683911
26
+
27
+ precision_macro: 0.5322711463107999
28
+
29
+ precision_micro: 0.5233333333333333
30
+
31
+ precision_weighted: 0.5447555045714482
32
+
33
+ recall_macro: 0.5207324546507
34
+
35
+ recall_micro: 0.5233333333333333
36
+
37
+ recall_weighted: 0.5233333333333333
38
+
39
+ accuracy: 0.5233333333333333
checkpoint-1250/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "_num_labels": 5,
4
+ "architectures": [
5
+ "LongformerForSequenceClassification"
6
+ ],
7
+ "attention_mode": "longformer",
8
+ "attention_probs_dropout_prob": 0.1,
9
+ "attention_window": [
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "bos_token_id": 0,
24
+ "eos_token_id": 2,
25
+ "gradient_checkpointing": false,
26
+ "hidden_act": "gelu",
27
+ "hidden_dropout_prob": 0.1,
28
+ "hidden_size": 768,
29
+ "id2label": {
30
+ "0": "very negative",
31
+ "1": "negative",
32
+ "2": "neutral",
33
+ "3": "positive",
34
+ "4": "very positive"
35
+ },
36
+ "ignore_attention_mask": false,
37
+ "initializer_range": 0.02,
38
+ "intermediate_size": 3072,
39
+ "label2id": {
40
+ "negative": 1,
41
+ "neutral": 2,
42
+ "positive": 3,
43
+ "very negative": 0,
44
+ "very positive": 4
45
+ },
46
+ "layer_norm_eps": 1e-05,
47
+ "max_position_embeddings": 4098,
48
+ "model_type": "longformer",
49
+ "num_attention_heads": 12,
50
+ "num_hidden_layers": 12,
51
+ "onnx_export": false,
52
+ "pad_token_id": 1,
53
+ "problem_type": "single_label_classification",
54
+ "sep_token_id": 2,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.48.0",
57
+ "type_vocab_size": 1,
58
+ "vocab_size": 50265
59
+ }
checkpoint-1250/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ea5a2958a9736d967d763abc8f3d7f4a1bebfee2308ff3232a133fdae09f851
3
+ size 594687412
checkpoint-1250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad0c96183a09d85452ddb4c7a1f516923b5c210a73709dcf2ec969e93ede113c
3
+ size 1189534778
checkpoint-1250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99a38c0987fd8f1754e2f11f40625624946104dfafd27ef602c27a3834a4cd57
3
+ size 14244
checkpoint-1250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88e1002e9313aa09f72c809ac17507e9e7c74ef6c0260b03690746a5ae7ee4ca
3
+ size 1064
checkpoint-1250/trainer_state.json ADDED
@@ -0,0 +1,1306 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.0463045835494995,
3
+ "best_model_checkpoint": "sentiment-analysis-longformer/checkpoint-1250",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1250,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0056,
13
+ "grad_norm": 4.3044047355651855,
14
+ "learning_rate": 8.000000000000001e-07,
15
+ "loss": 1.6185,
16
+ "step": 7
17
+ },
18
+ {
19
+ "epoch": 0.0112,
20
+ "grad_norm": 4.809700965881348,
21
+ "learning_rate": 1.7333333333333334e-06,
22
+ "loss": 1.6057,
23
+ "step": 14
24
+ },
25
+ {
26
+ "epoch": 0.0168,
27
+ "grad_norm": 2.63266921043396,
28
+ "learning_rate": 2.666666666666667e-06,
29
+ "loss": 1.6137,
30
+ "step": 21
31
+ },
32
+ {
33
+ "epoch": 0.0224,
34
+ "grad_norm": 3.832202196121216,
35
+ "learning_rate": 3.6e-06,
36
+ "loss": 1.6188,
37
+ "step": 28
38
+ },
39
+ {
40
+ "epoch": 0.028,
41
+ "grad_norm": 2.835721731185913,
42
+ "learning_rate": 4.533333333333334e-06,
43
+ "loss": 1.6015,
44
+ "step": 35
45
+ },
46
+ {
47
+ "epoch": 0.0336,
48
+ "grad_norm": 3.497213125228882,
49
+ "learning_rate": 5.466666666666667e-06,
50
+ "loss": 1.5799,
51
+ "step": 42
52
+ },
53
+ {
54
+ "epoch": 0.0392,
55
+ "grad_norm": 6.252377986907959,
56
+ "learning_rate": 6.4000000000000006e-06,
57
+ "loss": 1.6329,
58
+ "step": 49
59
+ },
60
+ {
61
+ "epoch": 0.0448,
62
+ "grad_norm": 3.252988576889038,
63
+ "learning_rate": 7.333333333333334e-06,
64
+ "loss": 1.6139,
65
+ "step": 56
66
+ },
67
+ {
68
+ "epoch": 0.0504,
69
+ "grad_norm": 3.3185133934020996,
70
+ "learning_rate": 8.266666666666667e-06,
71
+ "loss": 1.6002,
72
+ "step": 63
73
+ },
74
+ {
75
+ "epoch": 0.056,
76
+ "grad_norm": 1.9004288911819458,
77
+ "learning_rate": 9.2e-06,
78
+ "loss": 1.6223,
79
+ "step": 70
80
+ },
81
+ {
82
+ "epoch": 0.0616,
83
+ "grad_norm": 3.372849702835083,
84
+ "learning_rate": 1.0133333333333333e-05,
85
+ "loss": 1.6073,
86
+ "step": 77
87
+ },
88
+ {
89
+ "epoch": 0.0672,
90
+ "grad_norm": 5.070404052734375,
91
+ "learning_rate": 1.1066666666666667e-05,
92
+ "loss": 1.6172,
93
+ "step": 84
94
+ },
95
+ {
96
+ "epoch": 0.0728,
97
+ "grad_norm": 4.0468831062316895,
98
+ "learning_rate": 1.2e-05,
99
+ "loss": 1.6146,
100
+ "step": 91
101
+ },
102
+ {
103
+ "epoch": 0.0784,
104
+ "grad_norm": 3.557427167892456,
105
+ "learning_rate": 1.2933333333333334e-05,
106
+ "loss": 1.6143,
