bobox commited on
Commit
cc70ad0
1 Parent(s): 1c8da80

Training in progress, step 107, checkpoint

Browse files
checkpoint-107/1_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
+ }
checkpoint-107/README.md ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-107/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
checkpoint-107/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-v3-small",
3
+ "architectures": [
4
+ "DebertaV2Model"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 768,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 3072,
12
+ "layer_norm_eps": 1e-07,
13
+ "max_position_embeddings": 512,
14
+ "max_relative_positions": -1,
15
+ "model_type": "deberta-v2",
16
+ "norm_rel_ebd": "layer_norm",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 6,
19
+ "pad_token_id": 0,
20
+ "pooler_dropout": 0,
21
+ "pooler_hidden_act": "gelu",
22
+ "pooler_hidden_size": 768,
23
+ "pos_att_type": [
24
+ "p2c",
25
+ "c2p"
26
+ ],
27
+ "position_biased_input": false,
28
+ "position_buckets": 256,
29
+ "relative_attention": true,
30
+ "share_att_key": true,
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.42.3",
33
+ "type_vocab_size": 0,
34
+ "vocab_size": 128100
35
+ }
checkpoint-107/config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.42.3",
5
+ "pytorch": "2.1.2"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
checkpoint-107/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
+ ]
checkpoint-107/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d68669b06d21146c42d3ab8ac5fb5ecd13179c0bf7760c5eae496de3c2bedb9
3
+ size 1130520122
checkpoint-107/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:509380132a556aa5c5abd5ae7e2966c9aeeff6d3c17f413118c7e375b07b4a0c
3
+ size 565251810
checkpoint-107/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dba89f3e04d18dac9ee9bb8984f08a24cbbbd385d8c90bc6b32d99e0123f4094
3
+ size 14244
checkpoint-107/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb571e03d7f101ced903b4b06060b388f532cdee9c838a84c6bbeeb165f467db
3
+ size 1064
checkpoint-107/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
checkpoint-107/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": {
9
+ "content": "[UNK]",
10
+ "lstrip": false,
11
+ "normalized": true,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ }
15
+ }
checkpoint-107/spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
checkpoint-107/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-107/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
checkpoint-107/trainer_state.json ADDED
@@ -0,0 +1,719 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.6040931545518702,
5
+ "eval_steps": 27,
6
+ "global_step": 107,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016937191249117856,
13
+ "grad_norm": 34.22002029418945,
14
+ "learning_rate": 6.818181818181818e-07,
15
+ "loss": 7.2372,
16
+ "step": 3
17
+ },
18
+ {
19
+ "epoch": 0.03387438249823571,
20
+ "grad_norm": 21.76839828491211,
21
+ "learning_rate": 1.3636363636363636e-06,
22
+ "loss": 6.855,
23
+ "step": 6
24
+ },
25
+ {
26
+ "epoch": 0.05081157374735357,
27
+ "grad_norm": 21.260774612426758,
28
+ "learning_rate": 2.0454545454545457e-06,
29
+ "loss": 7.4707,
30
+ "step": 9
31
+ },
32
+ {
33
+ "epoch": 0.06774876499647142,
34
+ "grad_norm": 16.885921478271484,
35
+ "learning_rate": 2.7272727272727272e-06,
36
+ "loss": 7.0187,
37
+ "step": 12
38
+ },
39
+ {
40
+ "epoch": 0.08468595624558928,
41
+ "grad_norm": 19.509899139404297,
