bobox commited on
Commit
a33cdfa
·
verified ·
1 Parent(s): e3e3697

Training in progress, step 5166, checkpoint

Browse files
checkpoint-5166/1_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-5166/README.md ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5166/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
checkpoint-5166/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bobox/DeBERTa-ST-AllLayers-v3-checkpoints-tmp",
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-5166/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-5166/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-5166/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48dce9fac67f527c5cd15330176f1ed9f5d1112b836db760542dcebb4a58fc20
3
+ size 1130520122
checkpoint-5166/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25e1bad878927495b582acce5107edb86e060716f82f8eccb92dc12596e12fbc
3
+ size 565251810
checkpoint-5166/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25ab1e41e74ed8f024bb2c6e63aea8eda3721a6d4f4319bb800b079eb34bd1ca
3
+ size 14180
checkpoint-5166/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc253291aff0c40a90ac4c74ab0c743394b67fb8cd847ad95f915bb5a3ae2271
3
+ size 1064
checkpoint-5166/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
checkpoint-5166/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
checkpoint-5166/spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
checkpoint-5166/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5166/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "max_length": 512,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "pad_to_multiple_of": null,
53
+ "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
+ "sep_token": "[SEP]",
57
+ "sp_model_kwargs": {},
58
+ "split_by_punct": false,
59
+ "stride": 0,
60
+ "tokenizer_class": "DebertaV2Tokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
+ "unk_token": "[UNK]",
64
+ "vocab_type": "spm"
65
+ }
checkpoint-5166/trainer_state.json ADDED
@@ -0,0 +1,1237 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 1292,
6
+ "global_step": 5166,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.012582268679829655,
13
+ "grad_norm": 4.108283996582031,
14
+ "learning_rate": 3.7166085946573755e-07,
15
+ "loss": 0.4577,
16
+ "step": 65
17
+ },
18
+ {
19
+ "epoch": 0.02516453735965931,
20
+ "grad_norm": 0.5773646831512451,
21
+ "learning_rate": 7.491289198606272e-07,
22
+ "loss": 0.4707,
23
+ "step": 130
24
+ },
25
+ {
26
+ "epoch": 0.03774680603948897,
27
+ "grad_norm": 0.6127182841300964,
28
+ "learning_rate": 1.1265969802555168e-06,
29
+ "loss": 0.5259,
30
+ "step": 195
31
+ },
32
+ {
33
+ "epoch": 0.05032907471931862,
34
+ "grad_norm": 1.513021469116211,
35
+ "learning_rate": 1.5040650406504067e-06,
36
+ "loss": 0.5501,
37
+ "step": 260
38
+ },
39
+ {
40
+ "epoch": 0.06291134339914828,
41
+ "grad_norm": 11.750137329101562,
42
+ "learning_rate": 1.8815331010452962e-06,
43
+ "loss": 0.5089,
44
+ "step": 325
45
+ },
46
+ {
47
+ "epoch": 0.07549361207897794,
48
+ "grad_norm": 4.16257905960083,
49
+ "learning_rate": 2.259001161440186e-06,
50
+ "loss": 0.4816,
51
+ "step": 390
52
+ },
53
+ {
54
+ "epoch": 0.08807588075880758,
55
+ "grad_norm": 16.527780532836914,
56
+ "learning_rate": 2.6306620209059233e-06,
57
+ "loss": 0.5822,
58
+ "step": 455
59
+ },
60
+ {
61
+ "epoch": 0.10065814943863724,
62
+ "grad_norm": 0.1845797300338745,
63
+ "learning_rate": 3.0081300813008134e-06,
64
+ "loss": 0.5686,
65
+ "step": 520
66
+ },
67
+ {
68
+ "epoch": 0.1132404181184669,
69
+ "grad_norm": 12.172422409057617,
70
+ "learning_rate": 3.3855981416957026e-06,
71
+ "loss": 0.5686,
72
+ "step": 585
73
+ },
74
+ {
75
+ "epoch": 0.12582268679829656,
76
+ "grad_norm": 0.6270273923873901,
77
+ "learning_rate": 3.7630662020905923e-06,
78
+ "loss": 0.517,
79
+ "step": 650
80
+ },
81
+ {
82
+ "epoch": 0.1384049554781262,
83
+ "grad_norm": 3.6368539333343506,
84
+ "learning_rate": 4.140534262485482e-06,
85
+ "loss": 0.3615,
86
+ "step": 715
87
+ },
88
+ {
89
+ "epoch": 0.15098722415795587,
90
+ "grad_norm": 9.541145324707031,
91
+ "learning_rate": 4.518002322880372e-06,
92
+ "loss": 0.5978,
93
+ "step": 780
94
+ },
95
+ {
96
+ "epoch": 0.16356949283778552,
97
+ "grad_norm": 9.86439323425293,
98
+ "learning_rate": 4.895470383275261e-06,
99
+ "loss": 0.5153,
100
+ "step": 845
101
+ },
102
+ {
103
+ "epoch": 0.17615176151761516,
104
+ "grad_norm": 4.669048309326172,
105
+ "learning_rate": 5.272938443670151e-06,
106
+ "loss": 0.5059,
107
+ "step": 910
108
+ },
109
+ {
110
+ "epoch": 0.18873403019744484,
111
+ "grad_norm": 9.666926383972168,
112
+ "learning_rate": 5.650406504065041e-06,
113
+ "loss": 0.5624,
114
+ "step": 975
115
+ },
116
+ {
117
+ "epoch": 0.20131629887727448,
118
+ "grad_norm": 6.078874588012695,
119
+ "learning_rate": 6.02787456445993e-06,
120
+ "loss": 0.5201,
121
+ "step": 1040
122
+ },
123
+ {
124
+ "epoch": 0.21389856755710415,
125
+ "grad_norm": 1.1067451238632202,
126
+ "learning_rate": 6.4053426248548205e-06,
127
+ "loss": 0.6127,
128
+ "step": 1105
129
+ },
130
+ {
131
+ "epoch": 0.2264808362369338,
132
+ "grad_norm": 1.1589373350143433,
133
+ "learning_rate": 6.78281068524971e-06,
134
+ "loss": 0.5333,
135
+ "step": 1170
136
+ },
137
+ {
138
+ "epoch": 0.23906310491676344,
139
+ "grad_norm": 1.977501630783081,
140
+ "learning_rate": 7.1602787456446e-06,
141
+ "loss": 0.494,
142
+ "step": 1235
143
+ },
144
+ {
145
+ "epoch": 0.2500967866821525,
146
+ "eval_StS-test_pearson_cosine": 0.8821101738384596,
147
+ "eval_StS-test_pearson_dot": 0.8032893366124795,
148
+ "eval_StS-test_pearson_euclidean": 0.8697205121607111,
149
+ "eval_StS-test_pearson_manhattan": 0.8704995590187196,
150
+ "eval_StS-test_pearson_max": 0.8821101738384596,