107
+ "step": 98
108
+ },
109
+ {
110
+ "epoch": 0.084,
111
+ "grad_norm": 2.1088807582855225,
112
+ "learning_rate": 1.3866666666666667e-05,
113
+ "loss": 1.594,
114
+ "step": 105
115
+ },
116
+ {
117
+ "epoch": 0.0896,
118
+ "grad_norm": 7.924050807952881,
119
+ "learning_rate": 1.48e-05,
120
+ "loss": 1.5701,
121
+ "step": 112
122
+ },
123
+ {
124
+ "epoch": 0.0952,
125
+ "grad_norm": 11.886157989501953,
126
+ "learning_rate": 1.5733333333333334e-05,
127
+ "loss": 1.4665,
128
+ "step": 119
129
+ },
130
+ {
131
+ "epoch": 0.1008,
132
+ "grad_norm": NaN,
133
+ "learning_rate": 1.6533333333333333e-05,
134
+ "loss": 1.4903,
135
+ "step": 126
136
+ },
137
+ {
138
+ "epoch": 0.1064,
139
+ "grad_norm": 22.54196548461914,
140
+ "learning_rate": 1.7466666666666667e-05,
141
+ "loss": 1.3656,
142
+ "step": 133
143
+ },
144
+ {
145
+ "epoch": 0.112,
146
+ "grad_norm": 7.953355312347412,
147
+ "learning_rate": 1.826666666666667e-05,
148
+ "loss": 1.3593,
149
+ "step": 140
150
+ },
151
+ {
152
+ "epoch": 0.1176,
153
+ "grad_norm": 7.947963237762451,
154
+ "learning_rate": 1.9200000000000003e-05,
155
+ "loss": 1.3578,
156
+ "step": 147
157
+ },
158
+ {
159
+ "epoch": 0.1232,
160
+ "grad_norm": 37.732215881347656,
161
+ "learning_rate": 2.0133333333333336e-05,
162
+ "loss": 1.3341,
163
+ "step": 154
164
+ },
165
+ {
166
+ "epoch": 0.1288,
167
+ "grad_norm": 24.383142471313477,
168
+ "learning_rate": 2.0933333333333335e-05,
169
+ "loss": 1.3916,
170
+ "step": 161
171
+ },
172
+ {
173
+ "epoch": 0.1344,
174
+ "grad_norm": 10.203289031982422,
175
+ "learning_rate": 2.186666666666667e-05,
176
+ "loss": 1.1943,
177
+ "step": 168
178
+ },
179
+ {
180
+ "epoch": 0.14,
181
+ "grad_norm": 38.70049285888672,
182
+ "learning_rate": 2.2800000000000002e-05,
183
+ "loss": 1.115,
184
+ "step": 175
185
+ },
186
+ {
187
+ "epoch": 0.1456,
188
+ "grad_norm": 44.0554313659668,
189
+ "learning_rate": 2.3733333333333335e-05,
190
+ "loss": 1.1703,
191
+ "step": 182
192
+ },
193
+ {
194
+ "epoch": 0.1512,
195
+ "grad_norm": 44.12297821044922,
196
+ "learning_rate": 2.466666666666667e-05,
197
+ "loss": 1.2689,
198
+ "step": 189
199
+ },
200
+ {
201
+ "epoch": 0.1568,
202
+ "grad_norm": 24.104822158813477,
203
+ "learning_rate": 2.5600000000000002e-05,
204
+ "loss": 1.5482,
205
+ "step": 196
206
+ },
207
+ {
208
+ "epoch": 0.1624,
209
+ "grad_norm": 28.003192901611328,
210
+ "learning_rate": 2.6533333333333332e-05,
211
+ "loss": 1.1575,
212
+ "step": 203
213
+ },
214
+ {
215
+ "epoch": 0.168,
216
+ "grad_norm": 17.407148361206055,
217
+ "learning_rate": 2.746666666666667e-05,
218
+ "loss": 1.1556,
219
+ "step": 210
220
+ },
221
+ {
222
+ "epoch": 0.1736,
223
+ "grad_norm": 30.059938430786133,
224
+ "learning_rate": 2.84e-05,
225
+ "loss": 1.3255,
226
+ "step": 217
227
+ },
228
+ {
229
+ "epoch": 0.1792,
230
+ "grad_norm": 17.500478744506836,
231
+ "learning_rate": 2.9333333333333336e-05,
232
+ "loss": 1.1785,
233
+ "step": 224
234
+ },
235
+ {
236
+ "epoch": 0.1848,
237
+ "grad_norm": 18.938140869140625,
238
+ "learning_rate": 3.0266666666666666e-05,
239
+ "loss": 1.2427,
240
+ "step": 231
241
+ },
242
+ {
243
+ "epoch": 0.1904,
244
+ "grad_norm": 46.56889343261719,
245
+ "learning_rate": 3.12e-05,
246
+ "loss": 1.3003,
247
+ "step": 238
248
+ },
249
+ {
250
+ "epoch": 0.196,
251
+ "grad_norm": 14.280040740966797,
252
+ "learning_rate": 3.213333333333334e-05,
253
+ "loss": 1.3049,
254
+ "step": 245
255
+ },
256
+ {
257
+ "epoch": 0.2016,
258
+ "grad_norm": 20.890186309814453,
259
+ "learning_rate": 3.3066666666666666e-05,
260
+ "loss": 1.1882,
261
+ "step": 252
262
+ },
263
+ {
264
+ "epoch": 0.2072,
265
+ "grad_norm": 15.023652076721191,
266
+ "learning_rate": 3.4000000000000007e-05,
267
+ "loss": 1.3098,
268
+ "step": 259
269
+ },
270
+ {
271
+ "epoch": 0.2128,
272
+ "grad_norm": 15.437768936157227,
273
+ "learning_rate": 3.493333333333333e-05,
274
+ "loss": 1.1317,
275
+ "step": 266
276
+ },
277
+ {
278
+ "epoch": 0.2184,
279
+ "grad_norm": 17.70416831970215,
280
+ "learning_rate": 3.586666666666667e-05,
281
+ "loss": 1.2326,
282
+ "step": 273
283
+ },
284
+ {
285
+ "epoch": 0.224,
286
+ "grad_norm": 25.55316734313965,
287
+ "learning_rate": 3.68e-05,
288
+ "loss": 1.0633,
289
+ "step": 280
290
+ },
291
+ {
292
+ "epoch": 0.2296,
293
+ "grad_norm": 44.82810592651367,
294
+ "learning_rate": 3.773333333333334e-05,
295
+ "loss": 1.0947,
296
+ "step": 287
297
+ },
298
+ {
299
+ "epoch": 0.2352,
300
+ "grad_norm": 26.375022888183594,
301
+ "learning_rate": 3.866666666666667e-05,
302
+ "loss": 1.3421,
303
+ "step": 294
304
+ },
305
+ {
306
+ "epoch": 0.2408,
307