42
+ "learning_rate": 3.409090909090909e-06,
43
+ "loss": 6.6756,
44
+ "step": 15
45
+ },
46
+ {
47
+ "epoch": 0.10162314749470713,
48
+ "grad_norm": 7.9427289962768555,
49
+ "learning_rate": 4.0909090909090915e-06,
50
+ "loss": 6.0155,
51
+ "step": 18
52
+ },
53
+ {
54
+ "epoch": 0.11856033874382499,
55
+ "grad_norm": 7.325345039367676,
56
+ "learning_rate": 4.772727272727273e-06,
57
+ "loss": 6.1644,
58
+ "step": 21
59
+ },
60
+ {
61
+ "epoch": 0.13549752999294284,
62
+ "grad_norm": 7.544689655303955,
63
+ "learning_rate": 5.4545454545454545e-06,
64
+ "loss": 6.2158,
65
+ "step": 24
66
+ },
67
+ {
68
+ "epoch": 0.1524347212420607,
69
+ "grad_norm": 5.141758918762207,
70
+ "learning_rate": 6.136363636363637e-06,
71
+ "loss": 6.1369,
72
+ "step": 27
73
+ },
74
+ {
75
+ "epoch": 0.1524347212420607,
76
+ "eval_NLI-v2_cosine_accuracy": 1.0,
77
+ "eval_NLI-v2_dot_accuracy": 0.109375,
78
+ "eval_NLI-v2_euclidean_accuracy": 1.0,
79
+ "eval_NLI-v2_manhattan_accuracy": 1.0,
80
+ "eval_NLI-v2_max_accuracy": 1.0,
81
+ "eval_VitaminC_cosine_accuracy": 0.5546875,
82
+ "eval_VitaminC_cosine_accuracy_threshold": 0.9544724822044373,
83
+ "eval_VitaminC_cosine_ap": 0.5356492030729136,
84
+ "eval_VitaminC_cosine_f1": 0.6542553191489362,
85
+ "eval_VitaminC_cosine_f1_threshold": 0.7148199081420898,
86
+ "eval_VitaminC_cosine_precision": 0.48616600790513836,
87
+ "eval_VitaminC_cosine_recall": 1.0,
88
+ "eval_VitaminC_dot_accuracy": 0.55078125,
89
+ "eval_VitaminC_dot_accuracy_threshold": 414.4264831542969,
90
+ "eval_VitaminC_dot_ap": 0.5108219546857565,
91
+ "eval_VitaminC_dot_f1": 0.6507936507936508,
92
+ "eval_VitaminC_dot_f1_threshold": 271.6522521972656,
93
+ "eval_VitaminC_dot_precision": 0.4823529411764706,
94
+ "eval_VitaminC_dot_recall": 1.0,
95
+ "eval_VitaminC_euclidean_accuracy": 0.55078125,
96
+ "eval_VitaminC_euclidean_accuracy_threshold": 6.519885063171387,
97
+ "eval_VitaminC_euclidean_ap": 0.5226419655984281,
98
+ "eval_VitaminC_euclidean_f1": 0.6505376344086021,
99
+ "eval_VitaminC_euclidean_f1_threshold": 15.194067001342773,
100
+ "eval_VitaminC_euclidean_precision": 0.4859437751004016,
101
+ "eval_VitaminC_euclidean_recall": 0.983739837398374,
102
+ "eval_VitaminC_manhattan_accuracy": 0.546875,
103
+ "eval_VitaminC_manhattan_accuracy_threshold": 149.20114135742188,
104
+ "eval_VitaminC_manhattan_ap": 0.5237451656134715,
105
+ "eval_VitaminC_manhattan_f1": 0.6542553191489362,
106
+ "eval_VitaminC_manhattan_f1_threshold": 259.007080078125,
107
+ "eval_VitaminC_manhattan_precision": 0.48616600790513836,
108
+ "eval_VitaminC_manhattan_recall": 1.0,
109
+ "eval_VitaminC_max_accuracy": 0.5546875,
110
+ "eval_VitaminC_max_accuracy_threshold": 414.4264831542969,
111
+ "eval_VitaminC_max_ap": 0.5356492030729136,
112
+ "eval_VitaminC_max_f1": 0.6542553191489362,
113
+ "eval_VitaminC_max_f1_threshold": 271.6522521972656,
114
+ "eval_VitaminC_max_precision": 0.48616600790513836,
115
+ "eval_VitaminC_max_recall": 1.0,
116
+ "eval_sequential_score": 0.5356492030729136,
117
+ "eval_sts-test_pearson_cosine": 0.056062031998983373,
118
+ "eval_sts-test_pearson_dot": 0.2979259445723872,
119
+ "eval_sts-test_pearson_euclidean": 0.0498319208592713,
120
+ "eval_sts-test_pearson_manhattan": 0.07381429239121526,
121
+ "eval_sts-test_pearson_max": 0.2979259445723872,
122
+ "eval_sts-test_spearman_cosine": 0.1066788491614481,
123
+ "eval_sts-test_spearman_dot": 0.315952670306405,
124
