jstAnotherCapi commited on
Commit
d2a5f53
·
verified ·
1 Parent(s): 2003e5b

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ epoch2/error.log filter=lfs diff=lfs merge=lfs -text
epoch2/error.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e13044c2943acf7de381650f9b9e82eda0a8d8450ec29408f75290a3cdfe487
3
+ size 17162506
epoch2/logs.txt ADDED
@@ -0,0 +1,500 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-02-24 17:30:09 - Load pretrained SentenceTransformer: bert-base-arabertv02
2
+ 2026-02-24 17:30:19 - '[Errno -2] Name or service not known' thrown while requesting HEAD https://huggingface.co/bert-base-arabertv02/resolve/main/./modules.json
3
+ 2026-02-24 17:30:19 - Retrying in 1s [Retry 1/5].
4
+ 2026-02-24 17:30:20 - No sentence-transformers model found with name bert-base-arabertv02. Creating a new one with mean pooling.
5
+ Model is running on: cuda
6
+ 2026-02-24 17:30:25 - Reading the training and eval dataset
7
+ 2026-02-24 17:31:43 - DatasetDict({
8
+ train: Dataset({
9
+ features: ['anchor', 'positive', 'negative'],
10
+ num_rows: 3954179
11
+ })
12
+ })
13
+ 2026-02-24 17:31:43 - DatasetDict({
14
+ train: Dataset({
15
+ features: ['anchor', 'positive', 'negative'],
16
+ num_rows: 1129759
17
+ })
18
+ })
19
+ 2026-02-24 17:31:43 - DatasetDict({
20
+ train: Dataset({
21
+ features: ['anchor', 'positive', 'negative'],
22
+ num_rows: 564877
23
+ })
24
+ })
25
+ 2026-02-24 17:31:43 - TripletEvaluator: Evaluating the model on the dev-768 dataset (truncated to 768):
26
+ 2026-02-24 17:55:01 - Accuracy Cosine Similarity: 79.01%
27
+ 2026-02-24 17:55:01 - TripletEvaluator: Evaluating the model on the dev-512 dataset (truncated to 512):
28
+ 2026-02-24 18:17:34 - Accuracy Cosine Similarity: 77.91%
29
+ 2026-02-24 18:17:34 - TripletEvaluator: Evaluating the model on the dev-256 dataset (truncated to 256):
30
+ 2026-02-24 18:39:41 - Accuracy Cosine Similarity: 79.57%
31
+ 2026-02-24 18:39:41 - TripletEvaluator: Evaluating the model on the dev-128 dataset (truncated to 128):
32
+ 2026-02-24 19:02:02 - Accuracy Cosine Similarity: 78.63%
33
+ 2026-02-24 19:02:02 - TripletEvaluator: Evaluating the model on the dev-64 dataset (truncated to 64):
34
+ 2026-02-24 19:24:20 - Accuracy Cosine Similarity: 76.00%
35
+ {'loss': '13.01', 'grad_norm': '27.4', 'learning_rate': '6.441e-07', 'epoch': '0.006474'}
36
+ {'loss': '6.428', 'grad_norm': '19.47', 'learning_rate': '1.291e-06', 'epoch': '0.01295'}
37
+ {'loss': '4.365', 'grad_norm': '17.72', 'learning_rate': '1.939e-06', 'epoch': '0.01942'}
38
+ {'loss': '3.585', 'grad_norm': '14.64', 'learning_rate': '2.586e-06', 'epoch': '0.0259'}
39
+ {'loss': '3.183', 'grad_norm': '13.4', 'learning_rate': '3.234e-06', 'epoch': '0.03237'}
40
+ {'loss': '2.873', 'grad_norm': '10.47', 'learning_rate': '3.881e-06', 'epoch': '0.03884'}
41
+ {'loss': '2.634', 'grad_norm': '12.2', 'learning_rate': '4.528e-06', 'epoch': '0.04532'}
42
+ {'loss': '2.605', 'grad_norm': '12.74', 'learning_rate': '5.176e-06', 'epoch': '0.05179'}
43
+ {'loss': '2.31', 'grad_norm': '12.29', 'learning_rate': '5.823e-06', 'epoch': '0.05827'}
44
+ {'loss': '2.236', 'grad_norm': '10.22', 'learning_rate': '6.47e-06', 'epoch': '0.06474'}
45
+ {'loss': '2.155', 'grad_norm': '9.492', 'learning_rate': '7.118e-06', 'epoch': '0.07121'}
46
+ {'loss': '2.019', 'grad_norm': '9.747', 'learning_rate': '7.765e-06', 'epoch': '0.07769'}
47
+ {'loss': '1.926', 'grad_norm': '9.491', 'learning_rate': '8.412e-06', 'epoch': '0.08416'}
48
+ {'loss': '1.927', 'grad_norm': '8.972', 'learning_rate': '9.06e-06', 'epoch': '0.09064'}
49
+ {'loss': '1.866', 'grad_norm': '10.02', 'learning_rate': '9.707e-06', 'epoch': '0.09711'}
50
+ {'loss': '1.796', 'grad_norm': '9.32', 'learning_rate': '1.035e-05', 'epoch': '0.1036'}
51
+ {'loss': '1.731', 'grad_norm': '8.47', 'learning_rate': '1.1e-05', 'epoch': '0.1101'}
52
+ {'loss': '1.725', 'grad_norm': '7.867', 'learning_rate': '1.165e-05', 'epoch': '0.1165'}
53
+ {'loss': '1.619', 'grad_norm': '9.521', 'learning_rate': '1.23e-05', 'epoch': '0.123'}
54
+ {'loss': '1.634', 'grad_norm': '9.193', 'learning_rate': '1.294e-05', 'epoch': '0.1295'}
55
+ {'loss': '1.604', 'grad_norm': '9.547', 'learning_rate': '1.359e-05', 'epoch': '0.136'}
56
+ {'loss': '1.548', 'grad_norm': '8.025', 'learning_rate': '1.424e-05', 'epoch': '0.1424'}
57
+ {'loss': '1.542', 'grad_norm': '8.197', 'learning_rate': '1.489e-05', 'epoch': '0.1489'}
58
+ {'loss': '1.507', 'grad_norm': '7.923', 'learning_rate': '1.553e-05', 'epoch': '0.1554'}
59
+ {'loss': '1.48', 'grad_norm': '7.746', 'learning_rate': '1.618e-05', 'epoch': '0.1619'}
60
+ {'loss': '1.462', 'grad_norm': '10.12', 'learning_rate': '1.683e-05', 'epoch': '0.1683'}
61
+ {'loss': '1.446', 'grad_norm': '7.806', 'learning_rate': '1.748e-05', 'epoch': '0.1748'}
62
+ {'loss': '1.424', 'grad_norm': '6.649', 'learning_rate': '1.812e-05', 'epoch': '0.1813'}
63
+ {'loss': '1.393', 'grad_norm': '7.031', 'learning_rate': '1.877e-05', 'epoch': '0.1877'}
64
+ {'loss': '1.352', 'grad_norm': '6.146', 'learning_rate': '1.942e-05', 'epoch': '0.1942'}
65
+ 2026-02-24 20:40:33 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 0.19422189851905802 after 6000 steps (truncated to 768):
66
+ 2026-02-24 21:08:51 - Accuracy Cosine Similarity: 95.64%
67
+ 2026-02-24 21:08:51 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 0.19422189851905802 after 6000 steps (truncated to 512):
68
+ 2026-02-24 21:36:55 - Accuracy Cosine Similarity: 95.66%
69
+ 2026-02-24 21:36:55 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 0.19422189851905802 after 6000 steps (truncated to 256):
70
+ 2026-02-24 22:05:02 - Accuracy Cosine Similarity: 95.60%
71
+ 2026-02-24 22:05:02 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 0.19422189851905802 after 6000 steps (truncated to 128):
72
+ 2026-02-24 22:33:12 - Accuracy Cosine Similarity: 95.46%
73
+ 2026-02-24 22:33:12 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 0.19422189851905802 after 6000 steps (truncated to 64):
74
+ 2026-02-24 23:01:40 - Accuracy Cosine Similarity: 95.13%
75
+ {'eval_train_loss': '1.216', 'eval_dev-768_cosine_accuracy': '0.9564', 'eval_dev-512_cosine_accuracy': '0.9566', 'eval_dev-256_cosine_accuracy': '0.956', 'eval_dev-128_cosine_accuracy': '0.9546', 'eval_dev-64_cosine_accuracy': '0.9513', 'eval_sequential_score': '0.9564', 'eval_train_runtime': '9876', 'eval_train_samples_per_second': '114.4', 'eval_train_steps_per_second': '1.787', 'epoch': '0.1942'}
76
+ 2026-02-24 23:01:40 - Saving model checkpoint to output/arabert_20260224_1730/checkpoint-6000
77
+ 2026-02-24 23:01:40 - Save model to output/arabert_20260224_1730/checkpoint-6000
78
+ {'loss': '1.372', 'grad_norm': '8.833', 'learning_rate': '1.999e-05', 'epoch': '0.2007'}
79
+ {'loss': '1.37', 'grad_norm': '6.729', 'learning_rate': '1.992e-05', 'epoch': '0.2072'}
80
+ {'loss': '1.352', 'grad_norm': '6.586', 'learning_rate': '1.985e-05', 'epoch': '0.2136'}
81
+ {'loss': '1.288', 'grad_norm': '6.568', 'learning_rate': '1.978e-05', 'epoch': '0.2201'}
82
+ {'loss': '1.271', 'grad_norm': '6.606', 'learning_rate': '1.971e-05', 'epoch': '0.2266'}
83
+ {'loss': '1.243', 'grad_norm': '7.913', 'learning_rate': '1.963e-05', 'epoch': '0.2331'}
84
+ {'loss': '1.23', 'grad_norm': '6.931', 'learning_rate': '1.956e-05', 'epoch': '0.2395'}
85
+ {'loss': '1.212', 'grad_norm': '6.979', 'learning_rate': '1.949e-05', 'epoch': '0.246'}
86
+ {'loss':{'loss': '1.37', 'grad_norm': '6.732', 'learning_rate': '1.992e-05', 'epoch': '0.2072'}
87
+ {'loss': '1.352', 'grad_norm': '6.583', 'learning_rate': '1.985e-05'{'loss': '1.288', 'grad_norm': '6.564', 'learning_rate': '1.978e-05', 'epoch': '0.2201'}
88
+ ':{'loss': '1.271', 'grad_norm': '6.606', 'learning_rate': '1.971e-05', 'epoch': '0.2266'}
89
+ {'loss': '1.243', 'grad_norm': '7.906', 'learning_rate': '1.963e-05', 'epoch': '0.2331'}
90
+ {'loss': '1.23', 'grad_norm': '6.921', 'learning_rate': '1.956e-05', 'epoch': '0.2395'}
91
+ {'loss': '1.212', 'grad_norm': '6.985', 'learning_rate': '1.949e-05', 'epoch': '0.246'}
92
+ {'loss': '1.222', 'grad_norm': '7.511', 'learning_rate': '1.942e-05', 'epoch': '0.2525'}
93
+ {'loss': '1.212', 'grad_norm': '7.834', 'learning_rate': '1.935e-05', 'epoch': '0.259'}
94
+ {'loss': '1.199', 'grad_norm': '5.273', 'learning_rate': '1.927e-05', 'epoch': '0.2654'}
95
+ {'loss': '1.207', 'grad_norm': '7.454', 'learning_rate': '1.92e-05', 'epoch': '0.2719'}
96
+ {'loss': '1.197', 'grad_norm': '7.585', 'learning_rate': '1.913e-05', 'epoch': '0.2784'}
97
+ {'loss': '1.128', 'grad_norm': '7.16', 'learning_rate': '1.906e-05', 'epoch': '0.2849'}
98
+ {'loss': '1.139', 'grad_norm': '6.354', 'learning_rate': '1.899e-05', 'epoch': '0.2913'}
99
+ {'loss': '1.197', 'grad_norm': '8.433', 'learning_rate': '1.891e-05', 'epoch': '0.2978'}
100
+ {'loss': '1.099', 'grad_norm': '5.537', 'learning_rate': '1.884e-05', 'epoch': '0.3043'}
101
+ {'loss': '1.133', 'grad_norm': '7.014', 'learning_rate': '1.877e-05', 'epoch': '0.3108'}
102
+ {'loss': '1.104', 'grad_norm': '7.172', 'learning_rate': '1.87e-05', 'epoch': '0.3172'}
103
+ {'loss': '1.116', 'grad_norm': '7.246', 'learning_rate': '1.863e-05', 'epoch': '0.3237'}
104
+ {'loss': '1.083', 'grad_norm': '7.252', 'learning_rate': '1.855e-05', 'epoch': '0.3302'}
105
+ {'loss': '1.043', 'grad_norm': '9.376', 'learning_rate': '1.848e-05', 'epoch': '0.3367'}
106
+ {'loss': '1.104', 'grad_norm': '6.334', 'learning_rate': '1.841e-05', 'epoch': '0.3431'}
107
+ {'loss': '1.074', 'grad_norm': '8.013', 'learning_rate': '1.834e-05', 'epoch': '0.3496'}
108
+ {'loss': '1.075', 'grad_norm': '6.887', 'learning_rate': '1.827e-05', 'epoch': '0.3561'}
109
+ {'loss': '1.059', 'grad_norm': '6.77', 'learning_rate': '1.819e-05', 'epoch': '0.3625'}
110
+ {'loss': '1.063', 'grad_norm': '5.884', 'learning_rate': '1.812e-05', 'epoch': '0.369'}
111
+ {'loss': '1.025', 'grad_norm': '4.584', 'learning_rate': '1.805e-05', 'epoch': '0.3755'}
112
+ {'loss': '1.049', 'grad_norm': '5.729', 'learning_rate': '1.798e-05', 'epoch': '0.382'}
113
+ {'loss': '1.045', 'grad_norm': '4.304', 'learning_rate': '1.791e-05', 'epoch': '0.3884'}
114
+ 2026-02-25 16:31:58 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 0.38844379703811605 after 12000 steps (truncated to 768):
115
+ 2026-02-25 16:53:08 - Accuracy Cosine Similarity: 96.68%
116
+ 2026-02-25 16:53:08 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 0.38844379703811605 after 12000 steps (truncated to 512):
117
+ 2026-02-25 17:13:37 - Accuracy Cosine Similarity: 96.67%
118
+ 2026-02-25 17:13:37 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 0.38844379703811605 after 12000 steps (truncated to 256):
119
+ 2026-02-25 17:34:09 - Accuracy Cosine Similarity: 96.64%
120
+ 2026-02-25 17:34:09 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 0.38844379703811605 after 12000 steps (truncated to 128):
121
+ 2026-02-25 17:54:44 - Accuracy Cosine Similarity: 96.56%
122
+ 2026-02-25 17:54:44 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 0.38844379703811605 after 12000 steps (truncated to 64):
123
+ 2026-02-25 18:15:17 - Accuracy Cosine Similarity: 96.28%
124
+ {'eval_train_loss': '0.5325', 'eval_dev-768_cosine_accuracy': '0.9668', 'eval_dev-512_cosine_accuracy': '0.9667', 'eval_dev-256_cosine_accuracy': '0.9664', 'eval_dev-128_cosine_accuracy': '0.9656', 'eval_dev-64_cosine_accuracy': '0.9628', 'eval_sequential_score': '0.9668', 'eval_train_runtime': '9281', 'eval_train_samples_per_second': '121.7', 'eval_train_steps_per_second': '15.22', 'epoch': '0.3884'}
125
+ 2026-02-25 18:15:17 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-12000
126
+ 2026-02-25 18:15:17 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-12000
127
+ {'loss': '0.9883', 'grad_norm': '5.449', 'learning_rate': '1.783e-05', 'epoch': '0.3949'}
128
+ {'loss': '0.9907', 'grad_norm': '5.808', 'learning_rate': '1.776e-05', 'epoch': '0.4014'}
129
+ {'loss': '1.022', 'grad_norm': '6.168', 'learning_rate': '1.769e-05', 'epoch': '0.4079'}
130
+ {'loss': '0.987', 'grad_norm': '5.675', 'learning_rate': '1.762e-05', 'epoch': '0.4143'}
131
+ {'loss': '1.033', 'grad_norm': '5.961', 'learning_rate': '1.755e-05', 'epoch': '0.4208'}
132
+ {'loss': '0.9989', 'grad_norm': '7.682', 'learning_rate': '1.748e-05', 'epoch': '0.4273'}
133
+ {'loss': '0.9805', 'grad_norm': '6.046', 'learning_rate': '1.74e-05', 'epoch': '0.4338'}
134
+ {'loss': '0.9484', 'grad_norm': '6.129', 'learning_rate': '1.733e-05', 'epoch': '0.4402'}
135
+ {'loss': '0.9937', 'grad_norm': '4.2', 'learning_rate': '1.726e-05', 'epoch': '0.4467'}
136
+ {'loss': '1.016', 'grad_norm': '7.164', 'learning_rate': '1.719e-05', 'epoch': '0.4532'}
137
+ {'loss': '0.9726', 'grad_norm': '5.905', 'learning_rate': '1.712e-05', 'epoch': '0.4597'}
138
+ {'loss': '0.96', 'grad_norm': '5.824', 'learning_rate': '1.704e-05', 'epoch': '0.4661'}
139
+ {'loss': '0.9528', 'grad_norm': '6.058', 'learning_rate': '1.697e-05', 'epoch': '0.4726'}
140
+ {'loss': '0.9292', 'grad_norm': '4.911', 'learning_rate': '1.69e-05', 'epoch': '0.4791'}
141
+ {'loss': '0.9157', 'grad_norm': '6.308', 'learning_rate': '1.683e-05', 'epoch': '0.4856'}
142
+ {'loss': '0.9244', 'grad_norm': '5.036', 'learning_rate': '1.676e-05', 'epoch': '0.492'}
143
+ {'loss': '0.9192', 'grad_norm': '3.666', 'learning_rate': '1.668e-05', 'epoch': '0.4985'}
144
+ {'loss': '0.9424', 'grad_norm': '4.06', 'learning_rate': '1.661e-05', 'epoch': '0.505'}
145
+ {'loss': '0.9067', 'grad_norm': '5.872', 'learning_rate': '1.654e-05', 'epoch': '0.5115'}
146
+ {'loss': '0.9334', 'grad_norm': '5.868', 'learning_rate': '1.647e-05', 'epoch': '0.5179'}
147
+ {'loss': '0.8922', 'grad_norm': '6.681', 'learning_rate': '1.64e-05', 'epoch': '0.5244'}
148
+ {'loss': '0.9122', 'grad_norm': '7.907', 'learning_rate': '1.632e-05', 'epoch': '0.5309'}
149
+ {'loss': '0.8825', 'grad_norm': '6.144', 'learning_rate': '1.625e-05', 'epoch': '0.5373'}
150
+ {'loss': '0.9069', 'grad_norm': '6.138', 'learning_rate': '1.618e-05', 'epoch': '0.5438'}
151
+ {'loss': '0.894', 'grad_norm': '6.968', 'learning_rate': '1.611e-05', 'epoch': '0.5503'}
152
+ {'loss': '0.8898', 'grad_norm': '6.487', 'learning_rate': '1.604e-05', 'epoch': '0.5568'}
153
+ {'loss': '0.8735', 'grad_norm': '4.058', 'learning_rate': '1.596e-05', 'epoch': '0.5632'}
154
+ {'loss': '0.8694', 'grad_norm': '5.403', 'learning_rate': '1.589e-05', 'epoch': '0.5697'}
155
+ {'loss': '0.8776', 'grad_norm': '6.723', 'learning_rate': '1.582e-05', 'epoch': '0.5762'}
156
+ {'loss': '0.8664', 'grad_norm': '4.427', 'learning_rate': '1.575e-05', 'epoch': '0.5827'}
157
+ 2026-02-25 20:05:35 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 0.5826656955571741 after 18000 steps (truncated to 768):
158
+ 2026-02-25 20:26:21 - Accuracy Cosine Similarity: 97.12%
159
+ 2026-02-25 20:26:21 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 0.5826656955571741 after 18000 steps (truncated to 512):
160
+ 2026-02-25 20:46:59 - Accuracy Cosine Similarity: 97.11%
161
+ 2026-02-25 20:46:59 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 0.5826656955571741 after 18000 steps (truncated to 256):
162
+ 2026-02-25 21:07:40 - Accuracy Cosine Similarity: 97.09%
163
+ 2026-02-25 21:07:40 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 0.5826656955571741 after 18000 steps (truncated to 128):
164
+ 2026-02-25 21:28:20 - Accuracy Cosine Similarity: 97.03%
165
+ 2026-02-25 21:28:20 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 0.5826656955571741 after 18000 steps (truncated to 64):
166
+ 2026-02-25 21:49:02 - Accuracy Cosine Similarity: 96.82%
167
+ {'eval_train_loss': '0.4541', 'eval_dev-768_cosine_accuracy': '0.9712', 'eval_dev-512_cosine_accuracy': '0.9711', 'eval_dev-256_cosine_accuracy': '0.9709', 'eval_dev-128_cosine_accuracy': '0.9703', 'eval_dev-64_cosine_accuracy': '0.9682', 'eval_sequential_score': '0.9712', 'eval_train_runtime': '9294', 'eval_train_samples_per_second': '121.6', 'eval_train_steps_per_second': '15.19', 'epoch': '0.5827'}
168
+ 2026-02-25 21:49:02 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-18000
169
+ 2026-02-25 21:49:02 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-18000
170
+
171
+ 2026-02-26 14:22:01 - Load pretrained SentenceTransformer: bert-base-arabertv02
172
+ 2026-02-26 14:22:14 - '[Errno -2] Name or service not known' thrown while requesting HEAD https://huggingface.co/bert-base-arabertv02/resolve/main/./modules.json
173
+ 2026-02-26 14:22:14 - Retrying in 1s [Retry 1/5].