+ "grad_norm": 27.109973907470703,
308
+ "learning_rate": 3.960000000000001e-05,
309
+ "loss": 1.1407,
310
+ "step": 301
311
+ },
312
+ {
313
+ "epoch": 0.2464,
314
+ "grad_norm": 33.65099334716797,
315
+ "learning_rate": 4.0533333333333334e-05,
316
+ "loss": 0.9152,
317
+ "step": 308
318
+ },
319
+ {
320
+ "epoch": 0.252,
321
+ "grad_norm": 20.36009979248047,
322
+ "learning_rate": 4.146666666666667e-05,
323
+ "loss": 1.1515,
324
+ "step": 315
325
+ },
326
+ {
327
+ "epoch": 0.2576,
328
+ "grad_norm": 25.868627548217773,
329
+ "learning_rate": 4.24e-05,
330
+ "loss": 1.2575,
331
+ "step": 322
332
+ },
333
+ {
334
+ "epoch": 0.2632,
335
+ "grad_norm": 17.713809967041016,
336
+ "learning_rate": 4.3333333333333334e-05,
337
+ "loss": 1.0586,
338
+ "step": 329
339
+ },
340
+ {
341
+ "epoch": 0.2688,
342
+ "grad_norm": 12.709273338317871,
343
+ "learning_rate": 4.426666666666667e-05,
344
+ "loss": 1.2379,
345
+ "step": 336
346
+ },
347
+ {
348
+ "epoch": 0.2744,
349
+ "grad_norm": 62.326900482177734,
350
+ "learning_rate": 4.52e-05,
351
+ "loss": 1.0445,
352
+ "step": 343
353
+ },
354
+ {
355
+ "epoch": 0.28,
356
+ "grad_norm": 31.3813533782959,
357
+ "learning_rate": 4.6133333333333334e-05,
358
+ "loss": 1.2517,
359
+ "step": 350
360
+ },
361
+ {
362
+ "epoch": 0.2856,
363
+ "grad_norm": 31.736560821533203,
364
+ "learning_rate": 4.706666666666667e-05,
365
+ "loss": 1.2875,
366
+ "step": 357
367
+ },
368
+ {
369
+ "epoch": 0.2912,
370
+ "grad_norm": 9.466754913330078,
371
+ "learning_rate": 4.8e-05,
372
+ "loss": 1.323,
373
+ "step": 364
374
+ },
375
+ {
376
+ "epoch": 0.2968,
377
+ "grad_norm": 32.847103118896484,
378
+ "learning_rate": 4.8933333333333335e-05,
379
+ "loss": 1.3974,
380
+ "step": 371
381
+ },
382
+ {
383
+ "epoch": 0.3024,
384
+ "grad_norm": 10.510279655456543,
385
+ "learning_rate": 4.986666666666667e-05,
386
+ "loss": 1.17,
387
+ "step": 378
388
+ },
389
+ {
390
+ "epoch": 0.308,
391
+ "grad_norm": 37.245418548583984,
392
+ "learning_rate": 4.991111111111111e-05,
393
+ "loss": 1.1435,
394
+ "step": 385
395
+ },
396
+ {
397
+ "epoch": 0.3136,
398
+ "grad_norm": 29.977188110351562,
399
+ "learning_rate": 4.982222222222222e-05,
400
+ "loss": 1.2446,
401
+ "step": 392
402
+ },
403
+ {
404
+ "epoch": 0.3192,
405
+ "grad_norm": 21.97435188293457,
406
+ "learning_rate": 4.971851851851852e-05,
407
+ "loss": 1.2165,
408
+ "step": 399
409
+ },
410
+ {
411
+ "epoch": 0.3248,
412
+ "grad_norm": 15.939322471618652,
413
+ "learning_rate": 4.961481481481482e-05,
414
+ "loss": 1.4191,
415
+ "step": 406
416
+ },
417
+ {
418
+ "epoch": 0.3304,
419
+ "grad_norm": 14.586689949035645,
420
+ "learning_rate": 4.951111111111112e-05,
421
+ "loss": 1.2868,
422
+ "step": 413
423
+ },
424
+ {
425
+ "epoch": 0.336,
426
+ "grad_norm": 9.520950317382812,
427
+ "learning_rate": 4.940740740740741e-05,
428
+ "loss": 1.3862,
429
+ "step": 420
430
+ },
431
+ {
432
+ "epoch": 0.3416,
433
+ "grad_norm": 19.00330352783203,
434
+ "learning_rate": 4.9303703703703705e-05,
435
+ "loss": 1.3354,
436
+ "step": 427
437
+ },
438
+ {
439
+ "epoch": 0.3472,
440
+ "grad_norm": 6.289086818695068,
441
+ "learning_rate": 4.92e-05,
442
+ "loss": 1.3717,
443
+ "step": 434
444
+ },
445
+ {
446
+ "epoch": 0.3528,
447
+ "grad_norm": 66.89319610595703,
448
+ "learning_rate": 4.90962962962963e-05,
449
+ "loss": 1.2314,
450
+ "step": 441
451
+ },
452
+ {
453
+ "epoch": 0.3584,
454
+ "grad_norm": 14.694267272949219,
455
+ "learning_rate": 4.89925925925926e-05,
456
+ "loss": 1.2981,
457
+ "step": 448
458
+ },
459
+ {
460
+ "epoch": 0.364,
461
+ "grad_norm": 12.73987102508545,
462
+ "learning_rate": 4.888888888888889e-05,
463
+ "loss": 1.3561,
464
+ "step": 455
465
+ },
466
+ {
467
+ "epoch": 0.3696,
468
+ "grad_norm": 13.01091480255127,
469
+ "learning_rate": 4.878518518518519e-05,
470
+ "loss": 1.27,
471
+ "step": 462
472
+ },
473
+ {
474
+ "epoch": 0.3752,
475
+ "grad_norm": 13.623579978942871,
476
+ "learning_rate": 4.8681481481481485e-05,
477
+ "loss": 1.2522,
478
+ "step": 469
479
+ },
480
+ {
481
+ "epoch": 0.3808,
482
+ "grad_norm": 41.466163635253906,
483
+ "learning_rate": 4.8577777777777776e-05,
484
+ "loss": 1.4129,
485
+ "step": 476
486
+ },
487
+ {
488
+ "epoch": 0.3864,
489
+ "grad_norm": 28.096891403198242,
490
+ "learning_rate": 4.8474074074074074e-05,
491
+ "loss": 1.0454,
492
+ "step": 483
493
+ },
494
+ {
495
+ "epoch": 0.392,
496
+ "grad_norm": 26.944536209106445,
497
+ "learning_rate": 4.837037037037037e-05,
498
+ "loss": 1.2051,
499
+ "step": 490
500
+ },
501
+ {
502
+ "epoch": 0.3976,
503
+ "grad_norm": 18.42100715637207,
504
+ "learning_rate": 4.826666666666667e-05,
