Joemgu commited on
Commit
83b6b43
1 Parent(s): 63f42fc

Training in progress, step 300

Browse files
checkpoint-300/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/long-t5-tglobal-base",
3
+ "activation_dropout": 0.1,
4
+ "architectures": [
5
+ "LongT5ForConditionalGeneration"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "d_ff": 2048,
9
+ "d_kv": 64,
10
+ "d_model": 768,
11
+ "decoder_start_token_id": 0,
12
+ "dense_act_fn": "gelu_new",
13
+ "dropout": 0.1,
14
+ "dropout_rate": 0.1,
15
+ "encoder_attention_type": "transient-global",
16
+ "eos_token_id": 1,
17
+ "feed_forward_proj": "gated-gelu",
18
+ "global_block_size": 16,
19
+ "initializer_factor": 1.0,
20
+ "is_encoder_decoder": true,
21
+ "is_gated_act": true,
22
+ "layer_norm_epsilon": 1e-06,
23
+ "local_radius": 127,
24
+ "model_type": "longt5",
25
+ "n_positions": 4096,
26
+ "num_decoder_layers": 12,
27
+ "num_heads": 12,
28
+ "num_layers": 12,
29
+ "output_past": true,
30
+ "pad_token_id": 0,
31
+ "relative_attention_max_distance": 128,
32
+ "relative_attention_num_buckets": 32,
33
+ "tie_word_embeddings": false,
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.29.2",
36
+ "use_cache": false,
37
+ "vocab_size": 32128
38
+ }
checkpoint-300/generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 0,
4
+ "early_stopping": true,
5
+ "eos_token_id": 1,
6
+ "forced_eos_token_id": 1,
7
+ "max_length": 4096,
8
+ "pad_token_id": 0,
9
+ "transformers_version": "4.29.2"
10
+ }
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5475aafdab07f3220f03762f55c01401fd8d4fab415476a86c07a9e663c9e88
3
+ size 990512709
checkpoint-300/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00baa5820c6217c07a8cdbb56201f21a75a525e4fc15017519d42658135b0442
3
+ size 990452905
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:060622b8c626ccc68cc3de0404e9c20a9052e8224c4033c76f5afdc865a43bb4
3
+ size 14575
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fe71ead15f0032f66ab11812159b596d6c3bbb68b736d72f6959e9e316bfabc
3
+ size 627
checkpoint-300/special_tokens_map.json ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "eos_token": "</s>",
105
+ "pad_token": "<pad>",
106
+ "unk_token": "<unk>"
107
+ }
checkpoint-300/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
checkpoint-300/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-300/tokenizer_config.json ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "clean_up_tokenization_spaces": true,
105
+ "eos_token": "</s>",
106
+ "extra_ids": 100,
107
+ "model_max_length": 1000000000000000019884624838656,
108
+ "pad_token": "<pad>",
109
+ "tokenizer_class": "T5Tokenizer",
110
+ "unk_token": "<unk>"
111
+ }
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,1816 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.03,
5
+ "global_step": 300,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 1e-05,
13
+ "loss": 3.6346,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 2e-05,
19
+ "loss": 3.602,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 3e-05,
25
+ "loss": 3.5741,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.0,
30
+ "learning_rate": 4e-05,
31
+ "loss": 3.4904,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.0,
36
+ "learning_rate": 5e-05,
37
+ "loss": 3.4215,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.0,
42
+ "learning_rate": 6e-05,
43
+ "loss": 3.4568,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.0,
48
+ "learning_rate": 7.000000000000001e-05,
49
+ "loss": 3.3978,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.0,
54
+ "learning_rate": 8e-05,
55
+ "loss": 3.4991,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.0,
60
+ "learning_rate": 8.999999999999999e-05,
61
+ "loss": 3.5852,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.0,
66
+ "learning_rate": 0.0001,
67
+ "loss": 3.3132,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.0,
72
+ "learning_rate": 0.00011,
73
+ "loss": 3.2737,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.0,
78
+ "learning_rate": 0.00012,
79
+ "loss": 3.1399,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.0,
84
+ "learning_rate": 0.00013000000000000002,
85
+ "loss": 3.1828,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.0,
90
+ "learning_rate": 0.00014000000000000001,
91
+ "loss": 3.0638,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.0,
96
+ "learning_rate": 0.00015,
97
+ "loss": 3.185,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.0,
102
+ "learning_rate": 0.00016,
103
+ "loss": 3.0669,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.0,
108
+ "learning_rate": 0.00017,
109
+ "loss": 2.9026,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.0,
114
+ "learning_rate": 0.00017999999999999998,
115
+ "loss": 2.9591,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.0,
120
+ "learning_rate": 0.00019,
121
+ "loss": 2.8483,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.0,
126
+ "learning_rate": 0.0002,
127
+ "loss": 2.7707,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.0,
132
+ "learning_rate": 0.00021,
133