+ "eval_sts-test_spearman_euclidean": 0.07303394554435191,
125
+ "eval_sts-test_spearman_manhattan": 0.09039525717692232,
126
+ "eval_sts-test_spearman_max": 0.315952670306405,
127
+ "eval_vitaminc-pairs_loss": 2.698580741882324,
128
+ "eval_vitaminc-pairs_runtime": 1.4747,
129
+ "eval_vitaminc-pairs_samples_per_second": 73.236,
130
+ "eval_vitaminc-pairs_steps_per_second": 1.356,
131
+ "step": 27
132
+ },
133
+ {
134
+ "epoch": 0.1524347212420607,
135
+ "eval_negation-triplets_loss": 5.142906665802002,
136
+ "eval_negation-triplets_runtime": 0.2993,
137
+ "eval_negation-triplets_samples_per_second": 213.865,
138
+ "eval_negation-triplets_steps_per_second": 3.342,
139
+ "step": 27
140
+ },
141
+ {
142
+ "epoch": 0.1524347212420607,
143
+ "eval_scitail-pairs-pos_loss": 1.9216958284378052,
144
+ "eval_scitail-pairs-pos_runtime": 0.3834,
145
+ "eval_scitail-pairs-pos_samples_per_second": 140.842,
146
+ "eval_scitail-pairs-pos_steps_per_second": 2.608,
147
+ "step": 27
148
+ },
149
+ {
150
+ "epoch": 0.1524347212420607,
151
+ "eval_xsum-pairs_loss": 6.073049545288086,
152
+ "eval_xsum-pairs_runtime": 3.1587,
153
+ "eval_xsum-pairs_samples_per_second": 40.523,
154
+ "eval_xsum-pairs_steps_per_second": 0.633,
155
+ "step": 27
156
+ },
157
+ {
158
+ "epoch": 0.1524347212420607,
159
+ "eval_sciq_pairs_loss": 0.3449864387512207,
160
+ "eval_sciq_pairs_runtime": 3.3747,
161
+ "eval_sciq_pairs_samples_per_second": 37.93,
162
+ "eval_sciq_pairs_steps_per_second": 0.593,
163
+ "step": 27
164
+ },
165
+ {
166
+ "epoch": 0.1524347212420607,
167
+ "eval_qasc_pairs_loss": 3.2267842292785645,
168
+ "eval_qasc_pairs_runtime": 0.6576,
169
+ "eval_qasc_pairs_samples_per_second": 194.646,
170
+ "eval_qasc_pairs_steps_per_second": 3.041,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.1524347212420607,
175
+ "eval_openbookqa_pairs_loss": 4.405983924865723,
176
+ "eval_openbookqa_pairs_runtime": 0.6107,
177
+ "eval_openbookqa_pairs_samples_per_second": 209.594,
178
+ "eval_openbookqa_pairs_steps_per_second": 3.275,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 0.1524347212420607,
183
+ "eval_msmarco_pairs_loss": 6.937691688537598,
184
+ "eval_msmarco_pairs_runtime": 1.3091,
185
+ "eval_msmarco_pairs_samples_per_second": 97.779,
186
+ "eval_msmarco_pairs_steps_per_second": 1.528,
187
+ "step": 27
188
+ },
189
+ {
190
+ "epoch": 0.1524347212420607,
191
+ "eval_nq_pairs_loss": 6.794108867645264,
192
+ "eval_nq_pairs_runtime": 2.3968,
193
+ "eval_nq_pairs_samples_per_second": 53.404,
194
+ "eval_nq_pairs_steps_per_second": 0.834,
195
+ "step": 27
196
+ },
197
+ {
198
+ "epoch": 0.1524347212420607,
199
+ "eval_trivia_pairs_loss": 6.3355631828308105,
200
+ "eval_trivia_pairs_runtime": 4.4974,
201
+ "eval_trivia_pairs_samples_per_second": 28.461,
202
+ "eval_trivia_pairs_steps_per_second": 0.445,
203
+ "step": 27
204
+ },
205
+ {
206
+ "epoch": 0.1524347212420607,
207
+ "eval_gooaq_pairs_loss": 6.405998706817627,
208
+ "eval_gooaq_pairs_runtime": 0.8745,
209
+ "eval_gooaq_pairs_samples_per_second": 146.37,
210
+ "eval_gooaq_pairs_steps_per_second": 2.287,
211
+ "step": 27
212
+ },
213
+ {
214
+ "epoch": 0.1524347212420607,
215
+ "eval_paws-pos_loss": 2.2308223247528076,
216
+ "eval_paws-pos_runtime": 0.6998,
217
+ "eval_paws-pos_samples_per_second": 182.908,
218
+ "eval_paws-pos_steps_per_second": 2.858,
219
+ "step": 27
220
+ },
221
+ {
222
+ "epoch": 0.16937191249117856,
223
+ "grad_norm": 5.885251522064209,
224
+ "learning_rate": 6.818181818181818e-06,
