Hugging Face's logo

mhokchuekchuek
/
summarization_ja

TensorBoard
Model card Files Metrics Community
mhokchuekchuek commited on
Commit
96a42f2
·
1 Parent(s): 0ee618d

Upload 55 files

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +5 -0
  2. mt5-new-summarize-final/checkpoint-1000/config.json +35 -0
  3. mt5-new-summarize-final/checkpoint-1000/generation_config.json +11 -0
  4. mt5-new-summarize-final/checkpoint-1000/optimizer.pt +3 -0
  5. mt5-new-summarize-final/checkpoint-1000/rng_state.pth +3 -0
  6. mt5-new-summarize-final/checkpoint-1000/scheduler.pt +3 -0
  7. mt5-new-summarize-final/checkpoint-1000/special_tokens_map.json +5 -0
  8. mt5-new-summarize-final/checkpoint-1000/spiece.model +3 -0
  9. mt5-new-summarize-final/checkpoint-1000/tokenizer.json +3 -0
  10. mt5-new-summarize-final/checkpoint-1000/tokenizer_config.json +12 -0
  11. mt5-new-summarize-final/checkpoint-1000/trainer_state.json +628 -0
  12. mt5-new-summarize-final/checkpoint-1000/training_args.bin +3 -0
  13. mt5-new-summarize-final/checkpoint-1500/config.json +35 -0
  14. mt5-new-summarize-final/checkpoint-1500/generation_config.json +11 -0
  15. mt5-new-summarize-final/checkpoint-1500/optimizer.pt +3 -0
  16. mt5-new-summarize-final/checkpoint-1500/rng_state.pth +3 -0
  17. mt5-new-summarize-final/checkpoint-1500/scheduler.pt +3 -0
  18. mt5-new-summarize-final/checkpoint-1500/special_tokens_map.json +5 -0
  19. mt5-new-summarize-final/checkpoint-1500/spiece.model +3 -0
  20. mt5-new-summarize-final/checkpoint-1500/tokenizer.json +3 -0
  21. mt5-new-summarize-final/checkpoint-1500/tokenizer_config.json +12 -0
  22. mt5-new-summarize-final/checkpoint-1500/trainer_state.json +928 -0
  23. mt5-new-summarize-final/checkpoint-1500/training_args.bin +3 -0
  24. mt5-new-summarize-final/checkpoint-2000/config.json +35 -0
  25. mt5-new-summarize-final/checkpoint-2000/generation_config.json +11 -0
  26. mt5-new-summarize-final/checkpoint-2000/optimizer.pt +3 -0
  27. mt5-new-summarize-final/checkpoint-2000/rng_state.pth +3 -0
  28. mt5-new-summarize-final/checkpoint-2000/scheduler.pt +3 -0
  29. mt5-new-summarize-final/checkpoint-2000/special_tokens_map.json +5 -0
  30. mt5-new-summarize-final/checkpoint-2000/spiece.model +3 -0
  31. mt5-new-summarize-final/checkpoint-2000/tokenizer.json +3 -0
  32. mt5-new-summarize-final/checkpoint-2000/tokenizer_config.json +12 -0
  33. mt5-new-summarize-final/checkpoint-2000/trainer_state.json +1240 -0
  34. mt5-new-summarize-final/checkpoint-2000/training_args.bin +3 -0
  35. mt5-new-summarize-final/checkpoint-500/config.json +35 -0
  36. mt5-new-summarize-final/checkpoint-500/generation_config.json +7 -0
  37. mt5-new-summarize-final/checkpoint-500/optimizer.pt +3 -0
  38. mt5-new-summarize-final/checkpoint-500/rng_state.pth +3 -0
  39. mt5-new-summarize-final/checkpoint-500/scheduler.pt +3 -0
  40. mt5-new-summarize-final/checkpoint-500/special_tokens_map.json +5 -0
  41. mt5-new-summarize-final/checkpoint-500/spiece.model +3 -0
  42. mt5-new-summarize-final/checkpoint-500/tokenizer.json +3 -0
  43. mt5-new-summarize-final/checkpoint-500/tokenizer_config.json +12 -0
  44. mt5-new-summarize-final/checkpoint-500/trainer_state.json +316 -0
  45. mt5-new-summarize-final/checkpoint-500/training_args.bin +3 -0
  46. mt5-new-summarize-final/config.json +35 -0
  47. mt5-new-summarize-final/generation_config.json +11 -0
  48. mt5-new-summarize-final/pytorch_model.bin +3 -0
  49. mt5-new-summarize-final/runs/Jan29_13-58-29_d0795309bd3b/1675000709.4761653/events.out.tfevents.1675000709.d0795309bd3b.4670.1 +3 -0
  50. mt5-new-summarize-final/runs/Jan29_13-58-29_d0795309bd3b/events.out.tfevents.1675000709.d0795309bd3b.4670.0 +3 -0
.gitattributes CHANGED
@@ -32,3 +32,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ mt5-new-summarize-final/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
36
+ mt5-new-summarize-final/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ mt5-new-summarize-final/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ mt5-new-summarize-final/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ mt5-new-summarize-final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
mt5-new-summarize-final/checkpoint-1000/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mt5-small",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "length_penalty": 0.6,
19
+ "max_length": 128,
20
+ "model_type": "mt5",
21
+ "no_repeat_ngram_size": 2,
22
+ "num_beams": 15,
23
+ "num_decoder_layers": 8,
24
+ "num_heads": 6,
25
+ "num_layers": 8,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "tie_word_embeddings": false,
30
+ "tokenizer_class": "T5Tokenizer",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.26.0",
33
+ "use_cache": true,
34
+ "vocab_size": 250112
35
+ }
mt5-new-summarize-final/checkpoint-1000/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "length_penalty": 0.6,
6
+ "max_length": 128,
7
+ "no_repeat_ngram_size": 2,
8
+ "num_beams": 15,
9
+ "pad_token_id": 0,
10
+ "transformers_version": "4.26.0"
11
+ }
mt5-new-summarize-final/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3baa2f0acf6564b212a6bede34f1f382b1d244d1399dc6e36525e87f19aff8
3
+ size 2879429
mt5-new-summarize-final/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93b72fbffc25dd7aa7d70e3be24dbad99632d4072a14d2ccf2e19ebfaa6ee482
3
+ size 14575
mt5-new-summarize-final/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eb7cf2c3437ec60e3b841c26e0417054892b0d02ce922430502dd61a0091fd1
3
+ size 627
mt5-new-summarize-final/checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
mt5-new-summarize-final/checkpoint-1000/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
mt5-new-summarize-final/checkpoint-1000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93c3578052e1605d8332eb961bc08d72e246071974e4cc54aa6991826b802aa5
3
+ size 16330369
mt5-new-summarize-final/checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": null,
3
+ "eos_token": "</s>",
4
+ "extra_ids": 0,
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "name_or_path": "google/mt5-small",
7
+ "pad_token": "<pad>",
8
+ "sp_model_kwargs": {},
9
+ "special_tokens_map_file": "/root/.cache/huggingface/hub/models--google--mt5-small/snapshots/38f23af8ec210eb6c376d40e9c56bd25a80f195d/special_tokens_map.json",
10
+ "tokenizer_class": "T5Tokenizer",
11
+ "unk_token": "<unk>"
12
+ }
mt5-new-summarize-final/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,628 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.426928516156576,
5
+ "global_step": 1000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.555555555555555e-05,
13
+ "loss": 20.8315,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.0001111111111111111,
19
+ "loss": 16.4875,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 0.00016666666666666666,
25
+ "loss": 10.6503,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 0.0002222222222222222,
31
+ "loss": 6.6236,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 0.0002777777777777778,
37
+ "loss": 5.2548,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 0.0003333333333333333,
43
+ "loss": 4.6801,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "learning_rate": 0.0003888888888888889,
49
+ "loss": 4.3374,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.03,
54
+ "learning_rate": 0.0004444444444444444,
55
+ "loss": 3.9614,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.04,
60
+ "learning_rate": 0.0005,
61
+ "loss": 3.9704,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.04,
66
+ "learning_rate": 0.0004977797513321492,
67
+ "loss": 3.8112,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.05,
72
+ "learning_rate": 0.0004955595026642984,
73
+ "loss": 3.6272,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.05,
78
+ "learning_rate": 0.0004933392539964477,
79
+ "loss": 3.5572,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "learning_rate": 0.0004911190053285969,
85
+ "loss": 3.3966,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "learning_rate": 0.000488898756660746,
91
+ "loss": 3.4632,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.06,
96
+ "learning_rate": 0.0004866785079928952,
97
+ "loss": 3.4591,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.07,
102
+ "learning_rate": 0.0004844582593250444,
103
+ "loss": 3.2218,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.07,
108
+ "learning_rate": 0.0004822380106571936,
109
+ "loss": 3.4101,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.08,
114
+ "learning_rate": 0.00048001776198934283,
115
+ "loss": 3.2787,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.08,
120
+ "learning_rate": 0.000477797513321492,
121
+ "loss": 3.1043,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.09,
126
+ "learning_rate": 0.0004755772646536412,
127
+ "loss": 3.2435,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "learning_rate": 0.00047335701598579037,
133
+ "loss": 3.2286,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.09,
138
+ "learning_rate": 0.00047113676731793964,
139
+ "loss": 3.1484,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.1,
144
+ "learning_rate": 0.00046891651865008885,
145
+ "loss": 3.1817,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.1,
150
+ "learning_rate": 0.000466696269982238,
151
+ "loss": 3.1608,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.11,
156
+ "learning_rate": 0.00046447602131438723,
157
+ "loss": 3.1457,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.11,
162
+ "learning_rate": 0.0004622557726465364,
163
+ "loss": 3.1212,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.12,
168
+ "learning_rate": 0.00046003552397868566,
169
+ "loss": 2.9937,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "learning_rate": 0.0004578152753108348,
175
+ "loss": 3.2031,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.12,
180
+ "learning_rate": 0.00045559502664298403,
181