174
+ 2026-02-26 14:22:15 - No sentence-transformers model found with name bert-base-arabertv02. Creating a new one with mean pooling.
175
+ 2026-02-26 14:23:53 - Use pytorch device_name: cuda:0
176
+ 2026-02-26 14:23:53 - Load pretrained SentenceTransformer: /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-18000
177
+ {'loss': '0.8727', 'grad_norm': '5.438', 'learning_rate': '1.568e-05', 'epoch': '0.5891'}
178
+ {'loss': '0.8524', 'grad_norm': '5.458', 'learning_rate': '1.56e-05', 'epoch': '0.5956'}
179
+ {'loss': '0.8995', 'grad_norm': '6.666', 'learning_rate': '1.553e-05', 'epoch': '0.6021'}
180
+ {'loss': '0.836', 'grad_norm': '5.681', 'learning_rate': '1.546e-05', 'epoch': '0.6086'}
181
+ {'loss': '0.8628', 'grad_norm': '6.571', 'learning_rate': '1.539e-05', 'epoch': '0.615'}
182
+ {'loss': '0.8244', 'grad_norm': '6.389', 'learning_rate': '1.532e-05', 'epoch': '0.6215'}
183
+ {'loss': '0.8647', 'grad_norm': '4.987', 'learning_rate': '1.525e-05', 'epoch': '0.628'}
184
+ {'loss': '0.8479', 'grad_norm': '4.451', 'learning_rate': '1.517e-05', 'epoch': '0.6345'}
185
+ {'loss': '0.8204', 'grad_norm': '5.356', 'learning_rate': '1.51e-05', 'epoch': '0.6409'}
186
+ {'loss': '0.8359', 'grad_norm': '5.146', 'learning_rate': '1.503e-05', 'epoch': '0.6474'}
187
+ {'loss': '0.7952', 'grad_norm': '4.308', 'learning_rate': '1.496e-05', 'epoch': '0.6539'}
188
+ {'loss': '0.8375', 'grad_norm': '5.216', 'learning_rate': '1.489e-05', 'epoch': '0.6604'}
189
+ {'loss': '0.8364', 'grad_norm': '5.812', 'learning_rate': '1.481e-05', 'epoch': '0.6668'}
190
+ {'loss': '0.8131', 'grad_norm': '5.52', 'learning_rate': '1.474e-05', 'epoch': '0.6733'}
191
+ {'loss': '0.831', 'grad_norm': '6.452', 'learning_rate': '1.467e-05', 'epoch': '0.6798'}
192
+ {'loss': '0.8295', 'grad_norm': '4.274', 'learning_rate': '1.46e-05', 'epoch': '0.6863'}
193
+ {'loss': '0.7865', 'grad_norm': '4.77', 'learning_rate': '1.453e-05', 'epoch': '0.6927'}
194
+ {'loss': '0.796', 'grad_norm': '5.027', 'learning_rate': '1.445e-05', 'epoch': '0.6992'}
195
+ {'loss': '0.8287', 'grad_norm': '4.826', 'learning_rate': '1.438e-05', 'epoch': '0.7057'}
196
+ {'loss': '0.8214', 'grad_norm': '4.381', 'learning_rate': '1.431e-05', 'epoch': '0.7121'}
197
+ {'loss': '0.7879', 'grad_norm': '6.475', 'learning_rate': '1.424e-05', 'epoch': '0.7186'}
198
+ {'loss': '0.8139', 'grad_norm': '5.295', 'learning_rate': '1.417e-05', 'epoch': '0.7251'}
199
+ {'loss': '0.7849', 'grad_norm': '5.051', 'learning_rate': '1.409e-05', 'epoch': '0.7316'}
200
+ {'loss': '0.788', 'grad_norm': '5.113', 'learning_rate': '1.402e-05', 'epoch': '0.738'}
201
+ {'loss': '0.7725', 'grad_norm': '4.049', 'learning_rate': '1.395e-05', 'epoch': '0.7445'}
202
+ {'loss': '0.8086', 'grad_norm': '4.646', 'learning_rate': '1.388e-05', 'epoch': '0.751'}
203
+ {'loss': '0.7687', 'grad_norm': '5.049', 'learning_rate': '1.381e-05', 'epoch': '0.7575'}
204
+ {'loss': '0.7828', 'grad_norm': '6.568', 'learning_rate': '1.373e-05', 'epoch': '0.7639'}
205
+ {'loss': '0.7518', 'grad_norm': '5.9', 'learning_rate': '1.366e-05', 'epoch': '0.7704'}
206
+ {'loss': '0.7599', 'grad_norm': '6.338', 'learning_rate': '1.359e-05', 'epoch': '0.7769'}
207
+ 2026-02-26 16:19:09 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 0.7768875940762321 after 24000 steps (truncated to 768):
208
+ 2026-02-26 16:40:53 - Accuracy Cosine Similarity: 97.37%
209
+ 2026-02-26 16:40:53 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 0.7768875940762321 after 24000 steps (truncated to 512):
210
+ 2026-02-26 17:02:09 - Accuracy Cosine Similarity: 97.38%
211
+ 2026-02-26 17:02:09 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 0.7768875940762321 after 24000 steps (truncated to 256):
212
+ 2026-02-26 17:23:27 - Accuracy Cosine Similarity: 97.38%
213
+ 2026-02-26 17:23:27 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 0.7768875940762321 after 24000 steps (truncated to 128):
214
+ 2026-02-26 17:44:42 - Accuracy Cosine Similarity: 97.34%
215
+ 2026-02-26 17:44:42 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 0.7768875940762321 after 24000 steps (truncated to 64):
216
+ 2026-02-26 18:06:06 - Accuracy Cosine Similarity: 97.18%
217
+ {'eval_train_loss': '0.4041', 'eval_dev-768_cosine_accuracy': '0.9737', 'eval_dev-512_cosine_accuracy': '0.9738', 'eval_dev-256_cosine_accuracy': '0.9738', 'eval_dev-128_cosine_accuracy': '0.9734', 'eval_dev-64_cosine_accuracy': '0.9718', 'eval_sequential_score': '0.9737', 'eval_train_runtime': '9673', 'eval_train_samples_per_second': '116.8', 'eval_train_steps_per_second': '14.6', 'epoch': '0.7769'}
218
+ 2026-02-26 18:06:06 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-24000
219
+ 2026-02-26 18:06:06 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-24000
220
+ {'loss': '0.7332', 'grad_norm': '4.95', 'learning_rate': '1.352e-05', 'epoch': '0.7834'}
221
+ {'loss': '0.7476', 'grad_norm': '4.513', 'learning_rate': '1.345e-05', 'epoch': '0.7898'}
222
+ {'loss': '0.7806', 'grad_norm': '5.095', 'learning_rate': '1.337e-05', 'epoch': '0.7963'}
223
+ {'loss': '0.7511', 'grad_norm': '5.826', 'learning_rate': '1.33e-05', 'epoch': '0.8028'}
224
+ {'loss': '0.7652', 'grad_norm': '6.09', 'learning_rate': '1.323e-05', 'epoch': '0.8093'}
225
+ {'loss': '0.7883', 'grad_norm': '4.332', 'learning_rate': '1.316e-05', 'epoch': '0.8157'}
226
+ {'loss': '0.7305', 'grad_norm': '5.749', 'learning_rate': '1.309e-05', 'epoch': '0.8222'}
227
+ {'loss': '0.7308', 'grad_norm': '4.871', 'learning_rate': '1.302e-05', 'epoch': '0.8287'}
228
+ {'loss': '0.7368', 'grad_norm': '4.618', 'learning_rate': '1.294e-05', 'epoch': '0.8352'}
229
+ {'loss': '0.7432', 'grad_norm': '4.836', 'learning_rate': '1.287e-05', 'epoch': '0.8416'}
230
+ {'loss': '0.7046', 'grad_norm': '4.988', 'learning_rate': '1.28e-05', 'epoch': '0.8481'}
231
+ {'loss': '0.7476', 'grad_norm': '4.596', 'learning_rate': '1.273e-05', 'epoch': '0.8546'}
232
+ {'loss': '0.7212', 'grad_norm': '5.712', 'learning_rate': '1.266e-05', 'epoch': '0.8611'}
233
+ {'loss': '0.7335', 'grad_norm': '3.99', 'learning_rate': '1.258e-05', 'epoch': '0.8675'}
234
+ {'loss': '0.7415', 'grad_norm': '5.446', 'learning_rate': '1.251e-05', 'epoch': '0.874'}
235
+ {'loss': '0.6937', 'grad_norm': '5.257', 'learning_rate': '1.244e-05', 'epoch': '0.8805'}
236
+ {'loss': '0.7294', 'grad_norm': '5.302', 'learning_rate': '1.237e-05', 'epoch': '0.8869'}
237
+ {'loss': '0.7436', 'grad_norm': '3.847', 'learning_rate': '1.23e-05', 'epoch': '0.8934'}
238
+ {'loss': '0.7093', 'grad_norm': '6.182', 'learning_rate': '1.222e-05', 'epoch': '0.8999'}
239
+ {'loss': '0.748', 'grad_norm': '5.445', 'learning_rate': '1.215e-05', 'epoch': '0.9064'}
240
+ {'loss': '0.7039', 'grad_norm': '5.002', 'learning_rate': '1.208e-05', 'epoch': '0.9128'}
241
+ {'loss': '0.7091', 'grad_norm': '5.085', 'learning_rate': '1.201e-05', 'epoch': '0.9193'}
242
+ {'loss': '0.7019', 'grad_norm': '5.379', 'learning_rate': '1.194e-05', 'epoch': '0.9258'}
243
+ {'loss': '0.7081', 'grad_norm': '5.63', 'learning_rate': '1.186e-05', 'epoch': '0.9323'}
244
+ {'loss': '0.6833', 'grad_norm': '2.541', 'learning_rate': '1.179e-05', 'epoch': '0.9387'}
245
+ {'loss': '0.6982', 'grad_norm': '5.714', 'learning_rate': '1.172e-05', 'epoch': '0.9452'}
246
+ {'loss': '0.7249', 'grad_norm': '5.051', 'learning_rate': '1.165e-05', 'epoch': '0.9517'}
247
+ {'loss': '0.7282', 'grad_norm': '6.322', 'learning_rate': '1.158e-05', 'epoch': '0.9582'}
248
+ {'loss': '0.7147', 'grad_norm': '4.961', 'learning_rate': '1.15e-05', 'epoch': '0.9646'}
249
+ {'loss': '0.6742', 'grad_norm': '4.871', 'learning_rate': '1.143e-05', 'epoch': '0.9711'}
250
+ 2026-02-26 19:59:22 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 0.9711094925952901 after 30000 steps (truncated to 768):
251
+ 2026-02-26 20:20:41 - Accuracy Cosine Similarity: 97.58%
252
+ 2026-02-26 20:20:41 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 0.9711094925952901 after 30000 steps (truncated to 512):
253
+ 2026-02-26 20:42:02 - Accuracy Cosine Similarity: 97.59%
254
+ 2026-02-26 20:42:02 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 0.9711094925952901 after 30000 steps (truncated to 256):
255
+ 2026-02-26 21:03:17 - Accuracy Cosine Similarity: 97.61%
256
+ 2026-02-26 21:03:17 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 0.9711094925952901 after 30000 steps (truncated to 128):
257
+ 2026-02-26 21:24:42 - Accuracy Cosine Similarity: 97.57%
258
+ 2026-02-26 21:24:42 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 0.9711094925952901 after 30000 steps (truncated to 64):
259
+ 2026-02-26 21:46:03 - Accuracy Cosine Similarity: 97.42%
260
+ {'eval_train_loss': '0.364', 'eval_dev-768_cosine_accuracy': '0.9758', 'eval_dev-512_cosine_accuracy': '0.9759', 'eval_dev-256_cosine_accuracy': '0.9761', 'eval_dev-128_cosine_accuracy': '0.9757', 'eval_dev-64_cosine_accuracy': '0.9742', 'eval_sequential_score': '0.9758', 'eval_train_runtime': '9649', 'eval_train_samples_per_second': '117.1', 'eval_train_steps_per_second': '14.64', 'epoch': '0.9711'}
261
+ 2026-02-26 21:46:03 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-30000
262
+ 2026-02-26 21:46:03 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-30000
263
+ {'loss': '0.6901', 'grad_norm': '3.348', 'learning_rate': '1.136e-05', 'epoch': '0.9776'}
264
+ {'loss': '0.7067', 'grad_norm': '3.76', 'learning_rate': '1.129e-05', 'epoch': '0.9841'}
265
+ {'loss': '0.7166', 'grad_norm': '4.729', 'learning_rate': '1.122e-05', 'epoch': '0.9905'}
266
+ {'loss': '0.68', 'grad_norm': '4.648', 'learning_rate': '1.114e-05', 'epoch': '0.997'}
267
+ {'loss': '0.6846', 'grad_norm': '4.427', 'learning_rate': '1.107e-05', 'epoch': '1.003'}
268
+ {'loss': '0.6723', 'grad_norm': '4.459', 'learning_rate': '1.1e-05', 'epoch': '1.01'}
269
+ {'loss': '0.6573', 'grad_norm': '6.387', 'learning_rate': '1.093e-05', 'epoch': '1.016'}
270
+ {'loss': '0.6895', 'grad_norm': '4.1', 'learning_rate': '1.086e-05', 'epoch': '1.023'}
271
+ {'loss': '0.6588', 'grad_norm': '5.927', 'learning_rate': '1.079e-05', 'epoch': '1.029'}
272
+ {'loss': '0.6517', 'grad_norm': '5.9', 'learning_rate': '1.071e-05', 'epoch': '1.036'}
273
+ {'loss': '0.6498', 'grad_norm': '4.736', 'learning_rate': '1.064e-05', 'epoch': '1.042'}
274
+ {'loss': '0.6836', 'grad_norm': '5.029', 'learning_rate': '1.057e-05', 'epoch': '1.049'}
275
+ {'loss': '0.6819', 'grad_norm': '2.595', 'learning_rate': '1.05e-05', 'epoch': '1.055'}
276
+ {'loss': '0.6463', 'grad_norm': '4.963', 'learning_rate': '1.043e-05', 'epoch': '1.062'}
277
+ {'loss': '0.6645', 'grad_norm': '5.046', 'learning_rate': '1.035e-05', 'epoch': '1.068'}
278
+ {'loss': '0.6518', 'grad_norm': '3.307', 'learning_rate': '1.028e-05', 'epoch': '1.075'}
279
+ {'loss': '0.6235', 'grad_norm': '3.848', 'learning_rate': '1.021e-05', 'epoch': '1.081'}
280
+ {'loss': '0.6302', 'grad_norm': '4.664', 'learning_rate': '1.014e-05', 'epoch': '1.088'}
281
+ {'loss': '0.6452', 'grad_norm': '5.47', 'learning_rate': '1.007e-05', 'epoch': '1.094'}
282
+ {'loss': '0.6477', 'grad_norm': '5.26', 'learning_rate': '9.994e-06', 'epoch': '1.101'}
283
+ {'loss': '0.6084', 'grad_norm': '4.313', 'learning_rate': '9.922e-06', 'epoch': '1.107'}
284
+ {'loss': '0.6259', 'grad_norm': '6.499', 'learning_rate': '9.85e-06', 'epoch': '1.114'}
285
+ {'loss': '0.607', 'grad_norm': '3.922', 'learning_rate': '9.778e-06', 'epoch': '1.12'}
286
+ {'loss': '0.5977', 'grad_norm': '5.37', 'learning_rate': '9.706e-06', 'epoch': '1.126'}
287
+ {'loss': '0.6044', 'grad_norm': '5.068', 'learning_rate': '9.634e-06', 'epoch': '1.133'}
288
+ {'loss': '0.6007', 'grad_norm': '4.109', 'learning_rate': '9.562e-06', 'epoch': '1.139'}
289
+ {'loss': '0.5628', 'grad_norm': '4.954', 'learning_rate': '9.491e-06', 'epoch': '1.146'}
290
+ {'loss': '0.5732', 'grad_norm': '4.068', 'learning_rate': '9.419e-06', 'epoch': '1.152'}
291
+ {'loss': '0.5773', 'grad_norm': '4.939', 'learning_rate': '9.347e-06', 'epoch': '1.159'}
292
+ {'loss': '0.5719', 'grad_norm': '4.418', 'learning_rate': '9.275e-06', 'epoch': '1.165'}
293
+ 2026-02-26 23:38:18 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 1.1653152059561382 after 36000 steps (truncated to 768):
294
+ 2026-02-27 00:01:17 - Accuracy Cosine Similarity: 97.75%
295
+ 2026-02-27 00:01:17 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 1.1653152059561382 after 36000 steps (truncated to 512):
296
+ 2026-02-27 00:23:59 - Accuracy Cosine Similarity: 97.77%
297
+ 2026-02-27 00:23:59 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 1.1653152059561382 after 36000 steps (truncated to 256):
298
+ 2026-02-27 00:46:38 - Accuracy Cosine Similarity: 97.77%
299
+ 2026-02-27 00:46:38 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 1.1653152059561382 after 36000 steps (truncated to 128):