225
+ "loss": 5.7653,
226
+ "step": 30
227
+ },
228
+ {
229
+ "epoch": 0.1863091037402964,
230
+ "grad_norm": 7.357480049133301,
231
+ "learning_rate": 7.500000000000001e-06,
232
+ "loss": 6.1259,
233
+ "step": 33
234
+ },
235
+ {
236
+ "epoch": 0.20324629498941427,
237
+ "grad_norm": 7.321795463562012,
238
+ "learning_rate": 8.181818181818183e-06,
239
+ "loss": 5.7539,
240
+ "step": 36
241
+ },
242
+ {
243
+ "epoch": 0.22018348623853212,
244
+ "grad_norm": 4.239792346954346,
245
+ "learning_rate": 8.863636363636365e-06,
246
+ "loss": 6.0131,
247
+ "step": 39
248
+ },
249
+ {
250
+ "epoch": 0.23712067748764998,
251
+ "grad_norm": 3.9554407596588135,
252
+ "learning_rate": 9.545454545454547e-06,
253
+ "loss": 6.0074,
254
+ "step": 42
255
+ },
256
+ {
257
+ "epoch": 0.25405786873676783,
258
+ "grad_norm": 4.406026840209961,
259
+ "learning_rate": 1.0227272727272729e-05,
260
+ "loss": 5.7125,
261
+ "step": 45
262
+ },
263
+ {
264
+ "epoch": 0.2709950599858857,
265
+ "grad_norm": 7.235893249511719,
266
+ "learning_rate": 1.0909090909090909e-05,
267
+ "loss": 5.5634,
268
+ "step": 48
269
+ },
270
+ {
271
+ "epoch": 0.28793225123500354,
272
+ "grad_norm": 5.330288410186768,
273
+ "learning_rate": 1.1590909090909093e-05,
274
+ "loss": 5.2924,
275
+ "step": 51
276
+ },
277
+ {
278
+ "epoch": 0.3048694424841214,
279
+ "grad_norm": 7.216403961181641,
280
+ "learning_rate": 1.2272727272727274e-05,
281
+ "loss": 5.2286,
282
+ "step": 54
283
+ },
284
+ {
285
+ "epoch": 0.3048694424841214,
286
+ "eval_NLI-v2_cosine_accuracy": 1.0,
287
+ "eval_NLI-v2_dot_accuracy": 0.046875,
288
+ "eval_NLI-v2_euclidean_accuracy": 1.0,
289
+ "eval_NLI-v2_manhattan_accuracy": 1.0,
290
+ "eval_NLI-v2_max_accuracy": 1.0,
291
+ "eval_VitaminC_cosine_accuracy": 0.54296875,
292
+ "eval_VitaminC_cosine_accuracy_threshold": 0.9328227043151855,
293
+ "eval_VitaminC_cosine_ap": 0.5212059026196154,
294
+ "eval_VitaminC_cosine_f1": 0.6576819407008085,
295
+ "eval_VitaminC_cosine_f1_threshold": 0.7373804450035095,
296
+ "eval_VitaminC_cosine_precision": 0.49193548387096775,
297
+ "eval_VitaminC_cosine_recall": 0.991869918699187,
298
+ "eval_VitaminC_dot_accuracy": 0.55078125,
299
+ "eval_VitaminC_dot_accuracy_threshold": 418.2774658203125,
300
+ "eval_VitaminC_dot_ap": 0.5160594099493883,
301
+ "eval_VitaminC_dot_f1": 0.6521739130434782,
302
+ "eval_VitaminC_dot_f1_threshold": 291.5081481933594,
303
+ "eval_VitaminC_dot_precision": 0.4897959183673469,
304
+ "eval_VitaminC_dot_recall": 0.975609756097561,
305
+ "eval_VitaminC_euclidean_accuracy": 0.5390625,
306
+ "eval_VitaminC_euclidean_accuracy_threshold": 8.120429039001465,
307
+ "eval_VitaminC_euclidean_ap": 0.5224837623095228,
308
+ "eval_VitaminC_euclidean_f1": 0.6576819407008085,
309
+ "eval_VitaminC_euclidean_f1_threshold": 14.879999160766602,
310
+ "eval_VitaminC_euclidean_precision": 0.49193548387096775,
311
+ "eval_VitaminC_euclidean_recall": 0.991869918699187,
312
+ "eval_VitaminC_manhattan_accuracy": 0.53515625,
313
+ "eval_VitaminC_manhattan_accuracy_threshold": 137.40658569335938,
314
+ "eval_VitaminC_manhattan_ap": 0.5186382518671783,
315
+ "eval_VitaminC_manhattan_f1": 0.6576086956521738,
316
+ "eval_VitaminC_manhattan_f1_threshold": 263.32452392578125,
317
+ "eval_VitaminC_manhattan_precision": 0.49387755102040815,
318
+ "eval_VitaminC_manhattan_recall": 0.983739837398374,
319
+ "eval_VitaminC_max_accuracy": 0.55078125,
320
+ "eval_VitaminC_max_accuracy_threshold": 418.2774658203125,