505
+ "loss": 1.127,
506
+ "step": 497
507
+ },
508
+ {
509
+ "epoch": 0.4032,
510
+ "grad_norm": 32.047943115234375,
511
+ "learning_rate": 4.816296296296297e-05,
512
+ "loss": 1.2287,
513
+ "step": 504
514
+ },
515
+ {
516
+ "epoch": 0.4088,
517
+ "grad_norm": 36.99417495727539,
518
+ "learning_rate": 4.805925925925926e-05,
519
+ "loss": 1.2363,
520
+ "step": 511
521
+ },
522
+ {
523
+ "epoch": 0.4144,
524
+ "grad_norm": 19.844234466552734,
525
+ "learning_rate": 4.7955555555555556e-05,
526
+ "loss": 1.194,
527
+ "step": 518
528
+ },
529
+ {
530
+ "epoch": 0.42,
531
+ "grad_norm": 11.864848136901855,
532
+ "learning_rate": 4.7851851851851854e-05,
533
+ "loss": 1.2978,
534
+ "step": 525
535
+ },
536
+ {
537
+ "epoch": 0.4256,
538
+ "grad_norm": 13.68875789642334,
539
+ "learning_rate": 4.774814814814815e-05,
540
+ "loss": 1.0287,
541
+ "step": 532
542
+ },
543
+ {
544
+ "epoch": 0.4312,
545
+ "grad_norm": 21.006519317626953,
546
+ "learning_rate": 4.764444444444445e-05,
547
+ "loss": 1.1714,
548
+ "step": 539
549
+ },
550
+ {
551
+ "epoch": 0.4368,
552
+ "grad_norm": 14.085043907165527,
553
+ "learning_rate": 4.754074074074074e-05,
554
+ "loss": 1.3415,
555
+ "step": 546
556
+ },
557
+ {
558
+ "epoch": 0.4424,
559
+ "grad_norm": 6.128912925720215,
560
+ "learning_rate": 4.743703703703704e-05,
561
+ "loss": 1.2443,
562
+ "step": 553
563
+ },
564
+ {
565
+ "epoch": 0.448,
566
+ "grad_norm": 13.3624267578125,
567
+ "learning_rate": 4.7333333333333336e-05,
568
+ "loss": 1.0881,
569
+ "step": 560
570
+ },
571
+ {
572
+ "epoch": 0.4536,
573
+ "grad_norm": 20.717557907104492,
574
+ "learning_rate": 4.722962962962963e-05,
575
+ "loss": 1.4919,
576
+ "step": 567
577
+ },
578
+ {
579
+ "epoch": 0.4592,
580
+ "grad_norm": 9.526569366455078,
581
+ "learning_rate": 4.712592592592593e-05,
582
+ "loss": 1.1529,
583
+ "step": 574
584
+ },
585
+ {
586
+ "epoch": 0.4648,
587
+ "grad_norm": 21.219013214111328,
588
+ "learning_rate": 4.702222222222222e-05,
589
+ "loss": 1.1257,
590
+ "step": 581
591
+ },
592
+ {
593
+ "epoch": 0.4704,
594
+ "grad_norm": 16.09003448486328,
595
+ "learning_rate": 4.691851851851852e-05,
596
+ "loss": 0.9876,
597
+ "step": 588
598
+ },
599
+ {
600
+ "epoch": 0.476,
601
+ "grad_norm": 12.591224670410156,
602
+ "learning_rate": 4.681481481481482e-05,
603
+ "loss": 1.2323,
604
+ "step": 595
605
+ },
606
+ {
607
+ "epoch": 0.4816,
608
+ "grad_norm": 23.968263626098633,
609
+ "learning_rate": 4.671111111111111e-05,
610
+ "loss": 1.2817,
611
+ "step": 602
612
+ },
613
+ {
614
+ "epoch": 0.4872,
615
+ "grad_norm": 47.702857971191406,
616
+ "learning_rate": 4.660740740740741e-05,
617
+ "loss": 1.2608,
618
+ "step": 609
619
+ },
620
+ {
621
+ "epoch": 0.4928,
622
+ "grad_norm": 20.887250900268555,
623
+ "learning_rate": 4.6503703703703705e-05,
624
+ "loss": 1.3105,
625
+ "step": 616
626
+ },
627
+ {
628
+ "epoch": 0.4984,
629
+ "grad_norm": 12.29234790802002,
630
+ "learning_rate": 4.64e-05,
631
+ "loss": 1.1678,
632
+ "step": 623
633
+ },
634
+ {
635
+ "epoch": 0.504,
636
+ "grad_norm": 6.830883026123047,
637
+ "learning_rate": 4.62962962962963e-05,
638
+ "loss": 1.1546,
639
+ "step": 630
640
+ },
641
+ {
642
+ "epoch": 0.5096,
643
+ "grad_norm": 11.166626930236816,
644
+ "learning_rate": 4.619259259259259e-05,
645
+ "loss": 1.3306,
646
+ "step": 637
647
+ },
648
+ {
649
+ "epoch": 0.5152,
650
+ "grad_norm": 14.119309425354004,
651
+ "learning_rate": 4.608888888888889e-05,
652
+ "loss": 1.252,
653
+ "step": 644
654
+ },
655
+ {
656
+ "epoch": 0.5208,
657
+ "grad_norm": 18.480911254882812,
658
+ "learning_rate": 4.598518518518519e-05,
659
+ "loss": 1.2212,
660
+ "step": 651
661
+ },
662
+ {
663
+ "epoch": 0.5264,
664
+ "grad_norm": 14.240557670593262,
665
+ "learning_rate": 4.5881481481481485e-05,
666
+ "loss": 1.1538,
667
+ "step": 658
668
+ },
669
+ {
670
+ "epoch": 0.532,
671
+ "grad_norm": 26.27284812927246,
672
+ "learning_rate": 4.577777777777778e-05,
673
+ "loss": 1.3187,
674
+ "step": 665
675
+ },
676
+ {
677
+ "epoch": 0.5376,
678
+ "grad_norm": 15.894841194152832,
679
+ "learning_rate": 4.5674074074074074e-05,
680
+ "loss": 1.3132,
681
+ "step": 672
682
+ },
683
+ {
684
+ "epoch": 0.5432,
685
+ "grad_norm": 9.206047058105469,
686
+ "learning_rate": 4.557037037037037e-05,
687
+ "loss": 1.1607,
688
+ "step": 679
689
+ },
690
+ {
691
+ "epoch": 0.5488,
692
+ "grad_norm": 27.2408390045166,
693
+ "learning_rate": 4.546666666666667e-05,
694
+ "loss": 1.3794,
695
+ "step": 686
696
+ },
697
+ {
698
+ "epoch": 0.5544,
699
+ "grad_norm": 32.67033767700195,
700
+ "learning_rate": 4.536296296296296e-05,
701