+ "loss": 2.8564,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 0.0,
138
+ "learning_rate": 0.00022,
139
+ "loss": 2.6997,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 0.0,
144
+ "learning_rate": 0.00023,
145
+ "loss": 2.8052,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 0.0,
150
+ "learning_rate": 0.00024,
151
+ "loss": 2.5826,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 0.0,
156
+ "learning_rate": 0.00025,
157
+ "loss": 2.6266,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 0.0,
162
+ "learning_rate": 0.00026000000000000003,
163
+ "loss": 2.6477,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 0.0,
168
+ "learning_rate": 0.00027,
169
+ "loss": 2.6133,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 0.0,
174
+ "learning_rate": 0.00028000000000000003,
175
+ "loss": 2.5395,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 0.0,
180
+ "learning_rate": 0.00029,
181
+ "loss": 2.664,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 0.0,
186
+ "learning_rate": 0.0003,
187
+ "loss": 2.5568,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 0.0,
192
+ "learning_rate": 0.00031,
193
+ "loss": 2.4346,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 0.0,
198
+ "learning_rate": 0.00032,
199
+ "loss": 2.5765,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 0.0,
204
+ "learning_rate": 0.00033,
205
+ "loss": 2.4884,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 0.0,
210
+ "learning_rate": 0.00034,
211
+ "loss": 2.4257,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 0.0,
216
+ "learning_rate": 0.00035,
217
+ "loss": 2.4281,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 0.0,
222
+ "learning_rate": 0.00035999999999999997,
223
+ "loss": 2.4858,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 0.0,
228
+ "learning_rate": 0.00037,
229
+ "loss": 2.383,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 0.0,
234
+ "learning_rate": 0.00038,
235
+ "loss": 2.4865,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 0.0,
240
+ "learning_rate": 0.00039000000000000005,
241
+ "loss": 2.3564,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 0.0,
246
+ "learning_rate": 0.0004,
247
+ "loss": 2.4276,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 0.0,
252
+ "learning_rate": 0.00041,
253
+ "loss": 2.4225,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 0.0,
258
+ "learning_rate": 0.00042,
259
+ "loss": 2.2718,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 0.0,
264
+ "learning_rate": 0.00043,
265
+ "loss": 2.3566,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 0.0,
270
+ "learning_rate": 0.00044,
271
+ "loss": 2.3911,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 0.0,
276
+ "learning_rate": 0.00045000000000000004,
277
+ "loss": 2.3999,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 0.0,
282
+ "learning_rate": 0.00046,
283
+ "loss": 2.2776,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 0.0,
288
+ "learning_rate": 0.00047,
289
+ "loss": 2.3891,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 0.0,
294
+ "learning_rate": 0.00048,
295
+ "loss": 2.3132,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 0.0,
300
+ "learning_rate": 0.00049,
301
+ "loss": 2.3353,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 0.01,
306
+ "learning_rate": 0.0005,
307
+ "loss": 2.3691,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 0.01,
312
+ "learning_rate": 0.00051,
313
+ "loss": 2.3271,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 0.01,
318
+ "learning_rate": 0.0005200000000000001,
319
+ "loss": 2.2881,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 0.01,
324
+ "learning_rate": 0.0005300000000000001,
325
+ "loss": 2.2731,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 0.01,
330
+ "learning_rate": 0.00054,
331
+ "loss": 2.3186,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 0.01,
336
+ "learning_rate": 0.00055,
337
+ "loss": 2.3216,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 0.01,
342
+ "learning_rate": 0.0005600000000000001,
343
+ "loss": 2.3204,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 0.01,
348
+ "learning_rate": 0.00057,
349
+ "loss": 2.1832,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 0.01,
354
+ "learning_rate": 0.00058,
355
+ "loss": 2.4125,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 0.01,
360
+ "learning_rate": 0.00059,
361
+ "loss": 2.2649,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 0.01,
366
+ "learning_rate": 0.0006,
367
+ "loss": 2.1589,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 0.01,
372
+ "learning_rate": 0.00061,
373
+ "loss": 2.1431,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 0.01,
378
+ "learning_rate": 0.00062,
379
+ "loss": 2.3738,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 0.01,
384
+ "learning_rate": 0.00063,
385
+ "loss": 2.2525,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 0.01,
390
+ "learning_rate": 0.00064,
391
+ "loss": 2.2778,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 0.01,
396
+ "learning_rate": 0.0006500000000000001,
397