300
+ 2026-02-27 01:09:18 - Accuracy Cosine Similarity: 97.74%
301
+ 2026-02-27 01:09:18 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 1.1653152059561382 after 36000 steps (truncated to 64):
302
+ 2026-02-27 01:32:14 - Accuracy Cosine Similarity: 97.60%
303
+ {'eval_train_loss': '0.3356', 'eval_dev-768_cosine_accuracy': '0.9775', 'eval_dev-512_cosine_accuracy': '0.9777', 'eval_dev-256_cosine_accuracy': '0.9777', 'eval_dev-128_cosine_accuracy': '0.9774', 'eval_dev-64_cosine_accuracy': '0.976', 'eval_sequential_score': '0.9775', 'eval_train_runtime': '1.01e+04', 'eval_train_samples_per_second': '111.8', 'eval_train_steps_per_second': '13.98', 'epoch': '1.165'}
304
+ 2026-02-27 01:32:14 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-36000
305
+ 2026-02-27 01:32:14 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-36000
306
+ {'loss': '0.5471', 'grad_norm': '3.58', 'learning_rate': '9.203e-06', 'epoch': '1.172'}
307
+ {'loss': '0.5635', 'grad_norm': '5.198', 'learning_rate': '9.131e-06', 'epoch': '1.178'}
308
+ {'loss': '0.539', 'grad_norm': '4.468', 'learning_rate': '9.059e-06', 'epoch': '1.185'}
309
+ {'loss': '0.5428', 'grad_norm': '4.349', 'learning_rate': '8.987e-06', 'epoch': '1.191'}
310
+ {'loss': '0.5205', 'grad_norm': '2.936', 'learning_rate': '8.915e-06', 'epoch': '1.198'}
311
+ {'loss': '0.5362', 'grad_norm': '3.337', 'learning_rate': '8.843e-06', 'epoch': '1.204'}
312
+ {'loss': '0.5386', 'grad_norm': '5.76', 'learning_rate': '8.771e-06', 'epoch': '1.211'}
313
+ {'loss': '0.5203', 'grad_norm': '3.261', 'learning_rate': '8.699e-06', 'epoch': '1.217'}
314
+ {'loss': '0.5301', 'grad_norm': '3.732', 'learning_rate': '8.627e-06', 'epoch': '1.224'}
315
+ {'loss': '0.5232', 'grad_norm': '4.54', 'learning_rate': '8.555e-06', 'epoch': '1.23'}
316
+ {'loss': '0.4922', 'grad_norm': '4.291', 'learning_rate': '8.483e-06', 'epoch': '1.237'}
317
+ {'loss': '0.5029', 'grad_norm': '3.979', 'learning_rate': '8.412e-06', 'epoch': '1.243'}
318
+ {'loss': '0.4989', 'grad_norm': '7.829', 'learning_rate': '8.34e-06', 'epoch': '1.249'}
319
+ {'loss': '0.5053', 'grad_norm': '2.903', 'learning_rate': '8.268e-06', 'epoch': '1.256'}
320
+ {'loss': '0.5081', 'grad_norm': '5.471', 'learning_rate': '8.196e-06', 'epoch': '1.262'}
321
+ {'loss': '0.496', 'grad_norm': '5.204', 'learning_rate': '8.124e-06', 'epoch': '1.269'}
322
+ {'loss': '0.5052', 'grad_norm': '4.377', 'learning_rate': '8.052e-06', 'epoch': '1.275'}
323
+ {'loss': '0.4984', 'grad_norm': '4.184', 'learning_rate': '7.98e-06', 'epoch': '1.282'}
324
+ {'loss': '0.4909', 'grad_norm': '4.991', 'learning_rate': '7.908e-06', 'epoch': '1.288'}
325
+ {'loss': '0.512', 'grad_norm': '3.76', 'learning_rate': '7.836e-06', 'epoch': '1.295'}
326
+ {'loss': '0.4873', 'grad_norm': '3.844', 'learning_rate': '7.764e-06', 'epoch': '1.301'}
327
+ {'loss': '0.4896', 'grad_norm': '6.987', 'learning_rate': '7.692e-06', 'epoch': '1.308'}
328
+ {'loss': '0.49', 'grad_norm': '6.267', 'learning_rate': '7.62e-06', 'epoch': '1.314'}
329
+ {'loss': '0.5036', 'grad_norm': '3.776', 'learning_rate': '7.548e-06', 'epoch': '1.321'}
330
+ {'loss': '0.4876', 'grad_norm': '3.42', 'learning_rate': '7.476e-06', 'epoch': '1.327'}
331
+ {'loss': '0.4705', 'grad_norm': '5.478', 'learning_rate': '7.404e-06', 'epoch': '1.334'}
332
+ {'loss': '0.4786', 'grad_norm': '3.313', 'learning_rate': '7.333e-06', 'epoch': '1.34'}
333
+ {'loss': '0.4998', 'grad_norm': '3.13', 'learning_rate': '7.261e-06', 'epoch': '1.347'}
334
+ {'loss': '0.4692', 'grad_norm': '3.971', 'learning_rate': '7.189e-06', 'epoch': '1.353'}
335
+ {'loss': '0.5064', 'grad_norm': '6.238', 'learning_rate': '7.117e-06', 'epoch': '1.36'}
336
+ 2026-02-27 03:24:31 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 1.3595371044751963 after 42000 steps (truncated to 768):
337
+ 2026-02-27 03:47:24 - Accuracy Cosine Similarity: 97.88%
338
+ 2026-02-27 03:47:24 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 1.3595371044751963 after 42000 steps (truncated to 512):
339
+ 2026-02-27 04:10:18 - Accuracy Cosine Similarity: 97.90%
340
+ 2026-02-27 04:10:18 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 1.3595371044751963 after 42000 steps (truncated to 256):
341
+ 2026-02-27 04:33:12 - Accuracy Cosine Similarity: 97.90%
342
+ 2026-02-27 04:33:12 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 1.3595371044751963 after 42000 steps (truncated to 128):
343
+ 2026-02-27 04:56:07 - Accuracy Cosine Similarity: 97.85%
344
+ 2026-02-27 04:56:07 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 1.3595371044751963 after 42000 steps (truncated to 64):
345
+ 2026-02-27 05:19:11 - Accuracy Cosine Similarity: 97.74%
346
+ {'eval_train_loss': '0.316', 'eval_dev-768_cosine_accuracy': '0.9788', 'eval_dev-512_cosine_accuracy': '0.979', 'eval_dev-256_cosine_accuracy': '0.979', 'eval_dev-128_cosine_accuracy': '0.9785', 'eval_dev-64_cosine_accuracy': '0.9774', 'eval_sequential_score': '0.9788', 'eval_train_runtime': '1.014e+04', 'eval_train_samples_per_second': '111.5', 'eval_train_steps_per_second': '13.93', 'epoch': '1.36'}
347
+ 2026-02-27 05:19:11 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-42000
348
+ 2026-02-27 05:19:11 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-42000
349
+ {'loss': '0.4925', 'grad_norm': '5.158', 'learning_rate': '7.045e-06', 'epoch': '1.366'}
350
+ {'loss': '0.4601', 'grad_norm': '4.139', 'learning_rate': '6.973e-06', 'epoch': '1.372'}
351
+ {'loss': '0.4762', 'grad_norm': '3.411', 'learning_rate': '6.901e-06', 'epoch': '1.379'}
352
+ {'loss': '0.4986', 'grad_norm': '4.23', 'learning_rate': '6.829e-06', 'epoch': '1.385'}
353
+ {'loss': '0.4656', 'grad_norm': '5.326', 'learning_rate': '6.757e-06', 'epoch': '1.392'}
354
+ {'loss': '0.4507', 'grad_norm': '3.826', 'learning_rate': '6.685e-06', 'epoch': '1.398'}
355
+ {'loss': '0.4862', 'grad_norm': '3.509', 'learning_rate': '6.613e-06', 'epoch': '1.405'}
356
+ {'loss': '0.4596', 'grad_norm': '4.734', 'learning_rate': '6.541e-06', 'epoch': '1.411'}
357
+ {'loss': '0.4696', 'grad_norm': '4.799', 'learning_rate': '6.469e-06', 'epoch': '1.418'}
358
+ {'loss': '0.4925', 'grad_norm': '4.942', 'learning_rate': '6.397e-06', 'epoch': '1.424'}
359
+ {'loss': '0.4796', 'grad_norm': '4.147', 'learning_rate': '6.325e-06', 'epoch': '1.431'}
360
+ {'loss': '0.4525', 'grad_norm': '5.146', 'learning_rate': '6.254e-06', 'epoch': '1.437'}
361
+ {'loss': '0.4717', 'grad_norm': '3.52', 'learning_rate': '6.182e-06', 'epoch': '1.444'}
362
+ {'loss': '0.4803', 'grad_norm': '3.25', 'learning_rate': '6.11e-06', 'epoch': '1.45'}
363
+ {'loss': '0.4675', 'grad_norm': '7.35', 'learning_rate': '6.038e-06', 'epoch': '1.457'}
364
+ {'loss': '0.4631', 'grad_norm': '3.847', 'learning_rate': '5.966e-06', 'epoch': '1.463'}
365
+ {'loss': '0.4622', 'grad_norm': '4.57', 'learning_rate': '5.894e-06', 'epoch': '1.47'}
366
+ {'loss': '0.4496', 'grad_norm': '1.997', 'learning_rate': '5.822e-06', 'epoch': '1.476'}
367
+ {'loss': '0.4678', 'grad_norm': '4.266', 'learning_rate': '5.75e-06', 'epoch': '1.483'}
368
+ {'loss': '0.4495', 'grad_norm': '5.948', 'learning_rate': '5.678e-06', 'epoch': '1.489'}
369
+ {'loss': '0.4474', 'grad_norm': '3.7', 'learning_rate': '5.606e-06', 'epoch': '1.495'}
370
+ {'loss': '0.4587', 'grad_norm': '2.877', 'learning_rate': '5.534e-06', 'epoch': '1.502'}
371
+ {'loss': '0.4591', 'grad_norm': '4.245', 'learning_rate': '5.462e-06', 'epoch': '1.508'}
372
+ {'loss': '0.4573', 'grad_norm': '5.431', 'learning_rate': '5.39e-06', 'epoch': '1.515'}
373
+ {'loss': '0.4442', 'grad_norm': '3.338', 'learning_rate': '5.318e-06', 'epoch': '1.521'}
374
+ {'loss': '0.455', 'grad_norm': '4.723', 'learning_rate': '5.246e-06', 'epoch': '1.528'}
375
+ {'loss': '0.4493', 'grad_norm': '4.226', 'learning_rate': '5.175e-06', 'epoch': '1.534'}
376
+ {'loss': '0.4485', 'grad_norm': '4.451', 'learning_rate': '5.103e-06', 'epoch': '1.541'}
377
+ {'loss': '0.4569', 'grad_norm': '4.297', 'learning_rate': '5.031e-06', 'epoch': '1.547'}
378
+ {'loss': '0.4346', 'grad_norm': '4.199', 'learning_rate': '4.959e-06', 'epoch': '1.554'}
379
+ 2026-02-27 07:11:49 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 1.5537590029942543 after 48000 steps (truncated to 768):
380
+ 2026-02-27 07:34:37 - Accuracy Cosine Similarity: 97.99%
381
+ 2026-02-27 07:34:37 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 1.5537590029942543 after 48000 steps (truncated to 512):
382
+ 2026-02-27 07:57:13 - Accuracy Cosine Similarity: 98.02%
383
+ 2026-02-27 07:57:13 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 1.5537590029942543 after 48000 steps (truncated to 256):
384
+ 2026-02-27 08:20:07 - Accuracy Cosine Similarity: 98.02%
385
+ 2026-02-27 08:20:07 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 1.5537590029942543 after 48000 steps (truncated to 128):
386
+ 2026-02-27 08:42:52 - Accuracy Cosine Similarity: 97.98%
387
+ 2026-02-27 08:42:52 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 1.5537590029942543 after 48000 steps (truncated to 64):
388
+ 2026-02-27 09:05:32 - Accuracy Cosine Similarity: 97.88%
389
+ {'eval_train_loss': '0.3001', 'eval_dev-768_cosine_accuracy': '0.9799', 'eval_dev-512_cosine_accuracy': '0.9802', 'eval_dev-256_cosine_accuracy': '0.9802', 'eval_dev-128_cosine_accuracy': '0.9798', 'eval_dev-64_cosine_accuracy': '0.9788', 'eval_sequential_score': '0.9799', 'eval_train_runtime': '1.008e+04', 'eval_train_samples_per_second': '112.1', 'eval_train_steps_per_second': '14.02', 'epoch': '1.554'}
390
+ 2026-02-27 09:05:32 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-48000
391
+ 2026-02-27 09:05:32 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-48000
392
+ {'loss': '0.4469', 'grad_norm': '3.364', 'learning_rate': '4.887e-06', 'epoch': '1.56'}
393
+ {'loss': '0.4602', 'grad_norm': '5.309', 'learning_rate': '4.815e-06', 'epoch': '1.567'}
394
+ {'loss': '0.443', 'grad_norm': '3.875', 'learning_rate': '4.743e-06', 'epoch': '1.573'}
395
+ {'loss': '0.4524', 'grad_norm': '4.824', 'learning_rate': '4.671e-06', 'epoch': '1.58'}
396
+ {'loss': '0.4528', 'grad_norm': '4.996', 'learning_rate': '4.599e-06', 'epoch': '1.586'}
397
+ {'loss': '0.4348', 'grad_norm': '4.96', 'learning_rate': '4.527e-06', 'epoch': '1.593'}
398
+ {'loss': '0.4533', 'grad_norm': '5.219', 'learning_rate': '4.455e-06', 'epoch': '1.599'}
399
+ {'loss': '0.4523', 'grad_norm': '3.444', 'learning_rate': '4.383e-06', 'epoch': '1.606'}
400
+ {'loss': '0.4509', 'grad_norm': '5.647', 'learning_rate': '4.311e-06', 'epoch': '1.612'}
401
+ {'loss': '0.4365', 'grad_norm': '5.052', 'learning_rate': '4.239e-06', 'epoch': '1.618'}
402
+ {'loss': '0.4504', 'grad_norm': '5.786', 'learning_rate': '4.167e-06', 'epoch': '1.625'}
403
+ {'loss': '0.4292', 'grad_norm': '4.353', 'learning_rate': '4.096e-06', 'epoch': '1.631'}
404
+ {'loss': '0.4406', 'grad_norm': '2.976', 'learning_rate': '4.024e-06', 'epoch': '1.638'}
405
+ {'loss': '0.4333', 'grad_norm': '3.685', 'learning_rate': '3.952e-06', 'epoch': '1.644'}
406
+ {'loss': '0.4361', 'grad_norm': '4.107', 'learning_rate': '3.88e-06', 'epoch': '1.651'}
407
+ {'loss': '0.4065', 'grad_norm': '3.636', 'learning_rate': '3.808e-06', 'epoch': '1.657'}
408
+ {'loss': '0.4671', 'grad_norm': '3.464', 'learning_rate': '3.736e-06', 'epoch': '1.664'}
409
+ {'loss': '0.4328', 'grad_norm': '3.129', 'learning_rate': '3.664e-06', 'epoch': '1.67'}
410
+ {'loss': '0.431', 'grad_norm': '2.453', 'learning_rate': '3.592e-06', 'epoch': '1.677'}
411
+ {'loss': '0.4523', 'grad_norm': '3.727', 'learning_rate': '3.52e-06', 'epoch': '1.683'}
412
+ {'loss': '0.4232', 'grad_norm': '4.398', 'learning_rate': '3.448e-06', 'epoch': '1.69'}
413
+ {'loss': '0.4257', 'grad_norm': '2.861', 'learning_rate': '3.376e-06', 'epoch': '1.696'}
414
+ {'loss': '0.4448', 'grad_norm': '3.523', 'learning_rate': '3.304e-06', 'epoch': '1.703'}
415
+ {'loss': '0.4491', 'grad_norm': '3.893', 'learning_rate': '3.232e-06', 'epoch': '1.709'}
416
+ {'loss': '0.4224', 'grad_norm': '3.399', 'learning_rate': '3.16e-06', 'epoch': '1.716'}
417
+ {'loss': '0.4297', 'grad_norm': '4.703', 'learning_rate': '3.088e-06', 'epoch': '1.722'}
418
+ {'loss': '0.4522', 'grad_norm': '4.29', 'learning_rate': '3.017e-06', 'epoch': '1.729'}
419
+ {'loss': '0.4195', 'grad_norm': '4.29', 'learning_rate': '2.945e-06', 'epoch': '1.735'}
420
+ {'loss': '0.4227', 'grad_norm': '3.841', 'learning_rate': '2.873e-06', 'epoch': '1.742'}
421
+ {'loss': '0.4381', 'grad_norm': '4.086', 'learning_rate': '2.801e-06', 'epoch': '1.748'}
422
+ 2026-02-27 10:59:10 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 1.7479809015133123 after 54000 steps (truncated to 768):
423
+ 2026-02-27 11:22:08 - Accuracy Cosine Similarity: 98.07%