321
+ "eval_VitaminC_max_ap": 0.5224837623095228,
322
+ "eval_VitaminC_max_f1": 0.6576819407008085,
323
+ "eval_VitaminC_max_f1_threshold": 291.5081481933594,
324
+ "eval_VitaminC_max_precision": 0.49387755102040815,
325
+ "eval_VitaminC_max_recall": 0.991869918699187,
326
+ "eval_sequential_score": 0.5224837623095228,
327
+ "eval_sts-test_pearson_cosine": 0.14377091128453176,
328
+ "eval_sts-test_pearson_dot": 0.24728387094758872,
329
+ "eval_sts-test_pearson_euclidean": 0.14604155960515372,
330
+ "eval_sts-test_pearson_manhattan": 0.1446467532231986,
331
+ "eval_sts-test_pearson_max": 0.24728387094758872,
332
+ "eval_sts-test_spearman_cosine": 0.1968510434344728,
333
+ "eval_sts-test_spearman_dot": 0.29467218283745694,
334
+ "eval_sts-test_spearman_euclidean": 0.17218164683969664,
335
+ "eval_sts-test_spearman_manhattan": 0.17741843340856742,
336
+ "eval_sts-test_spearman_max": 0.29467218283745694,
337
+ "eval_vitaminc-pairs_loss": 2.664700746536255,
338
+ "eval_vitaminc-pairs_runtime": 1.4487,
339
+ "eval_vitaminc-pairs_samples_per_second": 74.551,
340
+ "eval_vitaminc-pairs_steps_per_second": 1.381,
341
+ "step": 54
342
+ },
343
+ {
344
+ "epoch": 0.3048694424841214,
345
+ "eval_negation-triplets_loss": 4.6218037605285645,
346
+ "eval_negation-triplets_runtime": 0.2971,
347
+ "eval_negation-triplets_samples_per_second": 215.438,
348
+ "eval_negation-triplets_steps_per_second": 3.366,
349
+ "step": 54
350
+ },
351
+ {
352
+ "epoch": 0.3048694424841214,
353
+ "eval_scitail-pairs-pos_loss": 1.2413936853408813,
354
+ "eval_scitail-pairs-pos_runtime": 0.372,
355
+ "eval_scitail-pairs-pos_samples_per_second": 145.175,
356
+ "eval_scitail-pairs-pos_steps_per_second": 2.688,
357
+ "step": 54
358
+ },
359
+ {
360
+ "epoch": 0.3048694424841214,
361
+ "eval_xsum-pairs_loss": 5.249766826629639,
362
+ "eval_xsum-pairs_runtime": 3.1506,
363
+ "eval_xsum-pairs_samples_per_second": 40.627,
364
+ "eval_xsum-pairs_steps_per_second": 0.635,
365
+ "step": 54
366
+ },
367
+ {
368
+ "epoch": 0.3048694424841214,
369
+ "eval_sciq_pairs_loss": 0.2961578667163849,
370
+ "eval_sciq_pairs_runtime": 3.2909,
371
+ "eval_sciq_pairs_samples_per_second": 38.895,
372
+ "eval_sciq_pairs_steps_per_second": 0.608,
373
+ "step": 54
374
+ },
375
+ {
376
+ "epoch": 0.3048694424841214,
377
+ "eval_qasc_pairs_loss": 2.530872344970703,
378
+ "eval_qasc_pairs_runtime": 0.6255,
379
+ "eval_qasc_pairs_samples_per_second": 204.63,
380
+ "eval_qasc_pairs_steps_per_second": 3.197,
381
+ "step": 54
382
+ },
383
+ {
384
+ "epoch": 0.3048694424841214,
385
+ "eval_openbookqa_pairs_loss": 3.8855104446411133,
386
+ "eval_openbookqa_pairs_runtime": 0.5742,
387
+ "eval_openbookqa_pairs_samples_per_second": 222.914,
388
+ "eval_openbookqa_pairs_steps_per_second": 3.483,
389
+ "step": 54
390
+ },
391
+ {
392
+ "epoch": 0.3048694424841214,
393
+ "eval_msmarco_pairs_loss": 5.246406555175781,
394
+ "eval_msmarco_pairs_runtime": 1.2872,
395
+ "eval_msmarco_pairs_samples_per_second": 99.442,
396
+ "eval_msmarco_pairs_steps_per_second": 1.554,
397
+ "step": 54
398
+ },
399
+ {
400
+ "epoch": 0.3048694424841214,
401
+ "eval_nq_pairs_loss": 5.332630157470703,
402
+ "eval_nq_pairs_runtime": 2.3739,
403
+ "eval_nq_pairs_samples_per_second": 53.92,
404
+ "eval_nq_pairs_steps_per_second": 0.843,
405
+ "step": 54
406
+ },
407
+ {
408
+ "epoch": 0.3048694424841214,
409
+ "eval_trivia_pairs_loss": 5.647429943084717,
410
+ "eval_trivia_pairs_runtime": 4.4729,