+ "loss": 2.2739,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 0.01,
402
+ "learning_rate": 0.00066,
403
+ "loss": 2.2878,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 0.01,
408
+ "learning_rate": 0.00067,
409
+ "loss": 2.196,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 0.01,
414
+ "learning_rate": 0.00068,
415
+ "loss": 2.2384,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 0.01,
420
+ "learning_rate": 0.00069,
421
+ "loss": 2.1638,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 0.01,
426
+ "learning_rate": 0.0007,
427
+ "loss": 2.1578,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 0.01,
432
+ "learning_rate": 0.00071,
433
+ "loss": 2.1781,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 0.01,
438
+ "learning_rate": 0.0007199999999999999,
439
+ "loss": 2.2486,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 0.01,
444
+ "learning_rate": 0.00073,
445
+ "loss": 2.2475,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 0.01,
450
+ "learning_rate": 0.00074,
451
+ "loss": 2.0713,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 0.01,
456
+ "learning_rate": 0.00075,
457
+ "loss": 2.1261,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 0.01,
462
+ "learning_rate": 0.00076,
463
+ "loss": 2.156,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 0.01,
468
+ "learning_rate": 0.0007700000000000001,
469
+ "loss": 2.2273,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 0.01,
474
+ "learning_rate": 0.0007800000000000001,
475
+ "loss": 2.2245,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 0.01,
480
+ "learning_rate": 0.00079,
481
+ "loss": 2.1527,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 0.01,
486
+ "learning_rate": 0.0008,
487
+ "loss": 2.0454,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 0.01,
492
+ "learning_rate": 0.0008100000000000001,
493
+ "loss": 2.2643,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 0.01,
498
+ "learning_rate": 0.00082,
499
+ "loss": 2.2447,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 0.01,
504
+ "learning_rate": 0.00083,
505
+ "loss": 2.1656,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 0.01,
510
+ "learning_rate": 0.00084,
511
+ "loss": 2.099,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 0.01,
516
+ "learning_rate": 0.00085,
517
+ "loss": 2.2078,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 0.01,
522
+ "learning_rate": 0.00086,
523
+ "loss": 2.1537,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 0.01,
528
+ "learning_rate": 0.00087,
529
+ "loss": 2.1013,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 0.01,
534
+ "learning_rate": 0.00088,
535
+ "loss": 2.0553,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 0.01,
540
+ "learning_rate": 0.0008900000000000001,
541
+ "loss": 2.1702,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 0.01,
546
+ "learning_rate": 0.0009000000000000001,
547
+ "loss": 2.0685,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 0.01,
552
+ "learning_rate": 0.00091,
553
+ "loss": 2.149,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 0.01,
558
+ "learning_rate": 0.00092,
559
+ "loss": 2.1102,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 0.01,
564
+ "learning_rate": 0.00093,
565
+ "loss": 2.2527,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 0.01,
570
+ "learning_rate": 0.00094,
571
+ "loss": 2.0719,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 0.01,
576
+ "learning_rate": 0.00095,
577
+ "loss": 2.1293,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 0.01,
582
+ "learning_rate": 0.00096,
583
+ "loss": 2.1333,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 0.01,
588
+ "learning_rate": 0.0009699999999999999,
589
+ "loss": 2.1497,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 0.01,
594
+ "learning_rate": 0.00098,
595
+ "loss": 2.0996,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 0.01,
600
+ "learning_rate": 0.00099,
601
+ "loss": 2.1576,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 0.01,
606
+ "learning_rate": 0.001,
607
+ "loss": 2.1483,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 0.01,
612
+ "learning_rate": 0.00099989898989899,
613
+ "loss": 2.1328,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 0.01,
618
+ "learning_rate": 0.00099979797979798,
619
+ "loss": 2.0475,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 0.01,
624
+ "learning_rate": 0.0009996969696969699,
625
+ "loss": 2.2059,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 0.01,
630
+ "learning_rate": 0.0009995959595959596,
631
+ "loss": 2.0895,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 0.01,
636
+ "learning_rate": 0.0009994949494949495,
637
+ "loss": 2.1077,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 0.01,
642
+ "learning_rate": 0.0009993939393939395,
643
+ "loss": 2.0621,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 0.01,
648
+ "learning_rate": 0.0009992929292929294,
649
+ "loss": 2.177,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 0.01,
654
+ "learning_rate": 0.0009991919191919192,
655