151
+ "eval_StS-test_spearman_cosine": 0.8943047751560564,
152
+ "eval_StS-test_spearman_dot": 0.8087424893555902,
153
+ "eval_StS-test_spearman_euclidean": 0.871583089708652,
154
+ "eval_StS-test_spearman_manhattan": 0.8737012027236009,
155
+ "eval_StS-test_spearman_max": 0.8943047751560564,
156
+ "eval_Vitaminc-test_cosine_accuracy": 0.5684210526315789,
157
+ "eval_Vitaminc-test_cosine_accuracy_threshold": 0.7028586268424988,
158
+ "eval_Vitaminc-test_cosine_ap": 0.5651043866206488,
159
+ "eval_Vitaminc-test_cosine_f1": 0.6755218216318786,
160
+ "eval_Vitaminc-test_cosine_f1_threshold": 0.5077509880065918,
161
+ "eval_Vitaminc-test_cosine_precision": 0.52046783625731,
162
+ "eval_Vitaminc-test_cosine_recall": 0.9621621621621622,
163
+ "eval_Vitaminc-test_dot_accuracy": 0.5684210526315789,
164
+ "eval_Vitaminc-test_dot_accuracy_threshold": 19.693286895751953,
165
+ "eval_Vitaminc-test_dot_ap": 0.5463931769790206,
166
+ "eval_Vitaminc-test_dot_f1": 0.6691449814126395,
167
+ "eval_Vitaminc-test_dot_f1_threshold": 13.839346885681152,
168
+ "eval_Vitaminc-test_dot_precision": 0.509915014164306,
169
+ "eval_Vitaminc-test_dot_recall": 0.972972972972973,
170
+ "eval_Vitaminc-test_euclidean_accuracy": 0.5894736842105263,
171
+ "eval_Vitaminc-test_euclidean_accuracy_threshold": 4.252468585968018,
172
+ "eval_Vitaminc-test_euclidean_ap": 0.5569049511912931,
173
+ "eval_Vitaminc-test_euclidean_f1": 0.6666666666666666,
174
+ "eval_Vitaminc-test_euclidean_f1_threshold": 6.922356128692627,
175
+ "eval_Vitaminc-test_euclidean_precision": 0.5041551246537396,
176
+ "eval_Vitaminc-test_euclidean_recall": 0.9837837837837838,
177
+ "eval_Vitaminc-test_manhattan_accuracy": 0.5815789473684211,
178
+ "eval_Vitaminc-test_manhattan_accuracy_threshold": 87.21337890625,
179
+ "eval_Vitaminc-test_manhattan_ap": 0.5572154085134091,
180
+ "eval_Vitaminc-test_manhattan_f1": 0.6666666666666667,
181
+ "eval_Vitaminc-test_manhattan_f1_threshold": 141.26380920410156,
182
+ "eval_Vitaminc-test_manhattan_precision": 0.505586592178771,
183
+ "eval_Vitaminc-test_manhattan_recall": 0.9783783783783784,
184
+ "eval_Vitaminc-test_max_accuracy": 0.5894736842105263,
185
+ "eval_Vitaminc-test_max_accuracy_threshold": 87.21337890625,
186
+ "eval_Vitaminc-test_max_ap": 0.5651043866206488,
187
+ "eval_Vitaminc-test_max_f1": 0.6755218216318786,
188
+ "eval_Vitaminc-test_max_f1_threshold": 141.26380920410156,
189
+ "eval_Vitaminc-test_max_precision": 0.52046783625731,
190
+ "eval_Vitaminc-test_max_recall": 0.9837837837837838,
191
+ "eval_mrpc-test_cosine_accuracy": 0.7473684210526316,
192
+ "eval_mrpc-test_cosine_accuracy_threshold": 0.7145693302154541,
193
+ "eval_mrpc-test_cosine_ap": 0.8563235829800693,
194
+ "eval_mrpc-test_cosine_f1": 0.8327645051194539,
195
+ "eval_mrpc-test_cosine_f1_threshold": 0.6522408723831177,
196
+ "eval_mrpc-test_cosine_precision": 0.7218934911242604,
197
+ "eval_mrpc-test_cosine_recall": 0.9838709677419355,
198
+ "eval_mrpc-test_dot_accuracy": 0.7026315789473684,
199
+ "eval_mrpc-test_dot_accuracy_threshold": 14.454626083374023,
200
+ "eval_mrpc-test_dot_ap": 0.796363256728503,
201
+ "eval_mrpc-test_dot_f1": 0.8054607508532423,
202
+ "eval_mrpc-test_dot_f1_threshold": 13.752894401550293,
203
+ "eval_mrpc-test_dot_precision": 0.6982248520710059,
204
+ "eval_mrpc-test_dot_recall": 0.9516129032258065,
205
+ "eval_mrpc-test_euclidean_accuracy": 0.7315789473684211,
206
+ "eval_mrpc-test_euclidean_accuracy_threshold": 3.890326499938965,
207
+ "eval_mrpc-test_euclidean_ap": 0.8252367395643119,
208
+ "eval_mrpc-test_euclidean_f1": 0.8165467625899281,
209
+ "eval_mrpc-test_euclidean_f1_threshold": 3.890326499938965,
210
+ "eval_mrpc-test_euclidean_precision": 0.737012987012987,
211
+ "eval_mrpc-test_euclidean_recall": 0.9153225806451613,
212
+ "eval_mrpc-test_manhattan_accuracy": 0.7289473684210527,
213
+ "eval_mrpc-test_manhattan_accuracy_threshold": 77.57926177978516,
214
+ "eval_mrpc-test_manhattan_ap": 0.8208816982117964,
215
+ "eval_mrpc-test_manhattan_f1": 0.815742397137746,
216
+ "eval_mrpc-test_manhattan_f1_threshold": 79.14703369140625,
217
+ "eval_mrpc-test_manhattan_precision": 0.7331189710610932,
218
+ "eval_mrpc-test_manhattan_recall": 0.9193548387096774,
219
+ "eval_mrpc-test_max_accuracy": 0.7473684210526316,
220
+ "eval_mrpc-test_max_accuracy_threshold": 77.57926177978516,
221
+ "eval_mrpc-test_max_ap": 0.8563235829800693,
222
+ "eval_mrpc-test_max_f1": 0.8327645051194539,
223
+ "eval_mrpc-test_max_f1_threshold": 79.14703369140625,
224
+ "eval_mrpc-test_max_precision": 0.737012987012987,
225
+ "eval_mrpc-test_max_recall": 0.9838709677419355,
226
+ "eval_nli-pairs_loss": 0.8093397617340088,
227
+ "eval_nli-pairs_runtime": 3.0363,
228
+ "eval_nli-pairs_samples_per_second": 52.696,
229
+ "eval_nli-pairs_steps_per_second": 1.647,
230
+ "eval_sequential_score": 0.5651043866206488,
231
+ "step": 1292
232
+ },
233
+ {
234
+ "epoch": 0.2500967866821525,
235
+ "eval_vitaminc-pairs_loss": 5.769770622253418,
236
+ "eval_vitaminc-pairs_runtime": 1.5488,
237
+ "eval_vitaminc-pairs_samples_per_second": 85.875,
238
+ "eval_vitaminc-pairs_steps_per_second": 3.228,
239
+ "step": 1292
240
+ },
241
+ {
242
+ "epoch": 0.2500967866821525,
243
+ "eval_qnli-contrastive_loss": 0.12109158933162689,
244
+ "eval_qnli-contrastive_runtime": 0.5097,
245
+ "eval_qnli-contrastive_samples_per_second": 313.889,
246
+ "eval_qnli-contrastive_steps_per_second": 9.809,
247
+ "step": 1292
248
+ },
249
+ {
250
+ "epoch": 0.2500967866821525,
251
+ "eval_scitail-pairs-qa_loss": 0.07553695887327194,
252
+ "eval_scitail-pairs-qa_runtime": 1.2071,
253
+ "eval_scitail-pairs-qa_samples_per_second": 132.548,
254
+ "eval_scitail-pairs-qa_steps_per_second": 4.142,
255
+ "step": 1292