411
+ "eval_trivia_pairs_samples_per_second": 28.617,
412
+ "eval_trivia_pairs_steps_per_second": 0.447,
413
+ "step": 54
414
+ },
415
+ {
416
+ "epoch": 0.3048694424841214,
417
+ "eval_gooaq_pairs_loss": 5.225871562957764,
418
+ "eval_gooaq_pairs_runtime": 0.8715,
419
+ "eval_gooaq_pairs_samples_per_second": 146.868,
420
+ "eval_gooaq_pairs_steps_per_second": 2.295,
421
+ "step": 54
422
+ },
423
+ {
424
+ "epoch": 0.3048694424841214,
425
+ "eval_paws-pos_loss": 0.8335962891578674,
426
+ "eval_paws-pos_runtime": 0.6844,
427
+ "eval_paws-pos_samples_per_second": 187.036,
428
+ "eval_paws-pos_steps_per_second": 2.922,
429
+ "step": 54
430
+ },
431
+ {
432
+ "epoch": 0.32180663373323926,
433
+ "grad_norm": 6.847682952880859,
434
+ "learning_rate": 1.2954545454545455e-05,
435
+ "loss": 4.4811,
436
+ "step": 57
437
+ },
438
+ {
439
+ "epoch": 0.3387438249823571,
440
+ "grad_norm": 8.383002281188965,
441
+ "learning_rate": 1.3636363636363637e-05,
442
+ "loss": 4.4239,
443
+ "step": 60
444
+ },
445
+ {
446
+ "epoch": 0.35568101623147497,
447
+ "grad_norm": 7.014843463897705,
448
+ "learning_rate": 1.431818181818182e-05,
449
+ "loss": 4.0273,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.3726182074805928,
454
+ "grad_norm": 5.9739885330200195,
455
+ "learning_rate": 1.5000000000000002e-05,
456
+ "loss": 3.4508,
457
+ "step": 66
458
+ },
459
+ {
460
+ "epoch": 0.3895553987297107,
461
+ "grad_norm": 11.202752113342285,
462
+ "learning_rate": 1.5681818181818182e-05,
463
+ "loss": 3.9702,
464
+ "step": 69
465
+ },
466
+ {
467
+ "epoch": 0.40649258997882853,
468
+ "grad_norm": 7.064818859100342,
469
+ "learning_rate": 1.6363636363636366e-05,
470
+ "loss": 3.5295,
471
+ "step": 72
472
+ },
473
+ {
474
+ "epoch": 0.4234297812279464,
475
+ "grad_norm": 5.912719249725342,
476
+ "learning_rate": 1.7045454545454546e-05,
477
+ "loss": 3.6395,
478
+ "step": 75
479
+ },
480
+ {
481
+ "epoch": 0.44036697247706424,
482
+ "grad_norm": 5.033207893371582,
483
+ "learning_rate": 1.772727272727273e-05,
484
+ "loss": 3.2398,
485
+ "step": 78
486
+ },
487
+ {
488
+ "epoch": 0.4573041637261821,
489
+ "grad_norm": 5.218384265899658,
490
+ "learning_rate": 1.840909090909091e-05,
491
+ "loss": 3.116,
492
+ "step": 81
493
+ },
494
+ {
495
+ "epoch": 0.4573041637261821,
496
+ "eval_NLI-v2_cosine_accuracy": 1.0,
497
+ "eval_NLI-v2_dot_accuracy": 0.0,
498
+ "eval_NLI-v2_euclidean_accuracy": 1.0,
499
+ "eval_NLI-v2_manhattan_accuracy": 1.0,
500
+ "eval_NLI-v2_max_accuracy": 1.0,
501
+ "eval_VitaminC_cosine_accuracy": 0.5546875,
502
+ "eval_VitaminC_cosine_accuracy_threshold": 0.9041332006454468,
503
+ "eval_VitaminC_cosine_ap": 0.5292859731465609,
504
+ "eval_VitaminC_cosine_f1": 0.6542553191489362,
505
+ "eval_VitaminC_cosine_f1_threshold": 0.452939510345459,
506
+ "eval_VitaminC_cosine_precision": 0.48616600790513836,
507
+ "eval_VitaminC_cosine_recall": 1.0,
508
+ "eval_VitaminC_dot_accuracy": 0.5546875,
509
+ "eval_VitaminC_dot_accuracy_threshold": 414.42559814453125,
510
+ "eval_VitaminC_dot_ap": 0.5222732504955002,
511
+ "eval_VitaminC_dot_f1": 0.6542553191489362,
512
+ "eval_VitaminC_dot_f1_threshold": 212.6934814453125,
513
+ "eval_VitaminC_dot_precision": 0.48616600790513836,
514
+ "eval_VitaminC_dot_recall": 1.0,
515
+ "eval_VitaminC_euclidean_accuracy": 0.5546875,
516
+ "eval_VitaminC_euclidean_accuracy_threshold": 9.18377685546875,
517
+ "eval_VitaminC_euclidean_ap": 0.5291787221346742,
518