+ "loss": 3.0713,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.13,
186
+ "learning_rate": 0.0004533747779751332,
187
+ "loss": 3.0352,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.13,
192
+ "learning_rate": 0.0004511545293072824,
193
+ "loss": 3.0872,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.14,
198
+ "learning_rate": 0.0004489342806394316,
199
+ "loss": 3.0049,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.14,
204
+ "learning_rate": 0.00044671403197158084,
205
+ "loss": 3.0659,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.15,
210
+ "learning_rate": 0.00044449378330373,
211
+ "loss": 3.0852,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "learning_rate": 0.0004422735346358792,
217
+ "loss": 2.9506,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.15,
222
+ "learning_rate": 0.00044005328596802843,
223
+ "loss": 3.027,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.16,
228
+ "learning_rate": 0.00043783303730017764,
229
+ "loss": 2.8812,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.16,
234
+ "learning_rate": 0.00043561278863232686,
235
+ "loss": 2.9738,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.17,
240
+ "learning_rate": 0.000433392539964476,
241
+ "loss": 3.0504,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.17,
246
+ "learning_rate": 0.00043117229129662523,
247
+ "loss": 3.0195,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.18,
252
+ "learning_rate": 0.0004289520426287744,
253
+ "loss": 2.968,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "learning_rate": 0.00042673179396092366,
259
+ "loss": 3.0864,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.18,
264
+ "learning_rate": 0.0004245115452930728,
265
+ "loss": 2.8735,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.19,
270
+ "learning_rate": 0.00042229129662522204,
271
+ "loss": 3.0391,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.19,
276
+ "learning_rate": 0.0004200710479573712,
277
+ "loss": 2.8288,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.2,
282
+ "learning_rate": 0.0004178507992895204,
283
+ "loss": 3.0182,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.2,
288
+ "learning_rate": 0.0004156305506216697,
289
+ "loss": 3.0606,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.2,
294
+ "learning_rate": 0.00041341030195381884,
295
+ "loss": 2.9467,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "learning_rate": 0.00041119005328596806,
301
+ "loss": 2.7842,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.21,
306
+ "learning_rate": 0.0004089698046181172,
307
+ "loss": 2.8472,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.22,
312
+ "learning_rate": 0.00040674955595026643,
313
+ "loss": 2.7883,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.22,
318
+ "learning_rate": 0.00040452930728241565,
319
+ "loss": 2.8592,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.23,
324
+ "learning_rate": 0.00040230905861456486,
325
+ "loss": 2.9186,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.23,
330
+ "learning_rate": 0.000400088809946714,
331
+ "loss": 2.911,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.23,
336
+ "learning_rate": 0.00039786856127886324,
337
+ "loss": 2.8355,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.24,
342
+ "learning_rate": 0.0003956483126110124,
343
+ "loss": 2.9385,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.24,
348
+ "learning_rate": 0.00039342806394316167,
349
+ "loss": 2.8465,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.25,
354
+ "learning_rate": 0.0003912078152753108,
355
+ "loss": 2.8408,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.25,
360
+ "learning_rate": 0.00038898756660746004,
361
+ "loss": 2.767,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.26,
366
+ "learning_rate": 0.00038676731793960926,
367
+ "loss": 2.9691,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.26,
372
+ "learning_rate": 0.0003845470692717584,
373
+ "loss": 2.7348,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.26,
378
+ "learning_rate": 0.0003823268206039077,
379
+ "loss": 2.7477,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.27,
384
+ "learning_rate": 0.00038010657193605685,
385
+ "loss": 2.8142,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.27,
390
+ "learning_rate": 0.00037788632326820606,
391
+ "loss": 2.7522,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.28,
396
+ "learning_rate": 0.0003756660746003552,
397
+ "loss": 2.9403,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.28,
402
+ "learning_rate": 0.00037344582593250444,
403
+ "loss": 2.8956,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.29,
408
+ "learning_rate": 0.00037122557726465365,
409
+ "loss": 2.7331,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.29,
414
+ "learning_rate": 0.00036900532859680287,
415
+ "loss": 2.8265,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.29,
420
+ "learning_rate": 0.000366785079928952,
421
+ "loss": 2.8919,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.3,
426
+ "learning_rate": 0.00036456483126110124,
427
+ "loss": 2.8362,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.3,
432
+ "learning_rate": 0.0003623445825932504,
433
+ "loss": 2.7282,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.31,
438
+ "learning_rate": 0.00036012433392539967,
439
+ "loss": 2.7628,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.31,
444
+ "learning_rate": 0.0003579040852575489,
445
+ "loss": 2.8508,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.32,
450
+ "learning_rate": 0.00035568383658969805,
451
+ "loss": 2.8603,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.32,
456
+ "learning_rate": 0.00035346358792184726,
457
+ "loss": 2.6765,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.32,
462
+ "learning_rate": 0.0003512433392539964,
463
+ "loss": 2.8767,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.33,
468
+ "learning_rate": 0.0003490230905861457,
469
+ "loss": 2.8446,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.33,
474
+ "learning_rate": 0.00034680284191829485,
475
+ "loss": 2.8524,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.34,
480
+ "learning_rate": 0.00034458259325044407,
481
+ "loss": 2.778,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.34,
486
+ "learning_rate": 0.0003423623445825932,
487
+ "loss": 2.7359,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.35,
492
+ "learning_rate": 0.00034014209591474244,
493
+ "loss": 2.7803,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.35,
498
+ "learning_rate": 0.00033792184724689166,
499
+ "loss": 2.7631,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.35,
504
+ "learning_rate": 0.00033570159857904087,
505
+ "loss": 2.7818,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.36,
510
+ "learning_rate": 0.00033348134991119003,
511
+ "loss": 2.6591,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.36,
516
+ "learning_rate": 0.00033126110124333925,
517
+ "loss": 2.6843,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.37,
522
+ "learning_rate": 0.00032904085257548846,
523
+ "loss": 2.7299,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.37,
528
+ "learning_rate": 0.0003268206039076377,
529
+ "loss": 2.6928,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.38,
534
+ "learning_rate": 0.0003246003552397869,
535
+ "loss": 2.8512,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.38,
540
+ "learning_rate": 0.00032238010657193605,
541
+ "loss": 2.7944,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.38,
546
+ "learning_rate": 0.00032015985790408526,
547
+ "loss": 2.7365,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.39,
552
+ "learning_rate": 0.0003179396092362344,
553
+ "loss": 2.6992,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.39,
558
+ "learning_rate": 0.0003157193605683837,
559
+ "loss": 2.743,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.4,
564
+ "learning_rate": 0.00031349911190053285,
565
+ "loss": 2.8022,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.4,
570
+ "learning_rate": 0.00031127886323268207,
571
+ "loss": 2.6603,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.41,
576
+ "learning_rate": 0.00030905861456483123,
577
+ "loss": 2.6277,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.41,
582
+ "learning_rate": 0.00030683836589698044,
583
+ "loss": 2.7652,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.41,
588
+ "learning_rate": 0.0003046181172291297,
589
+ "loss": 2.7063,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.42,
594
+ "learning_rate": 0.0003023978685612789,
595
+ "loss": 2.6602,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.42,
600
+ "learning_rate": 0.0003001776198934281,
601
+ "loss": 2.7592,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.43,
606
+ "learning_rate": 0.00029795737122557725,
607
+ "loss": 2.6628,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 0.43,
612
+ "eval_loss": 0.51519775390625,
613
+ "eval_rouge1": 0.8067071482496617,
614
+ "eval_rouge2": 0.7472387953801375,
615
+ "eval_rougeL": 0.7071393649432568,
616
+ "eval_rougeLsum": 0.8209607917685595,
617
+ "eval_runtime": 1807.8081,
618
+ "eval_samples_per_second": 0.553,
619
+ "eval_steps_per_second": 0.553,
620
+ "step": 1000
621
+ }
622
+ ],
623
+ "max_steps": 2342,
624
+ "num_train_epochs": 1,
625
+ "total_flos": 2.369574390079488e+16,
626
+ "trial_name": null,
627
+ "trial_params": null
628
+ }
mt5-new-summarize-final/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:233a568dcedd606fd72b432d2c6cc678ea8c5e086885a5642ed616a886f52eb9
3
+ size 3643
mt5-new-summarize-final/checkpoint-1500/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mt5-small",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "length_penalty": 0.6,
19
+ "max_length": 128,
20
+ "model_type": "mt5",
21
+ "no_repeat_ngram_size": 2,
22
+ "num_beams": 15,
23
+ "num_decoder_layers": 8,
24
+ "num_heads": 6,
25