256
+ },
257
+ {
258
+ "epoch": 0.2500967866821525,
259
+ "eval_scitail-pairs-pos_loss": 0.3979075253009796,
260
+ "eval_scitail-pairs-pos_runtime": 2.3649,
261
+ "eval_scitail-pairs-pos_samples_per_second": 67.656,
262
+ "eval_scitail-pairs-pos_steps_per_second": 2.114,
263
+ "step": 1292
264
+ },
265
+ {
266
+ "epoch": 0.2500967866821525,
267
+ "eval_xsum-pairs_loss": 0.313429057598114,
268
+ "eval_xsum-pairs_runtime": 1.4107,
269
+ "eval_xsum-pairs_samples_per_second": 113.419,
270
+ "eval_xsum-pairs_steps_per_second": 3.544,
271
+ "step": 1292
272
+ },
273
+ {
274
+ "epoch": 0.2500967866821525,
275
+ "eval_compression-pairs_loss": 0.08316509425640106,
276
+ "eval_compression-pairs_runtime": 0.3958,
277
+ "eval_compression-pairs_samples_per_second": 404.289,
278
+ "eval_compression-pairs_steps_per_second": 12.634,
279
+ "step": 1292
280
+ },
281
+ {
282
+ "epoch": 0.2500967866821525,
283
+ "eval_sciq_pairs_loss": 0.2692818343639374,
284
+ "eval_sciq_pairs_runtime": 7.8991,
285
+ "eval_sciq_pairs_samples_per_second": 20.255,
286
+ "eval_sciq_pairs_steps_per_second": 0.633,
287
+ "step": 1292
288
+ },
289
+ {
290
+ "epoch": 0.2500967866821525,
291
+ "eval_qasc_pairs_loss": 0.19870159029960632,
292
+ "eval_qasc_pairs_runtime": 1.4336,
293
+ "eval_qasc_pairs_samples_per_second": 111.608,
294
+ "eval_qasc_pairs_steps_per_second": 3.488,
295
+ "step": 1292
296
+ },
297
+ {
298
+ "epoch": 0.2500967866821525,
299
+ "eval_qasc_facts_sym_loss": 0.16445104777812958,
300
+ "eval_qasc_facts_sym_runtime": 0.3196,
301
+ "eval_qasc_facts_sym_samples_per_second": 500.598,
302
+ "eval_qasc_facts_sym_steps_per_second": 15.644,
303
+ "step": 1292
304
+ },
305
+ {
306
+ "epoch": 0.2500967866821525,
307
+ "eval_openbookqa_pairs_loss": 1.7182375192642212,
308
+ "eval_openbookqa_pairs_runtime": 1.2252,
309
+ "eval_openbookqa_pairs_samples_per_second": 130.592,
310
+ "eval_openbookqa_pairs_steps_per_second": 4.081,
311
+ "step": 1292
312
+ },
313
+ {
314
+ "epoch": 0.2500967866821525,
315
+ "eval_msmarco_pairs_loss": 0.4961338937282562,
316
+ "eval_msmarco_pairs_runtime": 3.1144,
317
+ "eval_msmarco_pairs_samples_per_second": 51.374,
318
+ "eval_msmarco_pairs_steps_per_second": 1.605,
319
+ "step": 1292
320
+ },
321
+ {
322
+ "epoch": 0.2500967866821525,
323
+ "eval_nq_pairs_loss": 0.4005078673362732,
324
+ "eval_nq_pairs_runtime": 7.7074,
325
+ "eval_nq_pairs_samples_per_second": 20.759,
326
+ "eval_nq_pairs_steps_per_second": 0.649,
327
+ "step": 1292
328
+ },
329
+ {
330
+ "epoch": 0.2500967866821525,
331
+ "eval_trivia_pairs_loss": 0.654505729675293,
332
+ "eval_trivia_pairs_runtime": 10.1383,
333
+ "eval_trivia_pairs_samples_per_second": 15.782,
334
+ "eval_trivia_pairs_steps_per_second": 0.493,
335
+ "step": 1292
336
+ },
337
+ {
338
+ "epoch": 0.2500967866821525,
339
+ "eval_quora_pairs_loss": 0.22833283245563507,
340
+ "eval_quora_pairs_runtime": 3.894,
341
+ "eval_quora_pairs_samples_per_second": 173.342,
342
+ "eval_quora_pairs_steps_per_second": 5.65,
343
+ "step": 1292
344
+ },
345
+ {
346
+ "epoch": 0.2500967866821525,
347
+ "eval_gooaq_pairs_loss": 0.4996432662010193,
348
+ "eval_gooaq_pairs_runtime": 2.2697,
349
+ "eval_gooaq_pairs_samples_per_second": 70.494,
350
+ "eval_gooaq_pairs_steps_per_second": 2.203,
351
+ "step": 1292
352
+ },
353
+ {
354
+ "epoch": 0.2500967866821525,
355
+ "eval_mrpc_pairs_loss": 0.04613902047276497,
356
+ "eval_mrpc_pairs_runtime": 0.3743,
357
+ "eval_mrpc_pairs_samples_per_second": 427.427,
358
+ "eval_mrpc_pairs_steps_per_second": 13.357,
359
+ "step": 1292
360
+ },
361
+ {
362
+ "epoch": 0.2516453735965931,
363
+ "grad_norm": 12.50217342376709,
364
+ "learning_rate": 7.53774680603949e-06,
365
+ "loss": 0.6236,
366
+ "step": 1300
367
+ },
368
+ {
369
+ "epoch": 0.26422764227642276,
370
+ "grad_norm": 1.8397300243377686,
371
+ "learning_rate": 7.91521486643438e-06,
372
+ "loss": 0.4947,
373
+ "step": 1365
374
+ },
375
+ {
376
+ "epoch": 0.2768099109562524,
377
+ "grad_norm": 5.885033130645752,
378
+ "learning_rate": 8.292682926829268e-06,
379
+ "loss": 0.5595,
380
+ "step": 1430
381
+ },
382
+ {
383
+ "epoch": 0.28939217963608205,
384
+ "grad_norm": 1.7783002853393555,
385
+ "learning_rate": 8.670150987224158e-06,
386
+ "loss": 0.641,
387
+ "step": 1495
388
+ },
389
+ {
390
+ "epoch": 0.30197444831591175,
391
+ "grad_norm": 5.305712699890137,
392
+ "learning_rate": 9.047619047619047e-06,
393
+ "loss": 0.5188,
394
+ "step": 1560
395
+ },
396
+ {
397
+ "epoch": 0.3145567169957414,
398
+ "grad_norm": 8.24319839477539,
399
+ "learning_rate": 9.425087108013936e-06,
400
+ "loss": 0.4927,
401
+ "step": 1625
402
+ },
403
+ {
404
+ "epoch": 0.32713898567557104,
405
+ "grad_norm": 11.07426929473877,
406
+ "learning_rate": 9.802555168408827e-06,
407
+ "loss": 0.657,
408
+ "step": 1690
409
+ },
410
+ {
411
+ "epoch": 0.3397212543554007,
412
+ "grad_norm": 9.04263687133789,
413
+ "learning_rate": 1.0180023228803716e-05,
414
+ "loss": 0.4665,
415
+ "step": 1755
416
+ },
417
+ {
418
+ "epoch": 0.3523035230352303,
419
+ "grad_norm": 1.4980370998382568,
420
+ "learning_rate": 1.0557491289198606e-05,
421
+ "loss": 0.4645,
422
+ "step": 1820
423
+ },
424
+ {
425
+ "epoch": 0.36488579171506,
426
+ "grad_norm": 16.511180877685547,
427
+ "learning_rate": 1.0934959349593495e-05,
428
+ "loss": 0.5887,
429
+ "step": 1885
430
+ },
431
+ {
432
+ "epoch": 0.37746806039488967,
433
+ "grad_norm": 5.706000804901123,
434
+ "learning_rate": 1.1312427409988386e-05,
435
+ "loss": 0.5308,
436
+ "step": 1950
437
+ },
438
+ {
439
+ "epoch": 0.3900503290747193,
440
+ "grad_norm": 1.0923340320587158,
441
+ "learning_rate": 1.1689895470383277e-05,
442
+ "loss": 0.536,
443
+ "step": 2015
444
+ },
445
+ {
446
+ "epoch": 0.40263259775454896,
447