+ "loss": 2.2142,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 0.01,
660
+ "learning_rate": 0.0009990909090909091,
661
+ "loss": 2.0881,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 0.01,
666
+ "learning_rate": 0.000998989898989899,
667
+ "loss": 2.1707,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 0.01,
672
+ "learning_rate": 0.000998888888888889,
673
+ "loss": 2.1403,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 0.01,
678
+ "learning_rate": 0.0009987878787878787,
679
+ "loss": 2.1171,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 0.01,
684
+ "learning_rate": 0.0009986868686868687,
685
+ "loss": 2.047,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 0.01,
690
+ "learning_rate": 0.0009985858585858586,
691
+ "loss": 2.0712,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 0.01,
696
+ "learning_rate": 0.0009984848484848486,
697
+ "loss": 2.1068,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 0.01,
702
+ "learning_rate": 0.0009983838383838383,
703
+ "loss": 2.1159,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 0.01,
708
+ "learning_rate": 0.0009982828282828283,
709
+ "loss": 2.0403,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 0.01,
714
+ "learning_rate": 0.0009981818181818182,
715
+ "loss": 2.0022,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 0.01,
720
+ "learning_rate": 0.0009980808080808081,
721
+ "loss": 2.0025,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 0.01,
726
+ "learning_rate": 0.000997979797979798,
727
+ "loss": 2.1265,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 0.01,
732
+ "learning_rate": 0.000997878787878788,
733
+ "loss": 2.1159,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 0.01,
738
+ "learning_rate": 0.0009977777777777778,
739
+ "loss": 1.958,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 0.01,
744
+ "learning_rate": 0.0009976767676767677,
745
+ "loss": 2.0857,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 0.01,
750
+ "learning_rate": 0.0009975757575757577,
751
+ "loss": 1.9986,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 0.01,
756
+ "learning_rate": 0.0009974747474747476,
757
+ "loss": 2.1304,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 0.01,
762
+ "learning_rate": 0.0009973737373737373,
763
+ "loss": 2.184,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 0.01,
768
+ "learning_rate": 0.0009972727272727273,
769
+ "loss": 2.0053,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 0.01,
774
+ "learning_rate": 0.0009971717171717172,
775
+ "loss": 2.028,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 0.01,
780
+ "learning_rate": 0.0009970707070707072,
781
+ "loss": 2.0992,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 0.01,
786
+ "learning_rate": 0.000996969696969697,
787
+ "loss": 2.0326,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 0.01,
792
+ "learning_rate": 0.0009968686868686869,
793
+ "loss": 2.0328,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 0.01,
798
+ "learning_rate": 0.0009967676767676768,
799
+ "loss": 2.0563,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 0.01,
804
+ "learning_rate": 0.0009966666666666668,
805
+ "loss": 2.178,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 0.01,
810
+ "learning_rate": 0.0009965656565656565,
811
+ "loss": 2.1038,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 0.01,
816
+ "learning_rate": 0.0009964646464646464,
817
+ "loss": 2.096,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 0.01,
822
+ "learning_rate": 0.0009963636363636364,
823
+ "loss": 2.1732,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 0.01,
828
+ "learning_rate": 0.0009962626262626263,
829
+ "loss": 2.1267,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 0.01,
834
+ "learning_rate": 0.000996161616161616,
835
+ "loss": 2.0657,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 0.01,
840
+ "learning_rate": 0.0009960606060606062,
841
+ "loss": 2.1286,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 0.01,
846
+ "learning_rate": 0.000995959595959596,
847
+ "loss": 2.0452,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 0.01,
852
+ "learning_rate": 0.000995858585858586,
853
+ "loss": 2.0499,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 0.01,
858
+ "learning_rate": 0.0009957575757575758,
859
+ "loss": 2.1283,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 0.01,
864
+ "learning_rate": 0.0009956565656565658,
865
+ "loss": 2.1229,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 0.01,
870
+ "learning_rate": 0.0009955555555555555,
871
+ "loss": 2.1071,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 0.01,
876
+ "learning_rate": 0.0009954545454545455,
877
+ "loss": 2.1224,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 0.01,
882
+ "learning_rate": 0.0009953535353535354,
883
+ "loss": 1.9851,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 0.01,
888
+ "learning_rate": 0.0009952525252525254,
889
+ "loss": 2.0053,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 0.01,
894
+ "learning_rate": 0.000995151515151515,
895
