Hugging Face's logo

Joemgu
/
long-t5-base-sumstew-8k

Text2Text Generation
Transformers
PyTorch
longt5
Inference Endpoints
Model card Files Community
Joemgu commited on
Commit
8df085b
•
1 Parent(s): dbefe8a

Training in progress, step 600

Browse files
Files changed (13) hide show
  1. checkpoint-600/config.json +38 -0
  2. checkpoint-600/generation_config.json +7 -0
  3. checkpoint-600/optimizer.pt +3 -0
  4. checkpoint-600/pytorch_model.bin +3 -0
  5. checkpoint-600/rng_state.pth +3 -0
  6. checkpoint-600/scheduler.pt +3 -0
  7. checkpoint-600/special_tokens_map.json +107 -0
  8. checkpoint-600/spiece.model +3 -0
  9. checkpoint-600/tokenizer.json +0 -0
  10. checkpoint-600/tokenizer_config.json +111 -0
  11. checkpoint-600/trainer_state.json +3616 -0
  12. checkpoint-600/training_args.bin +3 -0
  13. pytorch_model.bin +1 -1
checkpoint-600/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": true,
37
+ "vocab_size": 32128
38
+ }
checkpoint-600/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.29.2"
7
+ }
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2db47f8578f25df198334b9ad1efc14fc3ed6d2eb7d682cab20680bd4278eead
3
+ size 1980955781
checkpoint-600/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:874374987dbaa7df5e77d2d330b1bedbad6e692df7e14eef7d62092f8be4bf23
3
+ size 990452905
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02dc29a50b914e1c3cf67398390267460f17c36e11939faf6e945ff61142ea11
3
+ size 14575
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a596ad687419519463ce9a0bf96b915cb789c701622a5ed6531d8d08bc20226
3
+ size 627
checkpoint-600/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-600/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
checkpoint-600/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-600/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-600/trainer_state.json ADDED
@@ -0,0 +1,3616 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.1,
5
+ "global_step": 600,
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": 8.000000000000001e-06,
13
+ "loss": 4.0598,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 1.6000000000000003e-05,
19
+ "loss": 3.9538,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 2.4e-05,
25
+ "loss": 4.0434,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.0,
30
+ "learning_rate": 3.2000000000000005e-05,
31
+ "loss": 4.0516,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.0,
36
+ "learning_rate": 4e-05,
37
+ "loss": 3.998,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.0,
42
+ "learning_rate": 4.8e-05,
43
+ "loss": 4.0225,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.0,
48
+ "learning_rate": 5.6000000000000006e-05,
49
+ "loss": 3.904,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.0,
54
+ "learning_rate": 6.400000000000001e-05,
55
+ "loss": 3.792,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.0,
60
+ "learning_rate": 7.2e-05,
61
+ "loss": 3.5984,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.0,
66
+ "learning_rate": 8e-05,
67
+ "loss": 3.4579,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.0,
72
+ "learning_rate": 8.800000000000001e-05,
73
+ "loss": 3.6368,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.0,
78
+ "learning_rate": 9.6e-05,
79
+ "loss": 3.4327,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.0,
84
+ "learning_rate": 0.00010400000000000001,
85
+ "loss": 3.3701,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.0,
90
+ "learning_rate": 0.00011200000000000001,
91
+ "loss": 3.2466,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.0,
96
+ "learning_rate": 0.00012,
97
+ "loss": 3.2952,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.0,
102
+ "learning_rate": 0.00012800000000000002,
103
+ "loss": 3.1374,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.0,
108
+ "learning_rate": 0.00013600000000000003,
109
+ "loss": 3.139,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.0,
114
+ "learning_rate": 0.000144,
115
+ "loss": 3.0374,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.0,
120
+ "learning_rate": 0.000152,
121
+ "loss": 3.0814,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.0,
126
+ "learning_rate": 0.00016,
127
+ "loss": 3.0121,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.0,
132
+ "learning_rate": 0.000168,
133
+ "loss": 3.0238,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 0.0,
138