+ "loss": 1.3189,
702
+ "step": 693
703
+ },
704
+ {
705
+ "epoch": 0.56,
706
+ "grad_norm": 14.853004455566406,
707
+ "learning_rate": 4.5259259259259265e-05,
708
+ "loss": 1.2994,
709
+ "step": 700
710
+ },
711
+ {
712
+ "epoch": 0.5656,
713
+ "grad_norm": 17.385454177856445,
714
+ "learning_rate": 4.5155555555555556e-05,
715
+ "loss": 1.3544,
716
+ "step": 707
717
+ },
718
+ {
719
+ "epoch": 0.5712,
720
+ "grad_norm": 8.822907447814941,
721
+ "learning_rate": 4.5051851851851854e-05,
722
+ "loss": 1.0642,
723
+ "step": 714
724
+ },
725
+ {
726
+ "epoch": 0.5768,
727
+ "grad_norm": 10.728267669677734,
728
+ "learning_rate": 4.494814814814815e-05,
729
+ "loss": 1.2032,
730
+ "step": 721
731
+ },
732
+ {
733
+ "epoch": 0.5824,
734
+ "grad_norm": 18.303375244140625,
735
+ "learning_rate": 4.484444444444444e-05,
736
+ "loss": 1.0763,
737
+ "step": 728
738
+ },
739
+ {
740
+ "epoch": 0.588,
741
+ "grad_norm": 14.441327095031738,
742
+ "learning_rate": 4.474074074074075e-05,
743
+ "loss": 0.9293,
744
+ "step": 735
745
+ },
746
+ {
747
+ "epoch": 0.5936,
748
+ "grad_norm": 12.140674591064453,
749
+ "learning_rate": 4.463703703703704e-05,
750
+ "loss": 1.0541,
751
+ "step": 742
752
+ },
753
+ {
754
+ "epoch": 0.5992,
755
+ "grad_norm": 27.210546493530273,
756
+ "learning_rate": 4.4533333333333336e-05,
757
+ "loss": 1.408,
758
+ "step": 749
759
+ },
760
+ {
761
+ "epoch": 0.6048,
762
+ "grad_norm": 11.405797958374023,
763
+ "learning_rate": 4.4429629629629634e-05,
764
+ "loss": 1.1593,
765
+ "step": 756
766
+ },
767
+ {
768
+ "epoch": 0.6104,
769
+ "grad_norm": 16.38625144958496,
770
+ "learning_rate": 4.4325925925925925e-05,
771
+ "loss": 1.1842,
772
+ "step": 763
773
+ },
774
+ {
775
+ "epoch": 0.616,
776
+ "grad_norm": 20.51258659362793,
777
+ "learning_rate": 4.422222222222222e-05,
778
+ "loss": 1.2711,
779
+ "step": 770
780
+ },
781
+ {
782
+ "epoch": 0.6216,
783
+ "grad_norm": 27.440540313720703,
784
+ "learning_rate": 4.411851851851852e-05,
785
+ "loss": 1.1911,
786
+ "step": 777
787
+ },
788
+ {
789
+ "epoch": 0.6272,
790
+ "grad_norm": 24.75773048400879,
791
+ "learning_rate": 4.401481481481481e-05,
792
+ "loss": 1.046,
793
+ "step": 784
794
+ },
795
+ {
796
+ "epoch": 0.6328,
797
+ "grad_norm": 18.571340560913086,
798
+ "learning_rate": 4.3911111111111116e-05,
799
+ "loss": 1.2397,
800
+ "step": 791
801
+ },
802
+ {
803
+ "epoch": 0.6384,
804
+ "grad_norm": 27.95873260498047,
805
+ "learning_rate": 4.380740740740741e-05,
806
+ "loss": 1.3633,
807
+ "step": 798
808
+ },
809
+ {
810
+ "epoch": 0.644,
811
+ "grad_norm": 9.077492713928223,
812
+ "learning_rate": 4.3703703703703705e-05,
813
+ "loss": 1.1718,
814
+ "step": 805
815
+ },
816
+ {
817
+ "epoch": 0.6496,
818
+ "grad_norm": 12.240832328796387,
819
+ "learning_rate": 4.36e-05,
820
+ "loss": 1.1483,
821
+ "step": 812
822
+ },
823
+ {
824
+ "epoch": 0.6552,
825
+ "grad_norm": 28.757829666137695,
826
+ "learning_rate": 4.3496296296296294e-05,
827
+ "loss": 1.1503,
828
+ "step": 819
829
+ },
830
+ {
831
+ "epoch": 0.6608,
832
+ "grad_norm": 17.067678451538086,
833
+ "learning_rate": 4.33925925925926e-05,
834
+ "loss": 1.3015,
835
+ "step": 826
836
+ },
837
+ {
838
+ "epoch": 0.6664,
839
+ "grad_norm": 15.772140502929688,
840
+ "learning_rate": 4.328888888888889e-05,
841
+ "loss": 1.0159,
842
+ "step": 833
843
+ },
844
+ {
845
+ "epoch": 0.672,
846
+ "grad_norm": 9.903558731079102,
847
+ "learning_rate": 4.318518518518519e-05,
848
+ "loss": 1.2331,
849
+ "step": 840
850
+ },
851
+ {
852
+ "epoch": 0.6776,
853
+ "grad_norm": 10.482284545898438,
854
+ "learning_rate": 4.3081481481481485e-05,
855
+ "loss": 0.8856,
856
+ "step": 847
857
+ },
858
+ {
859
+ "epoch": 0.6832,
860
+ "grad_norm": 16.578685760498047,
861
+ "learning_rate": 4.2977777777777776e-05,
862
+ "loss": 1.1766,
863
+ "step": 854
864
+ },
865
+ {
866
+ "epoch": 0.6888,
867
+ "grad_norm": 22.36375617980957,
868
+ "learning_rate": 4.287407407407408e-05,
869
+ "loss": 1.1828,
870
+ "step": 861
871
+ },
872
+ {
873
+ "epoch": 0.6944,
874
+ "grad_norm": 51.31915283203125,
875
+ "learning_rate": 4.277037037037037e-05,
876
+ "loss": 1.2534,
877
+ "step": 868
878
+ },
879
+ {
880
+ "epoch": 0.7,
881
+ "grad_norm": 43.600711822509766,
882
+ "learning_rate": 4.266666666666667e-05,
883
+ "loss": 1.1778,
884
+ "step": 875
885
+ },
886
+ {
887
+ "epoch": 0.7056,
888
+ "grad_norm": 18.086084365844727,
889
+ "learning_rate": 4.256296296296297e-05,
890
+ "loss": 1.2067,
891
+ "step": 882
892
+ },
893
+ {
894
+ "epoch": 0.7112,
895
+ "grad_norm": 11.24244213104248,
896
+ "learning_rate": 4.245925925925926e-05,
897
+ "loss": 1.145,