+ "loss": 2.113,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 0.01,
900
+ "learning_rate": 0.000995050505050505,
901
+ "loss": 2.0945,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 0.01,
906
+ "learning_rate": 0.000994949494949495,
907
+ "loss": 2.0622,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 0.02,
912
+ "learning_rate": 0.000994848484848485,
913
+ "loss": 2.0936,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 0.02,
918
+ "learning_rate": 0.0009947474747474747,
919
+ "loss": 2.1155,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 0.02,
924
+ "learning_rate": 0.0009946464646464646,
925
+ "loss": 2.1981,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 0.02,
930
+ "learning_rate": 0.0009945454545454546,
931
+ "loss": 2.0479,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 0.02,
936
+ "learning_rate": 0.0009944444444444445,
937
+ "loss": 2.1307,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 0.02,
942
+ "learning_rate": 0.0009943434343434342,
943
+ "loss": 1.8631,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 0.02,
948
+ "learning_rate": 0.0009942424242424242,
949
+ "loss": 2.091,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 0.02,
954
+ "learning_rate": 0.0009941414141414141,
955
+ "loss": 2.1344,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 0.02,
960
+ "learning_rate": 0.000994040404040404,
961
+ "loss": 2.0833,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 0.02,
966
+ "learning_rate": 0.000993939393939394,
967
+ "loss": 1.9934,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 0.02,
972
+ "learning_rate": 0.000993838383838384,
973
+ "loss": 1.9833,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 0.02,
978
+ "learning_rate": 0.0009937373737373737,
979
+ "loss": 1.986,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 0.02,
984
+ "learning_rate": 0.0009936363636363636,
985
+ "loss": 2.0843,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 0.02,
990
+ "learning_rate": 0.0009935353535353536,
991
+ "loss": 2.0206,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 0.02,
996
+ "learning_rate": 0.0009934343434343435,
997
+ "loss": 2.0856,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 0.02,
1002
+ "learning_rate": 0.0009933333333333333,
1003
+ "loss": 2.0136,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 0.02,
1008
+ "learning_rate": 0.0009932323232323232,
1009
+ "loss": 1.9881,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 0.02,
1014
+ "learning_rate": 0.0009931313131313132,
1015
+ "loss": 2.1644,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 0.02,
1020
+ "learning_rate": 0.000993030303030303,
1021
+ "loss": 2.09,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 0.02,
1026
+ "learning_rate": 0.0009929292929292928,
1027
+ "loss": 2.1076,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 0.02,
1032
+ "learning_rate": 0.0009928282828282828,
1033
+ "loss": 1.9709,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 0.02,
1038
+ "learning_rate": 0.0009927272727272727,
1039
+ "loss": 2.0471,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 0.02,
1044
+ "learning_rate": 0.0009926262626262627,
1045
+ "loss": 2.0856,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 0.02,
1050
+ "learning_rate": 0.0009925252525252524,
1051
+ "loss": 2.0028,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 0.02,
1056
+ "learning_rate": 0.0009924242424242424,
1057
+ "loss": 2.0308,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 0.02,
1062
+ "learning_rate": 0.0009923232323232323,
1063
+ "loss": 2.0572,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 0.02,
1068
+ "learning_rate": 0.0009922222222222222,
1069
+ "loss": 2.0539,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 0.02,
1074
+ "learning_rate": 0.0009921212121212122,
1075
+ "loss": 2.0185,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 0.02,
1080
+ "learning_rate": 0.0009920202020202021,
1081
+ "loss": 2.0536,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 0.02,
1086
+ "learning_rate": 0.0009919191919191919,
1087
+ "loss": 2.0024,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 0.02,
1092
+ "learning_rate": 0.0009918181818181818,
1093
+ "loss": 2.1277,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 0.02,
1098
+ "learning_rate": 0.0009917171717171718,
1099
+ "loss": 2.0297,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 0.02,
1104
+ "learning_rate": 0.0009916161616161617,
1105
+ "loss": 2.0917,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 0.02,
1110
+ "learning_rate": 0.0009915151515151514,
1111
+ "loss": 2.1006,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 0.02,
1116
+ "learning_rate": 0.0009914141414141414,
1117
+ "loss": 2.0455,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 0.02,
1122
+ "learning_rate": 0.0009913131313131313,
1123
+ "loss": 2.0974,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 0.02,
1128
+ "learning_rate": 0.0009912121212121213,
1129
+ "loss": 2.0821,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 0.02,