+ "learning_rate": 0.00017600000000000002,
139
+ "loss": 2.84,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 0.0,
144
+ "learning_rate": 0.00018400000000000003,
145
+ "loss": 2.8892,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 0.0,
150
+ "learning_rate": 0.000192,
151
+ "loss": 2.8234,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 0.0,
156
+ "learning_rate": 0.0002,
157
+ "loss": 2.8687,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 0.0,
162
+ "learning_rate": 0.00020800000000000001,
163
+ "loss": 2.7765,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 0.0,
168
+ "learning_rate": 0.00021600000000000002,
169
+ "loss": 2.7827,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 0.0,
174
+ "learning_rate": 0.00022400000000000002,
175
+ "loss": 2.8096,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 0.0,
180
+ "learning_rate": 0.000232,
181
+ "loss": 2.7986,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 0.01,
186
+ "learning_rate": 0.00024,
187
+ "loss": 2.7335,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 0.01,
192
+ "learning_rate": 0.000248,
193
+ "loss": 2.7041,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 0.01,
198
+ "learning_rate": 0.00025600000000000004,
199
+ "loss": 2.6543,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 0.01,
204
+ "learning_rate": 0.000264,
205
+ "loss": 2.6889,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 0.01,
210
+ "learning_rate": 0.00027200000000000005,
211
+ "loss": 2.7073,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 0.01,
216
+ "learning_rate": 0.00028,
217
+ "loss": 2.7532,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 0.01,
222
+ "learning_rate": 0.000288,
223
+ "loss": 2.5479,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 0.01,
228
+ "learning_rate": 0.000296,
229
+ "loss": 2.5846,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 0.01,
234
+ "learning_rate": 0.000304,
235
+ "loss": 2.5425,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 0.01,
240
+ "learning_rate": 0.00031200000000000005,
241
+ "loss": 2.6203,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 0.01,
246
+ "learning_rate": 0.00032,
247
+ "loss": 2.578,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 0.01,
252
+ "learning_rate": 0.000328,
253
+ "loss": 2.6235,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 0.01,
258
+ "learning_rate": 0.000336,
259
+ "loss": 2.5181,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 0.01,
264
+ "learning_rate": 0.000344,
265
+ "loss": 2.5592,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 0.01,
270
+ "learning_rate": 0.00035200000000000005,
271
+ "loss": 2.4978,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 0.01,
276
+ "learning_rate": 0.00036,
277
+ "loss": 2.5523,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 0.01,
282
+ "learning_rate": 0.00036800000000000005,
283
+ "loss": 2.5996,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 0.01,
288
+ "learning_rate": 0.000376,
289
+ "loss": 2.5497,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 0.01,
294
+ "learning_rate": 0.000384,
295
+ "loss": 2.5293,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 0.01,
300
+ "learning_rate": 0.000392,
301
+ "loss": 2.417,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 0.01,
306
+ "learning_rate": 0.0004,
307
+ "loss": 2.4453,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 0.01,
312
+ "learning_rate": 0.00040800000000000005,
313
+ "loss": 2.4302,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 0.01,
318
+ "learning_rate": 0.00041600000000000003,
319
+ "loss": 2.3976,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 0.01,
324
+ "learning_rate": 0.00042400000000000006,
325
+ "loss": 2.5218,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 0.01,
330
+ "learning_rate": 0.00043200000000000004,
331
+ "loss": 2.4465,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 0.01,
336
+ "learning_rate": 0.00044000000000000007,
337
+ "loss": 2.4865,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 0.01,
342
+ "learning_rate": 0.00044800000000000005,
343
+ "loss": 2.5155,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 0.01,
348
+ "learning_rate": 0.00045599999999999997,
349
+ "loss": 2.502,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 0.01,
354
+ "learning_rate": 0.000464,
355
+ "loss": 2.4115,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 0.01,