+ "num_layers": 8,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "tie_word_embeddings": false,
30
+ "tokenizer_class": "T5Tokenizer",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.26.0",
33
+ "use_cache": true,
34
+ "vocab_size": 250112
35
+ }
mt5-new-summarize-final/checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "length_penalty": 0.6,
6
+ "max_length": 128,
7
+ "no_repeat_ngram_size": 2,
8
+ "num_beams": 15,
9
+ "pad_token_id": 0,
10
+ "transformers_version": "4.26.0"
11
+ }
mt5-new-summarize-final/checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32e68e0cee8d46a585d3814147b91f8b4bc012059656c68afb9a2a293be699d9
3
+ size 2879429
mt5-new-summarize-final/checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7e8b74df13be81372e8f28a3be8d00e9fa6155ca3f835bba73863ea1acbfd61
3
+ size 14575
mt5-new-summarize-final/checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07335cb72d3786e62311ca2b940c12fc5ac51a973658311a4a7d90cbf94853af
3
+ size 627
mt5-new-summarize-final/checkpoint-1500/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
mt5-new-summarize-final/checkpoint-1500/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
mt5-new-summarize-final/checkpoint-1500/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93c3578052e1605d8332eb961bc08d72e246071974e4cc54aa6991826b802aa5
3
+ size 16330369
mt5-new-summarize-final/checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": null,
3
+ "eos_token": "</s>",
4
+ "extra_ids": 0,
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "name_or_path": "google/mt5-small",
7
+ "pad_token": "<pad>",
8
+ "sp_model_kwargs": {},
9
+ "special_tokens_map_file": "/root/.cache/huggingface/hub/models--google--mt5-small/snapshots/38f23af8ec210eb6c376d40e9c56bd25a80f195d/special_tokens_map.json",
10
+ "tokenizer_class": "T5Tokenizer",
11
+ "unk_token": "<unk>"
12
+ }
mt5-new-summarize-final/checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,928 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.6403927742348641,
5
+ "global_step": 1500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.555555555555555e-05,
13
+ "loss": 20.8315,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.0001111111111111111,
19
+ "loss": 16.4875,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 0.00016666666666666666,
25
+ "loss": 10.6503,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 0.0002222222222222222,
31
+ "loss": 6.6236,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 0.0002777777777777778,
37
+ "loss": 5.2548,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 0.0003333333333333333,
43
+ "loss": 4.6801,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "learning_rate": 0.0003888888888888889,
49
+ "loss": 4.3374,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.03,
54
+ "learning_rate": 0.0004444444444444444,
55
+ "loss": 3.9614,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.04,
60
+ "learning_rate": 0.0005,
61
+ "loss": 3.9704,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.04,
66
+ "learning_rate": 0.0004977797513321492,
67
+ "loss": 3.8112,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.05,
72
+ "learning_rate": 0.0004955595026642984,
73
+ "loss": 3.6272,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.05,
78
+ "learning_rate": 0.0004933392539964477,
79
+ "loss": 3.5572,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "learning_rate": 0.0004911190053285969,
85
+ "loss": 3.3966,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "learning_rate": 0.000488898756660746,
91
+ "loss": 3.4632,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.06,
96
+ "learning_rate": 0.0004866785079928952,
97
+ "loss": 3.4591,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.07,
102
+ "learning_rate": 0.0004844582593250444,
103
+ "loss": 3.2218,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.07,
108
+ "learning_rate": 0.0004822380106571936,
109
+ "loss": 3.4101,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.08,
114
+ "learning_rate": 0.00048001776198934283,
115
+ "loss": 3.2787,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.08,
120
+ "learning_rate": 0.000477797513321492,
121
+ "loss": 3.1043,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.09,
126
+ "learning_rate": 0.0004755772646536412,
127
+ "loss": 3.2435,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "learning_rate": 0.00047335701598579037,
133
+ "loss": 3.2286,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.09,
138
+ "learning_rate": 0.00047113676731793964,
139
+ "loss": 3.1484,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.1,
144
+ "learning_rate": 0.00046891651865008885,
145
+ "loss": 3.1817,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.1,
150
+ "learning_rate": 0.000466696269982238,
151
+ "loss": 3.1608,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.11,
156
+ "learning_rate": 0.00046447602131438723,
157
+ "loss": 3.1457,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.11,
162
+ "learning_rate": 0.0004622557726465364,
163
+ "loss": 3.1212,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.12,
168
+ "learning_rate": 0.00046003552397868566,
169
+ "loss": 2.9937,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "learning_rate": 0.0004578152753108348,
175
+ "loss": 3.2031,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.12,
180
+ "learning_rate": 0.00045559502664298403,
181
+ "loss": 3.0713,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.13,
186
+ "learning_rate": 0.0004533747779751332,
187
+ "loss": 3.0352,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.13,
192
+ "learning_rate": 0.0004511545293072824,
193
+ "loss": 3.0872,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.14,
198
+ "learning_rate": 0.0004489342806394316,
199
+ "loss": 3.0049,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.14,
204
+ "learning_rate": 0.00044671403197158084,
205
+ "loss": 3.0659,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.15,
210
+ "learning_rate": 0.00044449378330373,
211
+ "loss": 3.0852,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "learning_rate": 0.0004422735346358792,
217
+ "loss": 2.9506,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.15,
222
+ "learning_rate": 0.00044005328596802843,
223
+ "loss": 3.027,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.16,
228
+ "learning_rate": 0.00043783303730017764,
229
+ "loss": 2.8812,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.16,
234
+ "learning_rate": 0.00043561278863232686,
235
+ "loss": 2.9738,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.17,
240
+ "learning_rate": 0.000433392539964476,
241
+ "loss": 3.0504,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.17,
246
+ "learning_rate": 0.00043117229129662523,
247
+ "loss": 3.0195,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.18,
252
+ "learning_rate": 0.0004289520426287744,
253
+ "loss": 2.968,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "learning_rate": 0.00042673179396092366,
259
+ "loss": 3.0864,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.18,
264
+ "learning_rate": 0.0004245115452930728,
265
+ "loss": 2.8735,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.19,
270
+ "learning_rate": 0.00042229129662522204,
271
+ "loss": 3.0391,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.19,
276
+ "learning_rate": 0.0004200710479573712,
277
+ "loss": 2.8288,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.2,
282
+ "learning_rate": 0.0004178507992895204,
283
+ "loss": 3.0182,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.2,
288
+ "learning_rate": 0.0004156305506216697,
289
+ "loss": 3.0606,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.2,
294
+ "learning_rate": 0.00041341030195381884,
295
+ "loss": 2.9467,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "learning_rate": 0.00041119005328596806,
301
+ "loss": 2.7842,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.21,
306
+ "learning_rate": 0.0004089698046181172,
307
+ "loss": 2.8472,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.22,
312
+ "learning_rate": 0.00040674955595026643,
313
+ "loss": 2.7883,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.22,
318
+ "learning_rate": 0.00040452930728241565,
319
+ "loss": 2.8592,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.23,
324
+ "learning_rate": 0.00040230905861456486,
325
+ "loss": 2.9186,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.23,
330
+ "learning_rate": 0.000400088809946714,
331
+ "loss": 2.911,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.23,
336
+ "learning_rate": 0.00039786856127886324,
337
+ "loss": 2.8355,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.24,
342
+ "learning_rate": 0.0003956483126110124,
343
+ "loss": 2.9385,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.24,
348
+ "learning_rate": 0.00039342806394316167,
349
+ "loss": 2.8465,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.25,
354
+ "learning_rate": 0.0003912078152753108,
355
+ "loss": 2.8408,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.25,
360
+ "learning_rate": 0.00038898756660746004,
361
+ "loss": 2.767,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.26,
366
+ "learning_rate": 0.00038676731793960926,
367
+ "loss": 2.9691,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.26,
372
+ "learning_rate": 0.0003845470692717584,
373
+ "loss": 2.7348,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.26,
378
+ "learning_rate": 0.0003823268206039077,
379
+ "loss": 2.7477,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.27,
384
+ "learning_rate": 0.00038010657193605685,
385
+ "loss": 2.8142,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.27,
390
+ "learning_rate": 0.00037788632326820606,
391
+ "loss": 2.7522,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.28,
396
+ "learning_rate": 0.0003756660746003552,
397
+ "loss": 2.9403,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.28,
402