424
+ 2026-02-27 11:22:08 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 1.7479809015133123 after 54000 steps (truncated to 512):
425
+ 2026-02-27 11:44:57 - Accuracy Cosine Similarity: 98.08%
426
+ 2026-02-27 11:44:57 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 1.7479809015133123 after 54000 steps (truncated to 256):
427
+ 2026-02-27 12:07:55 - Accuracy Cosine Similarity: 98.08%
428
+ 2026-02-27 12:07:55 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 1.7479809015133123 after 54000 steps (truncated to 128):
429
+ 2026-02-27 12:30:35 - Accuracy Cosine Similarity: 98.05%
430
+ 2026-02-27 12:30:35 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 1.7479809015133123 after 54000 steps (truncated to 64):
431
+ 2026-02-27 12:53:24 - Accuracy Cosine Similarity: 97.94%
432
+ {'eval_train_loss': '0.2875', 'eval_dev-768_cosine_accuracy': '0.9807', 'eval_dev-512_cosine_accuracy': '0.9808', 'eval_dev-256_cosine_accuracy': '0.9808', 'eval_dev-128_cosine_accuracy': '0.9805', 'eval_dev-64_cosine_accuracy': '0.9794', 'eval_sequential_score': '0.9807', 'eval_train_runtime': '1.012e+04', 'eval_train_samples_per_second': '111.6', 'eval_train_steps_per_second': '13.95', 'epoch': '1.748'}
433
+ 2026-02-27 12:53:24 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-54000
434
+ 2026-02-27 12:53:24 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-54000
435
+ {'loss': '0.446', 'grad_norm': '4.176', 'learning_rate': '2.729e-06', 'epoch': '1.754'}
436
+ {'loss': '0.426', 'grad_norm': '4.261', 'learning_rate': '2.657e-06', 'epoch': '1.761'}
437
+ {'loss': '0.4299', 'grad_norm': '4.676', 'learning_rate': '2.585e-06', 'epoch': '1.767'}
438
+ {'loss': '0.4247', 'grad_norm': '3.933', 'learning_rate': '2.513e-06', 'epoch': '1.774'}
439
+ {'loss': '0.4244', 'grad_norm': '4.853', 'learning_rate': '2.441e-06', 'epoch': '1.78'}
440
+ {'loss': '0.4185', 'grad_norm': '2.985', 'learning_rate': '2.369e-06', 'epoch': '1.787'}
441
+ {'loss': '0.4292', 'grad_norm': '3.804', 'learning_rate': '2.297e-06', 'epoch': '1.793'}
442
+ {'loss': '0.4468', 'grad_norm': '3.187', 'learning_rate': '2.225e-06', 'epoch': '1.8'}
443
+ {'loss': '0.4118', 'grad_norm': '4.004', 'learning_rate': '2.153e-06', 'epoch': '1.806'}
444
+ {'loss': '0.4306', 'grad_norm': '4.007', 'learning_rate': '2.081e-06', 'epoch': '1.813'}
445
+ {'loss': '0.4447', 'grad_norm': '4.323', 'learning_rate': '2.009e-06', 'epoch': '1.819'}
446
+ {'loss': '0.4147', 'grad_norm': '3.863', 'learning_rate': '1.938e-06', 'epoch': '1.826'}
447
+ {'loss': '0.4189', 'grad_norm': '4.788', 'learning_rate': '1.866e-06', 'epoch': '1.832'}
448
+ {'loss': '0.4167', 'grad_norm': '4.276', 'learning_rate': '1.794e-06', 'epoch': '1.839'}
449
+ {'loss': '0.4022', 'grad_norm': '3.887', 'learning_rate': '1.722e-06', 'epoch': '1.845'}
450
+ {'loss': '0.4158', 'grad_norm': '3.075', 'learning_rate': '1.65e-06', 'epoch': '1.852'}
451
+ {'loss': '0.4228', 'grad_norm': '3.993', 'learning_rate': '1.578e-06', 'epoch': '1.858'}
452
+ {'loss': '0.4256', 'grad_norm': '4.497', 'learning_rate': '1.506e-06', 'epoch': '1.865'}
453
+ {'loss': '0.4251', 'grad_norm': '4.539', 'learning_rate': '1.434e-06', 'epoch': '1.871'}
454
+ {'loss': '0.4232', 'grad_norm': '2.337', 'learning_rate': '1.362e-06', 'epoch': '1.877'}
455
+ {'loss': '0.4143', 'grad_norm': '3.389', 'learning_rate': '1.29e-06', 'epoch': '1.884'}
456
+ {'loss': '0.4331', 'grad_norm': '3.545', 'learning_rate': '1.218e-06', 'epoch': '1.89'}
457
+ {'loss': '0.4253', 'grad_norm': '5.606', 'learning_rate': '1.146e-06', 'epoch': '1.897'}
458
+ {'loss': '0.441', 'grad_norm': '4.453', 'learning_rate': '1.074e-06', 'epoch': '1.903'}
459
+ {'loss': '0.4337', 'grad_norm': '5.374', 'learning_rate': '1.002e-06', 'epoch': '1.91'}
460
+ {'loss': '0.4016', 'grad_norm': '2.246', 'learning_rate': '9.305e-07', 'epoch': '1.916'}
461
+ {'loss': '0.4249', 'grad_norm': '5.255', 'learning_rate': '8.585e-07', 'epoch': '1.923'}
462
+ {'loss': '0.4108', 'grad_norm': '3.59', 'learning_rate': '7.866e-07', 'epoch': '1.929'}
463
+ {'loss': '0.4272', 'grad_norm': '4.258', 'learning_rate': '7.147e-07', 'epoch': '1.936'}
464
+ {'loss': '0.3916', 'grad_norm': '3.476', 'learning_rate': '6.427e-07', 'epoch': '1.942'}
465
+ 2026-02-27 14:47:29 - TripletEvaluator: Evaluating the model on the dev-768 dataset in epoch 1.9422028000323703 after 60000 steps (truncated to 768):
466
+ 2026-02-27 15:10:59 - Accuracy Cosine Similarity: 98.10%
467
+ 2026-02-27 15:10:59 - TripletEvaluator: Evaluating the model on the dev-512 dataset in epoch 1.9422028000323703 after 60000 steps (truncated to 512):
468
+ 2026-02-27 15:34:30 - Accuracy Cosine Similarity: 98.11%
469
+ 2026-02-27 15:34:30 - TripletEvaluator: Evaluating the model on the dev-256 dataset in epoch 1.9422028000323703 after 60000 steps (truncated to 256):
470
+ 2026-02-27 15:58:10 - Accuracy Cosine Similarity: 98.13%
471
+ 2026-02-27 15:58:10 - TripletEvaluator: Evaluating the model on the dev-128 dataset in epoch 1.9422028000323703 after 60000 steps (truncated to 128):
472
+ 2026-02-27 16:21:18 - Accuracy Cosine Similarity: 98.11%
473
+ 2026-02-27 16:21:18 - TripletEvaluator: Evaluating the model on the dev-64 dataset in epoch 1.9422028000323703 after 60000 steps (truncated to 64):
474
+ 2026-02-27 16:44:14 - Accuracy Cosine Similarity: 97.97%
475
+ {'eval_train_loss': '0.2812', 'eval_dev-768_cosine_accuracy': '0.981', 'eval_dev-512_cosine_accuracy': '0.9811', 'eval_dev-256_cosine_accuracy': '0.9813', 'eval_dev-128_cosine_accuracy': '0.9811', 'eval_dev-64_cosine_accuracy': '0.9797', 'eval_sequential_score': '0.981', 'eval_train_runtime': '1.03e+04', 'eval_train_samples_per_second': '109.7', 'eval_train_steps_per_second': '13.71', 'epoch': '1.942'}
476
+ 2026-02-27 16:44:14 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-60000
477
+ 2026-02-27 16:44:14 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-60000
478
+ {'loss': '0.4334', 'grad_norm': '4.623', 'learning_rate': '5.708e-07', 'epoch': '1.949'}
479
+ {'loss': '0.4462', 'grad_norm': '5.31', 'learning_rate': '4.989e-07', 'epoch': '1.955'}
480
+ {'loss': '0.4436', 'grad_norm': '3.379', 'learning_rate': '4.269e-07', 'epoch': '1.962'}
481
+ {'loss': '0.4278', 'grad_norm': '5.471', 'learning_rate': '3.55e-07', 'epoch': '1.968'}
482
+ {'loss': '0.417', 'grad_norm': '3.435', 'learning_rate': '2.831e-07', 'epoch': '1.975'}
483
+ {'loss': '0.4376', 'grad_norm': '2.617', 'learning_rate': '2.111e-07', 'epoch': '1.981'}
484
+ {'loss': '0.4433', 'grad_norm': '3.465', 'learning_rate': '1.392e-07', 'epoch': '1.988'}
485
+ {'loss': '0.4292', 'grad_norm': '2.354', 'learning_rate': '6.726e-08', 'epoch': '1.994'}
486
+ 2026-02-27 17:01:56 - Saving model checkpoint to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-61786
487
+ 2026-02-27 17:01:56 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-61786
488
+ {'train_runtime': '9.588e+04', 'train_samples_per_second': '82.48', 'train_steps_per_second': '0.644', 'train_loss': '0.403', 'epoch': '2'}
489
+ 2026-02-27 17:01:58 - Save model to /home/skiredj.abderrahman/khalil/sbert_training/output/final
490
+ model saved successfully
491
+ 2026-02-27 17:01:59 - TripletEvaluator: Evaluating the model on the test-768 dataset (truncated to 768):
492
+ 2026-02-27 17:21:39 - Accuracy Cosine Similarity: 98.10%
493
+ 2026-02-27 17:21:39 - TripletEvaluator: Evaluating the model on the test-512 dataset (truncated to 512):
494
+ 2026-02-27 17:41:10 - Accuracy Cosine Similarity: 98.13%
495
+ 2026-02-27 17:41:10 - TripletEvaluator: Evaluating the model on the test-256 dataset (truncated to 256):
496
+ 2026-02-27 18:00:40 - Accuracy Cosine Similarity: 98.13%
497
+ 2026-02-27 18:00:40 - TripletEvaluator: Evaluating the model on the test-128 dataset (truncated to 128):
498
+ 2026-02-27 18:20:06 - Accuracy Cosine Similarity: 98.11%
499
+ 2026-02-27 18:20:06 - TripletEvaluator: Evaluating the model on the test-64 dataset (truncated to 64):
500
+ 2026-02-27 18:39:32 - Accuracy Cosine Similarity: 97.97%
epoch2/model/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
epoch2/model/README.md ADDED
@@ -0,0 +1,773 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - dense
7
+ - generated_from_trainer
8
+ - dataset_size:3954179
9
+ - loss:MatryoshkaLoss
10
+ - loss:MultipleNegativesRankingLoss
11
+ widget:
12
+ - source_sentence: إذا لم تكن قد فعلت ذلك بالفعل ، تحقق من تصميمنا الجديد للمراسلات
13
+ والحوارات واليوميات .
14
+ sentences:
15
+ - تم إعادة تصميم الرسائل والحوارات .
16
+ - يقدم مقهى بارج كل من وجبات الغداء والإفطار .
17
+ - قبل ان نعرف اسماء بعضنا او اي شيء قد تعانقنا وبكىنا .
18
+ - source_sentence: أين تقع جامعة واينزبرج
19
+ sentences:
20
+ - جامعة دالاس بابتيست ( DBU ) ، المعروفة سابقا باسم كلية دالاس بابتيست ، هي جامعة
21
+ فنون ليبرالية مسيحية تقع في دالاس ، تكساس . يقع الحرم الجامعي الرئيسي على بعد
22
+ حوالي 12 ميلا ( 19 كم ) جنوب غرب وسط مدينة دالاس ويطل على بحيرة ماونتين كريك .
23
+ تأسست جامعة دالاس بابتيست عام 1898 باسم كلية ديكاتور بابتيست ، وتدير حاليا حرما
24
+ جامعيا في دالاس وبلانو وهيرست .
25
+ - الزوجان معا
26
+ - تقع جامعة واينسبرغ في حرم جامعي معاصر في تلال جنوب غرب ولاية بنسلفانيا ، مع ثلاثة
27
+ مراكز للبالغين تقع في مناطق بيتسبرغ في ساوثبوينت وكرانبيري ومونروفيل . تم إدراج
28
+ Hanna Hall و Miller Hall في السجل الوطني للأماكن التاريخية .
29
+ - source_sentence: The isolated Russian forces resisted in several areas for two more
30
+ days .
31
+ sentences:
32
+ - 'ياهو : كيف يمكنني معرفة ما إذا كان البريد الإلكتروني الذي أرسلته قد تم استلامه
33
+ أو قراءته ؟'
34
+ - واستمرت الاشتباكات الحدودية خلال اليومين المقبلين ، حيث استهدفت المخافر الحدودية
35
+ من الجانبين والتي أسفرت عن وقوع عشرات الإصابات .
36
+ - قاومت القوات الروسية المعزولة في عة مناطق لمدة يومين آخرين .
37
+ - source_sentence: فتاة هيبي بشعر أشقر وأرجواني على الجانب يرتدي قميص أبيض وملابس
38
+ سوداء
39
+ sentences:
40
+ - فتاة " هيبي " ترتدي قميصا أبيضا وملابس سوداء شعرها أشقر وأحمر
41
+ - المرأة تضع يدها في جيب الرجل
42
+ - فتاة لديها سترة حمراء وسوداء
43
+ - source_sentence: رجل وامرأة يجلسان في سيارة ووجههما في الاتجاه المعاكس من الكاميرا
44
+ sentences:
45
+ - هناك شخصان وسيارة
46
+ - سيارة صدئة هي الشيء الوحيد المرئي
47
+ - كان أفضل حالا
48
+ pipeline_tag: sentence-similarity
49
+ library_name: sentence-transformers
50
+ metrics:
51
+ - cosine_accuracy
52
+ model-index:
53
+ - name: SentenceTransformer
54
+ results:
55
+ - task:
56
+ type: triplet
57
+ name: Triplet
58
+ dataset:
59
+ name: dev 768
60
+ type: dev-768
61
+ metrics:
62
+ - type: cosine_accuracy
63
+ value: 0.9809960126876831
64
+ name: Cosine Accuracy
65
+ - task:
66
+ type: triplet
67
+ name: Triplet
68
+ dataset:
69
+ name: dev 512
70
+ type: dev-512
71
+ metrics:
72
+ - type: cosine_accuracy
73
+ value: 0.9811199903488159
74
+ name: Cosine Accuracy
75
+ - task:
76
+ type: triplet
77
+ name: Triplet
78
+ dataset:
79
+ name: dev 256
80
+ type: dev-256
81
+ metrics:
82
+ - type: cosine_accuracy
83
+ value: 0.9813200235366821
84
+ name: Cosine Accuracy
85
+ - task:
86
+ type: triplet
87
+ name: Triplet
88
+ dataset:
89
+ name: dev 128
90
+ type: dev-128
91
+ metrics:
92
+ - type: cosine_accuracy
93
+ value: 0.9811360239982605
94
+ name: Cosine Accuracy
95
+ - task:
96
+ type: triplet
97
+ name: Triplet
98
+ dataset:
99
+ name: dev 64
100
+ type: dev-64
101
+ metrics:
102
+ - type: cosine_accuracy
103
+ value: 0.9796760082244873
104
+ name: Cosine Accuracy
105
+ ---
106
+
107
+ # SentenceTransformer
108
+
109
+ This is a [sentence-transformers](https://www.SBERT.net) model trained on the train dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
110
+
111
+ ## Model Details
112
+
113
+ ### Model Description
114
+ - **Model Type:** Sentence Transformer
115
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
116
+ - **Maximum Sequence Length:** 512 tokens
117
+ - **Output Dimensionality:** 768 dimensions
118
+ - **Similarity Function:** Cosine Similarity
119
+ - **Training Dataset:**
120
+ - train
121
+ <!-- - **Language:** Unknown -->
122
+ <!-- - **License:** Unknown -->
123
+
124
+ ### Model Sources
125
+
126
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
127
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers)
128
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
129
+
130
+ ### Full Model Architecture
131
+
132
+ ```
133
+ SentenceTransformer(
134
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
135
+ (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
136
+ )
137
+ ```
138
+
139
+ ## Usage
140
+
141
+ ### Direct Usage (Sentence Transformers)
142
+
143
+ First install the Sentence Transformers library:
144
+
145
+ ```bash
146
+ pip install -U sentence-transformers
147
+ ```
148
+
149
+ Then you can load this model and run inference.