360
+ "learning_rate": 0.000472,
361
+ "loss": 2.4561,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 0.01,
366
+ "learning_rate": 0.00048,
367
+ "loss": 2.4581,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 0.01,
372
+ "learning_rate": 0.000488,
373
+ "loss": 2.4264,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 0.01,
378
+ "learning_rate": 0.000496,
379
+ "loss": 2.4441,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 0.01,
384
+ "learning_rate": 0.000504,
385
+ "loss": 2.4781,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 0.01,
390
+ "learning_rate": 0.0005120000000000001,
391
+ "loss": 2.3868,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 0.01,
396
+ "learning_rate": 0.0005200000000000001,
397
+ "loss": 2.4542,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 0.01,
402
+ "learning_rate": 0.000528,
403
+ "loss": 2.4437,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 0.01,
408
+ "learning_rate": 0.000536,
409
+ "loss": 2.4007,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 0.01,
414
+ "learning_rate": 0.0005440000000000001,
415
+ "loss": 2.4116,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 0.01,
420
+ "learning_rate": 0.000552,
421
+ "loss": 2.4022,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 0.01,
426
+ "learning_rate": 0.00056,
427
+ "loss": 2.3727,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 0.01,
432
+ "learning_rate": 0.000568,
433
+ "loss": 2.38,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 0.01,
438
+ "learning_rate": 0.000576,
439
+ "loss": 2.3938,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 0.01,
444
+ "learning_rate": 0.000584,
445
+ "loss": 2.4948,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 0.01,
450
+ "learning_rate": 0.000592,
451
+ "loss": 2.3133,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 0.01,
456
+ "learning_rate": 0.0006000000000000001,
457
+ "loss": 2.4094,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 0.01,
462
+ "learning_rate": 0.000608,
463
+ "loss": 2.3753,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 0.01,
468
+ "learning_rate": 0.000616,
469
+ "loss": 2.4388,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 0.01,
474
+ "learning_rate": 0.0006240000000000001,
475
+ "loss": 2.4028,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 0.01,
480
+ "learning_rate": 0.0006320000000000001,
481
+ "loss": 2.3912,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 0.01,
486
+ "learning_rate": 0.00064,
487
+ "loss": 2.3202,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 0.01,
492
+ "learning_rate": 0.000648,
493
+ "loss": 2.404,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 0.01,
498
+ "learning_rate": 0.000656,
499
+ "loss": 2.3694,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 0.01,
504
+ "learning_rate": 0.000664,
505
+ "loss": 2.3869,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 0.01,
510
+ "learning_rate": 0.000672,
511
+ "loss": 2.3976,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 0.01,
516
+ "learning_rate": 0.00068,
517
+ "loss": 2.3407,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 0.01,
522
+ "learning_rate": 0.000688,
523
+ "loss": 2.35,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 0.01,
528
+ "learning_rate": 0.000696,
529
+ "loss": 2.3244,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 0.01,
534
+ "learning_rate": 0.0007040000000000001,
535
+ "loss": 2.2987,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 0.01,
540
+ "learning_rate": 0.0007120000000000001,
541
+ "loss": 2.403,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 0.01,
546
+ "learning_rate": 0.00072,
547
+ "loss": 2.4228,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 0.02,
552
+ "learning_rate": 0.000728,
553
+ "loss": 2.3729,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 0.02,
558
+ "learning_rate": 0.0007360000000000001,
559
+ "loss": 2.2986,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 0.02,
564
+ "learning_rate": 0.0007440000000000001,
565
+ "loss": 2.3624,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 0.02,
570
+ "learning_rate": 0.000752,
571
+ "loss": 2.3393,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 0.02,
576
+ "learning_rate": 0.00076,
577
+ "loss": 2.3151,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 0.02,
582
+ "learning_rate": 0.000768,