+ "learning_rate": 0.00037344582593250444,
403
+ "loss": 2.8956,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.29,
408
+ "learning_rate": 0.00037122557726465365,
409
+ "loss": 2.7331,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.29,
414
+ "learning_rate": 0.00036900532859680287,
415
+ "loss": 2.8265,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.29,
420
+ "learning_rate": 0.000366785079928952,
421
+ "loss": 2.8919,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.3,
426
+ "learning_rate": 0.00036456483126110124,
427
+ "loss": 2.8362,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.3,
432
+ "learning_rate": 0.0003623445825932504,
433
+ "loss": 2.7282,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.31,
438
+ "learning_rate": 0.00036012433392539967,
439
+ "loss": 2.7628,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.31,
444
+ "learning_rate": 0.0003579040852575489,
445
+ "loss": 2.8508,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.32,
450
+ "learning_rate": 0.00035568383658969805,
451
+ "loss": 2.8603,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.32,
456
+ "learning_rate": 0.00035346358792184726,
457
+ "loss": 2.6765,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.32,
462
+ "learning_rate": 0.0003512433392539964,
463
+ "loss": 2.8767,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.33,
468
+ "learning_rate": 0.0003490230905861457,
469
+ "loss": 2.8446,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.33,
474
+ "learning_rate": 0.00034680284191829485,
475
+ "loss": 2.8524,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.34,
480
+ "learning_rate": 0.00034458259325044407,
481
+ "loss": 2.778,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.34,
486
+ "learning_rate": 0.0003423623445825932,
487
+ "loss": 2.7359,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.35,
492
+ "learning_rate": 0.00034014209591474244,
493
+ "loss": 2.7803,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.35,
498
+ "learning_rate": 0.00033792184724689166,
499
+ "loss": 2.7631,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.35,
504
+ "learning_rate": 0.00033570159857904087,
505
+ "loss": 2.7818,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.36,
510
+ "learning_rate": 0.00033348134991119003,
511
+ "loss": 2.6591,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.36,
516
+ "learning_rate": 0.00033126110124333925,
517
+ "loss": 2.6843,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.37,
522
+ "learning_rate": 0.00032904085257548846,
523
+ "loss": 2.7299,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.37,
528
+ "learning_rate": 0.0003268206039076377,
529
+ "loss": 2.6928,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.38,
534
+ "learning_rate": 0.0003246003552397869,
535
+ "loss": 2.8512,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.38,
540
+ "learning_rate": 0.00032238010657193605,
541
+ "loss": 2.7944,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.38,
546
+ "learning_rate": 0.00032015985790408526,
547
+ "loss": 2.7365,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.39,
552
+ "learning_rate": 0.0003179396092362344,
553
+ "loss": 2.6992,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.39,
558
+ "learning_rate": 0.0003157193605683837,
559
+ "loss": 2.743,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.4,
564
+ "learning_rate": 0.00031349911190053285,
565
+ "loss": 2.8022,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.4,
570
+ "learning_rate": 0.00031127886323268207,
571
+ "loss": 2.6603,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.41,
576
+ "learning_rate": 0.00030905861456483123,
577
+ "loss": 2.6277,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.41,
582
+ "learning_rate": 0.00030683836589698044,
583
+ "loss": 2.7652,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.41,
588
+ "learning_rate": 0.0003046181172291297,
589
+ "loss": 2.7063,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.42,
594
+ "learning_rate": 0.0003023978685612789,
595
+ "loss": 2.6602,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.42,
600
+ "learning_rate": 0.0003001776198934281,
601
+ "loss": 2.7592,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.43,
606
+ "learning_rate": 0.00029795737122557725,
607
+ "loss": 2.6628,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 0.43,
612
+ "eval_loss": 0.51519775390625,
613
+ "eval_rouge1": 0.8067071482496617,
614
+ "eval_rouge2": 0.7472387953801375,
615
+ "eval_rougeL": 0.7071393649432568,
616
+ "eval_rougeLsum": 0.8209607917685595,
617
+ "eval_runtime": 1807.8081,
618
+ "eval_samples_per_second": 0.553,
619
+ "eval_steps_per_second": 0.553,
620
+ "step": 1000
621
+ },
622
+ {
623
+ "epoch": 0.43,
624
+ "learning_rate": 0.00029573712255772646,
625
+ "loss": 2.6164,
626
+ "step": 1010
627
+ },
628
+ {
629
+ "epoch": 0.44,
630
+ "learning_rate": 0.0002935168738898757,
631
+ "loss": 2.6947,
632
+ "step": 1020
633
+ },
634
+ {
635
+ "epoch": 0.44,
636
+ "learning_rate": 0.0002912966252220249,
637
+ "loss": 2.4547,
638
+ "step": 1030
639
+ },
640
+ {
641
+ "epoch": 0.44,
642
+ "learning_rate": 0.00028907637655417405,
643
+ "loss": 2.7347,
644
+ "step": 1040
645
+ },
646
+ {
647
+ "epoch": 0.45,
648
+ "learning_rate": 0.00028685612788632327,
649
+ "loss": 2.6491,
650
+ "step": 1050
651
+ },
652
+ {
653
+ "epoch": 0.45,
654
+ "learning_rate": 0.00028463587921847243,
655
+ "loss": 2.6778,
656
+ "step": 1060
657
+ },
658
+ {
659
+ "epoch": 0.46,
660
+ "learning_rate": 0.0002824156305506217,
661
+ "loss": 2.7157,
662
+ "step": 1070
663
+ },
664
+ {
665
+ "epoch": 0.46,
666
+ "learning_rate": 0.00028019538188277086,
667
+ "loss": 2.6445,
668
+ "step": 1080
669
+ },
670
+ {
671
+ "epoch": 0.47,
672
+ "learning_rate": 0.0002779751332149201,
673
+ "loss": 2.7732,
674
+ "step": 1090
675
+ },
676
+ {
677
+ "epoch": 0.47,
678
+ "learning_rate": 0.0002757548845470693,
679
+ "loss": 2.6777,
680
+ "step": 1100
681
+ },
682
+ {
683
+ "epoch": 0.47,
684
+ "learning_rate": 0.00027353463587921845,
685
+ "loss": 2.6231,
686
+ "step": 1110
687
+ },
688
+ {
689
+ "epoch": 0.48,
690
+ "learning_rate": 0.0002713143872113677,
691
+ "loss": 2.6003,
692
+ "step": 1120
693
+ },
694
+ {
695
+ "epoch": 0.48,
696
+ "learning_rate": 0.0002690941385435169,
697
+ "loss": 2.7226,
698
+ "step": 1130
699
+ },
700
+ {
701
+ "epoch": 0.49,
702
+ "learning_rate": 0.0002668738898756661,
703
+ "loss": 2.8623,
704
+ "step": 1140
705
+ },
706
+ {
707
+ "epoch": 0.49,
708
+ "learning_rate": 0.00026465364120781525,
709
+ "loss": 2.7391,
710
+ "step": 1150
711
+ },
712
+ {
713
+ "epoch": 0.5,
714
+ "learning_rate": 0.00026243339253996447,
715
+ "loss": 2.6836,
716
+ "step": 1160
717
+ },
718
+ {
719
+ "epoch": 0.5,
720
+ "learning_rate": 0.0002602131438721137,
721
+ "loss": 2.6357,
722
+ "step": 1170
723
+ },
724
+ {
725
+ "epoch": 0.5,
726
+ "learning_rate": 0.0002579928952042629,
727
+ "loss": 2.5891,
728
+ "step": 1180
729
+ },
730
+ {
731
+ "epoch": 0.51,
732
+ "learning_rate": 0.00025577264653641206,
733
+ "loss": 2.7383,
734
+ "step": 1190
735
+ },
736
+ {
737
+ "epoch": 0.51,
738
+ "learning_rate": 0.00025355239786856127,
739
+ "loss": 2.738,
740
+ "step": 1200
741
+ },
742
+ {
743
+ "epoch": 0.52,
744
+ "learning_rate": 0.00025133214920071043,
745
+ "loss": 2.62,
746
+ "step": 1210
747
+ },
748
+ {
749
+ "epoch": 0.52,
750
+ "learning_rate": 0.0002491119005328597,
751
+ "loss": 2.5796,
752
+ "step": 1220
753
+ },
754
+ {
755
+ "epoch": 0.53,
756
+ "learning_rate": 0.0002468916518650089,
757
+ "loss": 2.5744,
758
+ "step": 1230
759
+ },
760
+ {
761
+ "epoch": 0.53,
762
+ "learning_rate": 0.0002446714031971581,
763
+ "loss": 2.5047,
764
+ "step": 1240
765
+ },
766
+ {
767
+ "epoch": 0.53,
768
+ "learning_rate": 0.0002424511545293073,
769
+ "loss": 2.5233,
770
+ "step": 1250
771
+ },
772
+ {
773
+ "epoch": 0.54,
774
+ "learning_rate": 0.00024023090586145648,
775
+ "loss": 2.5248,
776
+ "step": 1260
777
+ },
778
+ {
779
+ "epoch": 0.54,
780
+ "learning_rate": 0.00023801065719360567,
781
+ "loss": 2.6073,
782
+ "step": 1270
783
+ },
784
+ {
785
+ "epoch": 0.55,
786
+ "learning_rate": 0.0002357904085257549,
787
+ "loss": 2.596,
788
+ "step": 1280
789
+ },
790
+ {
791
+ "epoch": 0.55,
792
+ "learning_rate": 0.0002335701598579041,
793
+ "loss": 2.6899,
794
+ "step": 1290
795
+ },
796
+ {
797
+ "epoch": 0.56,
798
+ "learning_rate": 0.0002313499111900533,
799
+ "loss": 2.6293,
800
+ "step": 1300
801
+ },
802
+ {
803
+ "epoch": 0.56,
804
+ "learning_rate": 0.0002291296625222025,
805
+ "loss": 2.639,
806
+ "step": 1310
807
+ },
808
+ {
809
+ "epoch": 0.56,
810
+ "learning_rate": 0.0002269094138543517,
811
+ "loss": 2.7105,
812
+ "step": 1320
813
+ },
814
+ {
815
+ "epoch": 0.57,
816
+ "learning_rate": 0.0002246891651865009,
817
+ "loss": 2.723,
818
+ "step": 1330
819
+ },
820
+ {
821
+ "epoch": 0.57,
822
+ "learning_rate": 0.0002224689165186501,
823
+ "loss": 2.5697,
824
+ "step": 1340
825
+ },
826
+ {
827
+ "epoch": 0.58,
828
+ "learning_rate": 0.0002202486678507993,
829
+ "loss": 2.715,
830
+ "step": 1350
831
+ },
832
+ {
833
+ "epoch": 0.58,
834
+ "learning_rate": 0.0002180284191829485,
835
+ "loss": 2.6387,
836
+ "step": 1360
837
+ },
838
+ {
839
+ "epoch": 0.58,
840
+ "learning_rate": 0.00021580817051509768,
841
+ "loss": 2.5276,
842
+ "step": 1370
843
+ },
844
+ {
845
+ "epoch": 0.59,
846
+ "learning_rate": 0.0002135879218472469,
847
+ "loss": 2.627,
848
+ "step": 1380
849
+ },
850
+ {
851
+ "epoch": 0.59,
852
+ "learning_rate": 0.00021136767317939608,
853
+ "loss": 2.5249,
854
+ "step": 1390
855
+ },
856
+ {
857
+ "epoch": 0.6,
858
+ "learning_rate": 0.0002091474245115453,
859
+ "loss": 2.629,
860
+ "step": 1400
861
+ },
862
+ {
863
+ "epoch": 0.6,
864
+ "learning_rate": 0.0002069271758436945,
865
+ "loss": 2.703,
866
+ "step": 1410
867
+ },
868
+ {
869
+ "epoch": 0.61,
870
+ "learning_rate": 0.0002047069271758437,