+ "grad_norm": 6.521665573120117,
448
+ "learning_rate": 1.2067363530778166e-05,
449
+ "loss": 0.4841,
450
+ "step": 2080
451
+ },
452
+ {
453
+ "epoch": 0.4152148664343786,
454
+ "grad_norm": 7.254842758178711,
455
+ "learning_rate": 1.2444831591173055e-05,
456
+ "loss": 0.6499,
457
+ "step": 2145
458
+ },
459
+ {
460
+ "epoch": 0.4277971351142083,
461
+ "grad_norm": 14.938628196716309,
462
+ "learning_rate": 1.2822299651567945e-05,
463
+ "loss": 0.5982,
464
+ "step": 2210
465
+ },
466
+ {
467
+ "epoch": 0.44037940379403795,
468
+ "grad_norm": 10.192171096801758,
469
+ "learning_rate": 1.3199767711962834e-05,
470
+ "loss": 0.5281,
471
+ "step": 2275
472
+ },
473
+ {
474
+ "epoch": 0.4529616724738676,
475
+ "grad_norm": 8.02379035949707,
476
+ "learning_rate": 1.3577235772357725e-05,
477
+ "loss": 0.6657,
478
+ "step": 2340
479
+ },
480
+ {
481
+ "epoch": 0.46554394115369724,
482
+ "grad_norm": 2.842752695083618,
483
+ "learning_rate": 1.3954703832752614e-05,
484
+ "loss": 0.5746,
485
+ "step": 2405
486
+ },
487
+ {
488
+ "epoch": 0.4781262098335269,
489
+ "grad_norm": 0.4373825788497925,
490
+ "learning_rate": 1.432636469221835e-05,
491
+ "loss": 0.5853,
492
+ "step": 2470
493
+ },
494
+ {
495
+ "epoch": 0.4907084785133566,
496
+ "grad_norm": 16.475561141967773,
497
+ "learning_rate": 1.4703832752613242e-05,
498
+ "loss": 0.5828,
499
+ "step": 2535
500
+ },
501
+ {
502
+ "epoch": 0.500193573364305,
503
+ "eval_StS-test_pearson_cosine": 0.8850112014664042,
504
+ "eval_StS-test_pearson_dot": 0.8059948281019231,
505
+ "eval_StS-test_pearson_euclidean": 0.8736480607586303,
506
+ "eval_StS-test_pearson_manhattan": 0.8746434402751778,
507
+ "eval_StS-test_pearson_max": 0.8850112014664042,
508
+ "eval_StS-test_spearman_cosine": 0.8964908683344393,
509
+ "eval_StS-test_spearman_dot": 0.8112940990649002,
510
+ "eval_StS-test_spearman_euclidean": 0.8728524906348676,
511
+ "eval_StS-test_spearman_manhattan": 0.8747700566942432,
512
+ "eval_StS-test_spearman_max": 0.8964908683344393,
513
+ "eval_Vitaminc-test_cosine_accuracy": 0.5842105263157895,
514
+ "eval_Vitaminc-test_cosine_accuracy_threshold": 0.6876633763313293,
515
+ "eval_Vitaminc-test_cosine_ap": 0.5620310838989798,
516
+ "eval_Vitaminc-test_cosine_f1": 0.6755725190839694,
517
+ "eval_Vitaminc-test_cosine_f1_threshold": 0.5227451324462891,
518
+ "eval_Vitaminc-test_cosine_precision": 0.5221238938053098,
519
+ "eval_Vitaminc-test_cosine_recall": 0.9567567567567568,
520
+ "eval_Vitaminc-test_dot_accuracy": 0.5552631578947368,
521
+ "eval_Vitaminc-test_dot_accuracy_threshold": 19.32275390625,
522
+ "eval_Vitaminc-test_dot_ap": 0.5459852333889619,
523
+ "eval_Vitaminc-test_dot_f1": 0.6715596330275228,
524
+ "eval_Vitaminc-test_dot_f1_threshold": 12.946126937866211,
525
+ "eval_Vitaminc-test_dot_precision": 0.5083333333333333,
526
+ "eval_Vitaminc-test_dot_recall": 0.9891891891891892,
527
+ "eval_Vitaminc-test_euclidean_accuracy": 0.5736842105263158,
528
+ "eval_Vitaminc-test_euclidean_accuracy_threshold": 3.9492130279541016,
529
+ "eval_Vitaminc-test_euclidean_ap": 0.5569069466677845,
530
+ "eval_Vitaminc-test_euclidean_f1": 0.6654343807763402,
531
+ "eval_Vitaminc-test_euclidean_f1_threshold": 6.641269207000732,
532
+ "eval_Vitaminc-test_euclidean_precision": 0.5056179775280899,
533
+ "eval_Vitaminc-test_euclidean_recall": 0.972972972972973,
534
+ "eval_Vitaminc-test_manhattan_accuracy": 0.5789473684210527,
535
+ "eval_Vitaminc-test_manhattan_accuracy_threshold": 80.14714050292969,
536
+ "eval_Vitaminc-test_manhattan_ap": 0.556199495550533,
537
+ "eval_Vitaminc-test_manhattan_f1": 0.6691312384473197,
538
+ "eval_Vitaminc-test_manhattan_f1_threshold": 138.50221252441406,
539
+ "eval_Vitaminc-test_manhattan_precision": 0.5084269662921348,
540
+ "eval_Vitaminc-test_manhattan_recall": 0.9783783783783784,
541
+ "eval_Vitaminc-test_max_accuracy": 0.5842105263157895,
542
+ "eval_Vitaminc-test_max_accuracy_threshold": 80.14714050292969,
543
+ "eval_Vitaminc-test_max_ap": 0.5620310838989798,
544
+ "eval_Vitaminc-test_max_f1": 0.6755725190839694,
545
+ "eval_Vitaminc-test_max_f1_threshold": 138.50221252441406,
546
+ "eval_Vitaminc-test_max_precision": 0.5221238938053098,
547
+ "eval_Vitaminc-test_max_recall": 0.9891891891891892,
548
+ "eval_mrpc-test_cosine_accuracy": 0.7368421052631579,
549
+ "eval_mrpc-test_cosine_accuracy_threshold": 0.7595815658569336,
550
+ "eval_mrpc-test_cosine_ap": 0.8489522026016108,
551
+ "eval_mrpc-test_cosine_f1": 0.8235294117647058,
552
+ "eval_mrpc-test_cosine_f1_threshold": 0.6731016635894775,
553
+ "eval_mrpc-test_cosine_precision": 0.7212121212121212,
554
+ "eval_mrpc-test_cosine_recall": 0.9596774193548387,
555
+ "eval_mrpc-test_dot_accuracy": 0.6894736842105263,
556
+ "eval_mrpc-test_dot_accuracy_threshold": 15.645252227783203,
557
+ "eval_mrpc-test_dot_ap": 0.7828717059557433,
558
+ "eval_mrpc-test_dot_f1": 0.8071895424836601,
559
+ "eval_mrpc-test_dot_f1_threshold": 11.141586303710938,
560
+ "eval_mrpc-test_dot_precision": 0.6785714285714286,
561
+ "eval_mrpc-test_dot_recall": 0.9959677419354839,
562
+ "eval_mrpc-test_euclidean_accuracy": 0.7368421052631579,
563
+ "eval_mrpc-test_euclidean_accuracy_threshold": 3.7453079223632812,
564
+ "eval_mrpc-test_euclidean_ap": 0.8190549606804299,
565
+ "eval_mrpc-test_euclidean_f1": 0.8172043010752688,
566
+ "eval_mrpc-test_euclidean_f1_threshold": 3.905801773071289,
567
+ "eval_mrpc-test_euclidean_precision": 0.7354838709677419,
568
+ "eval_mrpc-test_euclidean_recall": 0.9193548387096774,
569
+ "eval_mrpc-test_manhattan_accuracy": 0.7315789473684211,
570
+ "eval_mrpc-test_manhattan_accuracy_threshold": 74.24961853027344,
571
+ "eval_mrpc-test_manhattan_ap": 0.815561648496177,
572
+ "eval_mrpc-test_manhattan_f1": 0.81508078994614,