583
+ "loss": 2.3513,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 0.02,
588
+ "learning_rate": 0.000776,
589
+ "loss": 2.2885,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 0.02,
594
+ "learning_rate": 0.000784,
595
+ "loss": 2.3285,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 0.02,
600
+ "learning_rate": 0.0007920000000000001,
601
+ "loss": 2.3709,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 0.02,
606
+ "learning_rate": 0.0008,
607
+ "loss": 2.4014,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 0.02,
612
+ "learning_rate": 0.000799864406779661,
613
+ "loss": 2.3282,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 0.02,
618
+ "learning_rate": 0.0007997288135593221,
619
+ "loss": 2.3529,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 0.02,
624
+ "learning_rate": 0.0007995932203389831,
625
+ "loss": 2.3502,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 0.02,
630
+ "learning_rate": 0.0007994576271186441,
631
+ "loss": 2.2684,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 0.02,
636
+ "learning_rate": 0.000799322033898305,
637
+ "loss": 2.3121,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 0.02,
642
+ "learning_rate": 0.0007991864406779661,
643
+ "loss": 2.2489,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 0.02,
648
+ "learning_rate": 0.0007990508474576271,
649
+ "loss": 2.2469,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 0.02,
654
+ "learning_rate": 0.0007989152542372882,
655
+ "loss": 2.3522,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 0.02,
660
+ "learning_rate": 0.0007987796610169492,
661
+ "loss": 2.2518,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 0.02,
666
+ "learning_rate": 0.0007986440677966103,
667
+ "loss": 2.3545,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 0.02,
672
+ "learning_rate": 0.0007985084745762711,
673
+ "loss": 2.3159,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 0.02,
678
+ "learning_rate": 0.0007983728813559322,
679
+ "loss": 2.3242,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 0.02,
684
+ "learning_rate": 0.0007982372881355932,
685
+ "loss": 2.3071,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 0.02,
690
+ "learning_rate": 0.0007981016949152543,
691
+ "loss": 2.3003,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 0.02,
696
+ "learning_rate": 0.0007979661016949153,
697
+ "loss": 2.2448,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 0.02,
702
+ "learning_rate": 0.0007978305084745764,
703
+ "loss": 2.3383,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 0.02,
708
+ "learning_rate": 0.0007976949152542374,
709
+ "loss": 2.2184,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 0.02,
714
+ "learning_rate": 0.0007975593220338983,
715
+ "loss": 2.3312,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 0.02,
720
+ "learning_rate": 0.0007974237288135593,
721
+ "loss": 2.3349,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 0.02,
726
+ "learning_rate": 0.0007972881355932204,
727
+ "loss": 2.3324,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 0.02,
732
+ "learning_rate": 0.0007971525423728814,
733
+ "loss": 2.3189,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 0.02,
738
+ "learning_rate": 0.0007970169491525424,
739
+ "loss": 2.2545,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 0.02,
744
+ "learning_rate": 0.0007968813559322035,
745
+ "loss": 2.3608,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 0.02,
750
+ "learning_rate": 0.0007967457627118644,
751
+ "loss": 2.2787,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 0.02,
756
+ "learning_rate": 0.0007966101694915254,
757
+ "loss": 2.3173,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 0.02,
762
+ "learning_rate": 0.0007964745762711864,
763
+ "loss": 2.2544,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 0.02,
768
+ "learning_rate": 0.0007963389830508475,
769
+ "loss": 2.3787,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 0.02,
774
+ "learning_rate": 0.0007962033898305085,
775
+ "loss": 2.2431,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 0.02,
780
+ "learning_rate": 0.0007960677966101696,
781
+ "loss": 2.2985,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 0.02,
786
+ "learning_rate": 0.0007959322033898305,
787
+ "loss": 2.255,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 0.02,
792
+ "learning_rate": 0.0007957966101694916,