871
+ "loss": 2.544,
872
+ "step": 1420
873
+ },
874
+ {
875
+ "epoch": 0.61,
876
+ "learning_rate": 0.0002024866785079929,
877
+ "loss": 2.5992,
878
+ "step": 1430
879
+ },
880
+ {
881
+ "epoch": 0.61,
882
+ "learning_rate": 0.0002002664298401421,
883
+ "loss": 2.6988,
884
+ "step": 1440
885
+ },
886
+ {
887
+ "epoch": 0.62,
888
+ "learning_rate": 0.00019804618117229132,
889
+ "loss": 2.6195,
890
+ "step": 1450
891
+ },
892
+ {
893
+ "epoch": 0.62,
894
+ "learning_rate": 0.0001958259325044405,
895
+ "loss": 2.6247,
896
+ "step": 1460
897
+ },
898
+ {
899
+ "epoch": 0.63,
900
+ "learning_rate": 0.0001936056838365897,
901
+ "loss": 2.5853,
902
+ "step": 1470
903
+ },
904
+ {
905
+ "epoch": 0.63,
906
+ "learning_rate": 0.0001913854351687389,
907
+ "loss": 2.6298,
908
+ "step": 1480
909
+ },
910
+ {
911
+ "epoch": 0.64,
912
+ "learning_rate": 0.0001891651865008881,
913
+ "loss": 2.6378,
914
+ "step": 1490
915
+ },
916
+ {
917
+ "epoch": 0.64,
918
+ "learning_rate": 0.0001869449378330373,
919
+ "loss": 2.5682,
920
+ "step": 1500
921
+ }
922
+ ],
923
+ "max_steps": 2342,
924
+ "num_train_epochs": 1,
925
+ "total_flos": 3.54937584731136e+16,
926
+ "trial_name": null,
927
+ "trial_params": null
928
+ }
mt5-new-summarize-final/checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:233a568dcedd606fd72b432d2c6cc678ea8c5e086885a5642ed616a886f52eb9
3
+ size 3643
mt5-new-summarize-final/checkpoint-2000/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mt5-small",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "length_penalty": 0.6,
19
+ "max_length": 128,
20
+ "model_type": "mt5",
21
+ "no_repeat_ngram_size": 2,
22
+ "num_beams": 15,
23
+ "num_decoder_layers": 8,
24
+ "num_heads": 6,
25
+ "num_layers": 8,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "tie_word_embeddings": false,
30
+ "tokenizer_class": "T5Tokenizer",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.26.0",
33
+ "use_cache": true,
34
+ "vocab_size": 250112
35
+ }
mt5-new-summarize-final/checkpoint-2000/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "length_penalty": 0.6,
6
+ "max_length": 128,
7
+ "no_repeat_ngram_size": 2,
8
+ "num_beams": 15,
9
+ "pad_token_id": 0,
10
+ "transformers_version": "4.26.0"
11
+ }
mt5-new-summarize-final/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8398a38588a1d29c290e233a3aa7d80a655280542c57444f7f1ef146e8131d2e
3
+ size 2879429
mt5-new-summarize-final/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f116c931fa378a37a851637fc90de078ddb5e1fd5892542c59de11eeac2cd8f3
3
+ size 14575
mt5-new-summarize-final/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ed71672921938371fe751b4cf233d20e73dc5785b84b089e803a193211e35ae
3
+ size 627
mt5-new-summarize-final/checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
mt5-new-summarize-final/checkpoint-2000/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
mt5-new-summarize-final/checkpoint-2000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93c3578052e1605d8332eb961bc08d72e246071974e4cc54aa6991826b802aa5
3
+ size 16330369
mt5-new-summarize-final/checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": null,
3
+ "eos_token": "</s>",
4
+ "extra_ids": 0,
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "name_or_path": "google/mt5-small",
7
+ "pad_token": "<pad>",
8
+ "sp_model_kwargs": {},
9
+ "special_tokens_map_file": "/root/.cache/huggingface/hub/models--google--mt5-small/snapshots/38f23af8ec210eb6c376d40e9c56bd25a80f195d/special_tokens_map.json",
10
+ "tokenizer_class": "T5Tokenizer",
11
+ "unk_token": "<unk>"
12
+ }
mt5-new-summarize-final/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.853857032313152,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.555555555555555e-05,
13
+ "loss": 20.8315,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.0001111111111111111,
19
+ "loss": 16.4875,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 0.00016666666666666666,
25
+ "loss": 10.6503,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 0.0002222222222222222,
31
+ "loss": 6.6236,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 0.0002777777777777778,
37
+ "loss": 5.2548,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 0.0003333333333333333,
43
+ "loss": 4.6801,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "learning_rate": 0.0003888888888888889,
49
+ "loss": 4.3374,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.03,
54
+ "learning_rate": 0.0004444444444444444,
55
+ "loss": 3.9614,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.04,
60
+ "learning_rate": 0.0005,
61
+ "loss": 3.9704,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.04,
66
+ "learning_rate": 0.0004977797513321492,
67
+ "loss": 3.8112,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.05,
72
+ "learning_rate": 0.0004955595026642984,
73
+ "loss": 3.6272,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.05,
78
+ "learning_rate": 0.0004933392539964477,
79
+ "loss": 3.5572,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "learning_rate": 0.0004911190053285969,
85
+ "loss": 3.3966,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "learning_rate": 0.000488898756660746,
91
+ "loss": 3.4632,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.06,
96
+ "learning_rate": 0.0004866785079928952,
97
+ "loss": 3.4591,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.07,
102
+ "learning_rate": 0.0004844582593250444,
103
+ "loss": 3.2218,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.07,
108
+ "learning_rate": 0.0004822380106571936,
109
+ "loss": 3.4101,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.08,
114
+ "learning_rate": 0.00048001776198934283,
115
+ "loss": 3.2787,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.08,
120
+ "learning_rate": 0.000477797513321492,
121
+ "loss": 3.1043,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.09,
126
+ "learning_rate": 0.0004755772646536412,
127
+ "loss": 3.2435,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "learning_rate": 0.00047335701598579037,
133
+ "loss": 3.2286,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.09,
138
+ "learning_rate": 0.00047113676731793964,
139
+ "loss": 3.1484,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.1,
144
+ "learning_rate": 0.00046891651865008885,
145
+ "loss": 3.1817,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.1,
150
+ "learning_rate": 0.000466696269982238,
151
+ "loss": 3.1608,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.11,
156
+ "learning_rate": 0.00046447602131438723,
157
+ "loss": 3.1457,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.11,
162
+ "learning_rate": 0.0004622557726465364,
163
+ "loss": 3.1212,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.12,
168
+ "learning_rate": 0.00046003552397868566,
169
+ "loss": 2.9937,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "learning_rate": 0.0004578152753108348,
175
+ "loss": 3.2031,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.12,
180
+ "learning_rate": 0.00045559502664298403,
181
+ "loss": 3.0713,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.13,
186
+ "learning_rate": 0.0004533747779751332,
187
+ "loss": 3.0352,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.13,
192
+ "learning_rate": 0.0004511545293072824,
193
+ "loss": 3.0872,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.14,
198
+ "learning_rate": 0.0004489342806394316,
199
+ "loss": 3.0049,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.14,
204
+ "learning_rate": 0.00044671403197158084,
205
+ "loss": 3.0659,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.15,
210
+ "learning_rate": 0.00044449378330373,
211
+ "loss": 3.0852,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "learning_rate": 0.0004422735346358792,
217
+ "loss": 2.9506,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.15,
222
+ "learning_rate": 0.00044005328596802843,
223
+ "loss": 3.027,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.16,
228
+ "learning_rate": 0.00043783303730017764,
229
+ "loss": 2.8812,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.16,
234
+ "learning_rate": 0.00043561278863232686,
235
+ "loss": 2.9738,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.17,
240
+ "learning_rate": 0.000433392539964476,
241
+ "loss": 3.0504,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.17,
246
+ "learning_rate": 0.00043117229129662523,
247
+ "loss": 3.0195,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.18,
252
+ "learning_rate": 0.0004289520426287744,
253
+ "loss": 2.968,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "learning_rate": 0.00042673179396092366,
259
+ "loss": 3.0864,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.18,
264
+ "learning_rate": 0.0004245115452930728,
265
+ "loss": 2.8735,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.19,
270
+ "learning_rate": 0.00042229129662522204,
271
+ "loss": 3.0391,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.19,
276
+ "learning_rate": 0.0004200710479573712,
277
+ "loss": 2.8288,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.2,
282
+ "learning_rate": 0.0004178507992895204,
283
+ "loss": 3.0182,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.2,
288
+ "learning_rate": 0.0004156305506216697,
289
+ "loss": 3.0606,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.2,
294
+ "learning_rate": 0.00041341030195381884,
295
+ "loss": 2.9467,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "learning_rate": 0.00041119005328596806,
301
+ "loss": 2.7842,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.21,
306
+ "learning_rate": 0.0004089698046181172,
307
+ "loss": 2.8472,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.22,
312
+ "learning_rate": 0.00040674955595026643,
313
+ "loss": 2.7883,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.22,
318
+ "learning_rate": 0.00040452930728241565,
319
+ "loss": 2.8592,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.23,