573
+ "eval_mrpc-test_manhattan_f1_threshold": 78.01387023925781,
574
+ "eval_mrpc-test_manhattan_precision": 0.7346278317152104,
575
+ "eval_mrpc-test_manhattan_recall": 0.9153225806451613,
576
+ "eval_mrpc-test_max_accuracy": 0.7368421052631579,
577
+ "eval_mrpc-test_max_accuracy_threshold": 74.24961853027344,
578
+ "eval_mrpc-test_max_ap": 0.8489522026016108,
579
+ "eval_mrpc-test_max_f1": 0.8235294117647058,
580
+ "eval_mrpc-test_max_f1_threshold": 78.01387023925781,
581
+ "eval_mrpc-test_max_precision": 0.7354838709677419,
582
+ "eval_mrpc-test_max_recall": 0.9959677419354839,
583
+ "eval_nli-pairs_loss": 0.805642306804657,
584
+ "eval_nli-pairs_runtime": 3.1328,
585
+ "eval_nli-pairs_samples_per_second": 51.072,
586
+ "eval_nli-pairs_steps_per_second": 1.596,
587
+ "eval_sequential_score": 0.5620310838989798,
588
+ "step": 2584
589
+ },
590
+ {
591
+ "epoch": 0.500193573364305,
592
+ "eval_vitaminc-pairs_loss": 5.803024768829346,
593
+ "eval_vitaminc-pairs_runtime": 1.6236,
594
+ "eval_vitaminc-pairs_samples_per_second": 81.915,
595
+ "eval_vitaminc-pairs_steps_per_second": 3.08,
596
+ "step": 2584
597
+ },
598
+ {
599
+ "epoch": 0.500193573364305,
600
+ "eval_qnli-contrastive_loss": 0.1560516655445099,
601
+ "eval_qnli-contrastive_runtime": 0.5455,
602
+ "eval_qnli-contrastive_samples_per_second": 293.295,
603
+ "eval_qnli-contrastive_steps_per_second": 9.165,
604
+ "step": 2584
605
+ },
606
+ {
607
+ "epoch": 0.500193573364305,
608
+ "eval_scitail-pairs-qa_loss": 0.07059809565544128,
609
+ "eval_scitail-pairs-qa_runtime": 1.3304,
610
+ "eval_scitail-pairs-qa_samples_per_second": 120.266,
611
+ "eval_scitail-pairs-qa_steps_per_second": 3.758,
612
+ "step": 2584
613
+ },
614
+ {
615
+ "epoch": 0.500193573364305,
616
+ "eval_scitail-pairs-pos_loss": 0.40096405148506165,
617
+ "eval_scitail-pairs-pos_runtime": 2.6358,
618
+ "eval_scitail-pairs-pos_samples_per_second": 60.703,
619
+ "eval_scitail-pairs-pos_steps_per_second": 1.897,
620
+ "step": 2584
621
+ },
622
+ {
623
+ "epoch": 0.500193573364305,
624
+ "eval_xsum-pairs_loss": 0.3021017014980316,
625
+ "eval_xsum-pairs_runtime": 1.474,
626
+ "eval_xsum-pairs_samples_per_second": 108.547,
627
+ "eval_xsum-pairs_steps_per_second": 3.392,
628
+ "step": 2584
629
+ },
630
+ {
631
+ "epoch": 0.500193573364305,
632
+ "eval_compression-pairs_loss": 0.08080567419528961,
633
+ "eval_compression-pairs_runtime": 0.4395,
634
+ "eval_compression-pairs_samples_per_second": 364.011,
635
+ "eval_compression-pairs_steps_per_second": 11.375,
636
+ "step": 2584
637
+ },
638
+ {
639
+ "epoch": 0.500193573364305,
640
+ "eval_sciq_pairs_loss": 0.2838394045829773,
641
+ "eval_sciq_pairs_runtime": 8.1466,
642
+ "eval_sciq_pairs_samples_per_second": 19.64,
643
+ "eval_sciq_pairs_steps_per_second": 0.614,
644
+ "step": 2584
645
+ },
646
+ {
647
+ "epoch": 0.500193573364305,
648
+ "eval_qasc_pairs_loss": 0.19900637865066528,
649
+ "eval_qasc_pairs_runtime": 1.4833,
650
+ "eval_qasc_pairs_samples_per_second": 107.864,
651
+ "eval_qasc_pairs_steps_per_second": 3.371,
652
+ "step": 2584
653
+ },
654
+ {
655
+ "epoch": 0.500193573364305,
656
+ "eval_qasc_facts_sym_loss": 0.15245158970355988,
657
+ "eval_qasc_facts_sym_runtime": 0.3292,
658
+ "eval_qasc_facts_sym_samples_per_second": 485.978,
659
+ "eval_qasc_facts_sym_steps_per_second": 15.187,
660
+ "step": 2584
661
+ },
662
+ {
663
+ "epoch": 0.500193573364305,
664
+ "eval_openbookqa_pairs_loss": 1.7069475650787354,
665
+ "eval_openbookqa_pairs_runtime": 1.4335,
666
+ "eval_openbookqa_pairs_samples_per_second": 111.612,
667
+ "eval_openbookqa_pairs_steps_per_second": 3.488,
668
+ "step": 2584
669
+ },
670
+ {
671
+ "epoch": 0.500193573364305,
672
+ "eval_msmarco_pairs_loss": 0.5149500966072083,
673
+ "eval_msmarco_pairs_runtime": 3.2287,
674
+ "eval_msmarco_pairs_samples_per_second": 49.555,
675
+ "eval_msmarco_pairs_steps_per_second": 1.549,
676
+ "step": 2584
677
+ },
678
+ {
679
+ "epoch": 0.500193573364305,
680
+ "eval_nq_pairs_loss": 0.4162893295288086,
681
+ "eval_nq_pairs_runtime": 7.8277,
682
+ "eval_nq_pairs_samples_per_second": 20.44,
683
+ "eval_nq_pairs_steps_per_second": 0.639,
684
+ "step": 2584
685
+ },
686
+ {
687
+ "epoch": 0.500193573364305,
688
+ "eval_trivia_pairs_loss": 0.623663067817688,
689
+ "eval_trivia_pairs_runtime": 10.277,
690
+ "eval_trivia_pairs_samples_per_second": 15.569,
691
+ "eval_trivia_pairs_steps_per_second": 0.487,
692
+ "step": 2584
693
+ },
694
+ {
695
+ "epoch": 0.500193573364305,
696
+ "eval_quora_pairs_loss": 0.22309938073158264,
697
+ "eval_quora_pairs_runtime": 4.3318,
698
+ "eval_quora_pairs_samples_per_second": 155.824,
699
+ "eval_quora_pairs_steps_per_second": 5.079,
700
+ "step": 2584
701
+ },
702
+ {
703
+ "epoch": 0.500193573364305,
704
+ "eval_gooaq_pairs_loss": 0.49019113183021545,
705
+ "eval_gooaq_pairs_runtime": 2.2716,
706
+ "eval_gooaq_pairs_samples_per_second": 70.434,
707
+ "eval_gooaq_pairs_steps_per_second": 2.201,
708
+ "step": 2584
709
+ },
710
+ {
711
+ "epoch": 0.500193573364305,
712
+ "eval_mrpc_pairs_loss": 0.04494442045688629,
713
+ "eval_mrpc_pairs_runtime": 0.3756,
714
+ "eval_mrpc_pairs_samples_per_second": 425.958,
715
+ "eval_mrpc_pairs_steps_per_second": 13.311,
716
+ "step": 2584
717
+ },
718
+ {
719
+ "epoch": 0.5032907471931862,
720
+ "grad_norm": 0.5926259756088257,
721
+ "learning_rate": 1.5081300813008131e-05,
722
+ "loss": 0.4889,
723
+ "step": 2600
724
+ },
725
+ {
726
+ "epoch": 0.5158730158730159,
727
+ "grad_norm": 12.89693832397461,
728
+ "learning_rate": 1.545876887340302e-05,
729
+ "loss": 0.517,
730
+ "step": 2665
731
+ },
732
+ {
733
+ "epoch": 0.5284552845528455,
734
+ "grad_norm": 0.3751634359359741,
735
+ "learning_rate": 1.583623693379791e-05,