793
+ "loss": 2.2984,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 0.02,
798
+ "learning_rate": 0.0007956610169491526,
799
+ "loss": 2.2976,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 0.02,
804
+ "learning_rate": 0.0007955254237288136,
805
+ "loss": 2.2816,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 0.02,
810
+ "learning_rate": 0.0007953898305084746,
811
+ "loss": 2.2662,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 0.02,
816
+ "learning_rate": 0.0007952542372881357,
817
+ "loss": 2.2066,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 0.02,
822
+ "learning_rate": 0.0007951186440677966,
823
+ "loss": 2.2509,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 0.02,
828
+ "learning_rate": 0.0007949830508474577,
829
+ "loss": 2.2874,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 0.02,
834
+ "learning_rate": 0.0007948474576271187,
835
+ "loss": 2.2304,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 0.02,
840
+ "learning_rate": 0.0007947118644067797,
841
+ "loss": 2.2438,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 0.02,
846
+ "learning_rate": 0.0007945762711864407,
847
+ "loss": 2.293,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 0.02,
852
+ "learning_rate": 0.0007944406779661018,
853
+ "loss": 2.2882,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 0.02,
858
+ "learning_rate": 0.0007943050847457627,
859
+ "loss": 2.2687,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 0.02,
864
+ "learning_rate": 0.0007941694915254237,
865
+ "loss": 2.2579,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 0.02,
870
+ "learning_rate": 0.0007940338983050848,
871
+ "loss": 2.2091,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 0.02,
876
+ "learning_rate": 0.0007938983050847458,
877
+ "loss": 2.3201,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 0.02,
882
+ "learning_rate": 0.0007937627118644069,
883
+ "loss": 2.1964,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 0.02,
888
+ "learning_rate": 0.0007936271186440678,
889
+ "loss": 2.2587,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 0.02,
894
+ "learning_rate": 0.0007934915254237288,
895
+ "loss": 2.2987,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 0.02,
900
+ "learning_rate": 0.0007933559322033898,
901
+ "loss": 2.1963,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 0.03,
906
+ "learning_rate": 0.0007932203389830509,
907
+ "loss": 2.2045,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 0.03,
912
+ "learning_rate": 0.0007930847457627119,
913
+ "loss": 2.3174,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 0.03,
918
+ "learning_rate": 0.000792949152542373,
919
+ "loss": 2.3348,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 0.03,
924
+ "learning_rate": 0.000792813559322034,
925
+ "loss": 2.2241,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 0.03,
930
+ "learning_rate": 0.0007926779661016949,
931
+ "loss": 2.2472,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 0.03,
936
+ "learning_rate": 0.0007925423728813559,
937
+ "loss": 2.2352,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 0.03,
942
+ "learning_rate": 0.000792406779661017,
943
+ "loss": 2.2351,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 0.03,
948
+ "learning_rate": 0.000792271186440678,
949
+ "loss": 2.2205,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 0.03,
954
+ "learning_rate": 0.0007921355932203391,
955
+ "loss": 2.1175,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 0.03,
960
+ "learning_rate": 0.0007920000000000001,
961
+ "loss": 2.1801,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 0.03,
966
+ "learning_rate": 0.000791864406779661,
967
+ "loss": 2.2301,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 0.03,
972
+ "learning_rate": 0.000791728813559322,
973
+ "loss": 2.2605,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 0.03,
978
+ "learning_rate": 0.0007915932203389831,
979
+ "loss": 2.2462,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 0.03,
984
+ "learning_rate": 0.0007914576271186441,
985
+ "loss": 2.2029,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 0.03,
990
+ "learning_rate": 0.0007913220338983051,
991
+ "loss": 2.2551,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 0.03,
996
+ "learning_rate": 0.0007911864406779662,
997
+ "loss": 2.2825,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 0.03,