1134
+ "learning_rate": 0.0009911111111111112,
1135
+ "loss": 2.0931,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 0.02,
1140
+ "learning_rate": 0.000991010101010101,
1141
+ "loss": 2.0317,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 0.02,
1146
+ "learning_rate": 0.000990909090909091,
1147
+ "loss": 1.9059,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 0.02,
1152
+ "learning_rate": 0.0009908080808080809,
1153
+ "loss": 1.9258,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 0.02,
1158
+ "learning_rate": 0.0009907070707070708,
1159
+ "loss": 1.9566,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 0.02,
1164
+ "learning_rate": 0.0009906060606060605,
1165
+ "loss": 2.0572,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 0.02,
1170
+ "learning_rate": 0.0009905050505050505,
1171
+ "loss": 2.0917,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 0.02,
1176
+ "learning_rate": 0.0009904040404040404,
1177
+ "loss": 2.0177,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 0.02,
1182
+ "learning_rate": 0.0009903030303030304,
1183
+ "loss": 1.9265,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 0.02,
1188
+ "learning_rate": 0.0009902020202020203,
1189
+ "loss": 2.1021,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 0.02,
1194
+ "learning_rate": 0.0009901010101010103,
1195
+ "loss": 2.1158,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 0.02,
1200
+ "learning_rate": 0.00099,
1201
+ "loss": 1.9843,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 0.02,
1206
+ "learning_rate": 0.00098989898989899,
1207
+ "loss": 2.0997,
1208
+ "step": 200
1209
+ },
1210
+ {
1211
+ "epoch": 0.02,
1212
+ "learning_rate": 0.0009897979797979799,
1213
+ "loss": 2.0933,
1214
+ "step": 201
1215
+ },
1216
+ {
1217
+ "epoch": 0.02,
1218
+ "learning_rate": 0.0009896969696969698,
1219
+ "loss": 2.0834,
1220
+ "step": 202
1221
+ },
1222
+ {
1223
+ "epoch": 0.02,
1224
+ "learning_rate": 0.0009895959595959596,
1225
+ "loss": 2.1167,
1226
+ "step": 203
1227
+ },
1228
+ {
1229
+ "epoch": 0.02,
1230
+ "learning_rate": 0.0009894949494949495,
1231
+ "loss": 2.0517,
1232
+ "step": 204
1233
+ },
1234
+ {
1235
+ "epoch": 0.02,
1236
+ "learning_rate": 0.0009893939393939395,
1237
+ "loss": 1.9413,
1238
+ "step": 205
1239
+ },
1240
+ {
1241
+ "epoch": 0.02,
1242
+ "learning_rate": 0.0009892929292929294,
1243
+ "loss": 1.891,
1244
+ "step": 206
1245
+ },
1246
+ {
1247
+ "epoch": 0.02,
1248
+ "learning_rate": 0.0009891919191919191,
1249
+ "loss": 1.9382,
1250
+ "step": 207
1251
+ },
1252
+ {
1253
+ "epoch": 0.02,
1254
+ "learning_rate": 0.000989090909090909,
1255
+ "loss": 1.9736,
1256
+ "step": 208
1257
+ },
1258
+ {
1259
+ "epoch": 0.02,
1260
+ "learning_rate": 0.000988989898989899,
1261
+ "loss": 2.0332,
1262
+ "step": 209
1263
+ },
1264
+ {
1265
+ "epoch": 0.02,
1266
+ "learning_rate": 0.000988888888888889,
1267
+ "loss": 2.0221,
1268
+ "step": 210
1269
+ },
1270
+ {
1271
+ "epoch": 0.02,
1272
+ "learning_rate": 0.0009887878787878787,
1273
+ "loss": 2.0825,
1274
+ "step": 211
1275
+ },
1276
+ {
1277
+ "epoch": 0.02,
1278
+ "learning_rate": 0.0009886868686868687,
1279
+ "loss": 1.9693,
1280
+ "step": 212
1281
+ },
1282
+ {
1283
+ "epoch": 0.02,
1284
+ "learning_rate": 0.0009885858585858586,
1285
+ "loss": 1.9989,
1286
+ "step": 213
1287
+ },
1288
+ {
1289
+ "epoch": 0.02,
1290
+ "learning_rate": 0.0009884848484848485,
1291
+ "loss": 1.9897,
1292
+ "step": 214
1293
+ },
1294
+ {
1295
+ "epoch": 0.02,
1296
+ "learning_rate": 0.0009883838383838383,
1297
+ "loss": 2.0722,
1298
+ "step": 215
1299
+ },
1300
+ {
1301
+ "epoch": 0.02,
1302
+ "learning_rate": 0.0009882828282828284,
1303
+ "loss": 1.9653,
1304
+ "step": 216
1305
+ },
1306
+ {
1307
+ "epoch": 0.02,
1308
+ "learning_rate": 0.0009881818181818182,
1309
+ "loss": 2.0861,
1310
+ "step": 217
1311
+ },
1312
+ {
1313
+ "epoch": 0.02,
1314
+ "learning_rate": 0.0009880808080808081,
1315
+ "loss": 1.8935,
1316
+ "step": 218
1317
+ },
1318
+ {
1319
+ "epoch": 0.02,
1320
+ "learning_rate": 0.000987979797979798,
1321
+ "loss": 2.1295,
1322
+ "step": 219
1323
+ },
1324
+ {
1325
+ "epoch": 0.02,
1326
+ "learning_rate": 0.000987878787878788,
1327
+ "loss": 2.0555,
1328
+ "step": 220
1329
+ },
1330
+ {
1331
+ "epoch": 0.02,
1332
+ "learning_rate": 0.0009877777777777777,
1333
+ "loss": 2.0813,
1334
+ "step": 221
1335
+ },
1336
+ {
1337
+ "epoch": 0.02,
1338
+ "learning_rate": 0.0009876767676767677,
1339
+ "loss": 2.0426,
1340
+ "step": 222
1341
+ },
1342
+ {
1343
+ "epoch": 0.02,
1344
+ "learning_rate": 0.0009875757575757576,
1345
+ "loss": 1.9818,
1346
+ "step": 223
1347
+ },
1348
+ {
1349
+ "epoch": 0.02,
1350
+ "learning_rate": 0.0009874747474747476,
1351
+ "loss": 1.9744,
1352
+ "step": 224
1353
+ },
1354
+ {
1355
+ "epoch": 0.02,
1356
+ "learning_rate": 0.0009873737373737373,
1357
+ "loss": 1.9681,
1358
+ "step": 225
1359
+ },
1360
+ {
1361
+ "epoch": 0.02,
1362
+ "learning_rate": 0.0009872727272727273,
1363
+ "loss": 2.0538,
1364
+ "step": 226
1365
+ },
1366
+ {
1367
+ "epoch": 0.02,
1368