+ "eval_VitaminC_euclidean_f1": 0.6542553191489362,
519
+ "eval_VitaminC_euclidean_f1_threshold": 22.683509826660156,
520
+ "eval_VitaminC_euclidean_precision": 0.48616600790513836,
521
+ "eval_VitaminC_euclidean_recall": 1.0,
522
+ "eval_VitaminC_manhattan_accuracy": 0.55859375,
523
+ "eval_VitaminC_manhattan_accuracy_threshold": 173.8212127685547,
524
+ "eval_VitaminC_manhattan_ap": 0.5305698453165033,
525
+ "eval_VitaminC_manhattan_f1": 0.6542553191489362,
526
+ "eval_VitaminC_manhattan_f1_threshold": 415.5366516113281,
527
+ "eval_VitaminC_manhattan_precision": 0.48616600790513836,
528
+ "eval_VitaminC_manhattan_recall": 1.0,
529
+ "eval_VitaminC_max_accuracy": 0.55859375,
530
+ "eval_VitaminC_max_accuracy_threshold": 414.42559814453125,
531
+ "eval_VitaminC_max_ap": 0.5305698453165033,
532
+ "eval_VitaminC_max_f1": 0.6542553191489362,
533
+ "eval_VitaminC_max_f1_threshold": 415.5366516113281,
534
+ "eval_VitaminC_max_precision": 0.48616600790513836,
535
+ "eval_VitaminC_max_recall": 1.0,
536
+ "eval_sequential_score": 0.5305698453165033,
537
+ "eval_sts-test_pearson_cosine": 0.45494716382349193,
538
+ "eval_sts-test_pearson_dot": 0.44837123659858896,
539
+ "eval_sts-test_pearson_euclidean": 0.4480861256491879,
540
+ "eval_sts-test_pearson_manhattan": 0.4417008219313264,
541
+ "eval_sts-test_pearson_max": 0.45494716382349193,
542
+ "eval_sts-test_spearman_cosine": 0.48921418507251446,
543
+ "eval_sts-test_spearman_dot": 0.46707725062744593,
544
+ "eval_sts-test_spearman_euclidean": 0.4610824798409968,
545
+ "eval_sts-test_spearman_manhattan": 0.46068648052845956,
546
+ "eval_sts-test_spearman_max": 0.48921418507251446,
547
+ "eval_vitaminc-pairs_loss": 2.5043575763702393,
548
+ "eval_vitaminc-pairs_runtime": 1.4778,
549
+ "eval_vitaminc-pairs_samples_per_second": 73.079,
550
+ "eval_vitaminc-pairs_steps_per_second": 1.353,
551
+ "step": 81
552
+ },
553
+ {
554
+ "epoch": 0.4573041637261821,
555
+ "eval_negation-triplets_loss": 3.4229447841644287,
556
+ "eval_negation-triplets_runtime": 0.2991,
557
+ "eval_negation-triplets_samples_per_second": 213.954,
558
+ "eval_negation-triplets_steps_per_second": 3.343,
559
+ "step": 81
560
+ },
561
+ {
562
+ "epoch": 0.4573041637261821,
563
+ "eval_scitail-pairs-pos_loss": 0.2784869372844696,
564
+ "eval_scitail-pairs-pos_runtime": 0.3633,
565
+ "eval_scitail-pairs-pos_samples_per_second": 148.649,
566
+ "eval_scitail-pairs-pos_steps_per_second": 2.753,
567
+ "step": 81
568
+ },
569
+ {
570
+ "epoch": 0.4573041637261821,
571
+ "eval_xsum-pairs_loss": 2.428964614868164,
572
+ "eval_xsum-pairs_runtime": 3.1548,
573
+ "eval_xsum-pairs_samples_per_second": 40.573,
574
+ "eval_xsum-pairs_steps_per_second": 0.634,
575
+ "step": 81
576
+ },
577
+ {
578
+ "epoch": 0.4573041637261821,
579
+ "eval_sciq_pairs_loss": 0.15256048738956451,
580
+ "eval_sciq_pairs_runtime": 3.2432,
581
+ "eval_sciq_pairs_samples_per_second": 39.467,
582
+ "eval_sciq_pairs_steps_per_second": 0.617,
583
+ "step": 81
584
+ },
585
+ {
586
+ "epoch": 0.4573041637261821,
587
+ "eval_qasc_pairs_loss": 1.2902077436447144,
588
+ "eval_qasc_pairs_runtime": 0.6211,
589
+ "eval_qasc_pairs_samples_per_second": 206.085,
590
+ "eval_qasc_pairs_steps_per_second": 3.22,
591
+ "step": 81
592
+ },
593
+ {
594
+ "epoch": 0.4573041637261821,
595
+ "eval_openbookqa_pairs_loss": 2.4784862995147705,
596
+ "eval_openbookqa_pairs_runtime": 0.5758,
597
+ "eval_openbookqa_pairs_samples_per_second": 222.308,
598