150
+ ```python
151
+ from sentence_transformers import SentenceTransformer
152
+
153
+ # Download from the 🤗 Hub
154
+ model = SentenceTransformer("sentence_transformers_model_id")
155
+ # Run inference
156
+ sentences = [
157
+ 'رجل وامرأة يجلسان في سيارة ووجههما في الاتجاه المعاكس من الكاميرا',
158
+ 'هناك شخصان وسيارة',
159
+ 'سيارة صدئة هي الشيء الوحيد المرئي',
160
+ ]
161
+ embeddings = model.encode(sentences)
162
+ print(embeddings.shape)
163
+ # [3, 768]
164
+
165
+ # Get the similarity scores for the embeddings
166
+ similarities = model.similarity(embeddings, embeddings)
167
+ print(similarities)
168
+ # tensor([[1.0000, 0.6451, 0.3299],
169
+ # [0.6451, 1.0000, 0.4022],
170
+ # [0.3299, 0.4022, 1.0000]])
171
+ ```
172
+
173
+ <!--
174
+ ### Direct Usage (Transformers)
175
+
176
+ <details><summary>Click to see the direct usage in Transformers</summary>
177
+
178
+ </details>
179
+ -->
180
+
181
+ <!--
182
+ ### Downstream Usage (Sentence Transformers)
183
+
184
+ You can finetune this model on your own dataset.
185
+
186
+ <details><summary>Click to expand</summary>
187
+
188
+ </details>
189
+ -->
190
+
191
+ <!--
192
+ ### Out-of-Scope Use
193
+
194
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
195
+ -->
196
+
197
+ ## Evaluation
198
+
199
+ ### Metrics
200
+
201
+ #### Triplet
202
+
203
+ * Dataset: `dev-768`
204
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) with these parameters:
205
+ ```json
206
+ {
207
+ "truncate_dim": 768
208
+ }
209
+ ```
210
+
211
+ | Metric | Value |
212
+ |:--------------------|:----------|
213
+ | **cosine_accuracy** | **0.981** |
214
+
215
+ #### Triplet
216
+
217
+ * Dataset: `dev-512`
218
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) with these parameters:
219
+ ```json
220
+ {
221
+ "truncate_dim": 512
222
+ }
223
+ ```
224
+
225
+ | Metric | Value |
226
+ |:--------------------|:-----------|
227
+ | **cosine_accuracy** | **0.9811** |
228
+
229
+ #### Triplet
230
+
231
+ * Dataset: `dev-256`
232
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) with these parameters:
233
+ ```json
234
+ {
235
+ "truncate_dim": 256
236
+ }
237
+ ```
238
+
239
+ | Metric | Value |
240
+ |:--------------------|:-----------|
241
+ | **cosine_accuracy** | **0.9813** |
242
+
243
+ #### Triplet
244
+
245
+ * Dataset: `dev-128`
246
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) with these parameters:
247
+ ```json
248
+ {
249
+ "truncate_dim": 128
250
+ }
251
+ ```
252
+
253
+ | Metric | Value |
254
+ |:--------------------|:-----------|
255
+ | **cosine_accuracy** | **0.9811** |
256
+
257
+ #### Triplet
258
+
259
+ * Dataset: `dev-64`
260
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) with these parameters:
261
+ ```json
262
+ {
263
+ "truncate_dim": 64
264
+ }
265
+ ```
266
+
267
+ | Metric | Value |
268
+ |:--------------------|:-----------|
269
+ | **cosine_accuracy** | **0.9797** |
270
+
271
+ <!--
272
+ ## Bias, Risks and Limitations
273
+
274
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
275
+ -->
276
+
277
+ <!--
278
+ ### Recommendations
279
+
280
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
281
+ -->
282
+
283
+ ## Training Details
284
+
285
+ ### Training Dataset
286
+
287
+ #### train
288
+
289
+ * Dataset: train
290
+ * Size: 3,954,179 training samples
291
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
292
+ * Approximate statistics based on the first 1000 samples:
293
+ | | anchor | positive | negative |
294
+ |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
295
+ | type | string | string | string |
296
+ | details | <ul><li>min: 4 tokens</li><li>mean: 16.1 tokens</li><li>max: 113 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 41.85 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 41.99 tokens</li><li>max: 512 tokens</li></ul> |
297
+ * Samples:
298
+ | anchor | positive | negative |
299
+ |:----------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
300
+ | <code>في أي مقاطعة تقع لويسفيل أركنساس</code> | <code>لويسفيل هي بلدة في مقاطعة لافاييت ، أركنساس ، الولايات المتحدة . كان عدد السكان 1285 في تعداد عام 2000 . . المدينة هي مقر مقاطعة لافاييت .</code> | <code>ماونتن هوم ، أركنساس . ماونتن هوم هي مدينة صغيرة في مقاطعة باكستر ، أركنساس ، الولايات المتحدة ، في جبال أوزارك الجنوبية بالقرب من حدود الولاية الشمالية مع ميسوري . اعتبارا من تعداد عام 2010 ، بلغ عدد سكان المدينة 12448 نسمة .</code> |
301
+ | <code>متوسط سمك باب الخزانة</code> | <code>تتميز أبواب العالم القديم بميزات رائعة مثل السماكة المتزايدة ، والملامح الأعمق ، والأعمدة والقضبان الأوسع لإضفاء مظهر وإحساس أكثر دراماتيكية عند مقارنتها بأبواب الخزانة التقليدية . يبلغ عرض Stiles Rails القياسية 3 بوصات ويمكن تصنيعها في 1 و 1 1 - 8 و 1 سمك .</code> | <code>اعتمادا على الخطأ في اللوحة ، يبلغ متوسط أسعار الإصلاح 130 دولارا لإصلاح الأبواب الفولاذية و 190 دولارا للخشب و 170 دولارا للألمنيوم و 150 دولارا للألياف الزجاجية . مزيد من المعلومات حول كيفية استبدال لوحة باب المرآب . إذا تعطلت أداة فتح باب الجراج ، فقد تكون سلامتك في خطر . تريد التأكد من أن بابك يعمل بشكل صحيح حتى لا يغلق بطريق الخطأ على حيوان أليف أو شخص . تريد أيضا إغلاقها لإبعاد اللصوص عن منزلك .</code> |
302
+ | <code>ما هو تعريف الملء</code> | <code>اعادة تعبئه . اسم تخصيص ثان لوكيل الوصفات الطبية تم الحصول عليه من الصيدلية ، والذي يسمح به فعل الوصفة الأصلية علم الأدوية للحصول على المزيد من دواء معين ، بعد استخدام الكمية الموصوفة في البداية من الوكيل أو إعطائها . انظر الوصفة الطبية .</code> | <code>تعليمات إعادة الملء قم بإعادة الملء فقط باستخدام Spectracide ' Bug Stop Home Barrier Refill . قم بإزالة الغطاء . قم بقياس وصب 12 . 8 أونصة سائلة من المركز في حاوية فارغة سعة 1 جالون من Spectracide - Bug Stop - حاجز منزلي ، واملأه حتى 1 جالون بالماء ، استبدل الغطاء وأغلقه بإحكام . المنتج المنسكب قم بقياس 12 . 8 أونصة سائلة من المركز وصبها بحذر في حاوية فارغة سعة 1 جالون من Spectracide - حاجز منزلي من Spectracide - حاجز منزلي ، واملأه حتى 1 جالون بالماء . استبدل الغطاء وأغلقه بإحكام . امسح أي منتج مسكوب .</code> |
303
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
304
+ ```json
305
+ {
306
+ "loss": "MultipleNegativesRankingLoss",
307
+ "matryoshka_dims": [
308
+ 768,
309
+ 512,
310
+ 256,
311
+ 128,
312
+ 64
313
+ ],
314
+ "matryoshka_weights": [
315
+ 1,
316
+ 1,
317
+ 1,
318
+ 1,
319
+ 1
320
+ ],
321
+ "n_dims_per_step": -1
322
+ }
323
+ ```
324
+
325
+ ### Evaluation Dataset
326
+
327
+ #### train
328
+
329
+ * Dataset: train
330
+ * Size: 1,129,759 evaluation samples
331
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
332
+ * Approximate statistics based on the first 1000 samples:
333
+ | | anchor | positive | negative |
334
+ |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
335
+ | type | string | string | string |
336
+ | details | <ul><li>min: 4 tokens</li><li>mean: 16.7 tokens</li><li>max: 150 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 36.54 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 35.42 tokens</li><li>max: 512 tokens</li></ul> |
337
+ * Samples:
338
+ | anchor | positive | negative |
339
+ |:---------------------------------------------------------------------|:---------------------------------|:----------------------------------------------------------------------|
340
+ | <code>رجل يرتدي سروال تنس أزرق وقميص بولو أبيض يضرب كرة التنس</code> | <code>رجل يلعب رياضة</code> | <code>هناك رجل يرتدي زي البيسبول يضرب كرة البيسبول بمضرب التنس</code> |
341
+ | <code>امرأة في ثوب أسود تبدو متفاجئة</code> | <code>امرأة تغيرت مشاعرها</code> | <code>امرأة تسبح في المحيط</code> |
342
+ | <code>رجل يرتدي قميص أبيض يقفز على شيء ما على دراجته الصفراء</code> | <code>رجل يركب دراجته</code> | <code>رجل يركب لوح التزلج فوق المنحدر</code> |
343
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
344
+ ```json
345
+ {
346
+ "loss": "MultipleNegativesRankingLoss",
347
+ "matryoshka_dims": [
348
+ 768,
349
+ 512,
350
+ 256,
351
+ 128,
352
+ 64
353
+ ],
354
+ "matryoshka_weights": [
355
+ 1,
356
+ 1,
357
+ 1,
358
+ 1,
359
+ 1
360
+ ],
361
+ "n_dims_per_step": -1
362
+ }
363
+ ```
364
+
365
+ ### Training Hyperparameters
366
+ #### Non-Default Hyperparameters
367
+
368
+ - `per_device_train_batch_size`: 64
369
+ - `num_train_epochs`: 2
370
+ - `learning_rate`: 2e-05
371
+ - `warmup_steps`: 0.1
372
+ - `gradient_accumulation_steps`: 2
373
+ - `bf16`: True
374
+ - `eval_strategy`: steps
375
+ - `warmup_ratio`: 0.1
376
+ - `batch_sampler`: no_duplicates
377
+
378
+ #### All Hyperparameters
379
+ <details><summary>Click to expand</summary>
380
+
381
+ - `per_device_train_batch_size`: 64
382
+ - `num_train_epochs`: 2
383
+ - `max_steps`: -1
384
+ - `learning_rate`: 2e-05
385
+ - `lr_scheduler_type`: linear
386
+ - `lr_scheduler_kwargs`: None
387
+ - `warmup_steps`: 0.1
388
+ - `optim`: adamw_torch
389
+ - `optim_args`: None
390
+ - `weight_decay`: 0.0
391
+ - `adam_beta1`: 0.9
392
+ - `adam_beta2`: 0.999
393
+ - `adam_epsilon`: 1e-08
394
+ - `optim_target_modules`: None
395
+ - `gradient_accumulation_steps`: 2
396
+ - `average_tokens_across_devices`: True
397
+ - `max_grad_norm`: 1.0
398
+ - `label_smoothing_factor`: 0.0
399
+ - `bf16`: True
400
+ - `fp16`: False
401
+ - `bf16_full_eval`: False
402
+ - `fp16_full_eval`: False
403
+ - `tf32`: None
404
+ - `gradient_checkpointing`: False
405
+ - `gradient_checkpointing_kwargs`: None
406
+ - `torch_compile`: False
407
+ - `torch_compile_backend`: None
408
+ - `torch_compile_mode`: None
409
+ - `use_liger_kernel`: False
410
+ - `liger_kernel_config`: None
411
+ - `use_cache`: False
412
+ - `neftune_noise_alpha`: None
413
+ - `torch_empty_cache_steps`: None
414
+ - `auto_find_batch_size`: False
415
+ - `log_on_each_node`: True
416
+ - `logging_nan_inf_filter`: True
417
+ - `include_num_input_tokens_seen`: no
418
+ - `log_level`: passive
419
+ - `log_level_replica`: warning
420
+ - `disable_tqdm`: False
421
+ - `project`: huggingface
422
+ - `trackio_space_id`: trackio
423
+ - `eval_strategy`: steps
424
+ - `per_device_eval_batch_size`: 8
425
+ - `prediction_loss_only`: True
426
+ - `eval_on_start`: False
427
+ - `eval_do_concat_batches`: True
428
+ - `eval_use_gather_object`: False
429
+ - `eval_accumulation_steps`: None
430
+ - `include_for_metrics`: []
431
+ - `batch_eval_metrics`: False
432
+ - `save_only_model`: False
433
+ - `save_on_each_node`: False
434
+ - `enable_jit_checkpoint`: False
435
+ - `push_to_hub`: False
436
+ - `hub_private_repo`: None
437
+ - `hub_model_id`: None
438
+ - `hub_strategy`: every_save
439
+ - `hub_always_push`: False
440
+ - `hub_revision`: None
441
+ - `load_best_model_at_end`: False
442
+ - `ignore_data_skip`: False
443
+ - `restore_callback_states_from_checkpoint`: False
444
+ - `full_determinism`: False
445
+ - `seed`: 42
446
+ - `data_seed`: None
447
+ - `use_cpu`: False
448
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
449
+ - `parallelism_config`: None
450
+ - `dataloader_drop_last`: False
451
+ - `dataloader_num_workers`: 0
452
+ - `dataloader_pin_memory`: True
453
+ - `dataloader_persistent_workers`: False
454
+ - `dataloader_prefetch_factor`: None
455
+ - `remove_unused_columns`: True
456
+ - `label_names`: None
457
+ - `train_sampling_strategy`: random
458
+ - `length_column_name`: length
459
+ - `ddp_find_unused_parameters`: None
460
+ - `ddp_bucket_cap_mb`: None
461
+ - `ddp_broadcast_buffers`: False
462
+ - `ddp_backend`: None
463
+ - `ddp_timeout`: 1800
464
+ - `fsdp`: []
465
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
466
+ - `deepspeed`: None
467
+ - `debug`: []
468
+ - `skip_memory_metrics`: True
469
+ - `do_predict`: False
470
+ - `resume_from_checkpoint`: None
471
+ - `warmup_ratio`: 0.1
472
+ - `local_rank`: -1
473
+ - `prompts`: None
474
+ - `batch_sampler`: no_duplicates
475
+ - `multi_dataset_batch_sampler`: proportional
476
+ - `router_mapping`: {}
477
+ - `learning_rate_mapping`: {}
478
+
479
+ </details>
480
+
481
+ ### Training Logs
482
+ <details><summary>Click to expand</summary>
483
+
484
+ | Epoch | Step | Training Loss | train loss | dev-768_cosine_accuracy | dev-512_cosine_accuracy | dev-256_cosine_accuracy | dev-128_cosine_accuracy | dev-64_cosine_accuracy |
485
+ |:------:|:-----:|:-------------:|:----------:|:-----------------------:|:-----------------------:|:-----------------------:|:-----------------------:|:----------------------:|
486
+ | 0.5891 | 18200 | 0.8727 | - | - | - | - | - | - |
487
+ | 0.5956 | 18400 | 0.8524 | - | - | - | - | - | - |
488
+ | 0.6021 | 18600 | 0.8995 | - | - | - | - | - | - |
489
+ | 0.6086 | 18800 | 0.8360 | - | - | - | - | - | - |
490
+ | 0.6150 | 19000 | 0.8628 | - | - | - | - | - | - |
491
+ | 0.6215 | 19200 | 0.8244 | - | - | - | - | - | - |
492
+ | 0.6280 | 19400 | 0.8647 | - | - | - | - | - | - |
493
+ | 0.6345 | 19600 | 0.8479 | - | - | - | - | - | - |
494
+ | 0.6409 | 19800 | 0.8204 | - | - | - | - | - | - |
495
+ | 0.6474 | 20000 | 0.8359 | - | - | - | - | - | - |
496
+ | 0.6539 | 20200 | 0.7952 | - | - | - | - | - | - |
497
+ | 0.6604 | 20400 | 0.8375 | - | - | - | - | - | - |
498
+ | 0.6668 | 20600 | 0.8364 | - | - | - | - | - | - |
499
+ | 0.6733 | 20800 | 0.8131 | - | - | - | - | - | - |
500
+ | 0.6798 | 21000 | 0.8310 | - | - | - | - | - | - |