324
+ "learning_rate": 0.00040230905861456486,
325
+ "loss": 2.9186,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.23,
330
+ "learning_rate": 0.000400088809946714,
331
+ "loss": 2.911,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.23,
336
+ "learning_rate": 0.00039786856127886324,
337
+ "loss": 2.8355,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.24,
342
+ "learning_rate": 0.0003956483126110124,
343
+ "loss": 2.9385,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.24,
348
+ "learning_rate": 0.00039342806394316167,
349
+ "loss": 2.8465,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.25,
354
+ "learning_rate": 0.0003912078152753108,
355
+ "loss": 2.8408,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.25,
360
+ "learning_rate": 0.00038898756660746004,
361
+ "loss": 2.767,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.26,
366
+ "learning_rate": 0.00038676731793960926,
367
+ "loss": 2.9691,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.26,
372
+ "learning_rate": 0.0003845470692717584,
373
+ "loss": 2.7348,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.26,
378
+ "learning_rate": 0.0003823268206039077,
379
+ "loss": 2.7477,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.27,
384
+ "learning_rate": 0.00038010657193605685,
385
+ "loss": 2.8142,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.27,
390
+ "learning_rate": 0.00037788632326820606,
391
+ "loss": 2.7522,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.28,
396
+ "learning_rate": 0.0003756660746003552,
397
+ "loss": 2.9403,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.28,
402
+ "learning_rate": 0.00037344582593250444,
403
+ "loss": 2.8956,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.29,
408
+ "learning_rate": 0.00037122557726465365,
409
+ "loss": 2.7331,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.29,
414
+ "learning_rate": 0.00036900532859680287,
415
+ "loss": 2.8265,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.29,
420
+ "learning_rate": 0.000366785079928952,
421
+ "loss": 2.8919,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.3,
426
+ "learning_rate": 0.00036456483126110124,
427
+ "loss": 2.8362,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.3,
432
+ "learning_rate": 0.0003623445825932504,
433
+ "loss": 2.7282,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.31,
438
+ "learning_rate": 0.00036012433392539967,
439
+ "loss": 2.7628,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.31,
444
+ "learning_rate": 0.0003579040852575489,
445
+ "loss": 2.8508,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.32,
450
+ "learning_rate": 0.00035568383658969805,
451
+ "loss": 2.8603,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.32,
456
+ "learning_rate": 0.00035346358792184726,
457
+ "loss": 2.6765,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.32,
462
+ "learning_rate": 0.0003512433392539964,
463
+ "loss": 2.8767,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.33,
468
+ "learning_rate": 0.0003490230905861457,
469
+ "loss": 2.8446,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.33,
474
+ "learning_rate": 0.00034680284191829485,
475
+ "loss": 2.8524,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.34,
480
+ "learning_rate": 0.00034458259325044407,
481
+ "loss": 2.778,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.34,
486
+ "learning_rate": 0.0003423623445825932,
487
+ "loss": 2.7359,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.35,
492
+ "learning_rate": 0.00034014209591474244,
493
+ "loss": 2.7803,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.35,
498
+ "learning_rate": 0.00033792184724689166,
499
+ "loss": 2.7631,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.35,
504
+ "learning_rate": 0.00033570159857904087,
505
+ "loss": 2.7818,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.36,
510
+ "learning_rate": 0.00033348134991119003,
511
+ "loss": 2.6591,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.36,
516
+ "learning_rate": 0.00033126110124333925,
517
+ "loss": 2.6843,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.37,
522
+ "learning_rate": 0.00032904085257548846,
523
+ "loss": 2.7299,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.37,
528
+ "learning_rate": 0.0003268206039076377,
529
+ "loss": 2.6928,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.38,
534
+ "learning_rate": 0.0003246003552397869,
535
+ "loss": 2.8512,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.38,
540
+ "learning_rate": 0.00032238010657193605,
541
+ "loss": 2.7944,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.38,
546
+ "learning_rate": 0.00032015985790408526,
547
+ "loss": 2.7365,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.39,
552
+ "learning_rate": 0.0003179396092362344,
553
+ "loss": 2.6992,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.39,
558
+ "learning_rate": 0.0003157193605683837,
559
+ "loss": 2.743,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.4,
564
+ "learning_rate": 0.00031349911190053285,
565
+ "loss": 2.8022,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.4,
570
+ "learning_rate": 0.00031127886323268207,
571
+ "loss": 2.6603,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.41,
576
+ "learning_rate": 0.00030905861456483123,
577
+ "loss": 2.6277,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.41,
582
+ "learning_rate": 0.00030683836589698044,
583
+ "loss": 2.7652,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.41,
588
+ "learning_rate": 0.0003046181172291297,
589
+ "loss": 2.7063,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.42,
594
+ "learning_rate": 0.0003023978685612789,
595
+ "loss": 2.6602,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.42,
600
+ "learning_rate": 0.0003001776198934281,
601
+ "loss": 2.7592,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.43,
606
+ "learning_rate": 0.00029795737122557725,
607
+ "loss": 2.6628,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 0.43,
612
+ "eval_loss": 0.51519775390625,
613
+ "eval_rouge1": 0.8067071482496617,
614
+ "eval_rouge2": 0.7472387953801375,
615
+ "eval_rougeL": 0.7071393649432568,
616
+ "eval_rougeLsum": 0.8209607917685595,
617
+ "eval_runtime": 1807.8081,
618
+ "eval_samples_per_second": 0.553,
619
+ "eval_steps_per_second": 0.553,
620
+ "step": 1000
621
+ },
622
+ {
623
+ "epoch": 0.43,
624
+ "learning_rate": 0.00029573712255772646,
625
+ "loss": 2.6164,
626
+ "step": 1010
627
+ },
628
+ {
629
+ "epoch": 0.44,
630
+ "learning_rate": 0.0002935168738898757,
631
+ "loss": 2.6947,
632
+ "step": 1020
633
+ },
634
+ {
635
+ "epoch": 0.44,
636
+ "learning_rate": 0.0002912966252220249,
637
+ "loss": 2.4547,
638
+ "step": 1030
639
+ },
640
+ {
641
+ "epoch": 0.44,
642
+ "learning_rate": 0.00028907637655417405,
643
+ "loss": 2.7347,
644
+ "step": 1040
645
+ },
646
+ {
647
+ "epoch": 0.45,
648
+ "learning_rate": 0.00028685612788632327,
649
+ "loss": 2.6491,
650
+ "step": 1050
651
+ },
652
+ {
653
+ "epoch": 0.45,
654
+ "learning_rate": 0.00028463587921847243,
655
+ "loss": 2.6778,
656
+ "step": 1060
657
+ },
658
+ {
659
+ "epoch": 0.46,
660
+ "learning_rate": 0.0002824156305506217,
661
+ "loss": 2.7157,
662
+ "step": 1070
663
+ },
664
+ {
665
+ "epoch": 0.46,
666
+ "learning_rate": 0.00028019538188277086,
667
+ "loss": 2.6445,
668
+ "step": 1080
669
+ },
670
+ {
671
+ "epoch": 0.47,
672
+ "learning_rate": 0.0002779751332149201,
673
+ "loss": 2.7732,
674
+ "step": 1090
675
+ },
676
+ {
677
+ "epoch": 0.47,
678
+ "learning_rate": 0.0002757548845470693,
679
+ "loss": 2.6777,
680
+ "step": 1100
681
+ },
682
+ {
683
+ "epoch": 0.47,
684
+ "learning_rate": 0.00027353463587921845,
685
+ "loss": 2.6231,
686
+ "step": 1110
687
+ },
688
+ {
689
+ "epoch": 0.48,
690
+ "learning_rate": 0.0002713143872113677,
691
+ "loss": 2.6003,
692
+ "step": 1120
693
+ },
694
+ {
695
+ "epoch": 0.48,
696
+ "learning_rate": 0.0002690941385435169,
697
+ "loss": 2.7226,
698
+ "step": 1130
699
+ },
700
+ {
701
+ "epoch": 0.49,
702
+ "learning_rate": 0.0002668738898756661,
703
+ "loss": 2.8623,
704
+ "step": 1140
705
+ },
706
+ {
707
+ "epoch": 0.49,
708
+ "learning_rate": 0.00026465364120781525,
709
+ "loss": 2.7391,
710
+ "step": 1150
711
+ },
712
+ {
713
+ "epoch": 0.5,
714
+ "learning_rate": 0.00026243339253996447,
715
+ "loss": 2.6836,
716
+ "step": 1160
717
+ },
718
+ {
719
+ "epoch": 0.5,
720
+ "learning_rate": 0.0002602131438721137,
721
+ "loss": 2.6357,
722
+ "step": 1170
723
+ },
724
+ {
725
+ "epoch": 0.5,
726
+ "learning_rate": 0.0002579928952042629,
727
+ "loss": 2.5891,
728
+ "step": 1180
729
+ },
730
+ {
731
+ "epoch": 0.51,
732
+ "learning_rate": 0.00025577264653641206,
733
+ "loss": 2.7383,
734
+ "step": 1190
735
+ },
736
+ {
737
+ "epoch": 0.51,
738
+ "learning_rate": 0.00025355239786856127,
739
+ "loss": 2.738,
740
+ "step": 1200
741
+ },
742
+ {
743
+ "epoch": 0.52,
744
+ "learning_rate": 0.00025133214920071043,
745
+ "loss": 2.62,
746
+ "step": 1210
747
+ },
748
+ {
749
+ "epoch": 0.52,
750
+ "learning_rate": 0.0002491119005328597,
751
+ "loss": 2.5796,
752
+ "step": 1220
753
+ },
754
+ {
755
+ "epoch": 0.53,
756
+ "learning_rate": 0.0002468916518650089,
757
+ "loss": 2.5744,
758
+ "step": 1230
759
+ },
760
+ {
761
+ "epoch": 0.53,
762
+ "learning_rate": 0.0002446714031971581,
763
+ "loss": 2.5047,
764
+ "step": 1240
765
+ },
766
+ {
767
+ "epoch": 0.53,
768
+ "learning_rate": 0.0002424511545293073,
769
+ "loss": 2.5233,
770
+ "step": 1250
771
+ },
772
+ {
773
+ "epoch": 0.54,
774
+ "learning_rate": 0.00024023090586145648,
775
+ "loss": 2.5248,
776
+ "step": 1260
777
+ },
778
+ {
779
+ "epoch": 0.54,
780
+ "learning_rate": 0.00023801065719360567,
781
+ "loss": 2.6073,
782
+ "step": 1270
783
+ },
784
+ {
785
+ "epoch": 0.55,
786
+ "learning_rate": 0.0002357904085257549,
787
+ "loss": 2.596,
788
+ "step": 1280
789
+ },
790
+ {
791
+ "epoch": 0.55,
792
+ "learning_rate": 0.0002335701598579041,
793
+ "loss": 2.6899,
794
+ "step": 1290
795
+ },
796
+ {
797
+ "epoch": 0.56,
798
+ "learning_rate": 0.0002313499111900533,
799
+ "loss": 2.6293,
800
+ "step": 1300
801
+ },
802
+ {
803
+ "epoch": 0.56,
804
+ "learning_rate": 0.0002291296625222025,
805
+ "loss": 2.639,
806
+ "step": 1310
807
+ },
808
+ {
809
+ "epoch": 0.56,
810
+ "learning_rate": 0.0002269094138543517,
811
+ "loss": 2.7105,
812
+ "step": 1320
813
+ },
814
+ {
815
+ "epoch": 0.57,
816
+ "learning_rate": 0.0002246891651865009,
817
+ "loss": 2.723,
818
+ "step": 1330
819
+ },
820
+ {
821
+ "epoch": 0.57,
822
+ "learning_rate": 0.0002224689165186501,
823
+ "loss": 2.5697,
824
+ "step": 1340
825
+ },
826
+ {
827
+ "epoch": 0.58,
828
+ "learning_rate": 0.0002202486678507993,
829
+ "loss": 2.715,
830
+ "step": 1350
831
+ },
832
+ {
833
+ "epoch": 0.58,
834
+ "learning_rate": 0.0002180284191829485,
835
+ "loss": 2.6387,
836
+ "step": 1360
837
+ },
838
+ {
839
+ "epoch": 0.58,
840
+ "learning_rate": 0.00021580817051509768,
841
+ "loss": 2.5276,
842
+ "step": 1370
843
+ },
844
+ {
845
+ "epoch": 0.59,
846
+ "learning_rate": 0.0002135879218472469,
847
+ "loss": 2.627,
848
+ "step": 1380
849
+ },
850
+ {
851
+ "epoch": 0.59,
852
+ "learning_rate": 0.00021136767317939608,
853
+ "loss": 2.5249,
854
+ "step": 1390
855
+ },
856
+ {
857
+ "epoch": 0.6,
858
+ "learning_rate": 0.0002091474245115453,
859
+ "loss": 2.629,
860
+ "step": 1400
861
+ },
862
+ {
863
+ "epoch": 0.6,
864
+ "learning_rate": 0.0002069271758436945,
865
+ "loss": 2.703,
866
+ "step": 1410
867
+ },
868
+ {
869
+ "epoch": 0.61,
870
+ "learning_rate": 0.0002047069271758437,
871
+ "loss": 2.544,
872
+ "step": 1420
873
+ },
874
+ {
875
+ "epoch": 0.61,
876
+ "learning_rate": 0.0002024866785079929,
877
+ "loss": 2.5992,
878
+ "step": 1430
879
+ },
880
+ {
881
+ "epoch": 0.61,
882
+ "learning_rate": 0.0002002664298401421,
883
+ "loss": 2.6988,
884
+ "step": 1440
885
+ },
886
+ {
887
+ "epoch": 0.62,
888
+ "learning_rate": 0.00019804618117229132,
889
+ "loss": 2.6195,
890
+ "step": 1450
891
+ },
892
+ {
893
+ "epoch": 0.62,
894
+ "learning_rate": 0.0001958259325044405,
895
+ "loss": 2.6247,
896
+ "step": 1460
897
+ },
898
+ {
899
+ "epoch": 0.63,
900
+ "learning_rate": 0.0001936056838365897,
901
+ "loss": 2.5853,
902
+ "step": 1470
903
+ },
904
+ {
905
+ "epoch": 0.63,
906
+ "learning_rate": 0.0001913854351687389,
907
+ "loss": 2.6298,
908
+ "step": 1480
909
+ },
910
+ {
911
+ "epoch": 0.64,
912
+ "learning_rate": 0.0001891651865008881,
913
+ "loss": 2.6378,
914
+ "step": 1490
915
+ },
916
+ {
917
+ "epoch": 0.64,
918
+ "learning_rate": 0.0001869449378330373,
919
+ "loss": 2.5682,
920
+ "step": 1500
921
+ },
922
+ {
923
+ "epoch": 0.64,
924
+ "learning_rate": 0.0001847246891651865,
925
+ "loss": 2.4707,
926
+ "step": 1510
927
+ },
928
+ {
929
+ "epoch": 0.65,
930
+ "learning_rate": 0.00018250444049733568,
931
+ "loss": 2.6558,
932
+ "step": 1520
933
+ },
934
+ {
935
+ "epoch": 0.65,
936
+ "learning_rate": 0.00018028419182948492,
937
+ "loss": 2.5891,
938
+ "step": 1530
939
+ },
940
+ {
941
+ "epoch": 0.66,
942
+ "learning_rate": 0.0001780639431616341,
943
+ "loss": 2.6053,
944
+ "step": 1540
945
+ },
946
+ {
947
+ "epoch": 0.66,
948
+ "learning_rate": 0.00017584369449378333,
949
+ "loss": 2.7265,
950
+ "step": 1550
951
+ },
952
+ {
953
+ "epoch": 0.67,
954
+ "learning_rate": 0.00017362344582593251,
955
+ "loss": 2.604,
956
+ "step": 1560
957
+ },
958
+ {
959
+ "epoch": 0.67,
960
+ "learning_rate": 0.0001714031971580817,
961
+ "loss": 2.5988,
962
+ "step": 1570
963
+ },
964
+ {
965
+ "epoch": 0.67,
966
+ "learning_rate": 0.00016918294849023092,
967
+ "loss": 2.5128,
968
+ "step": 1580
969
+ },
970
+ {
971
+ "epoch": 0.68,
972
+ "learning_rate": 0.0001669626998223801,
973
+ "loss": 2.7477,
974
+ "step": 1590
975
+ },
976
+ {
977
+ "epoch": 0.68,
978
+ "learning_rate": 0.00016474245115452932,
979
+ "loss": 2.5823,
980
+ "step": 1600
981
+ },
982
+ {
983
+ "epoch": 0.69,
984
+ "learning_rate": 0.0001625222024866785,
985
+ "loss": 2.6041,
986
+ "step": 1610
987
+ },
988
+ {
989
+ "epoch": 0.69,
990
+ "learning_rate": 0.0001603019538188277,
991
+ "loss": 2.6499,
992
+ "step": 1620
993
+ },
994
+ {
995
+ "epoch": 0.7,
996
+ "learning_rate": 0.0001580817051509769,
997
+ "loss": 2.5358,
998
+ "step": 1630
999
+ },
1000
+ {
1001
+ "epoch": 0.7,
1002
+ "learning_rate": 0.0001558614564831261,
1003
+ "loss": 2.636,
1004
+ "step": 1640
1005
+ },
1006
+ {
1007
+ "epoch": 0.7,
1008
+ "learning_rate": 0.0001536412078152753,
1009
+ "loss": 2.6257,
1010
+ "step": 1650
1011
+ },
1012
+ {
1013
+ "epoch": 0.71,
1014
+ "learning_rate": 0.00015142095914742453,
1015
+ "loss": 2.6454,
1016
+ "step": 1660
1017
+ },
1018
+ {
1019
+ "epoch": 0.71,
1020
+ "learning_rate": 0.00014920071047957371,
1021
+ "loss": 2.563,
1022
+ "step": 1670
1023
+ },
1024
+ {
1025
+ "epoch": 0.72,
1026
+ "learning_rate": 0.00014698046181172293,
1027
+ "loss": 2.6789,
1028
+ "step": 1680
1029
+ },
1030
+ {
1031
+ "epoch": 0.72,
1032
+ "learning_rate": 0.00014476021314387212,
1033
+ "loss": 2.5691,
1034
+ "step": 1690
1035
+ },
1036
+ {
1037
+ "epoch": 0.73,
1038
+ "learning_rate": 0.00014253996447602133,
1039
+ "loss": 2.4895,
1040
+ "step": 1700
1041
+ },
1042
+ {
1043
+ "epoch": 0.73,
1044
+ "learning_rate": 0.00014031971580817052,
1045
+ "loss": 2.5423,
1046
+ "step": 1710
1047
+ },
1048
+ {
1049
+ "epoch": 0.73,
1050
+ "learning_rate": 0.0001380994671403197,
1051
+ "loss": 2.6112,
1052
+ "step": 1720
1053
+ },
1054
+ {
1055
+ "epoch": 0.74,
1056
+ "learning_rate": 0.00013587921847246892,
1057
+ "loss": 2.5571,
1058
+ "step": 1730
1059
+ },
1060
+ {
1061
+ "epoch": 0.74,
1062
+ "learning_rate": 0.0001336589698046181,
1063
+ "loss": 2.5789,
1064
+ "step": 1740
1065
+ },
1066
+ {
1067
+ "epoch": 0.75,
1068
+ "learning_rate": 0.00013143872113676732,
1069
+ "loss": 2.6101,
1070
+ "step": 1750
1071
+ },
1072
+ {
1073
+ "epoch": 0.75,
1074
+ "learning_rate": 0.0001292184724689165,
1075
+ "loss": 2.591,
1076
+ "step": 1760
1077
+ },
1078
+ {
1079
+ "epoch": 0.76,
1080
+ "learning_rate": 0.0001269982238010657,
1081
+ "loss": 2.6079,
1082
+ "step": 1770
1083
+ },
1084
+ {
1085
+ "epoch": 0.76,
1086
+ "learning_rate": 0.00012477797513321494,
1087
+ "loss": 2.6617,
1088
+ "step": 1780
1089
+ },
1090
+ {
1091
+ "epoch": 0.76,
1092
+ "learning_rate": 0.00012255772646536413,
1093
+ "loss": 2.579,
1094
+ "step": 1790
1095
+ },
1096
+ {
1097
+ "epoch": 0.77,
1098
+ "learning_rate": 0.00012033747779751332,
1099
+ "loss": 2.594,
1100
+ "step": 1800
1101
+ },
1102
+ {
1103
+ "epoch": 0.77,
1104
+ "learning_rate": 0.00011811722912966252,
1105
+ "loss": 2.6464,
1106
+ "step": 1810
1107
+ },
1108
+ {
1109
+ "epoch": 0.78,
1110
+ "learning_rate": 0.00011589698046181173,
1111
+ "loss": 2.6125,
1112
+ "step": 1820
1113
+ },
1114
+ {
1115
+ "epoch": 0.78,
1116
+ "learning_rate": 0.00011367673179396093,
1117
+ "loss": 2.6031,
1118
+ "step": 1830
1119
+ },
1120
+ {
1121
+ "epoch": 0.79,
1122
+ "learning_rate": 0.00011145648312611012,
1123
+ "loss": 2.5603,
1124
+ "step": 1840
1125
+ },
1126
+ {
1127
+ "epoch": 0.79,
1128
+ "learning_rate": 0.00010923623445825932,
1129
+ "loss": 2.5277,
1130
+ "step": 1850
1131
+ },
1132
+ {
1133
+ "epoch": 0.79,
1134
+ "learning_rate": 0.00010701598579040852,
1135
+ "loss": 2.6054,
1136
+ "step": 1860
1137
+ },
1138
+ {
1139
+ "epoch": 0.8,
1140
+ "learning_rate": 0.00010479573712255772,
1141
+ "loss": 2.4357,
1142
+ "step": 1870
1143
+ },
1144
+ {
1145
+ "epoch": 0.8,
1146
+ "learning_rate": 0.00010257548845470694,
1147
+ "loss": 2.5417,
1148
+ "step": 1880
1149
+ },
1150
+ {
1151
+ "epoch": 0.81,
1152
+ "learning_rate": 0.00010035523978685614,
1153
+ "loss": 2.5977,
1154
+ "step": 1890
1155
+ },
1156
+ {
1157
+ "epoch": 0.81,
1158
+ "learning_rate": 9.813499111900533e-05,
1159
+ "loss": 2.5966,
1160
+ "step": 1900
1161
+ },
1162
+ {
1163
+ "epoch": 0.82,
1164
+ "learning_rate": 9.591474245115453e-05,
1165
+ "loss": 2.5847,
1166
+ "step": 1910
1167
+ },
1168
+ {
1169
+ "epoch": 0.82,
1170
+ "learning_rate": 9.369449378330373e-05,
1171
+ "loss": 2.5082,
1172
+ "step": 1920
1173
+ },
1174
+ {
1175
+ "epoch": 0.82,
1176
+ "learning_rate": 9.147424511545293e-05,
1177
+ "loss": 2.6329,
1178
+ "step": 1930
1179
+ },
1180
+ {
1181
+ "epoch": 0.83,
1182
+ "learning_rate": 8.925399644760215e-05,
1183
+ "loss": 2.5748,
1184
+ "step": 1940
1185
+ },
1186
+ {
1187
+ "epoch": 0.83,
1188
+ "learning_rate": 8.703374777975133e-05,
1189
+ "loss": 2.7223,
1190
+ "step": 1950
1191
+ },
1192
+ {
1193
+ "epoch": 0.84,
1194
+ "learning_rate": 8.481349911190053e-05,
1195
+ "loss": 2.6406,
1196
+ "step": 1960
1197
+ },
1198
+ {
1199
+ "epoch": 0.84,
1200
+ "learning_rate": 8.259325044404974e-05,
1201
+ "loss": 2.5924,
1202
+ "step": 1970
1203
+ },
1204
+ {
1205
+ "epoch": 0.85,
1206
+ "learning_rate": 8.037300177619894e-05,
1207
+ "loss": 2.7957,
1208
+ "step": 1980
1209
+ },
1210
+ {
1211
+ "epoch": 0.85,
1212
+ "learning_rate": 7.815275310834814e-05,
1213
+ "loss": 2.5123,
1214
+ "step": 1990
1215
+ },
1216
+ {
1217
+ "epoch": 0.85,
1218
+ "learning_rate": 7.593250444049734e-05,
1219
+ "loss": 2.579,
1220
+ "step": 2000
1221
+ },
1222
+ {
1223
+ "epoch": 0.85,
1224
+ "eval_loss": 0.48173585534095764,
1225
+ "eval_rouge1": 0.8206909390534098,
1226
+ "eval_rouge2": 0.7623428554475744,
1227
+ "eval_rougeL": 0.7177491451029034,
1228
+ "eval_rougeLsum": 0.8341139788232226,
1229
+ "eval_runtime": 1978.7356,
1230
+ "eval_samples_per_second": 0.505,
1231
+ "eval_steps_per_second": 0.505,
1232
+ "step": 2000
1233
+ }
1234
+ ],
1235
+ "max_steps": 2342,
1236
+ "num_train_epochs": 1,
1237
+ "total_flos": 4.738034481494016e+16,
1238
+ "trial_name": null,
1239
+ "trial_params": null
1240
+ }
mt5-new-summarize-final/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:233a568dcedd606fd72b432d2c6cc678ea8c5e086885a5642ed616a886f52eb9
3
+ size 3643
mt5-new-summarize-final/checkpoint-500/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mt5-small",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "length_penalty": 0.6,
19
+ "max_length": 128,
20
+ "model_type": "mt5",
21
+ "no_repeat_ngram_size": 2,
22
+ "num_beams": 15,
23
+ "num_decoder_layers": 8,
24
+ "num_heads": 6,
25
+ "num_layers": 8,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "tie_word_embeddings": false,
30
+ "tokenizer_class": "T5Tokenizer",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.26.0",
33
+ "use_cache": true,
34
+ "vocab_size": 250112
35
+ }
mt5-new-summarize-final/checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.26.0"
7
+ }
mt5-new-summarize-final/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:862e91e1511f48c42446f90fddd26797161ae5d0f9789080e8f3ed1aac89b679
3
+ size 2879429
mt5-new-summarize-final/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e7569591bd924cae7a6b055850f4f744dea0b196b0796846f38580fc3438db6
3
+ size 14575
mt5-new-summarize-final/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e52e8159d86a2671fd8a649cd3c2e0d18aace8a5ba6181d11c8e8237ec341dcf
3
+ size 627
mt5-new-summarize-final/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
mt5-new-summarize-final/checkpoint-500/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
mt5-new-summarize-final/checkpoint-500/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93c3578052e1605d8332eb961bc08d72e246071974e4cc54aa6991826b802aa5
3
+ size 16330369
mt5-new-summarize-final/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": null,
3
+ "eos_token": "</s>",
4
+ "extra_ids": 0,
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "name_or_path": "google/mt5-small",
7
+ "pad_token": "<pad>",
8
+ "sp_model_kwargs": {},
9
+ "special_tokens_map_file": "/root/.cache/huggingface/hub/models--google--mt5-small/snapshots/38f23af8ec210eb6c376d40e9c56bd25a80f195d/special_tokens_map.json",
10
+ "tokenizer_class": "T5Tokenizer",
11
+ "unk_token": "<unk>"
12
+ }
mt5-new-summarize-final/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,316 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.213464258078288,
5
+ "global_step": 500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.555555555555555e-05,
13
+ "loss": 20.8315,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.0001111111111111111,
19
+ "loss": 16.4875,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 0.00016666666666666666,
25
+ "loss": 10.6503,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 0.0002222222222222222,
31
+ "loss": 6.6236,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 0.0002777777777777778,
37
+ "loss": 5.2548,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 0.0003333333333333333,
43
+ "loss": 4.6801,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "learning_rate": 0.0003888888888888889,
49
+ "loss": 4.3374,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.03,
54
+ "learning_rate": 0.0004444444444444444,
55
+ "loss": 3.9614,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.04,
60
+ "learning_rate": 0.0005,
61
+ "loss": 3.9704,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.04,
66
+ "learning_rate": 0.0004977797513321492,
67
+ "loss": 3.8112,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.05,
72
+ "learning_rate": 0.0004955595026642984,
73
+ "loss": 3.6272,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.05,
78
+ "learning_rate": 0.0004933392539964477,
79
+ "loss": 3.5572,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "learning_rate": 0.0004911190053285969,
85
+ "loss": 3.3966,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "learning_rate": 0.000488898756660746,
91
+ "loss": 3.4632,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.06,
96
+ "learning_rate": 0.0004866785079928952,
97
+ "loss": 3.4591,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.07,
102
+ "learning_rate": 0.0004844582593250444,
103
+ "loss": 3.2218,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.07,
108
+ "learning_rate": 0.0004822380106571936,
109
+ "loss": 3.4101,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.08,
114
+ "learning_rate": 0.00048001776198934283,
115
+ "loss": 3.2787,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.08,
120
+ "learning_rate": 0.000477797513321492,
121
+ "loss": 3.1043,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.09,
126
+ "learning_rate": 0.0004755772646536412,
127
+ "loss": 3.2435,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "learning_rate": 0.00047335701598579037,
133
+ "loss": 3.2286,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.09,
138
+ "learning_rate": 0.00047113676731793964,
139
+ "loss": 3.1484,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.1,
144
+ "learning_rate": 0.00046891651865008885,
145
+ "loss": 3.1817,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.1,
150
+ "learning_rate": 0.000466696269982238,
151
+ "loss": 3.1608,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.11,
156
+ "learning_rate": 0.00046447602131438723,
157
+ "loss": 3.1457,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.11,
162
+ "learning_rate": 0.0004622557726465364,
163
+ "loss": 3.1212,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.12,
168
+ "learning_rate": 0.00046003552397868566,
169
+ "loss": 2.9937,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "learning_rate": 0.0004578152753108348,
175
+ "loss": 3.2031,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.12,
180
+ "learning_rate": 0.00045559502664298403,
181
+ "loss": 3.0713,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.13,
186
+ "learning_rate": 0.0004533747779751332,
187
+ "loss": 3.0352,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.13,
192
+ "learning_rate": 0.0004511545293072824,
193
+ "loss": 3.0872,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.14,
198
+ "learning_rate": 0.0004489342806394316,
199
+ "loss": 3.0049,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.14,
204
+ "learning_rate": 0.00044671403197158084,
205
+ "loss": 3.0659,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.15,
210
+ "learning_rate": 0.00044449378330373,
211
+ "loss": 3.0852,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "learning_rate": 0.0004422735346358792,
217
+ "loss": 2.9506,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.15,
222
+ "learning_rate": 0.00044005328596802843,
223
+ "loss": 3.027,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.16,
228
+ "learning_rate": 0.00043783303730017764,
229
+ "loss": 2.8812,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.16,
234
+ "learning_rate": 0.00043561278863232686,
235
+ "loss": 2.9738,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.17,
240
+ "learning_rate": 0.000433392539964476,
241
+ "loss": 3.0504,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.17,
246
+ "learning_rate": 0.00043117229129662523,
247
+ "loss": 3.0195,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.18,
252
+ "learning_rate": 0.0004289520426287744,
253
+ "loss": 2.968,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "learning_rate": 0.00042673179396092366,
259
+ "loss": 3.0864,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.18,
264
+ "learning_rate": 0.0004245115452930728,
265
+ "loss": 2.8735,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.19,
270
+ "learning_rate": 0.00042229129662522204,
271
+ "loss": 3.0391,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.19,
276
+ "learning_rate": 0.0004200710479573712,
277
+ "loss": 2.8288,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.2,
282
+ "learning_rate": 0.0004178507992895204,
283
+ "loss": 3.0182,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.2,
288
+ "learning_rate": 0.0004156305506216697,
289
+ "loss": 3.0606,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.2,
294
+ "learning_rate": 0.00041341030195381884,
295
+ "loss": 2.9467,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "learning_rate": 0.00041119005328596806,
301
+ "loss": 2.7842,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.21,
306
+ "learning_rate": 0.0004089698046181172,
307
+ "loss": 2.8472,
308
+ "step": 500
309
+ }
310
+ ],
311
+ "max_steps": 2342,
312
+ "num_train_epochs": 1,
313
+ "total_flos": 1.18055554516992e+16,
314
+ "trial_name": null,
315
+ "trial_params": null
316
+ }
mt5-new-summarize-final/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:233a568dcedd606fd72b432d2c6cc678ea8c5e086885a5642ed616a886f52eb9
3
+ size 3643
mt5-new-summarize-final/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/mt5-small",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "length_penalty": 0.6,
19
+ "max_length": 128,
20
+ "model_type": "mt5",
21
+ "no_repeat_ngram_size": 2,
22
+ "num_beams": 15,
23
+ "num_decoder_layers": 8,
24
+ "num_heads": 6,
25
+ "num_layers": 8,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "tie_word_embeddings": false,
30
+ "tokenizer_class": "T5Tokenizer",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.26.0",
33
+ "use_cache": true,
34
+ "vocab_size": 250112
35
+ }
mt5-new-summarize-final/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "length_penalty": 0.6,
6
+ "max_length": 128,
7
+ "no_repeat_ngram_size": 2,
8
+ "num_beams": 15,
9
+ "pad_token_id": 0,
10
+ "transformers_version": "4.26.0"
11
+ }
mt5-new-summarize-final/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2902dce4d66552f797fdd41548dcbc0682bf2ab91938e503552557c0d0f7cb22
3
+ size 1200772485
mt5-new-summarize-final/runs/Jan29_13-58-29_d0795309bd3b/1675000709.4761653/events.out.tfevents.1675000709.d0795309bd3b.4670.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aad27ec134114b9f9f4fd9a2e390c618677f1ec53f9636eed716424805f0d91f
3
+ size 5988
mt5-new-summarize-final/runs/Jan29_13-58-29_d0795309bd3b/events.out.tfevents.1675000709.d0795309bd3b.4670.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7ef820c793c89b6e6c4e554c2a7eb554edfb58be6f33776d7ad79c242fa8da8
3
+ size 42224