898
+ "step": 889
899
+ },
900
+ {
901
+ "epoch": 0.7168,
902
+ "grad_norm": 12.832592964172363,
903
+ "learning_rate": 4.235555555555556e-05,
904
+ "loss": 1.0475,
905
+ "step": 896
906
+ },
907
+ {
908
+ "epoch": 0.7224,
909
+ "grad_norm": 33.09056854248047,
910
+ "learning_rate": 4.2251851851851854e-05,
911
+ "loss": 1.1055,
912
+ "step": 903
913
+ },
914
+ {
915
+ "epoch": 0.728,
916
+ "grad_norm": 16.331045150756836,
917
+ "learning_rate": 4.2148148148148145e-05,
918
+ "loss": 1.3999,
919
+ "step": 910
920
+ },
921
+ {
922
+ "epoch": 0.7336,
923
+ "grad_norm": 13.34793758392334,
924
+ "learning_rate": 4.204444444444445e-05,
925
+ "loss": 1.0609,
926
+ "step": 917
927
+ },
928
+ {
929
+ "epoch": 0.7392,
930
+ "grad_norm": 15.56451416015625,
931
+ "learning_rate": 4.194074074074074e-05,
932
+ "loss": 1.0921,
933
+ "step": 924
934
+ },
935
+ {
936
+ "epoch": 0.7448,
937
+ "grad_norm": 36.35686111450195,
938
+ "learning_rate": 4.183703703703704e-05,
939
+ "loss": 1.0468,
940
+ "step": 931
941
+ },
942
+ {
943
+ "epoch": 0.7504,
944
+ "grad_norm": 19.2562313079834,
945
+ "learning_rate": 4.1733333333333336e-05,
946
+ "loss": 1.3423,
947
+ "step": 938
948
+ },
949
+ {
950
+ "epoch": 0.756,
951
+ "grad_norm": 51.04462814331055,
952
+ "learning_rate": 4.162962962962963e-05,
953
+ "loss": 1.1937,
954
+ "step": 945
955
+ },
956
+ {
957
+ "epoch": 0.7616,
958
+ "grad_norm": 13.460155487060547,
959
+ "learning_rate": 4.152592592592593e-05,
960
+ "loss": 1.3615,
961
+ "step": 952
962
+ },
963
+ {
964
+ "epoch": 0.7672,
965
+ "grad_norm": 25.266849517822266,
966
+ "learning_rate": 4.142222222222222e-05,
967
+ "loss": 1.1688,
968
+ "step": 959
969
+ },
970
+ {
971
+ "epoch": 0.7728,
972
+ "grad_norm": 13.368788719177246,
973
+ "learning_rate": 4.131851851851852e-05,
974
+ "loss": 1.0765,
975
+ "step": 966
976
+ },
977
+ {
978
+ "epoch": 0.7784,
979
+ "grad_norm": 12.62553596496582,
980
+ "learning_rate": 4.121481481481482e-05,
981
+ "loss": 1.0952,
982
+ "step": 973
983
+ },
984
+ {
985
+ "epoch": 0.784,
986
+ "grad_norm": 17.647890090942383,
987
+ "learning_rate": 4.111111111111111e-05,
988
+ "loss": 1.08,
989
+ "step": 980
990
+ },
991
+ {
992
+ "epoch": 0.7896,
993
+ "grad_norm": 24.468290328979492,
994
+ "learning_rate": 4.1007407407407414e-05,
995
+ "loss": 1.3311,
996
+ "step": 987
997
+ },
998
+ {
999
+ "epoch": 0.7952,
1000
+ "grad_norm": 13.195050239562988,
1001
+ "learning_rate": 4.0903703703703705e-05,
1002
+ "loss": 1.2307,
1003
+ "step": 994
1004
+ },
1005
+ {
1006
+ "epoch": 0.8008,
1007
+ "grad_norm": 9.442852973937988,
1008
+ "learning_rate": 4.08e-05,
1009
+ "loss": 1.1655,
1010
+ "step": 1001
1011
+ },
1012
+ {
1013
+ "epoch": 0.8064,
1014
+ "grad_norm": 11.254621505737305,
1015
+ "learning_rate": 4.06962962962963e-05,
1016
+ "loss": 1.2512,
1017
+ "step": 1008
1018
+ },
1019
+ {
1020
+ "epoch": 0.812,
1021
+ "grad_norm": 11.979645729064941,
1022
+ "learning_rate": 4.059259259259259e-05,
1023
+ "loss": 1.2049,
1024
+ "step": 1015
1025
+ },
1026
+ {
1027
+ "epoch": 0.8176,
1028
+ "grad_norm": 13.316203117370605,
1029
+ "learning_rate": 4.0488888888888896e-05,
1030
+ "loss": 1.1102,
1031
+ "step": 1022
1032
+ },
1033
+ {
1034
+ "epoch": 0.8232,
1035
+ "grad_norm": 19.840560913085938,
1036
+ "learning_rate": 4.038518518518519e-05,
1037
+ "loss": 1.2102,
1038
+ "step": 1029
1039
+ },
1040
+ {
1041
+ "epoch": 0.8288,
1042
+ "grad_norm": 17.739408493041992,
1043
+ "learning_rate": 4.028148148148148e-05,
1044
+ "loss": 1.0162,
1045
+ "step": 1036
1046
+ },
1047
+ {
1048
+ "epoch": 0.8344,
1049
+ "grad_norm": 15.485058784484863,
1050
+ "learning_rate": 4.017777777777778e-05,
1051
+ "loss": 1.1379,
1052
+ "step": 1043
1053
+ },
1054
+ {
1055
+ "epoch": 0.84,
1056
+ "grad_norm": 29.864459991455078,
1057
+ "learning_rate": 4.007407407407407e-05,
1058
+ "loss": 1.322,
1059
+ "step": 1050
1060
+ },
1061
+ {
1062
+ "epoch": 0.8456,
1063
+ "grad_norm": 27.43077278137207,
1064
+ "learning_rate": 3.997037037037038e-05,
1065
+ "loss": 1.1343,
1066
+ "step": 1057
1067
+ },
1068
+ {
1069
+ "epoch": 0.8512,
1070
+ "grad_norm": 8.520703315734863,
1071
+ "learning_rate": 3.986666666666667e-05,
1072
+ "loss": 1.1657,
1073
+ "step": 1064
1074
+ },
1075
+ {
1076
+ "epoch": 0.8568,
1077
+ "grad_norm": 13.474251747131348,
1078
+ "learning_rate": 3.976296296296296e-05,
1079
+ "loss": 0.9206,
1080
+ "step": 1071
1081
+ },
1082
+ {
1083
+ "epoch": 0.8624,
1084
+ "grad_norm": 27.06108856201172,
1085
+ "learning_rate": 3.9659259259259265e-05,
1086
+ "loss": 0.9325,
1087
+ "step": 1078
1088
+ },
1089
+ {
1090
+ "epoch": 0.868,
1091
+ "grad_norm": 19.521345138549805,