501
+ | 0.6863 | 21200 | 0.8295 | - | - | - | - | - | - |
502
+ | 0.6927 | 21400 | 0.7865 | - | - | - | - | - | - |
503
+ | 0.6992 | 21600 | 0.7960 | - | - | - | - | - | - |
504
+ | 0.7057 | 21800 | 0.8287 | - | - | - | - | - | - |
505
+ | 0.7121 | 22000 | 0.8214 | - | - | - | - | - | - |
506
+ | 0.7186 | 22200 | 0.7879 | - | - | - | - | - | - |
507
+ | 0.7251 | 22400 | 0.8139 | - | - | - | - | - | - |
508
+ | 0.7316 | 22600 | 0.7849 | - | - | - | - | - | - |
509
+ | 0.7380 | 22800 | 0.7880 | - | - | - | - | - | - |
510
+ | 0.7445 | 23000 | 0.7725 | - | - | - | - | - | - |
511
+ | 0.7510 | 23200 | 0.8086 | - | - | - | - | - | - |
512
+ | 0.7575 | 23400 | 0.7687 | - | - | - | - | - | - |
513
+ | 0.7639 | 23600 | 0.7828 | - | - | - | - | - | - |
514
+ | 0.7704 | 23800 | 0.7518 | - | - | - | - | - | - |
515
+ | 0.7769 | 24000 | 0.7599 | 0.4041 | 0.9737 | 0.9738 | 0.9738 | 0.9734 | 0.9718 |
516
+ | 0.7834 | 24200 | 0.7332 | - | - | - | - | - | - |
517
+ | 0.7898 | 24400 | 0.7476 | - | - | - | - | - | - |
518
+ | 0.7963 | 24600 | 0.7806 | - | - | - | - | - | - |
519
+ | 0.8028 | 24800 | 0.7511 | - | - | - | - | - | - |
520
+ | 0.8093 | 25000 | 0.7652 | - | - | - | - | - | - |
521
+ | 0.8157 | 25200 | 0.7883 | - | - | - | - | - | - |
522
+ | 0.8222 | 25400 | 0.7305 | - | - | - | - | - | - |
523
+ | 0.8287 | 25600 | 0.7308 | - | - | - | - | - | - |
524
+ | 0.8352 | 25800 | 0.7368 | - | - | - | - | - | - |
525
+ | 0.8416 | 26000 | 0.7432 | - | - | - | - | - | - |
526
+ | 0.8481 | 26200 | 0.7046 | - | - | - | - | - | - |
527
+ | 0.8546 | 26400 | 0.7476 | - | - | - | - | - | - |
528
+ | 0.8611 | 26600 | 0.7212 | - | - | - | - | - | - |
529
+ | 0.8675 | 26800 | 0.7335 | - | - | - | - | - | - |
530
+ | 0.8740 | 27000 | 0.7415 | - | - | - | - | - | - |
531
+ | 0.8805 | 27200 | 0.6937 | - | - | - | - | - | - |
532
+ | 0.8869 | 27400 | 0.7294 | - | - | - | - | - | - |
533
+ | 0.8934 | 27600 | 0.7436 | - | - | - | - | - | - |
534
+ | 0.8999 | 27800 | 0.7093 | - | - | - | - | - | - |
535
+ | 0.9064 | 28000 | 0.7480 | - | - | - | - | - | - |
536
+ | 0.9128 | 28200 | 0.7039 | - | - | - | - | - | - |
537
+ | 0.9193 | 28400 | 0.7091 | - | - | - | - | - | - |
538
+ | 0.9258 | 28600 | 0.7019 | - | - | - | - | - | - |
539
+ | 0.9323 | 28800 | 0.7081 | - | - | - | - | - | - |
540
+ | 0.9387 | 29000 | 0.6833 | - | - | - | - | - | - |
541
+ | 0.9452 | 29200 | 0.6982 | - | - | - | - | - | - |
542
+ | 0.9517 | 29400 | 0.7249 | - | - | - | - | - | - |
543
+ | 0.9582 | 29600 | 0.7282 | - | - | - | - | - | - |
544
+ | 0.9646 | 29800 | 0.7147 | - | - | - | - | - | - |
545
+ | 0.9711 | 30000 | 0.6742 | 0.3640 | 0.9758 | 0.9759 | 0.9761 | 0.9757 | 0.9742 |
546
+ | 0.9776 | 30200 | 0.6901 | - | - | - | - | - | - |
547
+ | 0.9841 | 30400 | 0.7067 | - | - | - | - | - | - |
548
+ | 0.9905 | 30600 | 0.7166 | - | - | - | - | - | - |
549
+ | 0.9970 | 30800 | 0.6800 | - | - | - | - | - | - |
550
+ | 1.0035 | 31000 | 0.6846 | - | - | - | - | - | - |
551
+ | 1.0099 | 31200 | 0.6723 | - | - | - | - | - | - |
552
+ | 1.0164 | 31400 | 0.6573 | - | - | - | - | - | - |
553
+ | 1.0229 | 31600 | 0.6895 | - | - | - | - | - | - |
554
+ | 1.0294 | 31800 | 0.6588 | - | - | - | - | - | - |
555
+ | 1.0358 | 32000 | 0.6517 | - | - | - | - | - | - |
556
+ | 1.0423 | 32200 | 0.6498 | - | - | - | - | - | - |
557
+ | 1.0488 | 32400 | 0.6836 | - | - | - | - | - | - |
558
+ | 1.0553 | 32600 | 0.6819 | - | - | - | - | - | - |
559
+ | 1.0617 | 32800 | 0.6463 | - | - | - | - | - | - |
560
+ | 1.0682 | 33000 | 0.6645 | - | - | - | - | - | - |
561
+ | 1.0747 | 33200 | 0.6518 | - | - | - | - | - | - |
562
+ | 1.0812 | 33400 | 0.6235 | - | - | - | - | - | - |
563
+ | 1.0876 | 33600 | 0.6302 | - | - | - | - | - | - |
564
+ | 1.0941 | 33800 | 0.6452 | - | - | - | - | - | - |
565
+ | 1.1006 | 34000 | 0.6477 | - | - | - | - | - | - |
566
+ | 1.1070 | 34200 | 0.6084 | - | - | - | - | - | - |
567
+ | 1.1135 | 34400 | 0.6259 | - | - | - | - | - | - |
568
+ | 1.1200 | 34600 | 0.6070 | - | - | - | - | - | - |
569
+ | 1.1265 | 34800 | 0.5977 | - | - | - | - | - | - |
570
+ | 1.1329 | 35000 | 0.6044 | - | - | - | - | - | - |
571
+ | 1.1394 | 35200 | 0.6007 | - | - | - | - | - | - |
572
+ | 1.1459 | 35400 | 0.5628 | - | - | - | - | - | - |
573
+ | 1.1524 | 35600 | 0.5732 | - | - | - | - | - | - |
574
+ | 1.1588 | 35800 | 0.5773 | - | - | - | - | - | - |
575
+ | 1.1653 | 36000 | 0.5719 | 0.3356 | 0.9775 | 0.9777 | 0.9777 | 0.9774 | 0.9760 |
576
+ | 1.1718 | 36200 | 0.5471 | - | - | - | - | - | - |
577
+ | 1.1783 | 36400 | 0.5635 | - | - | - | - | - | - |
578
+ | 1.1847 | 36600 | 0.5390 | - | - | - | - | - | - |
579
+ | 1.1912 | 36800 | 0.5428 | - | - | - | - | - | - |
580
+ | 1.1977 | 37000 | 0.5205 | - | - | - | - | - | - |
581
+ | 1.2042 | 37200 | 0.5362 | - | - | - | - | - | - |
582
+ | 1.2106 | 37400 | 0.5386 | - | - | - | - | - | - |
583
+ | 1.2171 | 37600 | 0.5203 | - | - | - | - | - | - |
584
+ | 1.2236 | 37800 | 0.5301 | - | - | - | - | - | - |
585
+ | 1.2301 | 38000 | 0.5232 | - | - | - | - | - | - |
586
+ | 1.2365 | 38200 | 0.4922 | - | - | - | - | - | - |
587
+ | 1.2430 | 38400 | 0.5029 | - | - | - | - | - | - |
588
+ | 1.2495 | 38600 | 0.4989 | - | - | - | - | - | - |
589
+ | 1.2560 | 38800 | 0.5053 | - | - | - | - | - | - |
590
+ | 1.2624 | 39000 | 0.5081 | - | - | - | - | - | - |
591
+ | 1.2689 | 39200 | 0.4960 | - | - | - | - | - | - |
592
+ | 1.2754 | 39400 | 0.5052 | - | - | - | - | - | - |
593
+ | 1.2818 | 39600 | 0.4984 | - | - | - | - | - | - |
594
+ | 1.2883 | 39800 | 0.4909 | - | - | - | - | - | - |
595
+ | 1.2948 | 40000 | 0.5120 | - | - | - | - | - | - |
596
+ | 1.3013 | 40200 | 0.4873 | - | - | - | - | - | - |
597
+ | 1.3077 | 40400 | 0.4896 | - | - | - | - | - | - |
598
+ | 1.3142 | 40600 | 0.4900 | - | - | - | - | - | - |
599
+ | 1.3207 | 40800 | 0.5036 | - | - | - | - | - | - |
600
+ | 1.3272 | 41000 | 0.4876 | - | - | - | - | - | - |
601
+ | 1.3336 | 41200 | 0.4705 | - | - | - | - | - | - |
602
+ | 1.3401 | 41400 | 0.4786 | - | - | - | - | - | - |
603
+ | 1.3466 | 41600 | 0.4998 | - | - | - | - | - | - |
604
+ | 1.3531 | 41800 | 0.4692 | - | - | - | - | - | - |
605
+ | 1.3595 | 42000 | 0.5064 | 0.3160 | 0.9788 | 0.9790 | 0.9790 | 0.9785 | 0.9774 |
606
+ | 1.3660 | 42200 | 0.4925 | - | - | - | - | - | - |
607
+ | 1.3725 | 42400 | 0.4601 | - | - | - | - | - | - |
608
+ | 1.3790 | 42600 | 0.4762 | - | - | - | - | - | - |
609
+ | 1.3854 | 42800 | 0.4986 | - | - | - | - | - | - |
610
+ | 1.3919 | 43000 | 0.4656 | - | - | - | - | - | - |
611
+ | 1.3984 | 43200 | 0.4507 | - | - | - | - | - | - |
612
+ | 1.4049 | 43400 | 0.4862 | - | - | - | - | - | - |
613
+ | 1.4113 | 43600 | 0.4596 | - | - | - | - | - | - |
614
+ | 1.4178 | 43800 | 0.4696 | - | - | - | - | - | - |
615
+ | 1.4243 | 44000 | 0.4925 | - | - | - | - | - | - |
616
+ | 1.4308 | 44200 | 0.4796 | - | - | - | - | - | - |
617
+ | 1.4372 | 44400 | 0.4525 | - | - | - | - | - | - |
618
+ | 1.4437 | 44600 | 0.4717 | - | - | - | - | - | - |
619
+ | 1.4502 | 44800 | 0.4803 | - | - | - | - | - | - |
620
+ | 1.4566 | 45000 | 0.4675 | - | - | - | - | - | - |
621
+ | 1.4631 | 45200 | 0.4631 | - | - | - | - | - | - |
622
+ | 1.4696 | 45400 | 0.4622 | - | - | - | - | - | - |
623
+ | 1.4761 | 45600 | 0.4496 | - | - | - | - | - | - |
624
+ | 1.4825 | 45800 | 0.4678 | - | - | - | - | - | - |
625
+ | 1.4890 | 46000 | 0.4495 | - | - | - | - | - | - |
626
+ | 1.4955 | 46200 | 0.4474 | - | - | - | - | - | - |
627
+ | 1.5020 | 46400 | 0.4587 | - | - | - | - | - | - |
628
+ | 1.5084 | 46600 | 0.4591 | - | - | - | - | - | - |
629
+ | 1.5149 | 46800 | 0.4573 | - | - | - | - | - | - |
630
+ | 1.5214 | 47000 | 0.4442 | - | - | - | - | - | - |
631
+ | 1.5279 | 47200 | 0.4550 | - | - | - | - | - | - |
632
+ | 1.5343 | 47400 | 0.4493 | - | - | - | - | - | - |
633
+ | 1.5408 | 47600 | 0.4485 | - | - | - | - | - | - |
634
+ | 1.5473 | 47800 | 0.4569 | - | - | - | - | - | - |
635
+ | 1.5538 | 48000 | 0.4346 | 0.3001 | 0.9799 | 0.9802 | 0.9802 | 0.9798 | 0.9788 |
636
+ | 1.5602 | 48200 | 0.4469 | - | - | - | - | - | - |
637
+ | 1.5667 | 48400 | 0.4602 | - | - | - | - | - | - |
638
+ | 1.5732 | 48600 | 0.4430 | - | - | - | - | - | - |
639
+ | 1.5797 | 48800 | 0.4524 | - | - | - | - | - | - |
640
+ | 1.5861 | 49000 | 0.4528 | - | - | - | - | - | - |
641
+ | 1.5926 | 49200 | 0.4348 | - | - | - | - | - | - |
642
+ | 1.5991 | 49400 | 0.4533 | - | - | - | - | - | - |
643
+ | 1.6056 | 49600 | 0.4523 | - | - | - | - | - | - |
644
+ | 1.6120 | 49800 | 0.4509 | - | - | - | - | - | - |
645
+ | 1.6185 | 50000 | 0.4365 | - | - | - | - | - | - |
646
+ | 1.6250 | 50200 | 0.4504 | - | - | - | - | - | - |
647
+ | 1.6314 | 50400 | 0.4292 | - | - | - | - | - | - |
648
+ | 1.6379 | 50600 | 0.4406 | - | - | - | - | - | - |
649
+ | 1.6444 | 50800 | 0.4333 | - | - | - | - | - | - |
650
+ | 1.6509 | 51000 | 0.4361 | - | - | - | - | - | - |
651
+ | 1.6573 | 51200 | 0.4065 | - | - | - | - | - | - |
652
+ | 1.6638 | 51400 | 0.4671 | - | - | - | - | - | - |
653
+ | 1.6703 | 51600 | 0.4328 | - | - | - | - | - | - |
654
+ | 1.6768 | 51800 | 0.4310 | - | - | - | - | - | - |
655
+ | 1.6832 | 52000 | 0.4523 | - | - | - | - | - | - |
656
+ | 1.6897 | 52200 | 0.4232 | - | - | - | - | - | - |
657
+ | 1.6962 | 52400 | 0.4257 | - | - | - | - | - | - |
658
+ | 1.7027 | 52600 | 0.4448 | - | - | - | - | - | - |
659
+ | 1.7091 | 52800 | 0.4491 | - | - | - | - | - | - |
660
+ | 1.7156 | 53000 | 0.4224 | - | - | - | - | - | - |
661
+ | 1.7221 | 53200 | 0.4297 | - | - | - | - | - | - |
662
+ | 1.7286 | 53400 | 0.4522 | - | - | - | - | - | - |
663
+ | 1.7350 | 53600 | 0.4195 | - | - | - | - | - | - |
664
+ | 1.7415 | 53800 | 0.4227 | - | - | - | - | - | - |
665
+ | 1.7480 | 54000 | 0.4381 | 0.2875 | 0.9807 | 0.9808 | 0.9808 | 0.9805 | 0.9794 |
666
+ | 1.7545 | 54200 | 0.4460 | - | - | - | - | - | - |
667
+ | 1.7609 | 54400 | 0.4260 | - | - | - | - | - | - |
668
+ | 1.7674 | 54600 | 0.4299 | - | - | - | - | - | - |
669
+ | 1.7739 | 54800 | 0.4247 | - | - | - | - | - | - |
670
+ | 1.7804 | 55000 | 0.4244 | - | - | - | - | - | - |
671
+ | 1.7868 | 55200 | 0.4185 | - | - | - | - | - | - |
672
+ | 1.7933 | 55400 | 0.4292 | - | - | - | - | - | - |
673
+ | 1.7998 | 55600 | 0.4468 | - | - | - | - | - | - |
674
+ | 1.8062 | 55800 | 0.4118 | - | - | - | - | - | - |
675
+ | 1.8127 | 56000 | 0.4306 | - | - | - | - | - | - |
676
+ | 1.8192 | 56200 | 0.4447 | - | - | - | - | - | - |
677
+ | 1.8257 | 56400 | 0.4147 | - | - | - | - | - | - |
678
+ | 1.8321 | 56600 | 0.4189 | - | - | - | - | - | - |
679
+ | 1.8386 | 56800 | 0.4167 | - | - | - | - | - | - |
680
+ | 1.8451 | 57000 | 0.4022 | - | - | - | - | - | - |
681
+ | 1.8516 | 57200 | 0.4158 | - | - | - | - | - | - |
682
+ | 1.8580 | 57400 | 0.4228 | - | - | - | - | - | - |
683
+ | 1.8645 | 57600 | 0.4256 | - | - | - | - | - | - |
684
+ | 1.8710 | 57800 | 0.4251 | - | - | - | - | - | - |
685
+ | 1.8775 | 58000 | 0.4232 | - | - | - | - | - | - |
686
+ | 1.8839 | 58200 | 0.4143 | - | - | - | - | - | - |
687
+ | 1.8904 | 58400 | 0.4331 | - | - | - | - | - | - |
688
+ | 1.8969 | 58600 | 0.4253 | - | - | - | - | - | - |
689
+ | 1.9034 | 58800 | 0.4410 | - | - | - | - | - | - |
690
+ | 1.9098 | 59000 | 0.4337 | - | - | - | - | - | - |
691
+ | 1.9163 | 59200 | 0.4016 | - | - | - | - | - | - |
692
+ | 1.9228 | 59400 | 0.4249 | - | - | - | - | - | - |
693
+ | 1.9293 | 59600 | 0.4108 | - | - | - | - | - | - |
694
+ | 1.9357 | 59800 | 0.4272 | - | - | - | - | - | - |
695
+ | 1.9422 | 60000 | 0.3916 | 0.2812 | 0.9810 | 0.9811 | 0.9813 | 0.9811 | 0.9797 |
696
+ | 1.9487 | 60200 | 0.4334 | - | - | - | - | - | - |
697
+ | 1.9552 | 60400 | 0.4462 | - | - | - | - | - | - |
698
+ | 1.9616 | 60600 | 0.4436 | - | - | - | - | - | - |
699
+ | 1.9681 | 60800 | 0.4278 | - | - | - | - | - | - |
700
+ | 1.9746 | 61000 | 0.4170 | - | - | - | - | - | - |
701
+ | 1.9810 | 61200 | 0.4376 | - | - | - | - | - | - |
702
+ | 1.9875 | 61400 | 0.4433 | - | - | - | - | - | - |
703
+ | 1.9940 | 61600 | 0.4292 | - | - | - | - | - | - |
704
+
705
+ </details>
706
+
707
+ ### Framework Versions
708
+ - Python: 3.10.19
709
+ - Sentence Transformers: 5.2.3
710
+ - Transformers: 5.2.0
711
+ - PyTorch: 2.6.0+cu124
712
+ - Accelerate: 1.12.0
713
+ - Datasets: 4.5.0
714
+ - Tokenizers: 0.22.2
715
+
716
+ ## Citation
717
+
718
+ ### BibTeX
719
+
720
+ #### Sentence Transformers
721
+ ```bibtex
722
+ @inproceedings{reimers-2019-sentence-bert,
723
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
724
+ author = "Reimers, Nils and Gurevych, Iryna",
725
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
726
+ month = "11",
727
+ year = "2019",
728
+ publisher = "Association for Computational Linguistics",
729
+ url = "https://arxiv.org/abs/1908.10084",
730
+ }
731
+ ```
732
+
733
+ #### MatryoshkaLoss
734
+ ```bibtex
735
+ @misc{kusupati2024matryoshka,
736
+ title={Matryoshka Representation Learning},
737
+ author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
738
+ year={2024},
739
+ eprint={2205.13147},
740
+ archivePrefix={arXiv},
741
+ primaryClass={cs.LG}
742
+ }
743
+ ```
744
+
745
+ #### MultipleNegativesRankingLoss