+ "eval_openbookqa_pairs_steps_per_second": 3.474,
599
+ "step": 81
600
+ },
601
+ {
602
+ "epoch": 0.4573041637261821,
603
+ "eval_msmarco_pairs_loss": 2.967724084854126,
604
+ "eval_msmarco_pairs_runtime": 1.2944,
605
+ "eval_msmarco_pairs_samples_per_second": 98.885,
606
+ "eval_msmarco_pairs_steps_per_second": 1.545,
607
+ "step": 81
608
+ },
609
+ {
610
+ "epoch": 0.4573041637261821,
611
+ "eval_nq_pairs_loss": 3.358661413192749,
612
+ "eval_nq_pairs_runtime": 2.3827,
613
+ "eval_nq_pairs_samples_per_second": 53.722,
614
+ "eval_nq_pairs_steps_per_second": 0.839,
615
+ "step": 81
616
+ },
617
+ {
618
+ "epoch": 0.4573041637261821,
619
+ "eval_trivia_pairs_loss": 3.1391680240631104,
620
+ "eval_trivia_pairs_runtime": 4.4155,
621
+ "eval_trivia_pairs_samples_per_second": 28.989,
622
+ "eval_trivia_pairs_steps_per_second": 0.453,
623
+ "step": 81
624
+ },
625
+ {
626
+ "epoch": 0.4573041637261821,
627
+ "eval_gooaq_pairs_loss": 2.8774912357330322,
628
+ "eval_gooaq_pairs_runtime": 0.8746,
629
+ "eval_gooaq_pairs_samples_per_second": 146.346,
630
+ "eval_gooaq_pairs_steps_per_second": 2.287,
631
+ "step": 81
632
+ },
633
+ {
634
+ "epoch": 0.4573041637261821,
635
+ "eval_paws-pos_loss": 0.19754411280155182,
636
+ "eval_paws-pos_runtime": 0.684,
637
+ "eval_paws-pos_samples_per_second": 187.141,
638
+ "eval_paws-pos_steps_per_second": 2.924,
639
+ "step": 81
640
+ },
641
+ {
642
+ "epoch": 0.47424135497529996,
643
+ "grad_norm": 5.149569988250732,
644
+ "learning_rate": 1.9090909090909094e-05,
645
+ "loss": 2.6049,
646
+ "step": 84
647
+ },
648
+ {
649
+ "epoch": 0.4911785462244178,
650
+ "grad_norm": 5.012928009033203,
651
+ "learning_rate": 1.9772727272727274e-05,
652
+ "loss": 2.7738,
653
+ "step": 87
654
+ },
655
+ {
656
+ "epoch": 0.5081157374735357,
657
+ "grad_norm": 4.880725383758545,
658
+ "learning_rate": 2.0454545454545457e-05,
659
+ "loss": 2.5416,
660
+ "step": 90
661
+ },
662
+ {
663
+ "epoch": 0.5250529287226535,
664
+ "grad_norm": 5.618528366088867,
665
+ "learning_rate": 2.113636363636364e-05,
666
+ "loss": 2.3913,
667
+ "step": 93
668
+ },
669
+ {
670
+ "epoch": 0.5419901199717714,
671
+ "grad_norm": 5.020515441894531,
672
+ "learning_rate": 2.1818181818181818e-05,
673
+ "loss": 2.3144,
674
+ "step": 96
675
+ },
676
+ {
677
+ "epoch": 0.5589273112208892,
678
+ "grad_norm": 4.818451404571533,
679
+ "learning_rate": 2.25e-05,
680
+ "loss": 2.1857,
681
+ "step": 99
682
+ },
683
+ {
684
+ "epoch": 0.5758645024700071,
685
+ "grad_norm": 5.094771385192871,
686
+ "learning_rate": 2.3181818181818185e-05,
687
+ "loss": 1.8881,
688
+ "step": 102
689
+ },
690
+ {
691
+ "epoch": 0.592801693719125,
692
+ "grad_norm": 3.795962333679199,
693
+ "learning_rate": 2.3863636363636365e-05,
694
+ "loss": 2.2699,
695
+ "step": 105
696
+ }
697
+ ],
698
+ "logging_steps": 3,
699
+ "max_steps": 531,
700
+ "num_input_tokens_seen": 0,
701
+ "num_train_epochs": 3,
702
+ "save_steps": 107,
703
+ "stateful_callbacks": {
704
+ "TrainerControl": {
705
+ "args": {
706
+ "should_epoch_stop": false,
707
+ "should_evaluate": false,
708
+ "should_log": false,
709
+ "should_save": true,
710
+ "should_training_stop": false
711
+ },
712
+ "attributes": {}
713
+ }
714
+ },
715
+ "total_flos": 0.0,
716
+ "train_batch_size": 160,
717
+ "trial_name": null,
718
+ "trial_params": null
719
+ }
checkpoint-107/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:526057a01875662c89243f2f1101012515e0f154fca67a38aba3fb44dcb2d6d0
3
+ size 5688