1092
+ "learning_rate": 3.9555555555555556e-05,
1093
+ "loss": 1.1772,
1094
+ "step": 1085
1095
+ },
1096
+ {
1097
+ "epoch": 0.8736,
1098
+ "grad_norm": 21.125864028930664,
1099
+ "learning_rate": 3.945185185185185e-05,
1100
+ "loss": 1.2945,
1101
+ "step": 1092
1102
+ },
1103
+ {
1104
+ "epoch": 0.8792,
1105
+ "grad_norm": 20.886058807373047,
1106
+ "learning_rate": 3.934814814814815e-05,
1107
+ "loss": 1.2349,
1108
+ "step": 1099
1109
+ },
1110
+ {
1111
+ "epoch": 0.8848,
1112
+ "grad_norm": 12.874465942382812,
1113
+ "learning_rate": 3.924444444444444e-05,
1114
+ "loss": 1.2065,
1115
+ "step": 1106
1116
+ },
1117
+ {
1118
+ "epoch": 0.8904,
1119
+ "grad_norm": 12.321556091308594,
1120
+ "learning_rate": 3.914074074074075e-05,
1121
+ "loss": 0.944,
1122
+ "step": 1113
1123
+ },
1124
+ {
1125
+ "epoch": 0.896,
1126
+ "grad_norm": 30.77228355407715,
1127
+ "learning_rate": 3.903703703703704e-05,
1128
+ "loss": 1.055,
1129
+ "step": 1120
1130
+ },
1131
+ {
1132
+ "epoch": 0.9016,
1133
+ "grad_norm": 18.482967376708984,
1134
+ "learning_rate": 3.8933333333333336e-05,
1135
+ "loss": 1.2785,
1136
+ "step": 1127
1137
+ },
1138
+ {
1139
+ "epoch": 0.9072,
1140
+ "grad_norm": 11.217413902282715,
1141
+ "learning_rate": 3.882962962962963e-05,
1142
+ "loss": 1.0797,
1143
+ "step": 1134
1144
+ },
1145
+ {
1146
+ "epoch": 0.9128,
1147
+ "grad_norm": 18.9666690826416,
1148
+ "learning_rate": 3.8725925925925924e-05,
1149
+ "loss": 1.0066,
1150
+ "step": 1141
1151
+ },
1152
+ {
1153
+ "epoch": 0.9184,
1154
+ "grad_norm": 16.33734130859375,
1155
+ "learning_rate": 3.862222222222223e-05,
1156
+ "loss": 1.2794,
1157
+ "step": 1148
1158
+ },
1159
+ {
1160
+ "epoch": 0.924,
1161
+ "grad_norm": 10.510845184326172,
1162
+ "learning_rate": 3.851851851851852e-05,
1163
+ "loss": 0.9998,
1164
+ "step": 1155
1165
+ },
1166
+ {
1167
+ "epoch": 0.9296,
1168
+ "grad_norm": 16.016572952270508,
1169
+ "learning_rate": 3.841481481481482e-05,
1170
+ "loss": 0.967,
1171
+ "step": 1162
1172
+ },
1173
+ {
1174
+ "epoch": 0.9352,
1175
+ "grad_norm": 15.4122953414917,
1176
+ "learning_rate": 3.8311111111111115e-05,
1177
+ "loss": 0.867,
1178
+ "step": 1169
1179
+ },
1180
+ {
1181
+ "epoch": 0.9408,
1182
+ "grad_norm": 6.012162685394287,
1183
+ "learning_rate": 3.8207407407407407e-05,
1184
+ "loss": 1.125,
1185
+ "step": 1176
1186
+ },
1187
+ {
1188
+ "epoch": 0.9464,
1189
+ "grad_norm": 22.730403900146484,
1190
+ "learning_rate": 3.810370370370371e-05,
1191
+ "loss": 1.3726,
1192
+ "step": 1183
1193
+ },
1194
+ {
1195
+ "epoch": 0.952,
1196
+ "grad_norm": 9.818578720092773,
1197
+ "learning_rate": 3.8e-05,
1198
+ "loss": 1.1457,
1199
+ "step": 1190
1200
+ },
1201
+ {
1202
+ "epoch": 0.9576,
1203
+ "grad_norm": 22.369892120361328,
1204
+ "learning_rate": 3.789629629629629e-05,
1205
+ "loss": 1.1371,
1206
+ "step": 1197
1207
+ },
1208
+ {
1209
+ "epoch": 0.9632,
1210
+ "grad_norm": 17.22629737854004,
1211
+ "learning_rate": 3.77925925925926e-05,
1212
+ "loss": 1.039,
1213
+ "step": 1204
1214
+ },
1215
+ {
1216
+ "epoch": 0.9688,
1217
+ "grad_norm": 10.770013809204102,
1218
+ "learning_rate": 3.768888888888889e-05,
1219
+ "loss": 0.8811,
1220
+ "step": 1211
1221
+ },
1222
+ {
1223
+ "epoch": 0.9744,
1224
+ "grad_norm": 19.169052124023438,
1225
+ "learning_rate": 3.7585185185185186e-05,
1226
+ "loss": 1.282,
1227
+ "step": 1218
1228
+ },
1229
+ {
1230
+ "epoch": 0.98,
1231
+ "grad_norm": 9.597490310668945,
1232
+ "learning_rate": 3.7481481481481484e-05,
1233
+ "loss": 0.9561,
1234
+ "step": 1225
1235
+ },
1236
+ {
1237
+ "epoch": 0.9856,
1238
+ "grad_norm": 11.950583457946777,
1239
+ "learning_rate": 3.7377777777777775e-05,
1240
+ "loss": 1.1541,
1241
+ "step": 1232
1242
+ },
1243
+ {
1244
+ "epoch": 0.9912,
1245
+ "grad_norm": 13.237419128417969,
1246
+ "learning_rate": 3.727407407407408e-05,
1247
+ "loss": 1.0529,
1248
+ "step": 1239
1249
+ },
1250
+ {
1251
+ "epoch": 0.9968,
1252
+ "grad_norm": 23.540124893188477,
1253
+ "learning_rate": 3.717037037037037e-05,
1254
+ "loss": 1.2993,
1255
+ "step": 1246
1256
+ },
1257
+ {
1258
+ "epoch": 1.0,
1259
+ "eval_accuracy": 0.5233333333333333,
1260
+ "eval_f1_macro": 0.509019730829119,
1261
+ "eval_f1_micro": 0.5233333333333333,
1262
+ "eval_f1_weighted": 0.516464791683911,
1263
+ "eval_loss": 1.0463045835494995,
1264
+ "eval_precision_macro": 0.5322711463107999,
1265
+ "eval_precision_micro": 0.5233333333333333,
1266
+ "eval_precision_weighted": 0.5447555045714482,
1267
+ "eval_recall_macro": 0.5207324546507,
1268
+ "eval_recall_micro": 0.5233333333333333,
1269
+ "eval_recall_weighted": 0.5233333333333333,