1002
+ "learning_rate": 0.0007910508474576271,
1003
+ "loss": 2.2582,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 0.03,
1008
+ "learning_rate": 0.0007909152542372882,
1009
+ "loss": 2.2453,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 0.03,
1014
+ "learning_rate": 0.0007907796610169492,
1015
+ "loss": 2.2599,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 0.03,
1020
+ "learning_rate": 0.0007906440677966102,
1021
+ "loss": 2.2336,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 0.03,
1026
+ "learning_rate": 0.0007905084745762712,
1027
+ "loss": 2.1311,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 0.03,
1032
+ "learning_rate": 0.0007903728813559323,
1033
+ "loss": 2.2047,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 0.03,
1038
+ "learning_rate": 0.0007902372881355932,
1039
+ "loss": 2.2847,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 0.03,
1044
+ "learning_rate": 0.0007901016949152543,
1045
+ "loss": 2.1759,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 0.03,
1050
+ "learning_rate": 0.0007899661016949153,
1051
+ "loss": 2.1877,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 0.03,
1056
+ "learning_rate": 0.0007898305084745763,
1057
+ "loss": 2.2365,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 0.03,
1062
+ "learning_rate": 0.0007896949152542373,
1063
+ "loss": 2.1855,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 0.03,
1068
+ "learning_rate": 0.0007895593220338984,
1069
+ "loss": 2.2359,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 0.03,
1074
+ "learning_rate": 0.0007894237288135593,
1075
+ "loss": 2.2366,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 0.03,
1080
+ "learning_rate": 0.0007892881355932204,
1081
+ "loss": 2.2493,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 0.03,
1086
+ "learning_rate": 0.0007891525423728814,
1087
+ "loss": 2.2278,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 0.03,
1092
+ "learning_rate": 0.0007890169491525425,
1093
+ "loss": 2.2498,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 0.03,
1098
+ "learning_rate": 0.0007888813559322034,
1099
+ "loss": 2.2337,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 0.03,
1104
+ "learning_rate": 0.0007887457627118645,
1105
+ "loss": 2.2675,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 0.03,
1110
+ "learning_rate": 0.0007886101694915254,
1111
+ "loss": 2.2142,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 0.03,
1116
+ "learning_rate": 0.0007884745762711864,
1117
+ "loss": 2.2686,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 0.03,
1122
+ "learning_rate": 0.0007883389830508475,
1123
+ "loss": 2.1442,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 0.03,
1128
+ "learning_rate": 0.0007882033898305085,
1129
+ "loss": 2.2155,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 0.03,
1134
+ "learning_rate": 0.0007880677966101696,
1135
+ "loss": 2.2874,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 0.03,
1140
+ "learning_rate": 0.0007879322033898306,
1141
+ "loss": 2.1709,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 0.03,
1146
+ "learning_rate": 0.0007877966101694915,
1147
+ "loss": 2.2595,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 0.03,
1152
+ "learning_rate": 0.0007876610169491525,
1153
+ "loss": 2.2017,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 0.03,
1158
+ "learning_rate": 0.0007875254237288136,
1159
+ "loss": 2.1778,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 0.03,
1164
+ "learning_rate": 0.0007873898305084746,
1165
+ "loss": 2.2272,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 0.03,
1170
+ "learning_rate": 0.0007872542372881357,
1171
+ "loss": 2.2734,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 0.03,
1176
+ "learning_rate": 0.0007871186440677967,
1177
+ "loss": 2.1794,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 0.03,
1182
+ "learning_rate": 0.0007869830508474576,
1183
+ "loss": 2.2155,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 0.03,
1188
+ "learning_rate": 0.0007868474576271186,
1189
+ "loss": 2.2478,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 0.03,
1194
+ "learning_rate": 0.0007867118644067797,
1195
+ "loss": 2.1717,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 0.03,
1200
+ "learning_rate": 0.0007865762711864407,
1201
+ "loss": 2.2421,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 0.03,