736
+ "loss": 0.5479,
737
+ "step": 2730
738
+ },
739
+ {
740
+ "epoch": 0.5410375532326752,
741
+ "grad_norm": 10.321626663208008,
742
+ "learning_rate": 1.62137049941928e-05,
743
+ "loss": 0.6409,
744
+ "step": 2795
745
+ },
746
+ {
747
+ "epoch": 0.5536198219125048,
748
+ "grad_norm": 0.40425410866737366,
749
+ "learning_rate": 1.659117305458769e-05,
750
+ "loss": 0.6137,
751
+ "step": 2860
752
+ },
753
+ {
754
+ "epoch": 0.5662020905923345,
755
+ "grad_norm": 0.26469847559928894,
756
+ "learning_rate": 1.696864111498258e-05,
757
+ "loss": 0.5947,
758
+ "step": 2925
759
+ },
760
+ {
761
+ "epoch": 0.5787843592721641,
762
+ "grad_norm": 0.3376736640930176,
763
+ "learning_rate": 1.7346109175377468e-05,
764
+ "loss": 0.514,
765
+ "step": 2990
766
+ },
767
+ {
768
+ "epoch": 0.5913666279519938,
769
+ "grad_norm": 11.042977333068848,
770
+ "learning_rate": 1.7723577235772357e-05,
771
+ "loss": 0.5256,
772
+ "step": 3055
773
+ },
774
+ {
775
+ "epoch": 0.6039488966318235,
776
+ "grad_norm": 28.415281295776367,
777
+ "learning_rate": 1.810104529616725e-05,
778
+ "loss": 0.5574,
779
+ "step": 3120
780
+ },
781
+ {
782
+ "epoch": 0.6165311653116531,
783
+ "grad_norm": 1.138208270072937,
784
+ "learning_rate": 1.8478513356562136e-05,
785
+ "loss": 0.5669,
786
+ "step": 3185
787
+ },
788
+ {
789
+ "epoch": 0.6291134339914828,
790
+ "grad_norm": 0.38507771492004395,
791
+ "learning_rate": 1.885598141695703e-05,
792
+ "loss": 0.5185,
793
+ "step": 3250
794
+ },
795
+ {
796
+ "epoch": 0.6416957026713124,
797
+ "grad_norm": 0.4718106687068939,
798
+ "learning_rate": 1.9233449477351914e-05,
799
+ "loss": 0.4875,
800
+ "step": 3315
801
+ },
802
+ {
803
+ "epoch": 0.6542779713511421,
804
+ "grad_norm": 0.3544551432132721,
805
+ "learning_rate": 1.9610917537746807e-05,
806
+ "loss": 0.5773,
807
+ "step": 3380
808
+ },
809
+ {
810
+ "epoch": 0.6668602400309718,
811
+ "grad_norm": 15.193903923034668,
812
+ "learning_rate": 1.9988385598141696e-05,
813
+ "loss": 0.6102,
814
+ "step": 3445
815
+ },
816
+ {
817
+ "epoch": 0.6794425087108014,
818
+ "grad_norm": 14.930959701538086,
819
+ "learning_rate": 2.0365853658536586e-05,
820
+ "loss": 0.5882,
821
+ "step": 3510
822
+ },
823
+ {
824
+ "epoch": 0.6920247773906311,
825
+ "grad_norm": 0.7249612212181091,
826
+ "learning_rate": 2.0743321718931475e-05,
827
+ "loss": 0.5888,
828
+ "step": 3575
829
+ },
830
+ {
831
+ "epoch": 0.7046070460704607,
832
+ "grad_norm": 0.21552956104278564,
833
+ "learning_rate": 2.1120789779326364e-05,
834
+ "loss": 0.6088,
835
+ "step": 3640
836
+ },
837
+ {
838
+ "epoch": 0.7171893147502904,
839
+ "grad_norm": 3.176356554031372,
840
+ "learning_rate": 2.1498257839721253e-05,
841
+ "loss": 0.688,
842
+ "step": 3705
843
+ },
844
+ {
845
+ "epoch": 0.72977158343012,
846
+ "grad_norm": 4.0542378425598145,
847
+ "learning_rate": 2.1875725900116146e-05,
848
+ "loss": 0.6541,
849
+ "step": 3770
850
+ },
851
+ {
852
+ "epoch": 0.7423538521099496,
853
+ "grad_norm": 0.5486993193626404,
854
+ "learning_rate": 2.2253193960511032e-05,
855
+ "loss": 0.4526,
856
+ "step": 3835
857
+ },
858
+ {
859
+ "epoch": 0.7502903600464577,
860
+ "eval_StS-test_pearson_cosine": 0.8754429584344908,
861
+ "eval_StS-test_pearson_dot": 0.7812438977199179,
862
+ "eval_StS-test_pearson_euclidean": 0.863532080020405,
863
+ "eval_StS-test_pearson_manhattan": 0.8642206125042041,
864
+ "eval_StS-test_pearson_max": 0.8754429584344908,
865
+ "eval_StS-test_spearman_cosine": 0.8899322603927042,
866
+ "eval_StS-test_spearman_dot": 0.7897822634826679,
867
+ "eval_StS-test_spearman_euclidean": 0.8661349339052001,
868
+ "eval_StS-test_spearman_manhattan": 0.8677664578276145,
869
+ "eval_StS-test_spearman_max": 0.8899322603927042,
870
+ "eval_Vitaminc-test_cosine_accuracy": 0.5815789473684211,
871
+ "eval_Vitaminc-test_cosine_accuracy_threshold": 0.7150823473930359,
872
+ "eval_Vitaminc-test_cosine_ap": 0.5695623413042502,
873
+ "eval_Vitaminc-test_cosine_f1": 0.6690909090909091,
874
+ "eval_Vitaminc-test_cosine_f1_threshold": 0.4144631028175354,
875
+ "eval_Vitaminc-test_cosine_precision": 0.5041095890410959,
876
+ "eval_Vitaminc-test_cosine_recall": 0.9945945945945946,
877
+ "eval_Vitaminc-test_dot_accuracy": 0.5631578947368421,
878
+ "eval_Vitaminc-test_dot_accuracy_threshold": 25.451148986816406,
879
+ "eval_Vitaminc-test_dot_ap": 0.5500277455332443,
880
+ "eval_Vitaminc-test_dot_f1": 0.6691449814126395,
881
+ "eval_Vitaminc-test_dot_f1_threshold": 14.098217964172363,
882
+ "eval_Vitaminc-test_dot_precision": 0.509915014164306,
883
+ "eval_Vitaminc-test_dot_recall": 0.972972972972973,
884
+ "eval_Vitaminc-test_euclidean_accuracy": 0.5736842105263158,
885
+ "eval_Vitaminc-test_euclidean_accuracy_threshold": 3.945739269256592,
886
+ "eval_Vitaminc-test_euclidean_ap": 0.5590599048028368,
887
+ "eval_Vitaminc-test_euclidean_f1": 0.6654478976234005,
888
+ "eval_Vitaminc-test_euclidean_f1_threshold": 6.931303024291992,
889
+ "eval_Vitaminc-test_euclidean_precision": 0.5027624309392266,
890
+ "eval_Vitaminc-test_euclidean_recall": 0.9837837837837838,
891
+ "eval_Vitaminc-test_manhattan_accuracy": 0.5736842105263158,
892
+ "eval_Vitaminc-test_manhattan_accuracy_threshold": 84.99573516845703,
893
+ "eval_Vitaminc-test_manhattan_ap": 0.559747734383099,
894
+ "eval_Vitaminc-test_manhattan_f1": 0.6679035250463822,
895
+ "eval_Vitaminc-test_manhattan_f1_threshold": 139.30535888671875,
896
+ "eval_Vitaminc-test_manhattan_precision": 0.5084745762711864,
897
+ "eval_Vitaminc-test_manhattan_recall": 0.972972972972973,
898
+ "eval_Vitaminc-test_max_accuracy": 0.5815789473684211,
899
+ "eval_Vitaminc-test_max_accuracy_threshold": 84.99573516845703,
900
+ "eval_Vitaminc-test_max_ap": 0.5695623413042502,
901