746
+ ```bibtex
747
+ @misc{henderson2017efficient,
748
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
749
+ author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
750
+ year={2017},
751
+ eprint={1705.00652},
752
+ archivePrefix={arXiv},
753
+ primaryClass={cs.CL}
754
+ }
755
+ ```
756
+
757
+ <!--
758
+ ## Glossary
759
+
760
+ *Clearly define terms in order to be accessible across audiences.*
761
+ -->
762
+
763
+ <!--
764
+ ## Model Card Authors
765
+
766
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
767
+ -->
768
+
769
+ <!--
770
+ ## Model Card Contact
771
+
772
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
773
+ -->
epoch2/model/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": null,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "is_decoder": false,
17
+ "layer_norm_eps": 1e-12,
18
+ "max_position_embeddings": 512,
19
+ "model_type": "bert",
20
+ "num_attention_heads": 12,
21
+ "num_hidden_layers": 12,
22
+ "pad_token_id": 0,
23
+ "tie_word_embeddings": true,
24
+ "transformers_version": "5.2.0",
25
+ "type_vocab_size": 2,
26
+ "use_cache": true,
27
+ "vocab_size": 64000
28
+ }
epoch2/model/config_sentence_transformers.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "SentenceTransformer",
3
+ "__version__": {
4
+ "sentence_transformers": "5.2.3",
5
+ "transformers": "5.2.0",
6
+ "pytorch": "2.6.0+cu124"
7
+ },
8
+ "prompts": {
9
+ "query": "",
10
+ "document": ""
11
+ },
12
+ "default_prompt_name": null,
13
+ "similarity_fn_name": "cosine"
14
+ }
epoch2/model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:602c1647dc30e982f7f4b007ac82f3a97ef88b815b598d41e6c8d075be10730f
3
+ size 540795728
epoch2/model/modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ }
14
+ ]
epoch2/model/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
epoch2/model/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
epoch2/model/tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "cls_token": "[CLS]",
4
+ "do_basic_tokenize": true,
5
+ "do_lower_case": false,
6
+ "is_local": true,
7
+ "mask_token": "[MASK]",
8
+ "max_len": 512,
9
+ "model_max_length": 512,
10
+ "never_split": [
11
+ "[بريد]",
12
+ "[مستخدم]",
13
+ "[رابط]"
14
+ ],
15
+ "pad_token": "[PAD]",
16
+ "sep_token": "[SEP]",
17
+ "strip_accents": null,
18
+ "tokenize_chinese_chars": true,
19
+ "tokenizer_class": "BertTokenizer",
20
+ "unk_token": "[UNK]"
21
+ }
epoch2/output.log ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ --- RESUMING FROM: /home/skiredj.abderrahman/khalil/sbert_training/output/arabert_20260224_1730/checkpoint-18000 ---
2
+ {'loss': '0.8727', 'grad_norm': '5.438', 'learning_rate': '1.568e-05', 'epoch': '0.5891'}
3
+ {'loss': '0.8524', 'grad_norm': '5.458', 'learning_rate': '1.56e-05', 'epoch': '0.5956'}
4
+ {'loss': '0.8995', 'grad_norm': '6.666', 'learning_rate': '1.553e-05', 'epoch': '0.6021'}
5
+ {'loss': '0.836', 'grad_norm': '5.681', 'learning_rate': '1.546e-05', 'epoch': '0.6086'}
6
+ {'loss': '0.8628', 'grad_norm': '6.571', 'learning_rate': '1.539e-05', 'epoch': '0.615'}
7
+ {'loss': '0.8244', 'grad_norm': '6.389', 'learning_rate': '1.532e-05', 'epoch': '0.6215'}
8
+ {'loss': '0.8647', 'grad_norm': '4.987', 'learning_rate': '1.525e-05', 'epoch': '0.628'}
9
+ {'loss': '0.8479', 'grad_norm': '4.451', 'learning_rate': '1.517e-05', 'epoch': '0.6345'}
10
+ {'loss': '0.8204', 'grad_norm': '5.356', 'learning_rate': '1.51e-05', 'epoch': '0.6409'}
11
+ {'loss': '0.8359', 'grad_norm': '5.146', 'learning_rate': '1.503e-05', 'epoch': '0.6474'}
12
+ {'loss': '0.7952', 'grad_norm': '4.308', 'learning_rate': '1.496e-05', 'epoch': '0.6539'}
13
+ {'loss': '0.8375', 'grad_norm': '5.216', 'learning_rate': '1.489e-05', 'epoch': '0.6604'}
14
+ {'loss': '0.8364', 'grad_norm': '5.812', 'learning_rate': '1.481e-05', 'epoch': '0.6668'}
15
+ {'loss': '0.8131', 'grad_norm': '5.52', 'learning_rate': '1.474e-05', 'epoch': '0.6733'}
16
+ {'loss': '0.831', 'grad_norm': '6.452', 'learning_rate': '1.467e-05', 'epoch': '0.6798'}
17
+ {'loss': '0.8295', 'grad_norm': '4.274', 'learning_rate': '1.46e-05', 'epoch': '0.6863'}
18
+ {'loss': '0.7865', 'grad_norm': '4.77', 'learning_rate': '1.453e-05', 'epoch': '0.6927'}
19
+ {'loss': '0.796', 'grad_norm': '5.027', 'learning_rate': '1.445e-05', 'epoch': '0.6992'}
20
+ {'loss': '0.8287', 'grad_norm': '4.826', 'learning_rate': '1.438e-05', 'epoch': '0.7057'}
21
+ {'loss': '0.8214', 'grad_norm': '4.381', 'learning_rate': '1.431e-05', 'epoch': '0.7121'}
22
+ {'loss': '0.7879', 'grad_norm': '6.475', 'learning_rate': '1.424e-05', 'epoch': '0.7186'}
23
+ {'loss': '0.8139', 'grad_norm': '5.295', 'learning_rate': '1.417e-05', 'epoch': '0.7251'}
24
+ {'loss': '0.7849', 'grad_norm': '5.051', 'learning_rate': '1.409e-05', 'epoch': '0.7316'}
25
+ {'loss': '0.788', 'grad_norm': '5.113', 'learning_rate': '1.402e-05', 'epoch': '0.738'}
26
+ {'loss': '0.7725', 'grad_norm': '4.049', 'learning_rate': '1.395e-05', 'epoch': '0.7445'}
27
+ {'loss': '0.8086', 'grad_norm': '4.646', 'learning_rate': '1.388e-05', 'epoch': '0.751'}
28
+ {'loss': '0.7687', 'grad_norm': '5.049', 'learning_rate': '1.381e-05', 'epoch': '0.7575'}
29
+ {'loss': '0.7828', 'grad_norm': '6.568', 'learning_rate': '1.373e-05', 'epoch': '0.7639'}
30
+ {'loss': '0.7518', 'grad_norm': '5.9', 'learning_rate': '1.366e-05', 'epoch': '0.7704'}
31
+ {'loss': '0.7599', 'grad_norm': '6.338', 'learning_rate': '1.359e-05', 'epoch': '0.7769'}
32
+ {'eval_train_loss': '0.4041', 'eval_dev-768_cosine_accuracy': '0.9737', 'eval_dev-512_cosine_accuracy': '0.9738', 'eval_dev-256_cosine_accuracy': '0.9738', 'eval_dev-128_cosine_accuracy': '0.9734', 'eval_dev-64_cosine_accuracy': '0.9718', 'eval_sequential_score': '0.9737', 'eval_train_runtime': '9673', 'eval_train_samples_per_second': '116.8', 'eval_train_steps_per_second': '14.6', 'epoch': '0.7769'}
33
+ {'loss': '0.7332', 'grad_norm': '4.95', 'learning_rate': '1.352e-05', 'epoch': '0.7834'}
34
+ {'loss': '0.7476', 'grad_norm': '4.513', 'learning_rate': '1.345e-05', 'epoch': '0.7898'}
35
+ {'loss': '0.7806', 'grad_norm': '5.095', 'learning_rate': '1.337e-05', 'epoch': '0.7963'}
36
+ {'loss': '0.7511', 'grad_norm': '5.826', 'learning_rate': '1.33e-05', 'epoch': '0.8028'}
37
+ {'loss': '0.7652', 'grad_norm': '6.09', 'learning_rate': '1.323e-05', 'epoch': '0.8093'}
38
+ {'loss': '0.7883', 'grad_norm': '4.332', 'learning_rate': '1.316e-05', 'epoch': '0.8157'}
39
+ {'loss': '0.7305', 'grad_norm': '5.749', 'learning_rate': '1.309e-05', 'epoch': '0.8222'}
40
+ {'loss': '0.7308', 'grad_norm': '4.871', 'learning_rate': '1.302e-05', 'epoch': '0.8287'}
41
+ {'loss': '0.7368', 'grad_norm': '4.618', 'learning_rate': '1.294e-05', 'epoch': '0.8352'}
42
+ {'loss': '0.7432', 'grad_norm': '4.836', 'learning_rate': '1.287e-05', 'epoch': '0.8416'}
43
+ {'loss': '0.7046', 'grad_norm': '4.988', 'learning_rate': '1.28e-05', 'epoch': '0.8481'}
44
+ {'loss': '0.7476', 'grad_norm': '4.596', 'learning_rate': '1.273e-05', 'epoch': '0.8546'}
45
+ {'loss': '0.7212', 'grad_norm': '5.712', 'learning_rate': '1.266e-05', 'epoch': '0.8611'}
46
+ {'loss': '0.7335', 'grad_norm': '3.99', 'learning_rate': '1.258e-05', 'epoch': '0.8675'}
47
+ {'loss': '0.7415', 'grad_norm': '5.446', 'learning_rate': '1.251e-05', 'epoch': '0.874'}
48
+ {'loss': '0.6937', 'grad_norm': '5.257', 'learning_rate': '1.244e-05', 'epoch': '0.8805'}
49
+ {'loss': '0.7294', 'grad_norm': '5.302', 'learning_rate': '1.237e-05', 'epoch': '0.8869'}
50
+ {'loss': '0.7436', 'grad_norm': '3.847', 'learning_rate': '1.23e-05', 'epoch': '0.8934'}
51
+ {'loss': '0.7093', 'grad_norm': '6.182', 'learning_rate': '1.222e-05', 'epoch': '0.8999'}
52
+ {'loss': '0.748', 'grad_norm': '5.445', 'learning_rate': '1.215e-05', 'epoch': '0.9064'}
53
+ {'loss': '0.7039', 'grad_norm': '5.002', 'learning_rate': '1.208e-05', 'epoch': '0.9128'}
54
+ {'loss': '0.7091', 'grad_norm': '5.085', 'learning_rate': '1.201e-05', 'epoch': '0.9193'}
55
+ {'loss': '0.7019', 'grad_norm': '5.379', 'learning_rate': '1.194e-05', 'epoch': '0.9258'}
56
+ {'loss': '0.7081', 'grad_norm': '5.63', 'learning_rate': '1.186e-05', 'epoch': '0.9323'}
57
+ {'loss': '0.6833', 'grad_norm': '2.541', 'learning_rate': '1.179e-05', 'epoch': '0.9387'}
58
+ {'loss': '0.6982', 'grad_norm': '5.714', 'learning_rate': '1.172e-05', 'epoch': '0.9452'}
59
+ {'loss': '0.7249', 'grad_norm': '5.051', 'learning_rate': '1.165e-05', 'epoch': '0.9517'}
60
+ {'loss': '0.7282', 'grad_norm': '6.322', 'learning_rate': '1.158e-05', 'epoch': '0.9582'}
61
+ {'loss': '0.7147', 'grad_norm': '4.961', 'learning_rate': '1.15e-05', 'epoch': '0.9646'}
62
+ {'loss': '0.6742', 'grad_norm': '4.871', 'learning_rate': '1.143e-05', 'epoch': '0.9711'}
63
+ {'eval_train_loss': '0.364', 'eval_dev-768_cosine_accuracy': '0.9758', 'eval_dev-512_cosine_accuracy': '0.9759', 'eval_dev-256_cosine_accuracy': '0.9761', 'eval_dev-128_cosine_accuracy': '0.9757', 'eval_dev-64_cosine_accuracy': '0.9742', 'eval_sequential_score': '0.9758', 'eval_train_runtime': '9649', 'eval_train_samples_per_second': '117.1', 'eval_train_steps_per_second': '14.64', 'epoch': '0.9711'}
64
+ {'loss': '0.6901', 'grad_norm': '3.348', 'learning_rate': '1.136e-05', 'epoch': '0.9776'}
65
+ {'loss': '0.7067', 'grad_norm': '3.76', 'learning_rate': '1.129e-05', 'epoch': '0.9841'}
66
+ {'loss': '0.7166', 'grad_norm': '4.729', 'learning_rate': '1.122e-05', 'epoch': '0.9905'}
67
+ {'loss': '0.68', 'grad_norm': '4.648', 'learning_rate': '1.114e-05', 'epoch': '0.997'}
68
+ {'loss': '0.6846', 'grad_norm': '4.427', 'learning_rate': '1.107e-05', 'epoch': '1.003'}
69
+ {'loss': '0.6723', 'grad_norm': '4.459', 'learning_rate': '1.1e-05', 'epoch': '1.01'}
70
+ {'loss': '0.6573', 'grad_norm': '6.387', 'learning_rate': '1.093e-05', 'epoch': '1.016'}
71
+ {'loss': '0.6895', 'grad_norm': '4.1', 'learning_rate': '1.086e-05', 'epoch': '1.023'}
72
+ {'loss': '0.6588', 'grad_norm': '5.927', 'learning_rate': '1.079e-05', 'epoch': '1.029'}
73
+ {'loss': '0.6517', 'grad_norm': '5.9', 'learning_rate': '1.071e-05', 'epoch': '1.036'}
74
+ {'loss': '0.6498', 'grad_norm': '4.736', 'learning_rate': '1.064e-05', 'epoch': '1.042'}
75
+ {'loss': '0.6836', 'grad_norm': '5.029', 'learning_rate': '1.057e-05', 'epoch': '1.049'}
76
+ {'loss': '0.6819', 'grad_norm': '2.595', 'learning_rate': '1.05e-05', 'epoch': '1.055'}
77
+ {'loss': '0.6463', 'grad_norm': '4.963', 'learning_rate': '1.043e-05', 'epoch': '1.062'}
78
+ {'loss': '0.6645', 'grad_norm': '5.046', 'learning_rate': '1.035e-05', 'epoch': '1.068'}
79
+ {'loss': '0.6518', 'grad_norm': '3.307', 'learning_rate': '1.028e-05', 'epoch': '1.075'}
80
+ {'loss': '0.6235', 'grad_norm': '3.848', 'learning_rate': '1.021e-05', 'epoch': '1.081'}
81
+ {'loss': '0.6302', 'grad_norm': '4.664', 'learning_rate': '1.014e-05', 'epoch': '1.088'}
82
+ {'loss': '0.6452', 'grad_norm': '5.47', 'learning_rate': '1.007e-05', 'epoch': '1.094'}
83
+ {'loss': '0.6477', 'grad_norm': '5.26', 'learning_rate': '9.994e-06', 'epoch': '1.101'}
84
+ {'loss': '0.6084', 'grad_norm': '4.313', 'learning_rate': '9.922e-06', 'epoch': '1.107'}
85
+ {'loss': '0.6259', 'grad_norm': '6.499', 'learning_rate': '9.85e-06', 'epoch': '1.114'}
86
+ {'loss': '0.607', 'grad_norm': '3.922', 'learning_rate': '9.778e-06', 'epoch': '1.12'}
87
+ {'loss': '0.5977', 'grad_norm': '5.37', 'learning_rate': '9.706e-06', 'epoch': '1.126'}
88
+ {'loss': '0.6044', 'grad_norm': '5.068', 'learning_rate': '9.634e-06', 'epoch': '1.133'}
89
+ {'loss': '0.6007', 'grad_norm': '4.109', 'learning_rate': '9.562e-06', 'epoch': '1.139'}
90
+ {'loss': '0.5628', 'grad_norm': '4.954', 'learning_rate': '9.491e-06', 'epoch': '1.146'}
91
+ {'loss': '0.5732', 'grad_norm': '4.068', 'learning_rate': '9.419e-06', 'epoch': '1.152'}
92
+ {'loss': '0.5773', 'grad_norm': '4.939', 'learning_rate': '9.347e-06', 'epoch': '1.159'}
93
+ {'loss': '0.5719', 'grad_norm': '4.418', 'learning_rate': '9.275e-06', 'epoch': '1.165'}
94
+ {'eval_train_loss': '0.3356', 'eval_dev-768_cosine_accuracy': '0.9775', 'eval_dev-512_cosine_accuracy': '0.9777', 'eval_dev-256_cosine_accuracy': '0.9777', 'eval_dev-128_cosine_accuracy': '0.9774', 'eval_dev-64_cosine_accuracy': '0.976', 'eval_sequential_score': '0.9775', 'eval_train_runtime': '1.01e+04', 'eval_train_samples_per_second': '111.8', 'eval_train_steps_per_second': '13.98', 'epoch': '1.165'}
95
+ {'loss': '0.5471', 'grad_norm': '3.58', 'learning_rate': '9.203e-06', 'epoch': '1.172'}
96
+ {'loss': '0.5635', 'grad_norm': '5.198', 'learning_rate': '9.131e-06', 'epoch': '1.178'}
97
+ {'loss': '0.539', 'grad_norm': '4.468', 'learning_rate': '9.059e-06', 'epoch': '1.185'}
98
+ {'loss': '0.5428', 'grad_norm': '4.349', 'learning_rate': '8.987e-06', 'epoch': '1.191'}
99
+ {'loss': '0.5205', 'grad_norm': '2.936', 'learning_rate': '8.915e-06', 'epoch': '1.198'}
100
+ {'loss': '0.5362', 'grad_norm': '3.337', 'learning_rate': '8.843e-06', 'epoch': '1.204'}
101
+ {'loss': '0.5386', 'grad_norm': '5.76', 'learning_rate': '8.771e-06', 'epoch': '1.211'}
102
+ {'loss': '0.5203', 'grad_norm': '3.261', 'learning_rate': '8.699e-06', 'epoch': '1.217'}
103
+ {'loss': '0.5301', 'grad_norm': '3.732', 'learning_rate': '8.627e-06', 'epoch': '1.224'}
104
+ {'loss': '0.5232', 'grad_norm': '4.54', 'learning_rate': '8.555e-06', 'epoch': '1.23'}
105
+ {'loss': '0.4922', 'grad_norm': '4.291', 'learning_rate': '8.483e-06', 'epoch': '1.237'}
106
+ {'loss': '0.5029', 'grad_norm': '3.979', 'learning_rate': '8.412e-06', 'epoch': '1.243'}
107
+ {'loss': '0.4989', 'grad_norm': '7.829', 'learning_rate': '8.34e-06', 'epoch': '1.249'}
108
+ {'loss': '0.5053', 'grad_norm': '2.903', 'learning_rate': '8.268e-06', 'epoch': '1.256'}
109
+ {'loss': '0.5081', 'grad_norm': '5.471', 'learning_rate': '8.196e-06', 'epoch': '1.262'}
110
+ {'loss': '0.496', 'grad_norm': '5.204', 'learning_rate': '8.124e-06', 'epoch': '1.269'}
111
+ {'loss': '0.5052', 'grad_norm': '4.377', 'learning_rate': '8.052e-06', 'epoch': '1.275'}
112
+ {'loss': '0.4984', 'grad_norm': '4.184', 'learning_rate': '7.98e-06', 'epoch': '1.282'}
113
+ {'loss': '0.4909', 'grad_norm': '4.991', 'learning_rate': '7.908e-06', 'epoch': '1.288'}
114
+ {'loss': '0.512', 'grad_norm': '3.76', 'learning_rate': '7.836e-06', 'epoch': '1.295'}
115
+ {'loss': '0.4873', 'grad_norm': '3.844', 'learning_rate': '7.764e-06', 'epoch': '1.301'}
116
+ {'loss': '0.4896', 'grad_norm': '6.987', 'learning_rate': '7.692e-06', 'epoch': '1.308'}
117
+ {'loss': '0.49', 'grad_norm': '6.267', 'learning_rate': '7.62e-06', 'epoch': '1.314'}
118
+ {'loss': '0.5036', 'grad_norm': '3.776', 'learning_rate': '7.548e-06', 'epoch': '1.321'}
119
+ {'loss': '0.4876', 'grad_norm': '3.42', 'learning_rate': '7.476e-06', 'epoch': '1.327'}
120
+ {'loss': '0.4705', 'grad_norm': '5.478', 'learning_rate': '7.404e-06', 'epoch': '1.334'}
121
+ {'loss': '0.4786', 'grad_norm': '3.313', 'learning_rate': '7.333e-06', 'epoch': '1.34'}
122
+ {'loss': '0.4998', 'grad_norm': '3.13', 'learning_rate': '7.261e-06', 'epoch': '1.347'}
123
+ {'loss': '0.4692', 'grad_norm': '3.971', 'learning_rate': '7.189e-06', 'epoch': '1.353'}
124
+ {'loss': '0.5064', 'grad_norm': '6.238', 'learning_rate': '7.117e-06', 'epoch': '1.36'}
125
+ {'eval_train_loss': '0.316', 'eval_dev-768_cosine_accuracy': '0.9788', 'eval_dev-512_cosine_accuracy': '0.979', 'eval_dev-256_cosine_accuracy': '0.979', 'eval_dev-128_cosine_accuracy': '0.9785', 'eval_dev-64_cosine_accuracy': '0.9774', 'eval_sequential_score': '0.9788', 'eval_train_runtime': '1.014e+04', 'eval_train_samples_per_second': '111.5', 'eval_train_steps_per_second': '13.93', 'epoch': '1.36'}
126
+ {'loss': '0.4925', 'grad_norm': '5.158', 'learning_rate': '7.045e-06', 'epoch': '1.366'}
127
+ {'loss': '0.4601', 'grad_norm': '4.139', 'learning_rate': '6.973e-06', 'epoch': '1.372'}
128
+ {'loss': '0.4762', 'grad_norm': '3.411', 'learning_rate': '6.901e-06', 'epoch': '1.379'}
129
+ {'loss': '0.4986', 'grad_norm': '4.23', 'learning_rate': '6.829e-06', 'epoch': '1.385'}
130
+ {'loss': '0.4656', 'grad_norm': '5.326', 'learning_rate': '6.757e-06', 'epoch': '1.392'}
131
+ {'loss': '0.4507', 'grad_norm': '3.826', 'learning_rate': '6.685e-06', 'epoch': '1.398'}
132
+ {'loss': '0.4862', 'grad_norm': '3.509', 'learning_rate': '6.613e-06', 'epoch': '1.405'}
133
+ {'loss': '0.4596', 'grad_norm': '4.734', 'learning_rate': '6.541e-06', 'epoch': '1.411'}
134
+ {'loss': '0.4696', 'grad_norm': '4.799', 'learning_rate': '6.469e-06', 'epoch': '1.418'}
135
+ {'loss': '0.4925', 'grad_norm': '4.942', 'learning_rate': '6.397e-06', 'epoch': '1.424'}
136
+ {'loss': '0.4796', 'grad_norm': '4.147', 'learning_rate': '6.325e-06', 'epoch': '1.431'}
137
+ {'loss': '0.4525', 'grad_norm': '5.146', 'learning_rate': '6.254e-06', 'epoch': '1.437'}
138
+ {'loss': '0.4717', 'grad_norm': '3.52', 'learning_rate': '6.182e-06', 'epoch': '1.444'}
139
+ {'loss': '0.4803', 'grad_norm': '3.25', 'learning_rate': '6.11e-06', 'epoch': '1.45'}
140
+ {'loss': '0.4675', 'grad_norm': '7.35', 'learning_rate': '6.038e-06', 'epoch': '1.457'}
141
+ {'loss': '0.4631', 'grad_norm': '3.847', 'learning_rate': '5.966e-06', 'epoch': '1.463'}
142
+ {'loss': '0.4622', 'grad_norm': '4.57', 'learning_rate': '5.894e-06', 'epoch': '1.47'}
143
+ {'loss': '0.4496', 'grad_norm': '1.997', 'learning_rate': '5.822e-06', 'epoch': '1.476'}
144
+ {'loss': '0.4678', 'grad_norm': '4.266', 'learning_rate': '5.75e-06', 'epoch': '1.483'}
145
+ {'loss': '0.4495', 'grad_norm': '5.948', 'learning_rate': '5.678e-06', 'epoch': '1.489'}
146
+ {'loss': '0.4474', 'grad_norm': '3.7', 'learning_rate': '5.606e-06', 'epoch': '1.495'}
147
+ {'loss': '0.4587', 'grad_norm': '2.877', 'learning_rate': '5.534e-06', 'epoch': '1.502'}
148
+ {'loss': '0.4591', 'grad_norm': '4.245', 'learning_rate': '5.462e-06', 'epoch': '1.508'}
149
+ {'loss': '0.4573', 'grad_norm': '5.431', 'learning_rate': '5.39e-06', 'epoch': '1.515'}
150
+ {'loss': '0.4442', 'grad_norm': '3.338', 'learning_rate': '5.318e-06', 'epoch': '1.521'}
151
+ {'loss': '0.455', 'grad_norm': '4.723', 'learning_rate': '5.246e-06', 'epoch': '1.528'}
152
+ {'loss': '0.4493', 'grad_norm': '4.226', 'learning_rate': '5.175e-06', 'epoch': '1.534'}
153
+ {'loss': '0.4485', 'grad_norm': '4.451', 'learning_rate': '5.103e-06', 'epoch': '1.541'}
154
+ {'loss': '0.4569', 'grad_norm': '4.297', 'learning_rate': '5.031e-06', 'epoch': '1.547'}
155
+ {'loss': '0.4346', 'grad_norm': '4.199', 'learning_rate': '4.959e-06', 'epoch': '1.554'}
156
+ {'eval_train_loss': '0.3001', 'eval_dev-768_cosine_accuracy': '0.9799', 'eval_dev-512_cosine_accuracy': '0.9802', 'eval_dev-256_cosine_accuracy': '0.9802', 'eval_dev-128_cosine_accuracy': '0.9798', 'eval_dev-64_cosine_accuracy': '0.9788', 'eval_sequential_score': '0.9799', 'eval_train_runtime': '1.008e+04', 'eval_train_samples_per_second': '112.1', 'eval_train_steps_per_second': '14.02', 'epoch': '1.554'}
157
+ {'loss': '0.4469', 'grad_norm': '3.364', 'learning_rate': '4.887e-06', 'epoch': '1.56'}
158
+ {'loss': '0.4602', 'grad_norm': '5.309', 'learning_rate': '4.815e-06', 'epoch': '1.567'}
159
+ {'loss': '0.443', 'grad_norm': '3.875', 'learning_rate': '4.743e-06', 'epoch': '1.573'}
160
+ {'loss': '0.4524', 'grad_norm': '4.824', 'learning_rate': '4.671e-06', 'epoch': '1.58'}
161
+ {'loss': '0.4528', 'grad_norm': '4.996', 'learning_rate': '4.599e-06', 'epoch': '1.586'}
162
+ {'loss': '0.4348', 'grad_norm': '4.96', 'learning_rate': '4.527e-06', 'epoch': '1.593'}
163
+ {'loss': '0.4533', 'grad_norm': '5.219', 'learning_rate': '4.455e-06', 'epoch': '1.599'}
164
+ {'loss': '0.4523', 'grad_norm': '3.444', 'learning_rate': '4.383e-06', 'epoch': '1.606'}
165
+ {'loss': '0.4509', 'grad_norm': '5.647', 'learning_rate': '4.311e-06', 'epoch': '1.612'}
166
+ {'loss': '0.4365', 'grad_norm': '5.052', 'learning_rate': '4.239e-06', 'epoch': '1.618'}
167
+ {'loss': '0.4504', 'grad_norm': '5.786', 'learning_rate': '4.167e-06', 'epoch': '1.625'}
168
+ {'loss': '0.4292', 'grad_norm': '4.353', 'learning_rate': '4.096e-06', 'epoch': '1.631'}
169
+ {'loss': '0.4406', 'grad_norm': '2.976', 'learning_rate': '4.024e-06', 'epoch': '1.638'}
170
+ {'loss': '0.4333', 'grad_norm': '3.685', 'learning_rate': '3.952e-06', 'epoch': '1.644'}
171
+ {'loss': '0.4361', 'grad_norm': '4.107', 'learning_rate': '3.88e-06', 'epoch': '1.651'}
172
+ {'loss': '0.4065', 'grad_norm': '3.636', 'learning_rate': '3.808e-06', 'epoch': '1.657'}
173
+ {'loss': '0.4671', 'grad_norm': '3.464', 'learning_rate': '3.736e-06', 'epoch': '1.664'}
174
+ {'loss': '0.4328', 'grad_norm': '3.129', 'learning_rate': '3.664e-06', 'epoch': '1.67'}
175
+ {'loss': '0.431', 'grad_norm': '2.453', 'learning_rate': '3.592e-06', 'epoch': '1.677'}
176
+ {'loss': '0.4523', 'grad_norm': '3.727', 'learning_rate': '3.52e-06', 'epoch': '1.683'}
177
+ {'loss': '0.4232', 'grad_norm': '4.398', 'learning_rate': '3.448e-06', 'epoch': '1.69'}
178
+ {'loss': '0.4257', 'grad_norm': '2.861', 'learning_rate': '3.376e-06', 'epoch': '1.696'}
179
+ {'loss': '0.4448', 'grad_norm': '3.523', 'learning_rate': '3.304e-06', 'epoch': '1.703'}
180
+ {'loss': '0.4491', 'grad_norm': '3.893', 'learning_rate': '3.232e-06', 'epoch': '1.709'}
181
+ {'loss': '0.4224', 'grad_norm': '3.399', 'learning_rate': '3.16e-06', 'epoch': '1.716'}
182
+ {'loss': '0.4297', 'grad_norm': '4.703', 'learning_rate': '3.088e-06', 'epoch': '1.722'}
183
+ {'loss': '0.4522', 'grad_norm': '4.29', 'learning_rate': '3.017e-06', 'epoch': '1.729'}
184
+ {'loss': '0.4195', 'grad_norm': '4.29', 'learning_rate': '2.945e-06', 'epoch': '1.735'}
185
+ {'loss': '0.4227', 'grad_norm': '3.841', 'learning_rate': '2.873e-06', 'epoch': '1.742'}
186
+ {'loss': '0.4381', 'grad_norm': '4.086', 'learning_rate': '2.801e-06', 'epoch': '1.748'}
187
+ {'eval_train_loss': '0.2875', 'eval_dev-768_cosine_accuracy': '0.9807', 'eval_dev-512_cosine_accuracy': '0.9808', 'eval_dev-256_cosine_accuracy': '0.9808', 'eval_dev-128_cosine_accuracy': '0.9805', 'eval_dev-64_cosine_accuracy': '0.9794', 'eval_sequential_score': '0.9807', 'eval_train_runtime': '1.012e+04', 'eval_train_samples_per_second': '111.6', 'eval_train_steps_per_second': '13.95', 'epoch': '1.748'}
188
+ {'loss': '0.446', 'grad_norm': '4.176', 'learning_rate': '2.729e-06', 'epoch': '1.754'}
189
+ {'loss': '0.426', 'grad_norm': '4.261', 'learning_rate': '2.657e-06', 'epoch': '1.761'}
190
+ {'loss': '0.4299', 'grad_norm': '4.676', 'learning_rate': '2.585e-06', 'epoch': '1.767'}
191
+ {'loss': '0.4247', 'grad_norm': '3.933', 'learning_rate': '2.513e-06', 'epoch': '1.774'}
192
+ {'loss': '0.4244', 'grad_norm': '4.853', 'learning_rate': '2.441e-06', 'epoch': '1.78'}
193
+ {'loss': '0.4185', 'grad_norm': '2.985', 'learning_rate': '2.369e-06', 'epoch': '1.787'}
194
+ {'loss': '0.4292', 'grad_norm': '3.804', 'learning_rate': '2.297e-06', 'epoch': '1.793'}
195
+ {'loss': '0.4468', 'grad_norm': '3.187', 'learning_rate': '2.225e-06', 'epoch': '1.8'}
196
+ {'loss': '0.4118', 'grad_norm': '4.004', 'learning_rate': '2.153e-06', 'epoch': '1.806'}
197
+ {'loss': '0.4306', 'grad_norm': '4.007', 'learning_rate': '2.081e-06', 'epoch': '1.813'}
198
+ {'loss': '0.4447', 'grad_norm': '4.323', 'learning_rate': '2.009e-06', 'epoch': '1.819'}
199
+ {'loss': '0.4147', 'grad_norm': '3.863', 'learning_rate': '1.938e-06', 'epoch': '1.826'}
200
+ {'loss': '0.4189', 'grad_norm': '4.788', 'learning_rate': '1.866e-06', 'epoch': '1.832'}
201
+ {'loss': '0.4167', 'grad_norm': '4.276', 'learning_rate': '1.794e-06', 'epoch': '1.839'}
202
+ {'loss': '0.4022', 'grad_norm': '3.887', 'learning_rate': '1.722e-06', 'epoch': '1.845'}
203
+ {'loss': '0.4158', 'grad_norm': '3.075', 'learning_rate': '1.65e-06', 'epoch': '1.852'}
204
+ {'loss': '0.4228', 'grad_norm': '3.993', 'learning_rate': '1.578e-06', 'epoch': '1.858'}
205
+ {'loss': '0.4256', 'grad_norm': '4.497', 'learning_rate': '1.506e-06', 'epoch': '1.865'}
206
+ {'loss': '0.4251', 'grad_norm': '4.539', 'learning_rate': '1.434e-06', 'epoch': '1.871'}
207
+ {'loss': '0.4232', 'grad_norm': '2.337', 'learning_rate': '1.362e-06', 'epoch': '1.877'}
208
+ {'loss': '0.4143', 'grad_norm': '3.389', 'learning_rate': '1.29e-06', 'epoch': '1.884'}
209
+ {'loss': '0.4331', 'grad_norm': '3.545', 'learning_rate': '1.218e-06', 'epoch': '1.89'}
210
+ {'loss': '0.4253', 'grad_norm': '5.606', 'learning_rate': '1.146e-06', 'epoch': '1.897'}
211
+ {'loss': '0.441', 'grad_norm': '4.453', 'learning_rate': '1.074e-06', 'epoch': '1.903'}
212
+ {'loss': '0.4337', 'grad_norm': '5.374', 'learning_rate': '1.002e-06', 'epoch': '1.91'}
213
+ {'loss': '0.4016', 'grad_norm': '2.246', 'learning_rate': '9.305e-07', 'epoch': '1.916'}
214
+ {'loss': '0.4249', 'grad_norm': '5.255', 'learning_rate': '8.585e-07', 'epoch': '1.923'}
215
+ {'loss': '0.4108', 'grad_norm': '3.59', 'learning_rate': '7.866e-07', 'epoch': '1.929'}
216
+ {'loss': '0.4272', 'grad_norm': '4.258', 'learning_rate': '7.147e-07', 'epoch': '1.936'}
217
+ {'loss': '0.3916', 'grad_norm': '3.476', 'learning_rate': '6.427e-07', 'epoch': '1.942'}
218
+ {'eval_train_loss': '0.2812', 'eval_dev-768_cosine_accuracy': '0.981', 'eval_dev-512_cosine_accuracy': '0.9811', 'eval_dev-256_cosine_accuracy': '0.9813', 'eval_dev-128_cosine_accuracy': '0.9811', 'eval_dev-64_cosine_accuracy': '0.9797', 'eval_sequential_score': '0.981', 'eval_train_runtime': '1.03e+04', 'eval_train_samples_per_second': '109.7', 'eval_train_steps_per_second': '13.71', 'epoch': '1.942'}
219
+ {'loss': '0.4334', 'grad_norm': '4.623', 'learning_rate': '5.708e-07', 'epoch': '1.949'}
220
+ {'loss': '0.4462', 'grad_norm': '5.31', 'learning_rate': '4.989e-07', 'epoch': '1.955'}
221
+ {'loss': '0.4436', 'grad_norm': '3.379', 'learning_rate': '4.269e-07', 'epoch': '1.962'}
222
+ {'loss': '0.4278', 'grad_norm': '5.471', 'learning_rate': '3.55e-07', 'epoch': '1.968'}
223
+ {'loss': '0.417', 'grad_norm': '3.435', 'learning_rate': '2.831e-07', 'epoch': '1.975'}
224
+ {'loss': '0.4376', 'grad_norm': '2.617', 'learning_rate': '2.111e-07', 'epoch': '1.981'}
225
+ {'loss': '0.4433', 'grad_norm': '3.465', 'learning_rate': '1.392e-07', 'epoch': '1.988'}
226
+ {'loss': '0.4292', 'grad_norm': '2.354', 'learning_rate': '6.726e-08', 'epoch': '1.994'}
227
+ {'train_runtime': '9.588e+04', 'train_samples_per_second': '82.48', 'train_steps_per_second': '0.644', 'train_loss': '0.403', 'epoch': '2'}
228
+ model saved successfully