1206
+ "learning_rate": 0.0007864406779661018,
1207
+ "loss": 2.2584,
1208
+ "step": 200
1209
+ },
1210
+ {
1211
+ "epoch": 0.03,
1212
+ "learning_rate": 0.0007863050847457628,
1213
+ "loss": 2.2263,
1214
+ "step": 201
1215
+ },
1216
+ {
1217
+ "epoch": 0.03,
1218
+ "learning_rate": 0.0007861694915254239,
1219
+ "loss": 2.1951,
1220
+ "step": 202
1221
+ },
1222
+ {
1223
+ "epoch": 0.03,
1224
+ "learning_rate": 0.0007860338983050848,
1225
+ "loss": 2.1717,
1226
+ "step": 203
1227
+ },
1228
+ {
1229
+ "epoch": 0.03,
1230
+ "learning_rate": 0.0007858983050847457,
1231
+ "loss": 2.1705,
1232
+ "step": 204
1233
+ },
1234
+ {
1235
+ "epoch": 0.03,
1236
+ "learning_rate": 0.0007857627118644068,
1237
+ "loss": 2.199,
1238
+ "step": 205
1239
+ },
1240
+ {
1241
+ "epoch": 0.03,
1242
+ "learning_rate": 0.0007856271186440678,
1243
+ "loss": 2.3078,
1244
+ "step": 206
1245
+ },
1246
+ {
1247
+ "epoch": 0.03,
1248
+ "learning_rate": 0.0007854915254237289,
1249
+ "loss": 2.1449,
1250
+ "step": 207
1251
+ },
1252
+ {
1253
+ "epoch": 0.03,
1254
+ "learning_rate": 0.0007853559322033898,
1255
+ "loss": 2.1788,
1256
+ "step": 208
1257
+ },
1258
+ {
1259
+ "epoch": 0.03,
1260
+ "learning_rate": 0.0007852203389830509,
1261
+ "loss": 2.1998,
1262
+ "step": 209
1263
+ },
1264
+ {
1265
+ "epoch": 0.04,
1266
+ "learning_rate": 0.0007850847457627119,
1267
+ "loss": 2.2799,
1268
+ "step": 210
1269
+ },
1270
+ {
1271
+ "epoch": 0.04,
1272
+ "learning_rate": 0.0007849491525423729,
1273
+ "loss": 2.2501,
1274
+ "step": 211
1275
+ },
1276
+ {
1277
+ "epoch": 0.04,
1278
+ "learning_rate": 0.0007848135593220339,
1279
+ "loss": 2.2081,
1280
+ "step": 212
1281
+ },
1282
+ {
1283
+ "epoch": 0.04,
1284
+ "learning_rate": 0.000784677966101695,
1285
+ "loss": 2.1697,
1286
+ "step": 213
1287
+ },
1288
+ {
1289
+ "epoch": 0.04,
1290
+ "learning_rate": 0.0007845423728813559,
1291
+ "loss": 2.1713,
1292
+ "step": 214
1293
+ },
1294
+ {
1295
+ "epoch": 0.04,
1296
+ "learning_rate": 0.000784406779661017,
1297
+ "loss": 2.2226,
1298
+ "step": 215
1299
+ },
1300
+ {
1301
+ "epoch": 0.04,
1302
+ "learning_rate": 0.000784271186440678,
1303
+ "loss": 2.2084,
1304
+ "step": 216
1305
+ },
1306
+ {
1307
+ "epoch": 0.04,
1308
+ "learning_rate": 0.0007841355932203391,
1309
+ "loss": 2.1771,
1310
+ "step": 217
1311
+ },
1312
+ {
1313
+ "epoch": 0.04,
1314
+ "learning_rate": 0.000784,
1315
+ "loss": 2.2013,
1316
+ "step": 218
1317
+ },
1318
+ {
1319
+ "epoch": 0.04,
1320
+ "learning_rate": 0.0007838644067796611,
1321
+ "loss": 2.248,
1322
+ "step": 219
1323
+ },
1324
+ {
1325
+ "epoch": 0.04,
1326
+ "learning_rate": 0.000783728813559322,
1327
+ "loss": 2.2238,
1328
+ "step": 220
1329
+ },
1330
+ {
1331
+ "epoch": 0.04,
1332
+ "learning_rate": 0.0007835932203389831,
1333
+ "loss": 2.1268,
1334
+ "step": 221
1335
+ },
1336
+ {
1337
+ "epoch": 0.04,
1338
+ "learning_rate": 0.0007834576271186441,
1339
+ "loss": 2.161,
1340
+ "step": 222
1341
+ },
1342
+ {
1343
+ "epoch": 0.04,
1344
+ "learning_rate": 0.0007833220338983052,
1345
+ "loss": 2.1049,
1346
+ "step": 223
1347
+ },
1348
+ {
1349
+ "epoch": 0.04,
1350
+ "learning_rate": 0.0007831864406779662,
1351
+ "loss": 2.1929,
1352
+ "step": 224
1353
+ },
1354
+ {
1355
+ "epoch": 0.04,
1356
+ "learning_rate": 0.0007830508474576272,
1357
+ "loss": 2.2978,
1358
+ "step": 225
1359
+ },
1360
+ {
1361
+ "epoch": 0.04,
1362
+ "learning_rate": 0.0007829152542372881,
1363
+ "loss": 2.175,
1364
+ "step": 226
1365
+ },
1366
+ {
1367
+ "epoch": 0.04,
1368
+ "learning_rate": 0.0007827796610169491,
1369
+ "loss": 2.2298,
1370
+ "step": 227
1371
+ },
1372
+ {
1373
+ "epoch": 0.04,
1374
+ "learning_rate": 0.0007826440677966102,
1375
+ "loss": 2.1707,
1376
+ "step": 228
1377
+ },
1378
+ {
1379
+ "epoch": 0.04,
1380
+ "learning_rate": 0.0007825084745762712,
1381
+ "loss": 2.1759,
1382
+ "step": 229
1383
+ },
1384
+ {
1385
+ "epoch": 0.04,
1386
+ "learning_rate": 0.0007823728813559323,
1387
+ "loss": 2.182,
1388
+ "step": 230
1389
+ },
1390
+ {
1391
+ "epoch": 0.04,
1392
+ "learning_rate": 0.0007822372881355933,
1393
+ "loss": 2.219,
1394
+ "step": 231
1395
+ },
1396
+ {
1397
+ "epoch": 0.04,
1398
+ "learning_rate": 0.0007821016949152542,
1399
+ "loss": 2.2042,
1400
+ "step": 232
1401
+ },
1402
+ {
1403
+ "epoch": 0.04,
1404
+ "learning_rate": 0.0007819661016949152,
1405
+ "loss": 2.1706,
1406
+ "step": 233
1407
+ },
1408
+ {
1409
+ "epoch": 0.04,