+ "eval_Vitaminc-test_max_f1": 0.6691449814126395,
902
+ "eval_Vitaminc-test_max_f1_threshold": 139.30535888671875,
903
+ "eval_Vitaminc-test_max_precision": 0.509915014164306,
904
+ "eval_Vitaminc-test_max_recall": 0.9945945945945946,
905
+ "eval_mrpc-test_cosine_accuracy": 0.7447368421052631,
906
+ "eval_mrpc-test_cosine_accuracy_threshold": 0.7351576089859009,
907
+ "eval_mrpc-test_cosine_ap": 0.8492225227540071,
908
+ "eval_mrpc-test_cosine_f1": 0.8274647887323944,
909
+ "eval_mrpc-test_cosine_f1_threshold": 0.7097899913787842,
910
+ "eval_mrpc-test_cosine_precision": 0.734375,
911
+ "eval_mrpc-test_cosine_recall": 0.9475806451612904,
912
+ "eval_mrpc-test_dot_accuracy": 0.6947368421052632,
913
+ "eval_mrpc-test_dot_accuracy_threshold": 13.770221710205078,
914
+ "eval_mrpc-test_dot_ap": 0.7664430695917603,
915
+ "eval_mrpc-test_dot_f1": 0.8060200668896322,
916
+ "eval_mrpc-test_dot_f1_threshold": 13.664058685302734,
917
+ "eval_mrpc-test_dot_precision": 0.6885714285714286,
918
+ "eval_mrpc-test_dot_recall": 0.9717741935483871,
919
+ "eval_mrpc-test_euclidean_accuracy": 0.7421052631578947,
920
+ "eval_mrpc-test_euclidean_accuracy_threshold": 3.573061466217041,
921
+ "eval_mrpc-test_euclidean_ap": 0.8229315066081502,
922
+ "eval_mrpc-test_euclidean_f1": 0.8216216216216217,
923
+ "eval_mrpc-test_euclidean_f1_threshold": 3.7641119956970215,
924
+ "eval_mrpc-test_euclidean_precision": 0.742671009771987,
925
+ "eval_mrpc-test_euclidean_recall": 0.9193548387096774,
926
+ "eval_mrpc-test_manhattan_accuracy": 0.7394736842105263,
927
+ "eval_mrpc-test_manhattan_accuracy_threshold": 75.05320739746094,
928
+ "eval_mrpc-test_manhattan_ap": 0.8195074272882318,
929
+ "eval_mrpc-test_manhattan_f1": 0.8228980322003578,
930
+ "eval_mrpc-test_manhattan_f1_threshold": 76.21920013427734,
931
+ "eval_mrpc-test_manhattan_precision": 0.7395498392282959,
932
+ "eval_mrpc-test_manhattan_recall": 0.9274193548387096,
933
+ "eval_mrpc-test_max_accuracy": 0.7447368421052631,
934
+ "eval_mrpc-test_max_accuracy_threshold": 75.05320739746094,
935
+ "eval_mrpc-test_max_ap": 0.8492225227540071,
936
+ "eval_mrpc-test_max_f1": 0.8274647887323944,
937
+ "eval_mrpc-test_max_f1_threshold": 76.21920013427734,
938
+ "eval_mrpc-test_max_precision": 0.742671009771987,
939
+ "eval_mrpc-test_max_recall": 0.9717741935483871,
940
+ "eval_nli-pairs_loss": 0.8814637064933777,
941
+ "eval_nli-pairs_runtime": 2.9865,
942
+ "eval_nli-pairs_samples_per_second": 53.575,
943
+ "eval_nli-pairs_steps_per_second": 1.674,
944
+ "eval_sequential_score": 0.5695623413042502,
945
+ "step": 3876
946
+ },
947
+ {
948
+ "epoch": 0.7502903600464577,
949
+ "eval_vitaminc-pairs_loss": 5.706007957458496,
950
+ "eval_vitaminc-pairs_runtime": 1.5554,
951
+ "eval_vitaminc-pairs_samples_per_second": 85.507,
952
+ "eval_vitaminc-pairs_steps_per_second": 3.215,
953
+ "step": 3876
954
+ },
955
+ {
956
+ "epoch": 0.7502903600464577,
957
+ "eval_qnli-contrastive_loss": 0.14242754876613617,
958
+ "eval_qnli-contrastive_runtime": 0.5128,
959
+ "eval_qnli-contrastive_samples_per_second": 311.993,
960
+ "eval_qnli-contrastive_steps_per_second": 9.75,
961
+ "step": 3876
962
+ },
963
+ {
964
+ "epoch": 0.7502903600464577,
965
+ "eval_scitail-pairs-qa_loss": 0.0674210786819458,
966
+ "eval_scitail-pairs-qa_runtime": 1.2622,
967
+ "eval_scitail-pairs-qa_samples_per_second": 126.759,
968
+ "eval_scitail-pairs-qa_steps_per_second": 3.961,
969
+ "step": 3876
970
+ },
971
+ {
972
+ "epoch": 0.7502903600464577,
973
+ "eval_scitail-pairs-pos_loss": 0.36303356289863586,
974
+ "eval_scitail-pairs-pos_runtime": 2.3597,
975
+ "eval_scitail-pairs-pos_samples_per_second": 67.806,
976
+ "eval_scitail-pairs-pos_steps_per_second": 2.119,
977
+ "step": 3876
978
+ },
979
+ {
980
+ "epoch": 0.7502903600464577,
981
+ "eval_xsum-pairs_loss": 0.32611021399497986,
982
+ "eval_xsum-pairs_runtime": 1.4024,
983
+ "eval_xsum-pairs_samples_per_second": 114.091,
984
+ "eval_xsum-pairs_steps_per_second": 3.565,
985
+ "step": 3876
986
+ },
987
+ {
988
+ "epoch": 0.7502903600464577,
989
+ "eval_compression-pairs_loss": 0.08624111860990524,
990
+ "eval_compression-pairs_runtime": 0.3948,
991
+ "eval_compression-pairs_samples_per_second": 405.263,
992
+ "eval_compression-pairs_steps_per_second": 12.664,
993
+ "step": 3876
994
+ },
995
+ {
996
+ "epoch": 0.7502903600464577,
997
+ "eval_sciq_pairs_loss": 0.2929747402667999,
998
+ "eval_sciq_pairs_runtime": 7.9352,
999
+ "eval_sciq_pairs_samples_per_second": 20.163,
1000
+ "eval_sciq_pairs_steps_per_second": 0.63,
1001
+ "step": 3876
1002
+ },
1003
+ {
1004
+ "epoch": 0.7502903600464577,
1005
+ "eval_qasc_pairs_loss": 0.20694346725940704,
1006
+ "eval_qasc_pairs_runtime": 1.4441,
1007
+ "eval_qasc_pairs_samples_per_second": 110.798,
1008
+ "eval_qasc_pairs_steps_per_second": 3.462,
1009
+ "step": 3876
1010
+ },
1011
+ {
1012
+ "epoch": 0.7502903600464577,
1013
+ "eval_qasc_facts_sym_loss": 0.14476129412651062,
1014
+ "eval_qasc_facts_sym_runtime": 0.3295,
1015
+ "eval_qasc_facts_sym_samples_per_second": 485.621,
1016
+ "eval_qasc_facts_sym_steps_per_second": 15.176,
1017
+ "step": 3876
1018
+ },
1019
+ {
1020
+ "epoch": 0.7502903600464577,
1021
+ "eval_openbookqa_pairs_loss": 1.7119003534317017,
1022
+ "eval_openbookqa_pairs_runtime": 1.2433,
1023
+ "eval_openbookqa_pairs_samples_per_second": 128.687,
1024
+ "eval_openbookqa_pairs_steps_per_second": 4.021,
1025
+ "step": 3876
1026
+ },
1027
+ {
1028
+ "epoch": 0.7502903600464577,
1029
+ "eval_msmarco_pairs_loss": 0.4974798262119293,
1030
+ "eval_msmarco_pairs_runtime": 3.1059,
1031
+ "eval_msmarco_pairs_samples_per_second": 51.516,
1032
+ "eval_msmarco_pairs_steps_per_second": 1.61,
1033
+ "step": 3876
1034
+ },
1035
+ {
1036
+ "epoch": 0.7502903600464577,
1037