+ "learning_rate": 0.0009871717171717172,
1369
+ "loss": 2.0493,
1370
+ "step": 227
1371
+ },
1372
+ {
1373
+ "epoch": 0.02,
1374
+ "learning_rate": 0.0009870707070707072,
1375
+ "loss": 1.9484,
1376
+ "step": 228
1377
+ },
1378
+ {
1379
+ "epoch": 0.02,
1380
+ "learning_rate": 0.0009869696969696969,
1381
+ "loss": 2.0947,
1382
+ "step": 229
1383
+ },
1384
+ {
1385
+ "epoch": 0.02,
1386
+ "learning_rate": 0.0009868686868686868,
1387
+ "loss": 2.0193,
1388
+ "step": 230
1389
+ },
1390
+ {
1391
+ "epoch": 0.02,
1392
+ "learning_rate": 0.0009867676767676768,
1393
+ "loss": 2.0365,
1394
+ "step": 231
1395
+ },
1396
+ {
1397
+ "epoch": 0.02,
1398
+ "learning_rate": 0.0009866666666666667,
1399
+ "loss": 2.0624,
1400
+ "step": 232
1401
+ },
1402
+ {
1403
+ "epoch": 0.02,
1404
+ "learning_rate": 0.0009865656565656565,
1405
+ "loss": 1.9721,
1406
+ "step": 233
1407
+ },
1408
+ {
1409
+ "epoch": 0.02,
1410
+ "learning_rate": 0.0009864646464646464,
1411
+ "loss": 1.9658,
1412
+ "step": 234
1413
+ },
1414
+ {
1415
+ "epoch": 0.02,
1416
+ "learning_rate": 0.0009863636363636363,
1417
+ "loss": 2.0441,
1418
+ "step": 235
1419
+ },
1420
+ {
1421
+ "epoch": 0.02,
1422
+ "learning_rate": 0.0009862626262626263,
1423
+ "loss": 1.929,
1424
+ "step": 236
1425
+ },
1426
+ {
1427
+ "epoch": 0.02,
1428
+ "learning_rate": 0.0009861616161616162,
1429
+ "loss": 2.0288,
1430
+ "step": 237
1431
+ },
1432
+ {
1433
+ "epoch": 0.02,
1434
+ "learning_rate": 0.0009860606060606062,
1435
+ "loss": 1.9462,
1436
+ "step": 238
1437
+ },
1438
+ {
1439
+ "epoch": 0.02,
1440
+ "learning_rate": 0.000985959595959596,
1441
+ "loss": 1.9983,
1442
+ "step": 239
1443
+ },
1444
+ {
1445
+ "epoch": 0.02,
1446
+ "learning_rate": 0.0009858585858585859,
1447
+ "loss": 1.9697,
1448
+ "step": 240
1449
+ },
1450
+ {
1451
+ "epoch": 0.02,
1452
+ "learning_rate": 0.0009857575757575758,
1453
+ "loss": 2.035,
1454
+ "step": 241
1455
+ },
1456
+ {
1457
+ "epoch": 0.02,
1458
+ "learning_rate": 0.0009856565656565658,
1459
+ "loss": 2.0163,
1460
+ "step": 242
1461
+ },
1462
+ {
1463
+ "epoch": 0.02,
1464
+ "learning_rate": 0.0009855555555555555,
1465
+ "loss": 2.0443,
1466
+ "step": 243
1467
+ },
1468
+ {
1469
+ "epoch": 0.02,
1470
+ "learning_rate": 0.0009854545454545454,
1471
+ "loss": 1.9712,
1472
+ "step": 244
1473
+ },
1474
+ {
1475
+ "epoch": 0.02,
1476
+ "learning_rate": 0.0009853535353535354,
1477
+ "loss": 1.979,
1478
+ "step": 245
1479
+ },
1480
+ {
1481
+ "epoch": 0.02,
1482
+ "learning_rate": 0.0009852525252525253,
1483
+ "loss": 1.9346,
1484
+ "step": 246
1485
+ },
1486
+ {
1487
+ "epoch": 0.02,
1488
+ "learning_rate": 0.000985151515151515,
1489
+ "loss": 2.0026,
1490
+ "step": 247
1491
+ },
1492
+ {
1493
+ "epoch": 0.02,
1494
+ "learning_rate": 0.000985050505050505,
1495
+ "loss": 1.9729,
1496
+ "step": 248
1497
+ },
1498
+ {
1499
+ "epoch": 0.02,
1500
+ "learning_rate": 0.000984949494949495,
1501
+ "loss": 1.8462,
1502
+ "step": 249
1503
+ },
1504
+ {
1505
+ "epoch": 0.03,
1506
+ "learning_rate": 0.000984848484848485,
1507
+ "loss": 1.9336,
1508
+ "step": 250
1509
+ },
1510
+ {
1511
+ "epoch": 0.03,
1512
+ "learning_rate": 0.0009847474747474746,
1513
+ "loss": 1.9938,
1514
+ "step": 251
1515
+ },
1516
+ {
1517
+ "epoch": 0.03,
1518
+ "learning_rate": 0.0009846464646464646,
1519
+ "loss": 1.9834,
1520
+ "step": 252
1521
+ },
1522
+ {
1523
+ "epoch": 0.03,
1524
+ "learning_rate": 0.0009845454545454545,
1525
+ "loss": 1.9892,
1526
+ "step": 253
1527
+ },
1528
+ {
1529
+ "epoch": 0.03,
1530
+ "learning_rate": 0.0009844444444444445,
1531
+ "loss": 1.9013,
1532
+ "step": 254
1533
+ },
1534
+ {
1535
+ "epoch": 0.03,
1536
+ "learning_rate": 0.0009843434343434344,
1537
+ "loss": 2.0017,
1538
+ "step": 255
1539
+ },
1540
+ {
1541
+ "epoch": 0.03,
1542
+ "learning_rate": 0.0009842424242424244,
1543
+ "loss": 1.9771,
1544
+ "step": 256
1545
+ },
1546
+ {
1547
+ "epoch": 0.03,
1548
+ "learning_rate": 0.000984141414141414,
1549
+ "loss": 2.0135,
1550
+ "step": 257
1551
+ },
1552
+ {
1553
+ "epoch": 0.03,
1554
+ "learning_rate": 0.000984040404040404,
1555
+ "loss": 1.9034,
1556
+ "step": 258
1557
+ },
1558
+ {
1559
+ "epoch": 0.03,
1560
+ "learning_rate": 0.000983939393939394,
1561
+ "loss": 1.9586,
1562
+ "step": 259
1563
+ },
1564
+ {
1565
+ "epoch": 0.03,
1566
+ "learning_rate": 0.000983838383838384,
1567
+ "loss": 1.9966,
1568
+ "step": 260
1569
+ },
1570
+ {
1571
+ "epoch": 0.03,
1572
+ "learning_rate": 0.0009837373737373737,
1573
+ "loss": 1.9742,
1574
+ "step": 261
1575
+ },
1576
+ {
1577
+ "epoch": 0.03,
1578
+ "learning_rate": 0.0009836363636363636,
1579
+ "loss": 1.9376,
1580
+ "step": 262
1581
+ },
1582
+ {
1583
+ "epoch": 0.03,
1584
+ "learning_rate": 0.0009835353535353536,
1585
+ "loss": 1.9006,
1586
+ "step": 263
1587
+ },
1588
+ {
1589
+ "epoch": 0.03,
1590
+ "learning_rate": 0.0009834343434343435,
1591
+ "loss": 2.028,
1592
+ "step": 264
1593
+ },
1594
+ {
1595
+ "epoch": 0.03,
1596
+ "learning_rate": 0.0009833333333333332,
1597
+ "loss": 2.0231,
1598
+ "step": 265
1599
+ },
1600
+ {
1601
+ "epoch": 0.03,
1602