1270
+ "eval_runtime": 8.5444,
1271
+ "eval_samples_per_second": 35.111,
1272
+ "eval_steps_per_second": 2.224,
1273
+ "step": 1250
1274
+ }
1275
+ ],
1276
+ "logging_steps": 7,
1277
+ "max_steps": 3750,
1278
+ "num_input_tokens_seen": 0,
1279
+ "num_train_epochs": 3,
1280
+ "save_steps": 500,
1281
+ "stateful_callbacks": {
1282
+ "EarlyStoppingCallback": {
1283
+ "args": {
1284
+ "early_stopping_patience": 5,
1285
+ "early_stopping_threshold": 0.01
1286
+ },
1287
+ "attributes": {
1288
+ "early_stopping_patience_counter": 0
1289
+ }
1290
+ },
1291
+ "TrainerControl": {
1292
+ "args": {
1293
+ "should_epoch_stop": false,
1294
+ "should_evaluate": false,
1295
+ "should_log": false,
1296
+ "should_save": true,
1297
+ "should_training_stop": false
1298
+ },
1299
+ "attributes": {}
1300
+ }
1301
+ },
1302
+ "total_flos": 821082232320000.0,
1303
+ "train_batch_size": 8,
1304
+ "trial_name": null,
1305
+ "trial_params": null
1306
+ }
checkpoint-1250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e5bbc0c904e355fa1698548e96cd8e3199960f68b14173c93d658bffc65f533
3
+ size 5432
config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "_num_labels": 5,
4
+ "architectures": [
5
+ "LongformerForSequenceClassification"
6
+ ],
7
+ "attention_mode": "longformer",
8
+ "attention_probs_dropout_prob": 0.1,
9
+ "attention_window": [
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "bos_token_id": 0,
24
+ "eos_token_id": 2,
25
+ "gradient_checkpointing": false,
26
+ "hidden_act": "gelu",
27
+ "hidden_dropout_prob": 0.1,
28
+ "hidden_size": 768,
29
+ "id2label": {
30
+ "0": "very negative",
31
+ "1": "negative",
32
+ "2": "neutral",
33
+ "3": "positive",
34
+ "4": "very positive"
35
+ },
36
+ "ignore_attention_mask": false,
37
+ "initializer_range": 0.02,
38
+ "intermediate_size": 3072,
39
+ "label2id": {
40
+ "negative": 1,
41
+ "neutral": 2,
42
+ "positive": 3,
43
+ "very negative": 0,
44
+ "very positive": 4
45
+ },
46
+ "layer_norm_eps": 1e-05,
47
+ "max_position_embeddings": 4098,
48
+ "model_type": "longformer",
49
+ "num_attention_heads": 12,
50
+ "num_hidden_layers": 12,
51
+ "onnx_export": false,
52
+ "pad_token_id": 1,
53
+ "problem_type": "single_label_classification",
54
+ "sep_token_id": 2,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.48.0",
57
+ "type_vocab_size": 1,
58
+ "vocab_size": 50265
59
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ea5a2958a9736d967d763abc8f3d7f4a1bebfee2308ff3232a133fdae09f851
3
+ size 594687412
runs/Mar19_20-59-07_r-spacesedan-longformer-sentiment-analysis-training-l-c848c-35h/events.out.tfevents.1742417949.r-spacesedan-longformer-sentiment-analysis-training-l-c848c-35h.71.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa6d2b087c036f25451e4d92a84255e66156ddc132ad701843f345de7a0a7355
3
- size 109073
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25b457dee90156719b88255eb9dc54ca2c2294d6880f6b10561201aa6f130c3c
3
+ size 121232
runs/Mar19_20-59-07_r-spacesedan-longformer-sentiment-analysis-training-l-c848c-35h/events.out.tfevents.1742421321.r-spacesedan-longformer-sentiment-analysis-training-l-c848c-35h.71.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:524cd65ce443699f2dcc729855d349e4c6b38a65565cb9462721666a8357fdfd
3
+ size 921
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "<mask>",
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "<pad>",
54
+ "sep_token": "</s>",
55
+ "tokenizer_class": "LongformerTokenizer",
56
+ "trim_offsets": true,
57
+ "unk_token": "<unk>"
58
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e5bbc0c904e355fa1698548e96cd8e3199960f68b14173c93d658bffc65f533
3
+ size 5432
training_params.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "data_path": "spacesedan/Amazon_Reviews_Sentiment_10K",
3
+ "model": "allenai/longformer-base-4096",
4
+ "lr": 5e-05,
5
+ "epochs": 3,
6
+ "max_seq_length": 128,
7
+ "batch_size": 8,
8
+ "warmup_ratio": 0.1,
9
+ "gradient_accumulation": 1,
10
+ "optimizer": "adamw_torch",
11
+ "scheduler": "linear",
12
+ "weight_decay": 0.0,
13
+ "max_grad_norm": 1.0,
14
+ "seed": 42,
15
+ "train_split": "train",
16
+ "valid_split": "test",
17
+ "text_column": "text",
18
+ "target_column": "label",
19
+ "logging_steps": -1,
20
+ "project_name": "sentiment-analysis-longformer",
21
+ "auto_find_batch_size": false,
22
+ "mixed_precision": "fp16",
23
+ "save_total_limit": 1,
24
+ "push_to_hub": true,
25
+ "eval_strategy": "epoch",
26
+ "username": "spacesedan",
27
+ "log": "tensorboard",
28
+ "early_stopping_patience": 5,
29
+ "early_stopping_threshold": 0.01
30
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff