sheepy928 commited on
Commit
6274a68
·
verified ·
1 Parent(s): 3fad7f2

Upload 11 files

Browse files
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "../../save/marinda-default-ep0.4/checkpoint-25000",
3
+ "architectures": [
4
+ "MarindaForBinSim"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 1,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "ep_add_linear_projection": true,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "margin": 0.5,
18
+ "max_position_embeddings": 600,
19
+ "max_relative_position_embeddings": 8,
20
+ "model_type": "marinda",
21
+ "num_attention_heads": 12,
22
+ "num_hidden_layers": 12,
23
+ "pad_token_id": 3,
24
+ "position_embedding_type": "mixed",
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.30.2",
27
+ "type_vocab_size": 2,
28
+ "use_cache": true,
29
+ "vocab_size": 30523
30
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc2981c0aff3bc162cfa4b008337107242bed314e13b9e5758867ec1f87a3719
3
+ size 876537733
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c3083dcee35f85f211e4f435b4787b7192be533e0a723c28ff619c1af34332f
3
+ size 438279705
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc10561f746cc3d82626462b49655b214f0fbd3b805d6ecd04eaed895f02bae4
3
+ size 15607
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3b3b9b2984a1b445d8091e56e1495c5c61aaa9efc9d45ede235f629723cb4b9
3
+ size 15607
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3b0e375f13ad09230466e9e0755c40b14cd6140cf94b8450198d9a2521a810b
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "clean_up_tokenization_spaces": true,
3
+ "model_max_length": 512,
4
+ "tokenizer_class": "MarindaTokenizer"
5
+ }
trainer_state.json ADDED
@@ -0,0 +1,2464 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.7905919557268505,
5
+ "global_step": 4000,
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": 9.998023324767742e-06,
13
+ "loss": 0.101,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 9.996046649535482e-06,
19
+ "loss": 0.0803,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 9.994069974303224e-06,
25
+ "loss": 0.0877,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.01,
30
+ "learning_rate": 9.992093299070964e-06,
31
+ "loss": 0.0749,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.01,
36
+ "learning_rate": 9.990116623838703e-06,
37
+ "loss": 0.0682,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "learning_rate": 9.988139948606445e-06,
43
+ "loss": 0.0606,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "learning_rate": 9.986163273374185e-06,
49
+ "loss": 0.0546,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.02,
54
+ "learning_rate": 9.984186598141926e-06,
55
+ "loss": 0.0534,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.02,
60
+ "learning_rate": 9.982209922909668e-06,
61
+ "loss": 0.0682,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.02,
66
+ "learning_rate": 9.980233247677408e-06,
67
+ "loss": 0.0521,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.02,
72
+ "learning_rate": 9.978256572445149e-06,
73
+ "loss": 0.0556,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.02,
78
+ "learning_rate": 9.976279897212889e-06,
79
+ "loss": 0.0432,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.03,
84
+ "learning_rate": 9.974303221980629e-06,
85
+ "loss": 0.0506,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.03,
90
+ "learning_rate": 9.97232654674837e-06,
91
+ "loss": 0.0491,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.03,
96
+ "learning_rate": 9.97034987151611e-06,
97
+ "loss": 0.0402,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.03,
102
+ "learning_rate": 9.968373196283852e-06,
103
+ "loss": 0.0453,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.03,
108
+ "learning_rate": 9.966396521051591e-06,
109
+ "loss": 0.0435,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.04,
114
+ "learning_rate": 9.964419845819333e-06,
115
+ "loss": 0.0402,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.04,
120
+ "learning_rate": 9.962443170587075e-06,
121
+ "loss": 0.0434,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.04,
126
+ "learning_rate": 9.960466495354814e-06,
127
+ "loss": 0.0385,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.04,
132
+ "learning_rate": 9.958489820122554e-06,
133
+ "loss": 0.0413,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.04,
138
+ "learning_rate": 9.956513144890296e-06,
139
+ "loss": 0.0426,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.05,
144
+ "learning_rate": 9.954536469658036e-06,
145
+ "loss": 0.0384,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.05,
150
+ "learning_rate": 9.952559794425777e-06,
151
+ "loss": 0.0371,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.05,
156
+ "learning_rate": 9.950583119193517e-06,
157
+ "loss": 0.0366,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.05,
162
+ "learning_rate": 9.948606443961258e-06,
163
+ "loss": 0.0359,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.05,
168
+ "learning_rate": 9.946629768728998e-06,
169
+ "loss": 0.0436,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.06,
174
+ "learning_rate": 9.944653093496738e-06,
175
+ "loss": 0.0331,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.06,
180
+ "learning_rate": 9.94267641826448e-06,
181
+ "loss": 0.0339,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.06,
186
+ "learning_rate": 9.940699743032221e-06,
187
+ "loss": 0.0329,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.06,
192
+ "learning_rate": 9.938723067799961e-06,
193
+ "loss": 0.0345,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.06,
198
+ "learning_rate": 9.936746392567702e-06,
199
+ "loss": 0.0367,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.07,
204
+ "learning_rate": 9.934769717335442e-06,
205
+ "loss": 0.0354,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.07,
210
+ "learning_rate": 9.932793042103184e-06,
211
+ "loss": 0.0283,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.07,
216
+ "learning_rate": 9.930816366870924e-06,
217
+ "loss": 0.0293,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.07,
222
+ "learning_rate": 9.928839691638663e-06,
223
+ "loss": 0.0317,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.07,
228
+ "learning_rate": 9.926863016406405e-06,
229
+ "loss": 0.0264,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.08,
234
+ "learning_rate": 9.924886341174147e-06,
235
+ "loss": 0.0362,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.08,
240
+ "learning_rate": 9.922909665941886e-06,
241
+ "loss": 0.0296,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.08,
246
+ "learning_rate": 9.920932990709628e-06,
247
+ "loss": 0.0274,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.08,
252
+ "learning_rate": 9.918956315477368e-06,
253
+ "loss": 0.0282,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.08,
258
+ "learning_rate": 9.91697964024511e-06,
259
+ "loss": 0.033,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.08,
264
+ "learning_rate": 9.915002965012849e-06,
265
+ "loss": 0.0259,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.09,
270
+ "learning_rate": 9.913026289780589e-06,
271
+ "loss": 0.0351,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.09,
276
+ "learning_rate": 9.91104961454833e-06,
277
+ "loss": 0.0309,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.09,
282
+ "learning_rate": 9.909072939316072e-06,
283
+ "loss": 0.0321,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.09,
288
+ "learning_rate": 9.907096264083812e-06,
289
+ "loss": 0.0277,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.09,
294
+ "learning_rate": 9.905119588851553e-06,
295
+ "loss": 0.0273,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.1,
300
+ "learning_rate": 9.903142913619293e-06,
301
+ "loss": 0.0248,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.1,
306
+ "learning_rate": 9.901166238387033e-06,
307
+ "loss": 0.0338,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.1,
312
+ "learning_rate": 9.899189563154774e-06,
313
+ "loss": 0.0278,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.1,
318
+ "learning_rate": 9.897212887922514e-06,
319
+ "loss": 0.0315,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.1,
324
+ "learning_rate": 9.895236212690256e-06,
325
+ "loss": 0.0299,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.11,
330
+ "learning_rate": 9.893259537457996e-06,
331
+ "loss": 0.0228,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.11,
336
+ "learning_rate": 9.891282862225737e-06,
337
+ "loss": 0.0271,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.11,
342
+ "learning_rate": 9.889306186993479e-06,
343
+ "loss": 0.023,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.11,
348
+ "learning_rate": 9.887329511761219e-06,
349
+ "loss": 0.024,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.11,
354
+ "learning_rate": 9.885352836528958e-06,
355
+ "loss": 0.0334,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.12,
360
+ "learning_rate": 9.8833761612967e-06,
361
+ "loss": 0.026,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.12,
366
+ "learning_rate": 9.88139948606444e-06,
367
+ "loss": 0.0263,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.12,
372
+ "learning_rate": 9.879422810832181e-06,
373
+ "loss": 0.0285,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.12,
378
+ "learning_rate": 9.877446135599921e-06,
379
+ "loss": 0.0296,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.12,
384
+ "learning_rate": 9.875469460367663e-06,
385
+ "loss": 0.0253,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.13,
390
+ "learning_rate": 9.873492785135404e-06,
391
+ "loss": 0.0299,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.13,
396
+ "learning_rate": 9.871516109903144e-06,
397
+ "loss": 0.0179,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.13,
402
+ "learning_rate": 9.869539434670884e-06,
403
+ "loss": 0.0234,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.13,
408
+ "learning_rate": 9.867562759438625e-06,
409
+ "loss": 0.0256,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.13,
414
+ "learning_rate": 9.865586084206365e-06,
415
+ "loss": 0.0215,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.14,
420
+ "learning_rate": 9.863609408974107e-06,
421
+ "loss": 0.0285,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.14,
426
+ "learning_rate": 9.861632733741846e-06,
427
+ "loss": 0.0247,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.14,
432
+ "learning_rate": 9.859656058509588e-06,
433
+ "loss": 0.0257,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.14,
438
+ "learning_rate": 9.85767938327733e-06,
439
+ "loss": 0.0277,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.14,
444
+ "learning_rate": 9.855702708045068e-06,
445
+ "loss": 0.0211,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.15,
450
+ "learning_rate": 9.85372603281281e-06,
451
+ "loss": 0.0231,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.15,
456
+ "learning_rate": 9.85174935758055e-06,
457
+ "loss": 0.0231,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.15,
462
+ "learning_rate": 9.84977268234829e-06,
463
+ "loss": 0.0244,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.15,
468
+ "learning_rate": 9.847796007116032e-06,
469
+ "loss": 0.0227,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.15,
474
+ "learning_rate": 9.845819331883772e-06,
475
+ "loss": 0.0256,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.16,
480
+ "learning_rate": 9.843842656651513e-06,
481
+ "loss": 0.0244,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.16,
486
+ "learning_rate": 9.841865981419253e-06,
487
+ "loss": 0.0242,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.16,
492
+ "learning_rate": 9.839889306186993e-06,
493
+ "loss": 0.0241,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.16,
498
+ "learning_rate": 9.837912630954735e-06,
499
+ "loss": 0.0186,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.16,
504
+ "learning_rate": 9.835935955722476e-06,
505
+ "loss": 0.0262,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.17,
510
+ "learning_rate": 9.833959280490216e-06,
511
+ "loss": 0.0216,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.17,
516
+ "learning_rate": 9.831982605257957e-06,
517
+ "loss": 0.0192,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.17,
522
+ "learning_rate": 9.830005930025697e-06,
523
+ "loss": 0.0284,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.17,
528
+ "learning_rate": 9.828029254793439e-06,
529
+ "loss": 0.0258,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.17,
534
+ "learning_rate": 9.826052579561179e-06,
535
+ "loss": 0.0205,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.18,
540
+ "learning_rate": 9.824075904328918e-06,
541
+ "loss": 0.0246,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.18,
546
+ "learning_rate": 9.82209922909666e-06,
547
+ "loss": 0.0214,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.18,
552
+ "learning_rate": 9.820122553864402e-06,
553
+ "loss": 0.0222,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.18,
558
+ "learning_rate": 9.818145878632141e-06,
559
+ "loss": 0.0179,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.18,
564
+ "learning_rate": 9.816169203399883e-06,
565
+ "loss": 0.0201,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.19,
570
+ "learning_rate": 9.814192528167623e-06,
571
+ "loss": 0.0199,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.19,
576
+ "learning_rate": 9.812215852935364e-06,
577
+ "loss": 0.0238,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.19,
582
+ "learning_rate": 9.810239177703104e-06,
583
+ "loss": 0.0136,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.19,
588
+ "learning_rate": 9.808262502470844e-06,
589
+ "loss": 0.0196,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.19,
594
+ "learning_rate": 9.806285827238585e-06,
595
+ "loss": 0.0228,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.2,
600
+ "learning_rate": 9.804309152006325e-06,
601
+ "loss": 0.02,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.2,
606
+ "learning_rate": 9.802332476774067e-06,
607
+ "loss": 0.0212,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "batch_size": 128,
612
+ "epoch": 0.2,
613
+ "step": 1000,
614
+ "valid_mrr": 0.713063593146979,
615
+ "valid_mrr_wo_pooler": 0.7165689455071271,
616
+ "valid_pr1": 0.6296875,
617
+ "valid_pr1_wo_pooler": 0.6328125,
618
+ "valid_runtime": 19.0638,
619
+ "valid_samples_per_second": 60.376,
620
+ "valid_steps_per_second": 0.21
621
+ },
622
+ {
623
+ "epoch": 0.2,
624
+ "learning_rate": 9.800355801541808e-06,
625
+ "loss": 0.017,
626
+ "step": 1010
627
+ },
628
+ {
629
+ "epoch": 0.2,
630
+ "learning_rate": 9.798379126309548e-06,
631
+ "loss": 0.0231,
632
+ "step": 1020
633
+ },
634
+ {
635
+ "epoch": 0.2,
636
+ "learning_rate": 9.79640245107729e-06,
637
+ "loss": 0.0206,
638
+ "step": 1030
639
+ },
640
+ {
641
+ "epoch": 0.21,
642
+ "learning_rate": 9.79442577584503e-06,
643
+ "loss": 0.0228,
644
+ "step": 1040
645
+ },
646
+ {
647
+ "epoch": 0.21,
648
+ "learning_rate": 9.79244910061277e-06,
649
+ "loss": 0.0122,
650
+ "step": 1050
651
+ },
652
+ {
653
+ "epoch": 0.21,
654
+ "learning_rate": 9.79047242538051e-06,
655
+ "loss": 0.016,
656
+ "step": 1060
657
+ },
658
+ {
659
+ "epoch": 0.21,
660
+ "learning_rate": 9.78849575014825e-06,
661
+ "loss": 0.0199,
662
+ "step": 1070
663
+ },
664
+ {
665
+ "epoch": 0.21,
666
+ "learning_rate": 9.786519074915992e-06,
667
+ "loss": 0.0144,
668
+ "step": 1080
669
+ },
670
+ {
671
+ "epoch": 0.22,
672
+ "learning_rate": 9.784542399683734e-06,
673
+ "loss": 0.0179,
674
+ "step": 1090
675
+ },
676
+ {
677
+ "epoch": 0.22,
678
+ "learning_rate": 9.782565724451474e-06,
679
+ "loss": 0.0159,
680
+ "step": 1100
681
+ },
682
+ {
683
+ "epoch": 0.22,
684
+ "learning_rate": 9.780589049219215e-06,
685
+ "loss": 0.0216,
686
+ "step": 1110
687
+ },
688
+ {
689
+ "epoch": 0.22,
690
+ "learning_rate": 9.778612373986955e-06,
691
+ "loss": 0.0176,
692
+ "step": 1120
693
+ },
694
+ {
695
+ "epoch": 0.22,
696
+ "learning_rate": 9.776635698754695e-06,
697
+ "loss": 0.0198,
698
+ "step": 1130
699
+ },
700
+ {
701
+ "epoch": 0.23,
702
+ "learning_rate": 9.774659023522436e-06,
703
+ "loss": 0.0176,
704
+ "step": 1140
705
+ },
706
+ {
707
+ "epoch": 0.23,
708
+ "learning_rate": 9.772682348290176e-06,
709
+ "loss": 0.0224,
710
+ "step": 1150
711
+ },
712
+ {
713
+ "epoch": 0.23,
714
+ "learning_rate": 9.770705673057918e-06,
715
+ "loss": 0.0172,
716
+ "step": 1160
717
+ },
718
+ {
719
+ "epoch": 0.23,
720
+ "learning_rate": 9.768728997825659e-06,
721
+ "loss": 0.0147,
722
+ "step": 1170
723
+ },
724
+ {
725
+ "epoch": 0.23,
726
+ "learning_rate": 9.766752322593399e-06,
727
+ "loss": 0.023,
728
+ "step": 1180
729
+ },
730
+ {
731
+ "epoch": 0.24,
732
+ "learning_rate": 9.764775647361139e-06,
733
+ "loss": 0.0247,
734
+ "step": 1190
735
+ },
736
+ {
737
+ "epoch": 0.24,
738
+ "learning_rate": 9.76279897212888e-06,
739
+ "loss": 0.0203,
740
+ "step": 1200
741
+ },
742
+ {
743
+ "epoch": 0.24,
744
+ "learning_rate": 9.76082229689662e-06,
745
+ "loss": 0.016,
746
+ "step": 1210
747
+ },
748
+ {
749
+ "epoch": 0.24,
750
+ "learning_rate": 9.758845621664362e-06,
751
+ "loss": 0.0179,
752
+ "step": 1220
753
+ },
754
+ {
755
+ "epoch": 0.24,
756
+ "learning_rate": 9.756868946432102e-06,
757
+ "loss": 0.0179,
758
+ "step": 1230
759
+ },
760
+ {
761
+ "epoch": 0.25,
762
+ "learning_rate": 9.754892271199843e-06,
763
+ "loss": 0.0207,
764
+ "step": 1240
765
+ },
766
+ {
767
+ "epoch": 0.25,
768
+ "learning_rate": 9.752915595967583e-06,
769
+ "loss": 0.0204,
770
+ "step": 1250
771
+ },
772
+ {
773
+ "epoch": 0.25,
774
+ "learning_rate": 9.750938920735324e-06,
775
+ "loss": 0.0208,
776
+ "step": 1260
777
+ },
778
+ {
779
+ "epoch": 0.25,
780
+ "learning_rate": 9.748962245503064e-06,
781
+ "loss": 0.0161,
782
+ "step": 1270
783
+ },
784
+ {
785
+ "epoch": 0.25,
786
+ "learning_rate": 9.746985570270806e-06,
787
+ "loss": 0.0188,
788
+ "step": 1280
789
+ },
790
+ {
791
+ "epoch": 0.25,
792
+ "learning_rate": 9.745008895038546e-06,
793
+ "loss": 0.0179,
794
+ "step": 1290
795
+ },
796
+ {
797
+ "epoch": 0.26,
798
+ "learning_rate": 9.743032219806287e-06,
799
+ "loss": 0.0164,
800
+ "step": 1300
801
+ },
802
+ {
803
+ "epoch": 0.26,
804
+ "learning_rate": 9.741055544574027e-06,
805
+ "loss": 0.013,
806
+ "step": 1310
807
+ },
808
+ {
809
+ "epoch": 0.26,
810
+ "learning_rate": 9.739078869341768e-06,
811
+ "loss": 0.0174,
812
+ "step": 1320
813
+ },
814
+ {
815
+ "epoch": 0.26,
816
+ "learning_rate": 9.737102194109508e-06,
817
+ "loss": 0.0197,
818
+ "step": 1330
819
+ },
820
+ {
821
+ "epoch": 0.26,
822
+ "learning_rate": 9.73512551887725e-06,
823
+ "loss": 0.02,
824
+ "step": 1340
825
+ },
826
+ {
827
+ "epoch": 0.27,
828
+ "learning_rate": 9.73314884364499e-06,
829
+ "loss": 0.0218,
830
+ "step": 1350
831
+ },
832
+ {
833
+ "epoch": 0.27,
834
+ "learning_rate": 9.73117216841273e-06,
835
+ "loss": 0.0137,
836
+ "step": 1360
837
+ },
838
+ {
839
+ "epoch": 0.27,
840
+ "learning_rate": 9.729195493180471e-06,
841
+ "loss": 0.0199,
842
+ "step": 1370
843
+ },
844
+ {
845
+ "epoch": 0.27,
846
+ "learning_rate": 9.727218817948212e-06,
847
+ "loss": 0.0143,
848
+ "step": 1380
849
+ },
850
+ {
851
+ "epoch": 0.27,
852
+ "learning_rate": 9.725242142715952e-06,
853
+ "loss": 0.017,
854
+ "step": 1390
855
+ },
856
+ {
857
+ "epoch": 0.28,
858
+ "learning_rate": 9.723265467483694e-06,
859
+ "loss": 0.0169,
860
+ "step": 1400
861
+ },
862
+ {
863
+ "epoch": 0.28,
864
+ "learning_rate": 9.721288792251434e-06,
865
+ "loss": 0.0175,
866
+ "step": 1410
867
+ },
868
+ {
869
+ "epoch": 0.28,
870
+ "learning_rate": 9.719312117019175e-06,
871
+ "loss": 0.0162,
872
+ "step": 1420
873
+ },
874
+ {
875
+ "epoch": 0.28,
876
+ "learning_rate": 9.717335441786915e-06,
877
+ "loss": 0.018,
878
+ "step": 1430
879
+ },
880
+ {
881
+ "epoch": 0.28,
882
+ "learning_rate": 9.715358766554655e-06,
883
+ "loss": 0.0226,
884
+ "step": 1440
885
+ },
886
+ {
887
+ "epoch": 0.29,
888
+ "learning_rate": 9.713382091322396e-06,
889
+ "loss": 0.0153,
890
+ "step": 1450
891
+ },
892
+ {
893
+ "epoch": 0.29,
894
+ "learning_rate": 9.711405416090138e-06,
895
+ "loss": 0.0146,
896
+ "step": 1460
897
+ },
898
+ {
899
+ "epoch": 0.29,
900
+ "learning_rate": 9.709428740857878e-06,
901
+ "loss": 0.0171,
902
+ "step": 1470
903
+ },
904
+ {
905
+ "epoch": 0.29,
906
+ "learning_rate": 9.70745206562562e-06,
907
+ "loss": 0.0195,
908
+ "step": 1480
909
+ },
910
+ {
911
+ "epoch": 0.29,
912
+ "learning_rate": 9.705475390393359e-06,
913
+ "loss": 0.0185,
914
+ "step": 1490
915
+ },
916
+ {
917
+ "epoch": 0.3,
918
+ "learning_rate": 9.703498715161099e-06,
919
+ "loss": 0.0171,
920
+ "step": 1500
921
+ },
922
+ {
923
+ "epoch": 0.3,
924
+ "learning_rate": 9.70152203992884e-06,
925
+ "loss": 0.017,
926
+ "step": 1510
927
+ },
928
+ {
929
+ "epoch": 0.3,
930
+ "learning_rate": 9.69954536469658e-06,
931
+ "loss": 0.0127,
932
+ "step": 1520
933
+ },
934
+ {
935
+ "epoch": 0.3,
936
+ "learning_rate": 9.697568689464322e-06,
937
+ "loss": 0.0178,
938
+ "step": 1530
939
+ },
940
+ {
941
+ "epoch": 0.3,
942
+ "learning_rate": 9.695592014232063e-06,
943
+ "loss": 0.016,
944
+ "step": 1540
945
+ },
946
+ {
947
+ "epoch": 0.31,
948
+ "learning_rate": 9.693615338999803e-06,
949
+ "loss": 0.0171,
950
+ "step": 1550
951
+ },
952
+ {
953
+ "epoch": 0.31,
954
+ "learning_rate": 9.691638663767545e-06,
955
+ "loss": 0.0141,
956
+ "step": 1560
957
+ },
958
+ {
959
+ "epoch": 0.31,
960
+ "learning_rate": 9.689661988535285e-06,
961
+ "loss": 0.0173,
962
+ "step": 1570
963
+ },
964
+ {
965
+ "epoch": 0.31,
966
+ "learning_rate": 9.687685313303024e-06,
967
+ "loss": 0.0148,
968
+ "step": 1580
969
+ },
970
+ {
971
+ "epoch": 0.31,
972
+ "learning_rate": 9.685708638070766e-06,
973
+ "loss": 0.0139,
974
+ "step": 1590
975
+ },
976
+ {
977
+ "epoch": 0.32,
978
+ "learning_rate": 9.683731962838506e-06,
979
+ "loss": 0.0186,
980
+ "step": 1600
981
+ },
982
+ {
983
+ "epoch": 0.32,
984
+ "learning_rate": 9.681755287606247e-06,
985
+ "loss": 0.0206,
986
+ "step": 1610
987
+ },
988
+ {
989
+ "epoch": 0.32,
990
+ "learning_rate": 9.679778612373987e-06,
991
+ "loss": 0.0201,
992
+ "step": 1620
993
+ },
994
+ {
995
+ "epoch": 0.32,
996
+ "learning_rate": 9.677801937141729e-06,
997
+ "loss": 0.0121,
998
+ "step": 1630
999
+ },
1000
+ {
1001
+ "epoch": 0.32,
1002
+ "learning_rate": 9.67582526190947e-06,
1003
+ "loss": 0.0195,
1004
+ "step": 1640
1005
+ },
1006
+ {
1007
+ "epoch": 0.33,
1008
+ "learning_rate": 9.67384858667721e-06,
1009
+ "loss": 0.0212,
1010
+ "step": 1650
1011
+ },
1012
+ {
1013
+ "epoch": 0.33,
1014
+ "learning_rate": 9.67187191144495e-06,
1015
+ "loss": 0.0164,
1016
+ "step": 1660
1017
+ },
1018
+ {
1019
+ "epoch": 0.33,
1020
+ "learning_rate": 9.669895236212691e-06,
1021
+ "loss": 0.0196,
1022
+ "step": 1670
1023
+ },
1024
+ {
1025
+ "epoch": 0.33,
1026
+ "learning_rate": 9.667918560980431e-06,
1027
+ "loss": 0.0187,
1028
+ "step": 1680
1029
+ },
1030
+ {
1031
+ "epoch": 0.33,
1032
+ "learning_rate": 9.665941885748173e-06,
1033
+ "loss": 0.0169,
1034
+ "step": 1690
1035
+ },
1036
+ {
1037
+ "epoch": 0.34,
1038
+ "learning_rate": 9.663965210515912e-06,
1039
+ "loss": 0.0181,
1040
+ "step": 1700
1041
+ },
1042
+ {
1043
+ "epoch": 0.34,
1044
+ "learning_rate": 9.661988535283654e-06,
1045
+ "loss": 0.0184,
1046
+ "step": 1710
1047
+ },
1048
+ {
1049
+ "epoch": 0.34,
1050
+ "learning_rate": 9.660011860051396e-06,
1051
+ "loss": 0.0196,
1052
+ "step": 1720
1053
+ },
1054
+ {
1055
+ "epoch": 0.34,
1056
+ "learning_rate": 9.658035184819135e-06,
1057
+ "loss": 0.0158,
1058
+ "step": 1730
1059
+ },
1060
+ {
1061
+ "epoch": 0.34,
1062
+ "learning_rate": 9.656058509586875e-06,
1063
+ "loss": 0.0147,
1064
+ "step": 1740
1065
+ },
1066
+ {
1067
+ "epoch": 0.35,
1068
+ "learning_rate": 9.654081834354617e-06,
1069
+ "loss": 0.0102,
1070
+ "step": 1750
1071
+ },
1072
+ {
1073
+ "epoch": 0.35,
1074
+ "learning_rate": 9.652105159122357e-06,
1075
+ "loss": 0.0212,
1076
+ "step": 1760
1077
+ },
1078
+ {
1079
+ "epoch": 0.35,
1080
+ "learning_rate": 9.650128483890098e-06,
1081
+ "loss": 0.0144,
1082
+ "step": 1770
1083
+ },
1084
+ {
1085
+ "epoch": 0.35,
1086
+ "learning_rate": 9.648151808657838e-06,
1087
+ "loss": 0.0168,
1088
+ "step": 1780
1089
+ },
1090
+ {
1091
+ "epoch": 0.35,
1092
+ "learning_rate": 9.64617513342558e-06,
1093
+ "loss": 0.0133,
1094
+ "step": 1790
1095
+ },
1096
+ {
1097
+ "epoch": 0.36,
1098
+ "learning_rate": 9.644198458193321e-06,
1099
+ "loss": 0.0178,
1100
+ "step": 1800
1101
+ },
1102
+ {
1103
+ "epoch": 0.36,
1104
+ "learning_rate": 9.642221782961059e-06,
1105
+ "loss": 0.0135,
1106
+ "step": 1810
1107
+ },
1108
+ {
1109
+ "epoch": 0.36,
1110
+ "learning_rate": 9.6402451077288e-06,
1111
+ "loss": 0.0177,
1112
+ "step": 1820
1113
+ },
1114
+ {
1115
+ "epoch": 0.36,
1116
+ "learning_rate": 9.638268432496542e-06,
1117
+ "loss": 0.0173,
1118
+ "step": 1830
1119
+ },
1120
+ {
1121
+ "epoch": 0.36,
1122
+ "learning_rate": 9.636291757264282e-06,
1123
+ "loss": 0.0176,
1124
+ "step": 1840
1125
+ },
1126
+ {
1127
+ "epoch": 0.37,
1128
+ "learning_rate": 9.634315082032023e-06,
1129
+ "loss": 0.0164,
1130
+ "step": 1850
1131
+ },
1132
+ {
1133
+ "epoch": 0.37,
1134
+ "learning_rate": 9.632338406799763e-06,
1135
+ "loss": 0.0196,
1136
+ "step": 1860
1137
+ },
1138
+ {
1139
+ "epoch": 0.37,
1140
+ "learning_rate": 9.630361731567505e-06,
1141
+ "loss": 0.0215,
1142
+ "step": 1870
1143
+ },
1144
+ {
1145
+ "epoch": 0.37,
1146
+ "learning_rate": 9.628385056335245e-06,
1147
+ "loss": 0.0183,
1148
+ "step": 1880
1149
+ },
1150
+ {
1151
+ "epoch": 0.37,
1152
+ "learning_rate": 9.626408381102984e-06,
1153
+ "loss": 0.0107,
1154
+ "step": 1890
1155
+ },
1156
+ {
1157
+ "epoch": 0.38,
1158
+ "learning_rate": 9.624431705870726e-06,
1159
+ "loss": 0.019,
1160
+ "step": 1900
1161
+ },
1162
+ {
1163
+ "epoch": 0.38,
1164
+ "learning_rate": 9.622455030638468e-06,
1165
+ "loss": 0.0157,
1166
+ "step": 1910
1167
+ },
1168
+ {
1169
+ "epoch": 0.38,
1170
+ "learning_rate": 9.620478355406207e-06,
1171
+ "loss": 0.0171,
1172
+ "step": 1920
1173
+ },
1174
+ {
1175
+ "epoch": 0.38,
1176
+ "learning_rate": 9.618501680173949e-06,
1177
+ "loss": 0.0153,
1178
+ "step": 1930
1179
+ },
1180
+ {
1181
+ "epoch": 0.38,
1182
+ "learning_rate": 9.616525004941689e-06,
1183
+ "loss": 0.0164,
1184
+ "step": 1940
1185
+ },
1186
+ {
1187
+ "epoch": 0.39,
1188
+ "learning_rate": 9.61454832970943e-06,
1189
+ "loss": 0.0161,
1190
+ "step": 1950
1191
+ },
1192
+ {
1193
+ "epoch": 0.39,
1194
+ "learning_rate": 9.61257165447717e-06,
1195
+ "loss": 0.0178,
1196
+ "step": 1960
1197
+ },
1198
+ {
1199
+ "epoch": 0.39,
1200
+ "learning_rate": 9.61059497924491e-06,
1201
+ "loss": 0.0111,
1202
+ "step": 1970
1203
+ },
1204
+ {
1205
+ "epoch": 0.39,
1206
+ "learning_rate": 9.608618304012651e-06,
1207
+ "loss": 0.0175,
1208
+ "step": 1980
1209
+ },
1210
+ {
1211
+ "epoch": 0.39,
1212
+ "learning_rate": 9.606641628780393e-06,
1213
+ "loss": 0.0122,
1214
+ "step": 1990
1215
+ },
1216
+ {
1217
+ "epoch": 0.4,
1218
+ "learning_rate": 9.604664953548133e-06,
1219
+ "loss": 0.0132,
1220
+ "step": 2000
1221
+ },
1222
+ {
1223
+ "batch_size": 128,
1224
+ "epoch": 0.4,
1225
+ "step": 2000,
1226
+ "valid_mrr": 0.7407695795617386,
1227
+ "valid_mrr_wo_pooler": 0.7474766604446019,
1228
+ "valid_pr1": 0.6546875,
1229
+ "valid_pr1_wo_pooler": 0.66875,
1230
+ "valid_runtime": 17.8534,
1231
+ "valid_samples_per_second": 64.47,
1232
+ "valid_steps_per_second": 0.224
1233
+ },
1234
+ {
1235
+ "epoch": 0.4,
1236
+ "learning_rate": 9.602688278315874e-06,
1237
+ "loss": 0.0163,
1238
+ "step": 2010
1239
+ },
1240
+ {
1241
+ "epoch": 0.4,
1242
+ "learning_rate": 9.600711603083614e-06,
1243
+ "loss": 0.0128,
1244
+ "step": 2020
1245
+ },
1246
+ {
1247
+ "epoch": 0.4,
1248
+ "learning_rate": 9.598734927851356e-06,
1249
+ "loss": 0.0171,
1250
+ "step": 2030
1251
+ },
1252
+ {
1253
+ "epoch": 0.4,
1254
+ "learning_rate": 9.596758252619095e-06,
1255
+ "loss": 0.0118,
1256
+ "step": 2040
1257
+ },
1258
+ {
1259
+ "epoch": 0.41,
1260
+ "learning_rate": 9.594781577386835e-06,
1261
+ "loss": 0.0199,
1262
+ "step": 2050
1263
+ },
1264
+ {
1265
+ "epoch": 0.41,
1266
+ "learning_rate": 9.592804902154577e-06,
1267
+ "loss": 0.0107,
1268
+ "step": 2060
1269
+ },
1270
+ {
1271
+ "epoch": 0.41,
1272
+ "learning_rate": 9.590828226922317e-06,
1273
+ "loss": 0.0151,
1274
+ "step": 2070
1275
+ },
1276
+ {
1277
+ "epoch": 0.41,
1278
+ "learning_rate": 9.588851551690058e-06,
1279
+ "loss": 0.0154,
1280
+ "step": 2080
1281
+ },
1282
+ {
1283
+ "epoch": 0.41,
1284
+ "learning_rate": 9.5868748764578e-06,
1285
+ "loss": 0.0187,
1286
+ "step": 2090
1287
+ },
1288
+ {
1289
+ "epoch": 0.42,
1290
+ "learning_rate": 9.58489820122554e-06,
1291
+ "loss": 0.0147,
1292
+ "step": 2100
1293
+ },
1294
+ {
1295
+ "epoch": 0.42,
1296
+ "learning_rate": 9.582921525993281e-06,
1297
+ "loss": 0.0166,
1298
+ "step": 2110
1299
+ },
1300
+ {
1301
+ "epoch": 0.42,
1302
+ "learning_rate": 9.580944850761021e-06,
1303
+ "loss": 0.0166,
1304
+ "step": 2120
1305
+ },
1306
+ {
1307
+ "epoch": 0.42,
1308
+ "learning_rate": 9.57896817552876e-06,
1309
+ "loss": 0.0129,
1310
+ "step": 2130
1311
+ },
1312
+ {
1313
+ "epoch": 0.42,
1314
+ "learning_rate": 9.576991500296502e-06,
1315
+ "loss": 0.0122,
1316
+ "step": 2140
1317
+ },
1318
+ {
1319
+ "epoch": 0.42,
1320
+ "learning_rate": 9.575014825064242e-06,
1321
+ "loss": 0.012,
1322
+ "step": 2150
1323
+ },
1324
+ {
1325
+ "epoch": 0.43,
1326
+ "learning_rate": 9.573038149831984e-06,
1327
+ "loss": 0.0152,
1328
+ "step": 2160
1329
+ },
1330
+ {
1331
+ "epoch": 0.43,
1332
+ "learning_rate": 9.571061474599725e-06,
1333
+ "loss": 0.0121,
1334
+ "step": 2170
1335
+ },
1336
+ {
1337
+ "epoch": 0.43,
1338
+ "learning_rate": 9.569084799367465e-06,
1339
+ "loss": 0.0126,
1340
+ "step": 2180
1341
+ },
1342
+ {
1343
+ "epoch": 0.43,
1344
+ "learning_rate": 9.567108124135205e-06,
1345
+ "loss": 0.0161,
1346
+ "step": 2190
1347
+ },
1348
+ {
1349
+ "epoch": 0.43,
1350
+ "learning_rate": 9.565131448902946e-06,
1351
+ "loss": 0.0208,
1352
+ "step": 2200
1353
+ },
1354
+ {
1355
+ "epoch": 0.44,
1356
+ "learning_rate": 9.563154773670686e-06,
1357
+ "loss": 0.0127,
1358
+ "step": 2210
1359
+ },
1360
+ {
1361
+ "epoch": 0.44,
1362
+ "learning_rate": 9.561178098438428e-06,
1363
+ "loss": 0.0192,
1364
+ "step": 2220
1365
+ },
1366
+ {
1367
+ "epoch": 0.44,
1368
+ "learning_rate": 9.559201423206167e-06,
1369
+ "loss": 0.0156,
1370
+ "step": 2230
1371
+ },
1372
+ {
1373
+ "epoch": 0.44,
1374
+ "learning_rate": 9.557224747973909e-06,
1375
+ "loss": 0.0091,
1376
+ "step": 2240
1377
+ },
1378
+ {
1379
+ "epoch": 0.44,
1380
+ "learning_rate": 9.55524807274165e-06,
1381
+ "loss": 0.0136,
1382
+ "step": 2250
1383
+ },
1384
+ {
1385
+ "epoch": 0.45,
1386
+ "learning_rate": 9.55327139750939e-06,
1387
+ "loss": 0.0183,
1388
+ "step": 2260
1389
+ },
1390
+ {
1391
+ "epoch": 0.45,
1392
+ "learning_rate": 9.55129472227713e-06,
1393
+ "loss": 0.0164,
1394
+ "step": 2270
1395
+ },
1396
+ {
1397
+ "epoch": 0.45,
1398
+ "learning_rate": 9.549318047044872e-06,
1399
+ "loss": 0.0194,
1400
+ "step": 2280
1401
+ },
1402
+ {
1403
+ "epoch": 0.45,
1404
+ "learning_rate": 9.547341371812612e-06,
1405
+ "loss": 0.0156,
1406
+ "step": 2290
1407
+ },
1408
+ {
1409
+ "epoch": 0.45,
1410
+ "learning_rate": 9.545364696580353e-06,
1411
+ "loss": 0.0158,
1412
+ "step": 2300
1413
+ },
1414
+ {
1415
+ "epoch": 0.46,
1416
+ "learning_rate": 9.543388021348093e-06,
1417
+ "loss": 0.0095,
1418
+ "step": 2310
1419
+ },
1420
+ {
1421
+ "epoch": 0.46,
1422
+ "learning_rate": 9.541411346115834e-06,
1423
+ "loss": 0.0158,
1424
+ "step": 2320
1425
+ },
1426
+ {
1427
+ "epoch": 0.46,
1428
+ "learning_rate": 9.539434670883574e-06,
1429
+ "loss": 0.013,
1430
+ "step": 2330
1431
+ },
1432
+ {
1433
+ "epoch": 0.46,
1434
+ "learning_rate": 9.537457995651316e-06,
1435
+ "loss": 0.0134,
1436
+ "step": 2340
1437
+ },
1438
+ {
1439
+ "epoch": 0.46,
1440
+ "learning_rate": 9.535481320419056e-06,
1441
+ "loss": 0.0175,
1442
+ "step": 2350
1443
+ },
1444
+ {
1445
+ "epoch": 0.47,
1446
+ "learning_rate": 9.533504645186797e-06,
1447
+ "loss": 0.0122,
1448
+ "step": 2360
1449
+ },
1450
+ {
1451
+ "epoch": 0.47,
1452
+ "learning_rate": 9.531527969954537e-06,
1453
+ "loss": 0.0161,
1454
+ "step": 2370
1455
+ },
1456
+ {
1457
+ "epoch": 0.47,
1458
+ "learning_rate": 9.529551294722278e-06,
1459
+ "loss": 0.0125,
1460
+ "step": 2380
1461
+ },
1462
+ {
1463
+ "epoch": 0.47,
1464
+ "learning_rate": 9.527574619490018e-06,
1465
+ "loss": 0.0157,
1466
+ "step": 2390
1467
+ },
1468
+ {
1469
+ "epoch": 0.47,
1470
+ "learning_rate": 9.52559794425776e-06,
1471
+ "loss": 0.0142,
1472
+ "step": 2400
1473
+ },
1474
+ {
1475
+ "epoch": 0.48,
1476
+ "learning_rate": 9.5236212690255e-06,
1477
+ "loss": 0.0143,
1478
+ "step": 2410
1479
+ },
1480
+ {
1481
+ "epoch": 0.48,
1482
+ "learning_rate": 9.52164459379324e-06,
1483
+ "loss": 0.0161,
1484
+ "step": 2420
1485
+ },
1486
+ {
1487
+ "epoch": 0.48,
1488
+ "learning_rate": 9.519667918560981e-06,
1489
+ "loss": 0.0188,
1490
+ "step": 2430
1491
+ },
1492
+ {
1493
+ "epoch": 0.48,
1494
+ "learning_rate": 9.51769124332872e-06,
1495
+ "loss": 0.0111,
1496
+ "step": 2440
1497
+ },
1498
+ {
1499
+ "epoch": 0.48,
1500
+ "learning_rate": 9.515714568096462e-06,
1501
+ "loss": 0.0137,
1502
+ "step": 2450
1503
+ },
1504
+ {
1505
+ "epoch": 0.49,
1506
+ "learning_rate": 9.513737892864204e-06,
1507
+ "loss": 0.0138,
1508
+ "step": 2460
1509
+ },
1510
+ {
1511
+ "epoch": 0.49,
1512
+ "learning_rate": 9.511761217631944e-06,
1513
+ "loss": 0.0134,
1514
+ "step": 2470
1515
+ },
1516
+ {
1517
+ "epoch": 0.49,
1518
+ "learning_rate": 9.509784542399685e-06,
1519
+ "loss": 0.0139,
1520
+ "step": 2480
1521
+ },
1522
+ {
1523
+ "epoch": 0.49,
1524
+ "learning_rate": 9.507807867167425e-06,
1525
+ "loss": 0.0098,
1526
+ "step": 2490
1527
+ },
1528
+ {
1529
+ "epoch": 0.49,
1530
+ "learning_rate": 9.505831191935165e-06,
1531
+ "loss": 0.0116,
1532
+ "step": 2500
1533
+ },
1534
+ {
1535
+ "epoch": 0.5,
1536
+ "learning_rate": 9.503854516702906e-06,
1537
+ "loss": 0.0177,
1538
+ "step": 2510
1539
+ },
1540
+ {
1541
+ "epoch": 0.5,
1542
+ "learning_rate": 9.501877841470646e-06,
1543
+ "loss": 0.0125,
1544
+ "step": 2520
1545
+ },
1546
+ {
1547
+ "epoch": 0.5,
1548
+ "learning_rate": 9.499901166238388e-06,
1549
+ "loss": 0.01,
1550
+ "step": 2530
1551
+ },
1552
+ {
1553
+ "epoch": 0.5,
1554
+ "learning_rate": 9.49792449100613e-06,
1555
+ "loss": 0.0116,
1556
+ "step": 2540
1557
+ },
1558
+ {
1559
+ "epoch": 0.5,
1560
+ "learning_rate": 9.495947815773869e-06,
1561
+ "loss": 0.0099,
1562
+ "step": 2550
1563
+ },
1564
+ {
1565
+ "epoch": 0.51,
1566
+ "learning_rate": 9.49397114054161e-06,
1567
+ "loss": 0.0137,
1568
+ "step": 2560
1569
+ },
1570
+ {
1571
+ "epoch": 0.51,
1572
+ "learning_rate": 9.49199446530935e-06,
1573
+ "loss": 0.0133,
1574
+ "step": 2570
1575
+ },
1576
+ {
1577
+ "epoch": 0.51,
1578
+ "learning_rate": 9.49001779007709e-06,
1579
+ "loss": 0.0153,
1580
+ "step": 2580
1581
+ },
1582
+ {
1583
+ "epoch": 0.51,
1584
+ "learning_rate": 9.488041114844832e-06,
1585
+ "loss": 0.0142,
1586
+ "step": 2590
1587
+ },
1588
+ {
1589
+ "epoch": 0.51,
1590
+ "learning_rate": 9.486064439612572e-06,
1591
+ "loss": 0.0118,
1592
+ "step": 2600
1593
+ },
1594
+ {
1595
+ "epoch": 0.52,
1596
+ "learning_rate": 9.484087764380313e-06,
1597
+ "loss": 0.0138,
1598
+ "step": 2610
1599
+ },
1600
+ {
1601
+ "epoch": 0.52,
1602
+ "learning_rate": 9.482111089148055e-06,
1603
+ "loss": 0.0098,
1604
+ "step": 2620
1605
+ },
1606
+ {
1607
+ "epoch": 0.52,
1608
+ "learning_rate": 9.480134413915795e-06,
1609
+ "loss": 0.0172,
1610
+ "step": 2630
1611
+ },
1612
+ {
1613
+ "epoch": 0.52,
1614
+ "learning_rate": 9.478157738683536e-06,
1615
+ "loss": 0.015,
1616
+ "step": 2640
1617
+ },
1618
+ {
1619
+ "epoch": 0.52,
1620
+ "learning_rate": 9.476181063451276e-06,
1621
+ "loss": 0.0147,
1622
+ "step": 2650
1623
+ },
1624
+ {
1625
+ "epoch": 0.53,
1626
+ "learning_rate": 9.474204388219016e-06,
1627
+ "loss": 0.0126,
1628
+ "step": 2660
1629
+ },
1630
+ {
1631
+ "epoch": 0.53,
1632
+ "learning_rate": 9.472227712986757e-06,
1633
+ "loss": 0.0115,
1634
+ "step": 2670
1635
+ },
1636
+ {
1637
+ "epoch": 0.53,
1638
+ "learning_rate": 9.470251037754497e-06,
1639
+ "loss": 0.0137,
1640
+ "step": 2680
1641
+ },
1642
+ {
1643
+ "epoch": 0.53,
1644
+ "learning_rate": 9.468274362522239e-06,
1645
+ "loss": 0.0145,
1646
+ "step": 2690
1647
+ },
1648
+ {
1649
+ "epoch": 0.53,
1650
+ "learning_rate": 9.466297687289978e-06,
1651
+ "loss": 0.012,
1652
+ "step": 2700
1653
+ },
1654
+ {
1655
+ "epoch": 0.54,
1656
+ "learning_rate": 9.46432101205772e-06,
1657
+ "loss": 0.0154,
1658
+ "step": 2710
1659
+ },
1660
+ {
1661
+ "epoch": 0.54,
1662
+ "learning_rate": 9.462344336825461e-06,
1663
+ "loss": 0.0133,
1664
+ "step": 2720
1665
+ },
1666
+ {
1667
+ "epoch": 0.54,
1668
+ "learning_rate": 9.460367661593201e-06,
1669
+ "loss": 0.0108,
1670
+ "step": 2730
1671
+ },
1672
+ {
1673
+ "epoch": 0.54,
1674
+ "learning_rate": 9.458390986360941e-06,
1675
+ "loss": 0.0137,
1676
+ "step": 2740
1677
+ },
1678
+ {
1679
+ "epoch": 0.54,
1680
+ "learning_rate": 9.456414311128683e-06,
1681
+ "loss": 0.0146,
1682
+ "step": 2750
1683
+ },
1684
+ {
1685
+ "epoch": 0.55,
1686
+ "learning_rate": 9.454437635896422e-06,
1687
+ "loss": 0.0122,
1688
+ "step": 2760
1689
+ },
1690
+ {
1691
+ "epoch": 0.55,
1692
+ "learning_rate": 9.452460960664164e-06,
1693
+ "loss": 0.0135,
1694
+ "step": 2770
1695
+ },
1696
+ {
1697
+ "epoch": 0.55,
1698
+ "learning_rate": 9.450484285431904e-06,
1699
+ "loss": 0.0121,
1700
+ "step": 2780
1701
+ },
1702
+ {
1703
+ "epoch": 0.55,
1704
+ "learning_rate": 9.448507610199645e-06,
1705
+ "loss": 0.0167,
1706
+ "step": 2790
1707
+ },
1708
+ {
1709
+ "epoch": 0.55,
1710
+ "learning_rate": 9.446530934967387e-06,
1711
+ "loss": 0.0143,
1712
+ "step": 2800
1713
+ },
1714
+ {
1715
+ "epoch": 0.56,
1716
+ "learning_rate": 9.444554259735125e-06,
1717
+ "loss": 0.0081,
1718
+ "step": 2810
1719
+ },
1720
+ {
1721
+ "epoch": 0.56,
1722
+ "learning_rate": 9.442577584502867e-06,
1723
+ "loss": 0.0107,
1724
+ "step": 2820
1725
+ },
1726
+ {
1727
+ "epoch": 0.56,
1728
+ "learning_rate": 9.440600909270608e-06,
1729
+ "loss": 0.0177,
1730
+ "step": 2830
1731
+ },
1732
+ {
1733
+ "epoch": 0.56,
1734
+ "learning_rate": 9.438624234038348e-06,
1735
+ "loss": 0.0114,
1736
+ "step": 2840
1737
+ },
1738
+ {
1739
+ "epoch": 0.56,
1740
+ "learning_rate": 9.43664755880609e-06,
1741
+ "loss": 0.0146,
1742
+ "step": 2850
1743
+ },
1744
+ {
1745
+ "epoch": 0.57,
1746
+ "learning_rate": 9.43467088357383e-06,
1747
+ "loss": 0.0088,
1748
+ "step": 2860
1749
+ },
1750
+ {
1751
+ "epoch": 0.57,
1752
+ "learning_rate": 9.43269420834157e-06,
1753
+ "loss": 0.0123,
1754
+ "step": 2870
1755
+ },
1756
+ {
1757
+ "epoch": 0.57,
1758
+ "learning_rate": 9.43071753310931e-06,
1759
+ "loss": 0.0122,
1760
+ "step": 2880
1761
+ },
1762
+ {
1763
+ "epoch": 0.57,
1764
+ "learning_rate": 9.42874085787705e-06,
1765
+ "loss": 0.013,
1766
+ "step": 2890
1767
+ },
1768
+ {
1769
+ "epoch": 0.57,
1770
+ "learning_rate": 9.426764182644792e-06,
1771
+ "loss": 0.0181,
1772
+ "step": 2900
1773
+ },
1774
+ {
1775
+ "epoch": 0.58,
1776
+ "learning_rate": 9.424787507412533e-06,
1777
+ "loss": 0.0122,
1778
+ "step": 2910
1779
+ },
1780
+ {
1781
+ "epoch": 0.58,
1782
+ "learning_rate": 9.422810832180273e-06,
1783
+ "loss": 0.0157,
1784
+ "step": 2920
1785
+ },
1786
+ {
1787
+ "epoch": 0.58,
1788
+ "learning_rate": 9.420834156948015e-06,
1789
+ "loss": 0.0127,
1790
+ "step": 2930
1791
+ },
1792
+ {
1793
+ "epoch": 0.58,
1794
+ "learning_rate": 9.418857481715755e-06,
1795
+ "loss": 0.0121,
1796
+ "step": 2940
1797
+ },
1798
+ {
1799
+ "epoch": 0.58,
1800
+ "learning_rate": 9.416880806483496e-06,
1801
+ "loss": 0.0126,
1802
+ "step": 2950
1803
+ },
1804
+ {
1805
+ "epoch": 0.59,
1806
+ "learning_rate": 9.414904131251236e-06,
1807
+ "loss": 0.0171,
1808
+ "step": 2960
1809
+ },
1810
+ {
1811
+ "epoch": 0.59,
1812
+ "learning_rate": 9.412927456018976e-06,
1813
+ "loss": 0.0099,
1814
+ "step": 2970
1815
+ },
1816
+ {
1817
+ "epoch": 0.59,
1818
+ "learning_rate": 9.410950780786717e-06,
1819
+ "loss": 0.0183,
1820
+ "step": 2980
1821
+ },
1822
+ {
1823
+ "epoch": 0.59,
1824
+ "learning_rate": 9.408974105554459e-06,
1825
+ "loss": 0.0073,
1826
+ "step": 2990
1827
+ },
1828
+ {
1829
+ "epoch": 0.59,
1830
+ "learning_rate": 9.406997430322199e-06,
1831
+ "loss": 0.0121,
1832
+ "step": 3000
1833
+ },
1834
+ {
1835
+ "batch_size": 128,
1836
+ "epoch": 0.59,
1837
+ "step": 3000,
1838
+ "valid_mrr": 0.7545951753143683,
1839
+ "valid_mrr_wo_pooler": 0.7653977925713823,
1840
+ "valid_pr1": 0.6734375,
1841
+ "valid_pr1_wo_pooler": 0.6921875,
1842
+ "valid_runtime": 17.9025,
1843
+ "valid_samples_per_second": 64.293,
1844
+ "valid_steps_per_second": 0.223
1845
+ },
1846
+ {
1847
+ "epoch": 0.59,
1848
+ "learning_rate": 9.40502075508994e-06,
1849
+ "loss": 0.0185,
1850
+ "step": 3010
1851
+ },
1852
+ {
1853
+ "epoch": 0.6,
1854
+ "learning_rate": 9.40304407985768e-06,
1855
+ "loss": 0.0102,
1856
+ "step": 3020
1857
+ },
1858
+ {
1859
+ "epoch": 0.6,
1860
+ "learning_rate": 9.401067404625422e-06,
1861
+ "loss": 0.0105,
1862
+ "step": 3030
1863
+ },
1864
+ {
1865
+ "epoch": 0.6,
1866
+ "learning_rate": 9.399090729393161e-06,
1867
+ "loss": 0.0119,
1868
+ "step": 3040
1869
+ },
1870
+ {
1871
+ "epoch": 0.6,
1872
+ "learning_rate": 9.397114054160901e-06,
1873
+ "loss": 0.0096,
1874
+ "step": 3050
1875
+ },
1876
+ {
1877
+ "epoch": 0.6,
1878
+ "learning_rate": 9.395137378928643e-06,
1879
+ "loss": 0.0131,
1880
+ "step": 3060
1881
+ },
1882
+ {
1883
+ "epoch": 0.61,
1884
+ "learning_rate": 9.393160703696383e-06,
1885
+ "loss": 0.0183,
1886
+ "step": 3070
1887
+ },
1888
+ {
1889
+ "epoch": 0.61,
1890
+ "learning_rate": 9.391184028464124e-06,
1891
+ "loss": 0.01,
1892
+ "step": 3080
1893
+ },
1894
+ {
1895
+ "epoch": 0.61,
1896
+ "learning_rate": 9.389207353231866e-06,
1897
+ "loss": 0.0135,
1898
+ "step": 3090
1899
+ },
1900
+ {
1901
+ "epoch": 0.61,
1902
+ "learning_rate": 9.387230677999605e-06,
1903
+ "loss": 0.0141,
1904
+ "step": 3100
1905
+ },
1906
+ {
1907
+ "epoch": 0.61,
1908
+ "learning_rate": 9.385254002767345e-06,
1909
+ "loss": 0.0132,
1910
+ "step": 3110
1911
+ },
1912
+ {
1913
+ "epoch": 0.62,
1914
+ "learning_rate": 9.383277327535087e-06,
1915
+ "loss": 0.0108,
1916
+ "step": 3120
1917
+ },
1918
+ {
1919
+ "epoch": 0.62,
1920
+ "learning_rate": 9.381300652302827e-06,
1921
+ "loss": 0.0127,
1922
+ "step": 3130
1923
+ },
1924
+ {
1925
+ "epoch": 0.62,
1926
+ "learning_rate": 9.379323977070568e-06,
1927
+ "loss": 0.0121,
1928
+ "step": 3140
1929
+ },
1930
+ {
1931
+ "epoch": 0.62,
1932
+ "learning_rate": 9.377347301838308e-06,
1933
+ "loss": 0.0145,
1934
+ "step": 3150
1935
+ },
1936
+ {
1937
+ "epoch": 0.62,
1938
+ "learning_rate": 9.37537062660605e-06,
1939
+ "loss": 0.0116,
1940
+ "step": 3160
1941
+ },
1942
+ {
1943
+ "epoch": 0.63,
1944
+ "learning_rate": 9.373393951373791e-06,
1945
+ "loss": 0.0142,
1946
+ "step": 3170
1947
+ },
1948
+ {
1949
+ "epoch": 0.63,
1950
+ "learning_rate": 9.371417276141531e-06,
1951
+ "loss": 0.0109,
1952
+ "step": 3180
1953
+ },
1954
+ {
1955
+ "epoch": 0.63,
1956
+ "learning_rate": 9.36944060090927e-06,
1957
+ "loss": 0.0175,
1958
+ "step": 3190
1959
+ },
1960
+ {
1961
+ "epoch": 0.63,
1962
+ "learning_rate": 9.367463925677012e-06,
1963
+ "loss": 0.0134,
1964
+ "step": 3200
1965
+ },
1966
+ {
1967
+ "epoch": 0.63,
1968
+ "learning_rate": 9.365487250444752e-06,
1969
+ "loss": 0.0123,
1970
+ "step": 3210
1971
+ },
1972
+ {
1973
+ "epoch": 0.64,
1974
+ "learning_rate": 9.363510575212494e-06,
1975
+ "loss": 0.0112,
1976
+ "step": 3220
1977
+ },
1978
+ {
1979
+ "epoch": 0.64,
1980
+ "learning_rate": 9.361533899980233e-06,
1981
+ "loss": 0.009,
1982
+ "step": 3230
1983
+ },
1984
+ {
1985
+ "epoch": 0.64,
1986
+ "learning_rate": 9.359557224747975e-06,
1987
+ "loss": 0.0131,
1988
+ "step": 3240
1989
+ },
1990
+ {
1991
+ "epoch": 0.64,
1992
+ "learning_rate": 9.357580549515716e-06,
1993
+ "loss": 0.0116,
1994
+ "step": 3250
1995
+ },
1996
+ {
1997
+ "epoch": 0.64,
1998
+ "learning_rate": 9.355603874283456e-06,
1999
+ "loss": 0.0115,
2000
+ "step": 3260
2001
+ },
2002
+ {
2003
+ "epoch": 0.65,
2004
+ "learning_rate": 9.353627199051196e-06,
2005
+ "loss": 0.0138,
2006
+ "step": 3270
2007
+ },
2008
+ {
2009
+ "epoch": 0.65,
2010
+ "learning_rate": 9.351650523818938e-06,
2011
+ "loss": 0.0089,
2012
+ "step": 3280
2013
+ },
2014
+ {
2015
+ "epoch": 0.65,
2016
+ "learning_rate": 9.349673848586677e-06,
2017
+ "loss": 0.0102,
2018
+ "step": 3290
2019
+ },
2020
+ {
2021
+ "epoch": 0.65,
2022
+ "learning_rate": 9.347697173354419e-06,
2023
+ "loss": 0.0157,
2024
+ "step": 3300
2025
+ },
2026
+ {
2027
+ "epoch": 0.65,
2028
+ "learning_rate": 9.345720498122159e-06,
2029
+ "loss": 0.0101,
2030
+ "step": 3310
2031
+ },
2032
+ {
2033
+ "epoch": 0.66,
2034
+ "learning_rate": 9.3437438228899e-06,
2035
+ "loss": 0.0091,
2036
+ "step": 3320
2037
+ },
2038
+ {
2039
+ "epoch": 0.66,
2040
+ "learning_rate": 9.34176714765764e-06,
2041
+ "loss": 0.01,
2042
+ "step": 3330
2043
+ },
2044
+ {
2045
+ "epoch": 0.66,
2046
+ "learning_rate": 9.33979047242538e-06,
2047
+ "loss": 0.0081,
2048
+ "step": 3340
2049
+ },
2050
+ {
2051
+ "epoch": 0.66,
2052
+ "learning_rate": 9.337813797193122e-06,
2053
+ "loss": 0.0098,
2054
+ "step": 3350
2055
+ },
2056
+ {
2057
+ "epoch": 0.66,
2058
+ "learning_rate": 9.335837121960863e-06,
2059
+ "loss": 0.0153,
2060
+ "step": 3360
2061
+ },
2062
+ {
2063
+ "epoch": 0.67,
2064
+ "learning_rate": 9.333860446728603e-06,
2065
+ "loss": 0.012,
2066
+ "step": 3370
2067
+ },
2068
+ {
2069
+ "epoch": 0.67,
2070
+ "learning_rate": 9.331883771496344e-06,
2071
+ "loss": 0.0159,
2072
+ "step": 3380
2073
+ },
2074
+ {
2075
+ "epoch": 0.67,
2076
+ "learning_rate": 9.329907096264084e-06,
2077
+ "loss": 0.0129,
2078
+ "step": 3390
2079
+ },
2080
+ {
2081
+ "epoch": 0.67,
2082
+ "learning_rate": 9.327930421031826e-06,
2083
+ "loss": 0.0122,
2084
+ "step": 3400
2085
+ },
2086
+ {
2087
+ "epoch": 0.67,
2088
+ "learning_rate": 9.325953745799566e-06,
2089
+ "loss": 0.0132,
2090
+ "step": 3410
2091
+ },
2092
+ {
2093
+ "epoch": 0.68,
2094
+ "learning_rate": 9.323977070567305e-06,
2095
+ "loss": 0.0137,
2096
+ "step": 3420
2097
+ },
2098
+ {
2099
+ "epoch": 0.68,
2100
+ "learning_rate": 9.322000395335047e-06,
2101
+ "loss": 0.0113,
2102
+ "step": 3430
2103
+ },
2104
+ {
2105
+ "epoch": 0.68,
2106
+ "learning_rate": 9.320023720102788e-06,
2107
+ "loss": 0.0144,
2108
+ "step": 3440
2109
+ },
2110
+ {
2111
+ "epoch": 0.68,
2112
+ "learning_rate": 9.318047044870528e-06,
2113
+ "loss": 0.01,
2114
+ "step": 3450
2115
+ },
2116
+ {
2117
+ "epoch": 0.68,
2118
+ "learning_rate": 9.31607036963827e-06,
2119
+ "loss": 0.0118,
2120
+ "step": 3460
2121
+ },
2122
+ {
2123
+ "epoch": 0.69,
2124
+ "learning_rate": 9.31409369440601e-06,
2125
+ "loss": 0.0112,
2126
+ "step": 3470
2127
+ },
2128
+ {
2129
+ "epoch": 0.69,
2130
+ "learning_rate": 9.312117019173751e-06,
2131
+ "loss": 0.0115,
2132
+ "step": 3480
2133
+ },
2134
+ {
2135
+ "epoch": 0.69,
2136
+ "learning_rate": 9.310140343941491e-06,
2137
+ "loss": 0.0147,
2138
+ "step": 3490
2139
+ },
2140
+ {
2141
+ "epoch": 0.69,
2142
+ "learning_rate": 9.308163668709231e-06,
2143
+ "loss": 0.0179,
2144
+ "step": 3500
2145
+ },
2146
+ {
2147
+ "epoch": 0.69,
2148
+ "learning_rate": 9.306186993476972e-06,
2149
+ "loss": 0.0149,
2150
+ "step": 3510
2151
+ },
2152
+ {
2153
+ "epoch": 0.7,
2154
+ "learning_rate": 9.304210318244712e-06,
2155
+ "loss": 0.0098,
2156
+ "step": 3520
2157
+ },
2158
+ {
2159
+ "epoch": 0.7,
2160
+ "learning_rate": 9.302233643012454e-06,
2161
+ "loss": 0.0099,
2162
+ "step": 3530
2163
+ },
2164
+ {
2165
+ "epoch": 0.7,
2166
+ "learning_rate": 9.300256967780195e-06,
2167
+ "loss": 0.0132,
2168
+ "step": 3540
2169
+ },
2170
+ {
2171
+ "epoch": 0.7,
2172
+ "learning_rate": 9.298280292547935e-06,
2173
+ "loss": 0.0088,
2174
+ "step": 3550
2175
+ },
2176
+ {
2177
+ "epoch": 0.7,
2178
+ "learning_rate": 9.296303617315677e-06,
2179
+ "loss": 0.0102,
2180
+ "step": 3560
2181
+ },
2182
+ {
2183
+ "epoch": 0.71,
2184
+ "learning_rate": 9.294326942083416e-06,
2185
+ "loss": 0.0135,
2186
+ "step": 3570
2187
+ },
2188
+ {
2189
+ "epoch": 0.71,
2190
+ "learning_rate": 9.292350266851156e-06,
2191
+ "loss": 0.0126,
2192
+ "step": 3580
2193
+ },
2194
+ {
2195
+ "epoch": 0.71,
2196
+ "learning_rate": 9.290373591618898e-06,
2197
+ "loss": 0.0088,
2198
+ "step": 3590
2199
+ },
2200
+ {
2201
+ "epoch": 0.71,
2202
+ "learning_rate": 9.288396916386638e-06,
2203
+ "loss": 0.0089,
2204
+ "step": 3600
2205
+ },
2206
+ {
2207
+ "epoch": 0.71,
2208
+ "learning_rate": 9.286420241154379e-06,
2209
+ "loss": 0.0149,
2210
+ "step": 3610
2211
+ },
2212
+ {
2213
+ "epoch": 0.72,
2214
+ "learning_rate": 9.28444356592212e-06,
2215
+ "loss": 0.0115,
2216
+ "step": 3620
2217
+ },
2218
+ {
2219
+ "epoch": 0.72,
2220
+ "learning_rate": 9.28246689068986e-06,
2221
+ "loss": 0.0105,
2222
+ "step": 3630
2223
+ },
2224
+ {
2225
+ "epoch": 0.72,
2226
+ "learning_rate": 9.280490215457602e-06,
2227
+ "loss": 0.0133,
2228
+ "step": 3640
2229
+ },
2230
+ {
2231
+ "epoch": 0.72,
2232
+ "learning_rate": 9.278513540225342e-06,
2233
+ "loss": 0.0145,
2234
+ "step": 3650
2235
+ },
2236
+ {
2237
+ "epoch": 0.72,
2238
+ "learning_rate": 9.276536864993082e-06,
2239
+ "loss": 0.014,
2240
+ "step": 3660
2241
+ },
2242
+ {
2243
+ "epoch": 0.73,
2244
+ "learning_rate": 9.274560189760823e-06,
2245
+ "loss": 0.01,
2246
+ "step": 3670
2247
+ },
2248
+ {
2249
+ "epoch": 0.73,
2250
+ "learning_rate": 9.272583514528563e-06,
2251
+ "loss": 0.0131,
2252
+ "step": 3680
2253
+ },
2254
+ {
2255
+ "epoch": 0.73,
2256
+ "learning_rate": 9.270606839296305e-06,
2257
+ "loss": 0.0133,
2258
+ "step": 3690
2259
+ },
2260
+ {
2261
+ "epoch": 0.73,
2262
+ "learning_rate": 9.268630164064046e-06,
2263
+ "loss": 0.0123,
2264
+ "step": 3700
2265
+ },
2266
+ {
2267
+ "epoch": 0.73,
2268
+ "learning_rate": 9.266653488831786e-06,
2269
+ "loss": 0.0113,
2270
+ "step": 3710
2271
+ },
2272
+ {
2273
+ "epoch": 0.74,
2274
+ "learning_rate": 9.264676813599527e-06,
2275
+ "loss": 0.01,
2276
+ "step": 3720
2277
+ },
2278
+ {
2279
+ "epoch": 0.74,
2280
+ "learning_rate": 9.262700138367267e-06,
2281
+ "loss": 0.0124,
2282
+ "step": 3730
2283
+ },
2284
+ {
2285
+ "epoch": 0.74,
2286
+ "learning_rate": 9.260723463135007e-06,
2287
+ "loss": 0.0147,
2288
+ "step": 3740
2289
+ },
2290
+ {
2291
+ "epoch": 0.74,
2292
+ "learning_rate": 9.258746787902749e-06,
2293
+ "loss": 0.0096,
2294
+ "step": 3750
2295
+ },
2296
+ {
2297
+ "epoch": 0.74,
2298
+ "learning_rate": 9.256770112670488e-06,
2299
+ "loss": 0.0101,
2300
+ "step": 3760
2301
+ },
2302
+ {
2303
+ "epoch": 0.75,
2304
+ "learning_rate": 9.25479343743823e-06,
2305
+ "loss": 0.0124,
2306
+ "step": 3770
2307
+ },
2308
+ {
2309
+ "epoch": 0.75,
2310
+ "learning_rate": 9.25281676220597e-06,
2311
+ "loss": 0.0132,
2312
+ "step": 3780
2313
+ },
2314
+ {
2315
+ "epoch": 0.75,
2316
+ "learning_rate": 9.250840086973711e-06,
2317
+ "loss": 0.0112,
2318
+ "step": 3790
2319
+ },
2320
+ {
2321
+ "epoch": 0.75,
2322
+ "learning_rate": 9.248863411741453e-06,
2323
+ "loss": 0.0119,
2324
+ "step": 3800
2325
+ },
2326
+ {
2327
+ "epoch": 0.75,
2328
+ "learning_rate": 9.246886736509193e-06,
2329
+ "loss": 0.0141,
2330
+ "step": 3810
2331
+ },
2332
+ {
2333
+ "epoch": 0.76,
2334
+ "learning_rate": 9.244910061276933e-06,
2335
+ "loss": 0.0102,
2336
+ "step": 3820
2337
+ },
2338
+ {
2339
+ "epoch": 0.76,
2340
+ "learning_rate": 9.242933386044674e-06,
2341
+ "loss": 0.0123,
2342
+ "step": 3830
2343
+ },
2344
+ {
2345
+ "epoch": 0.76,
2346
+ "learning_rate": 9.240956710812414e-06,
2347
+ "loss": 0.0146,
2348
+ "step": 3840
2349
+ },
2350
+ {
2351
+ "epoch": 0.76,
2352
+ "learning_rate": 9.238980035580155e-06,
2353
+ "loss": 0.0052,
2354
+ "step": 3850
2355
+ },
2356
+ {
2357
+ "epoch": 0.76,
2358
+ "learning_rate": 9.237003360347895e-06,
2359
+ "loss": 0.0113,
2360
+ "step": 3860
2361
+ },
2362
+ {
2363
+ "epoch": 0.76,
2364
+ "learning_rate": 9.235026685115637e-06,
2365
+ "loss": 0.0126,
2366
+ "step": 3870
2367
+ },
2368
+ {
2369
+ "epoch": 0.77,
2370
+ "learning_rate": 9.233050009883377e-06,
2371
+ "loss": 0.0103,
2372
+ "step": 3880
2373
+ },
2374
+ {
2375
+ "epoch": 0.77,
2376
+ "learning_rate": 9.231073334651116e-06,
2377
+ "loss": 0.0117,
2378
+ "step": 3890
2379
+ },
2380
+ {
2381
+ "epoch": 0.77,
2382
+ "learning_rate": 9.229096659418858e-06,
2383
+ "loss": 0.0144,
2384
+ "step": 3900
2385
+ },
2386
+ {
2387
+ "epoch": 0.77,
2388
+ "learning_rate": 9.2271199841866e-06,
2389
+ "loss": 0.0149,
2390
+ "step": 3910
2391
+ },
2392
+ {
2393
+ "epoch": 0.77,
2394
+ "learning_rate": 9.22514330895434e-06,
2395
+ "loss": 0.0112,
2396
+ "step": 3920
2397
+ },
2398
+ {
2399
+ "epoch": 0.78,
2400
+ "learning_rate": 9.22316663372208e-06,
2401
+ "loss": 0.0111,
2402
+ "step": 3930
2403
+ },
2404
+ {
2405
+ "epoch": 0.78,
2406
+ "learning_rate": 9.22118995848982e-06,
2407
+ "loss": 0.0091,
2408
+ "step": 3940
2409
+ },
2410
+ {
2411
+ "epoch": 0.78,
2412
+ "learning_rate": 9.219213283257562e-06,
2413
+ "loss": 0.0142,
2414
+ "step": 3950
2415
+ },
2416
+ {
2417
+ "epoch": 0.78,
2418
+ "learning_rate": 9.217236608025302e-06,
2419
+ "loss": 0.0099,
2420
+ "step": 3960
2421
+ },
2422
+ {
2423
+ "epoch": 0.78,
2424
+ "learning_rate": 9.215259932793042e-06,
2425
+ "loss": 0.0108,
2426
+ "step": 3970
2427
+ },
2428
+ {
2429
+ "epoch": 0.79,
2430
+ "learning_rate": 9.213283257560783e-06,
2431
+ "loss": 0.0128,
2432
+ "step": 3980
2433
+ },
2434
+ {
2435
+ "epoch": 0.79,
2436
+ "learning_rate": 9.211306582328525e-06,
2437
+ "loss": 0.0134,
2438
+ "step": 3990
2439
+ },
2440
+ {
2441
+ "epoch": 0.79,
2442
+ "learning_rate": 9.209329907096265e-06,
2443
+ "loss": 0.0151,
2444
+ "step": 4000
2445
+ },
2446
+ {
2447
+ "batch_size": 128,
2448
+ "epoch": 0.79,
2449
+ "step": 4000,
2450
+ "valid_mrr": 0.7539704627359516,
2451
+ "valid_mrr_wo_pooler": 0.7610087812537554,
2452
+ "valid_pr1": 0.671875,
2453
+ "valid_pr1_wo_pooler": 0.6859375,
2454
+ "valid_runtime": 17.9577,
2455
+ "valid_samples_per_second": 64.095,
2456
+ "valid_steps_per_second": 0.223
2457
+ }
2458
+ ],
2459
+ "max_steps": 50590,
2460
+ "num_train_epochs": 10,
2461
+ "total_flos": 2.02065661919232e+17,
2462
+ "trial_name": null,
2463
+ "trial_params": null
2464
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adb073c2a5b234f3bc54d972a5640ab2a194b7b21ab9faecd750cd095f8a1e6b
3
+ size 3963