+ "eval_nq_pairs_loss": 0.46937060356140137,
1038
+ "eval_nq_pairs_runtime": 7.6944,
1039
+ "eval_nq_pairs_samples_per_second": 20.794,
1040
+ "eval_nq_pairs_steps_per_second": 0.65,
1041
+ "step": 3876
1042
+ },
1043
+ {
1044
+ "epoch": 0.7502903600464577,
1045
+ "eval_trivia_pairs_loss": 0.6990238428115845,
1046
+ "eval_trivia_pairs_runtime": 10.1876,
1047
+ "eval_trivia_pairs_samples_per_second": 15.705,
1048
+ "eval_trivia_pairs_steps_per_second": 0.491,
1049
+ "step": 3876
1050
+ },
1051
+ {
1052
+ "epoch": 0.7502903600464577,
1053
+ "eval_quora_pairs_loss": 0.21933521330356598,
1054
+ "eval_quora_pairs_runtime": 3.8953,
1055
+ "eval_quora_pairs_samples_per_second": 173.287,
1056
+ "eval_quora_pairs_steps_per_second": 5.648,
1057
+ "step": 3876
1058
+ },
1059
+ {
1060
+ "epoch": 0.7502903600464577,
1061
+ "eval_gooaq_pairs_loss": 0.5000605583190918,
1062
+ "eval_gooaq_pairs_runtime": 2.2758,
1063
+ "eval_gooaq_pairs_samples_per_second": 70.306,
1064
+ "eval_gooaq_pairs_steps_per_second": 2.197,
1065
+ "step": 3876
1066
+ },
1067
+ {
1068
+ "epoch": 0.7502903600464577,
1069
+ "eval_mrpc_pairs_loss": 0.044196613132953644,
1070
+ "eval_mrpc_pairs_runtime": 0.3742,
1071
+ "eval_mrpc_pairs_samples_per_second": 427.588,
1072
+ "eval_mrpc_pairs_steps_per_second": 13.362,
1073
+ "step": 3876
1074
+ },
1075
+ {
1076
+ "epoch": 0.7549361207897793,
1077
+ "grad_norm": 1.587725281715393,
1078
+ "learning_rate": 2.2630662020905924e-05,
1079
+ "loss": 0.4466,
1080
+ "step": 3900
1081
+ },
1082
+ {
1083
+ "epoch": 0.7675183894696089,
1084
+ "grad_norm": 9.404407501220703,
1085
+ "learning_rate": 2.3008130081300814e-05,
1086
+ "loss": 0.6293,
1087
+ "step": 3965
1088
+ },
1089
+ {
1090
+ "epoch": 0.7801006581494386,
1091
+ "grad_norm": 3.672903537750244,
1092
+ "learning_rate": 2.3385598141695703e-05,
1093
+ "loss": 0.5591,
1094
+ "step": 4030
1095
+ },
1096
+ {
1097
+ "epoch": 0.7926829268292683,
1098
+ "grad_norm": 1.3529752492904663,
1099
+ "learning_rate": 2.3763066202090596e-05,
1100
+ "loss": 0.5702,
1101
+ "step": 4095
1102
+ },
1103
+ {
1104
+ "epoch": 0.8052651955090979,
1105
+ "grad_norm": 21.519527435302734,
1106
+ "learning_rate": 2.414053426248548e-05,
1107
+ "loss": 0.4645,
1108
+ "step": 4160
1109
+ },
1110
+ {
1111
+ "epoch": 0.8178474641889276,
1112
+ "grad_norm": 5.5179362297058105,
1113
+ "learning_rate": 2.4518002322880374e-05,
1114
+ "loss": 0.5175,
1115
+ "step": 4225
1116
+ },
1117
+ {
1118
+ "epoch": 0.8304297328687572,
1119
+ "grad_norm": 10.127123832702637,
1120
+ "learning_rate": 2.489547038327526e-05,
1121
+ "loss": 0.5188,
1122
+ "step": 4290
1123
+ },
1124
+ {
1125
+ "epoch": 0.8430120015485869,
1126
+ "grad_norm": 5.518251895904541,
1127
+ "learning_rate": 2.5272938443670153e-05,
1128
+ "loss": 0.4908,
1129
+ "step": 4355
1130
+ },
1131
+ {
1132
+ "epoch": 0.8555942702284166,
1133
+ "grad_norm": 0.7371014356613159,
1134
+ "learning_rate": 2.5650406504065042e-05,
1135
+ "loss": 0.5075,
1136
+ "step": 4420
1137
+ },
1138
+ {
1139
+ "epoch": 0.8681765389082462,
1140
+ "grad_norm": 22.15941619873047,
1141
+ "learning_rate": 2.602206736353078e-05,
1142
+ "loss": 0.5625,
1143
+ "step": 4485
1144
+ },
1145
+ {
1146
+ "epoch": 0.8807588075880759,
1147
+ "grad_norm": 1.0873440504074097,
1148
+ "learning_rate": 2.6399535423925668e-05,
1149
+ "loss": 0.4952,
1150
+ "step": 4550
1151
+ },
1152
+ {
1153
+ "epoch": 0.8933410762679055,
1154
+ "grad_norm": 19.877098083496094,
1155
+ "learning_rate": 2.6777003484320557e-05,
1156
+ "loss": 0.5818,
1157
+ "step": 4615
1158
+ },
1159
+ {
1160
+ "epoch": 0.9059233449477352,
1161
+ "grad_norm": 9.559219360351562,
1162
+ "learning_rate": 2.715447154471545e-05,
1163
+ "loss": 0.6346,
1164
+ "step": 4680
1165
+ },
1166
+ {
1167
+ "epoch": 0.9185056136275649,
1168
+ "grad_norm": 2.896815299987793,
1169
+ "learning_rate": 2.7531939605110336e-05,
1170
+ "loss": 0.4971,
1171
+ "step": 4745
1172
+ },
1173
+ {
1174
+ "epoch": 0.9310878823073945,
1175
+ "grad_norm": 4.5045318603515625,
1176
+ "learning_rate": 2.7909407665505228e-05,
1177
+ "loss": 0.4834,
1178
+ "step": 4810
1179
+ },
1180
+ {
1181
+ "epoch": 0.9436701509872242,
1182
+ "grad_norm": 1.2259025573730469,
1183
+ "learning_rate": 2.8286875725900114e-05,
1184
+ "loss": 0.5992,
1185
+ "step": 4875
1186
+ },
1187
+ {
1188
+ "epoch": 0.9562524196670538,
1189
+ "grad_norm": 9.3356294631958,
1190
+ "learning_rate": 2.8664343786295007e-05,
1191
+ "loss": 0.5244,
1192
+ "step": 4940
1193
+ },
1194
+ {
1195
+ "epoch": 0.9688346883468835,
1196
+ "grad_norm": 2.221039056777954,
1197
+ "learning_rate": 2.9041811846689896e-05,
1198
+ "loss": 0.6051,
1199
+ "step": 5005
1200
+ },
1201
+ {
1202
+ "epoch": 0.9814169570267132,
1203
+ "grad_norm": 1.3200041055679321,
1204
+ "learning_rate": 2.9419279907084785e-05,
1205
+ "loss": 0.4323,
1206
+ "step": 5070
1207
+ },
1208
+ {
1209
+ "epoch": 0.9939992257065428,
1210
+ "grad_norm": 8.689266204833984,
1211
+ "learning_rate": 2.9796747967479674e-05,
1212
+ "loss": 0.5347,
1213
+ "step": 5135
1214
+ }
1215
+ ],
1216
+ "logging_steps": 65,
1217
+ "max_steps": 25830,
1218
+ "num_input_tokens_seen": 0,
1219
+ "num_train_epochs": 5,
1220
+ "save_steps": 2583,
1221
+ "stateful_callbacks": {
1222
+ "TrainerControl": {
1223
+ "args": {
1224
+ "should_epoch_stop": false,
1225
+ "should_evaluate": false,
1226
+ "should_log": false,
1227
+ "should_save": true,
1228
+ "should_training_stop": false
1229
+ },
1230
+ "attributes": {}
1231
+ }
1232
+ },
1233
+ "total_flos": 0.0,
1234
+ "train_batch_size": 32,
1235
+ "trial_name": null,
1236
+ "trial_params": null
1237
+ }
checkpoint-5166/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2519118bce6ce43cd8ca9ca2ad2ad4642dfcfe5f9a3c6dd2eeb5a81e8223693e
3
+ size 5624