+ "learning_rate": 0.0009832323232323232,
1603
+ "loss": 2.0239,
1604
+ "step": 266
1605
+ },
1606
+ {
1607
+ "epoch": 0.03,
1608
+ "learning_rate": 0.0009831313131313131,
1609
+ "loss": 1.9672,
1610
+ "step": 267
1611
+ },
1612
+ {
1613
+ "epoch": 0.03,
1614
+ "learning_rate": 0.000983030303030303,
1615
+ "loss": 2.098,
1616
+ "step": 268
1617
+ },
1618
+ {
1619
+ "epoch": 0.03,
1620
+ "learning_rate": 0.0009829292929292928,
1621
+ "loss": 2.0253,
1622
+ "step": 269
1623
+ },
1624
+ {
1625
+ "epoch": 0.03,
1626
+ "learning_rate": 0.0009828282828282828,
1627
+ "loss": 1.8544,
1628
+ "step": 270
1629
+ },
1630
+ {
1631
+ "epoch": 0.03,
1632
+ "learning_rate": 0.0009827272727272727,
1633
+ "loss": 1.9616,
1634
+ "step": 271
1635
+ },
1636
+ {
1637
+ "epoch": 0.03,
1638
+ "learning_rate": 0.0009826262626262626,
1639
+ "loss": 1.9871,
1640
+ "step": 272
1641
+ },
1642
+ {
1643
+ "epoch": 0.03,
1644
+ "learning_rate": 0.0009825252525252526,
1645
+ "loss": 2.0377,
1646
+ "step": 273
1647
+ },
1648
+ {
1649
+ "epoch": 0.03,
1650
+ "learning_rate": 0.0009824242424242425,
1651
+ "loss": 1.9943,
1652
+ "step": 274
1653
+ },
1654
+ {
1655
+ "epoch": 0.03,
1656
+ "learning_rate": 0.0009823232323232325,
1657
+ "loss": 1.9859,
1658
+ "step": 275
1659
+ },
1660
+ {
1661
+ "epoch": 0.03,
1662
+ "learning_rate": 0.0009822222222222222,
1663
+ "loss": 1.94,
1664
+ "step": 276
1665
+ },
1666
+ {
1667
+ "epoch": 0.03,
1668
+ "learning_rate": 0.0009821212121212122,
1669
+ "loss": 1.9271,
1670
+ "step": 277
1671
+ },
1672
+ {
1673
+ "epoch": 0.03,
1674
+ "learning_rate": 0.0009820202020202021,
1675
+ "loss": 1.924,
1676
+ "step": 278
1677
+ },
1678
+ {
1679
+ "epoch": 0.03,
1680
+ "learning_rate": 0.000981919191919192,
1681
+ "loss": 2.0862,
1682
+ "step": 279
1683
+ },
1684
+ {
1685
+ "epoch": 0.03,
1686
+ "learning_rate": 0.0009818181818181818,
1687
+ "loss": 2.1038,
1688
+ "step": 280
1689
+ },
1690
+ {
1691
+ "epoch": 0.03,
1692
+ "learning_rate": 0.0009817171717171717,
1693
+ "loss": 1.9944,
1694
+ "step": 281
1695
+ },
1696
+ {
1697
+ "epoch": 0.03,
1698
+ "learning_rate": 0.0009816161616161617,
1699
+ "loss": 1.9266,
1700
+ "step": 282
1701
+ },
1702
+ {
1703
+ "epoch": 0.03,
1704
+ "learning_rate": 0.0009815151515151516,
1705
+ "loss": 2.0527,
1706
+ "step": 283
1707
+ },
1708
+ {
1709
+ "epoch": 0.03,
1710
+ "learning_rate": 0.0009814141414141414,
1711
+ "loss": 1.9245,
1712
+ "step": 284
1713
+ },
1714
+ {
1715
+ "epoch": 0.03,
1716
+ "learning_rate": 0.0009813131313131313,
1717
+ "loss": 2.0017,
1718
+ "step": 285
1719
+ },
1720
+ {
1721
+ "epoch": 0.03,
1722
+ "learning_rate": 0.0009812121212121213,
1723
+ "loss": 1.998,
1724
+ "step": 286
1725
+ },
1726
+ {
1727
+ "epoch": 0.03,
1728
+ "learning_rate": 0.0009811111111111112,
1729
+ "loss": 1.9133,
1730
+ "step": 287
1731
+ },
1732
+ {
1733
+ "epoch": 0.03,
1734
+ "learning_rate": 0.000981010101010101,
1735
+ "loss": 2.0358,
1736
+ "step": 288
1737
+ },
1738
+ {
1739
+ "epoch": 0.03,
1740
+ "learning_rate": 0.0009809090909090909,
1741
+ "loss": 1.939,
1742
+ "step": 289
1743
+ },
1744
+ {
1745
+ "epoch": 0.03,
1746
+ "learning_rate": 0.0009808080808080808,
1747
+ "loss": 1.9883,
1748
+ "step": 290
1749
+ },
1750
+ {
1751
+ "epoch": 0.03,
1752
+ "learning_rate": 0.0009807070707070708,
1753
+ "loss": 2.061,
1754
+ "step": 291
1755
+ },
1756
+ {
1757
+ "epoch": 0.03,
1758
+ "learning_rate": 0.0009806060606060607,
1759
+ "loss": 1.9319,
1760
+ "step": 292
1761
+ },
1762
+ {
1763
+ "epoch": 0.03,
1764
+ "learning_rate": 0.0009805050505050507,
1765
+ "loss": 1.9661,
1766
+ "step": 293
1767
+ },
1768
+ {
1769
+ "epoch": 0.03,
1770
+ "learning_rate": 0.0009804040404040404,
1771
+ "loss": 1.9547,
1772
+ "step": 294
1773
+ },
1774
+ {
1775
+ "epoch": 0.03,
1776
+ "learning_rate": 0.0009803030303030303,
1777
+ "loss": 1.9286,
1778
+ "step": 295
1779
+ },
1780
+ {
1781
+ "epoch": 0.03,
1782
+ "learning_rate": 0.0009802020202020203,
1783
+ "loss": 2.0367,
1784
+ "step": 296
1785
+ },
1786
+ {
1787
+ "epoch": 0.03,
1788
+ "learning_rate": 0.0009801010101010102,
1789
+ "loss": 1.9614,
1790
+ "step": 297
1791
+ },
1792
+ {
1793
+ "epoch": 0.03,
1794
+ "learning_rate": 0.00098,
1795
+ "loss": 1.9543,
1796
+ "step": 298
1797
+ },
1798
+ {
1799
+ "epoch": 0.03,
1800
+ "learning_rate": 0.00097989898989899,
1801
+ "loss": 1.9144,
1802
+ "step": 299
1803
+ },
1804
+ {
1805
+ "epoch": 0.03,
1806
+ "learning_rate": 0.0009797979797979799,
1807
+ "loss": 1.9562,
1808
+ "step": 300
1809
+ }
1810
+ ],
1811
+ "max_steps": 10000,
1812
+ "num_train_epochs": 9223372036854775807,
1813
+ "total_flos": 2.88769529654784e+17,
1814
+ "trial_name": null,
1815
+ "trial_params": null
1816
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7cacb874b42486b1f942a28a638e5d6ce7bd58efdd79ac522bbcd8667bed6d7
3
+ size 5371
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f2af1dd490d18a7b423acfc0fc38368d669310b597ca5c29d328e6bb621c241
3
  size 990452905
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00baa5820c6217c07a8cdbb56201f21a75a525e4fc15017519d42658135b0442
3
  size 990452905