1410
+ "learning_rate": 0.0007818305084745763,
1411
+ "loss": 2.171,
1412
+ "step": 234
1413
+ },
1414
+ {
1415
+ "epoch": 0.04,
1416
+ "learning_rate": 0.0007816949152542373,
1417
+ "loss": 2.0928,
1418
+ "step": 235
1419
+ },
1420
+ {
1421
+ "epoch": 0.04,
1422
+ "learning_rate": 0.0007815593220338984,
1423
+ "loss": 2.2218,
1424
+ "step": 236
1425
+ },
1426
+ {
1427
+ "epoch": 0.04,
1428
+ "learning_rate": 0.0007814237288135594,
1429
+ "loss": 2.1285,
1430
+ "step": 237
1431
+ },
1432
+ {
1433
+ "epoch": 0.04,
1434
+ "learning_rate": 0.0007812881355932205,
1435
+ "loss": 2.2702,
1436
+ "step": 238
1437
+ },
1438
+ {
1439
+ "epoch": 0.04,
1440
+ "learning_rate": 0.0007811525423728813,
1441
+ "loss": 2.2371,
1442
+ "step": 239
1443
+ },
1444
+ {
1445
+ "epoch": 0.04,
1446
+ "learning_rate": 0.0007810169491525424,
1447
+ "loss": 2.1952,
1448
+ "step": 240
1449
+ },
1450
+ {
1451
+ "epoch": 0.04,
1452
+ "learning_rate": 0.0007808813559322034,
1453
+ "loss": 2.228,
1454
+ "step": 241
1455
+ },
1456
+ {
1457
+ "epoch": 0.04,
1458
+ "learning_rate": 0.0007807457627118645,
1459
+ "loss": 2.0863,
1460
+ "step": 242
1461
+ },
1462
+ {
1463
+ "epoch": 0.04,
1464
+ "learning_rate": 0.0007806101694915255,
1465
+ "loss": 2.1102,
1466
+ "step": 243
1467
+ },
1468
+ {
1469
+ "epoch": 0.04,
1470
+ "learning_rate": 0.0007804745762711866,
1471
+ "loss": 2.0992,
1472
+ "step": 244
1473
+ },
1474
+ {
1475
+ "epoch": 0.04,
1476
+ "learning_rate": 0.0007803389830508475,
1477
+ "loss": 2.231,
1478
+ "step": 245
1479
+ },
1480
+ {
1481
+ "epoch": 0.04,
1482
+ "learning_rate": 0.0007802033898305085,
1483
+ "loss": 2.1459,
1484
+ "step": 246
1485
+ },
1486
+ {
1487
+ "epoch": 0.04,
1488
+ "learning_rate": 0.0007800677966101695,
1489
+ "loss": 2.2173,
1490
+ "step": 247
1491
+ },
1492
+ {
1493
+ "epoch": 0.04,
1494
+ "learning_rate": 0.0007799322033898305,
1495
+ "loss": 2.1938,
1496
+ "step": 248
1497
+ },
1498
+ {
1499
+ "epoch": 0.04,
1500
+ "learning_rate": 0.0007797966101694916,
1501
+ "loss": 2.1128,
1502
+ "step": 249
1503
+ },
1504
+ {
1505
+ "epoch": 0.04,
1506
+ "learning_rate": 0.0007796610169491525,
1507
+ "loss": 2.1401,
1508
+ "step": 250
1509
+ },
1510
+ {
1511
+ "epoch": 0.04,
1512
+ "learning_rate": 0.0007795254237288136,
1513
+ "loss": 2.1503,
1514
+ "step": 251
1515
+ },
1516
+ {
1517
+ "epoch": 0.04,
1518
+ "learning_rate": 0.0007793898305084746,
1519
+ "loss": 2.0771,
1520
+ "step": 252
1521
+ },
1522
+ {
1523
+ "epoch": 0.04,
1524
+ "learning_rate": 0.0007792542372881356,
1525
+ "loss": 2.0993,
1526
+ "step": 253
1527
+ },
1528
+ {
1529
+ "epoch": 0.04,
1530
+ "learning_rate": 0.0007791186440677966,
1531
+ "loss": 2.24,
1532
+ "step": 254
1533
+ },
1534
+ {
1535
+ "epoch": 0.04,
1536
+ "learning_rate": 0.0007789830508474577,
1537
+ "loss": 2.154,
1538
+ "step": 255
1539
+ },
1540
+ {
1541
+ "epoch": 0.04,
1542
+ "learning_rate": 0.0007788474576271186,
1543
+ "loss": 2.1774,
1544
+ "step": 256
1545
+ },
1546
+ {
1547
+ "epoch": 0.04,
1548
+ "learning_rate": 0.0007787118644067797,
1549
+ "loss": 2.0888,
1550
+ "step": 257
1551
+ },
1552
+ {
1553
+ "epoch": 0.04,
1554
+ "learning_rate": 0.0007785762711864407,
1555
+ "loss": 2.216,
1556
+ "step": 258
1557
+ },
1558
+ {
1559
+ "epoch": 0.04,
1560
+ "learning_rate": 0.0007784406779661018,
1561
+ "loss": 2.1042,
1562
+ "step": 259
1563
+ },
1564
+ {
1565
+ "epoch": 0.04,
1566
+ "learning_rate": 0.0007783050847457628,
1567
+ "loss": 2.2299,
1568
+ "step": 260
1569
+ },
1570
+ {
1571
+ "epoch": 0.04,
1572
+ "learning_rate": 0.0007781694915254238,
1573
+ "loss": 2.201,
1574
+ "step": 261
1575
+ },
1576
+ {
1577
+ "epoch": 0.04,
1578
+ "learning_rate": 0.0007780338983050847,
1579
+ "loss": 2.1315,
1580
+ "step": 262
1581
+ },
1582
+ {
1583
+ "epoch": 0.04,
1584
+ "learning_rate": 0.0007778983050847458,
1585
+ "loss": 2.1977,
1586
+ "step": 263
1587
+ },
1588
+ {
1589
+ "epoch": 0.04,
1590
+ "learning_rate": 0.0007777627118644068,
1591
+ "loss": 2.2053,
1592
+ "step": 264
1593
+ },
1594
+ {
1595
+ "epoch": 0.04,
1596
+ "learning_rate": 0.0007776271186440679,
1597
+ "loss": 2.1809,
1598
+ "step": 265
1599
+ },
1600
+ {
1601
+ "epoch": 0.04,
1602
+ "learning_rate": 0.0007774915254237289,
1603
+ "loss": 2.2412,
1604
+ "step": 266
1605
+ },
1606
+ {
1607
+ "epoch": 0.04,
1608
+ "learning_rate": 0.0007773559322033899,
1609
+ "loss": 2.2178,
1610
+ "step": 267
1611
+ },
1612
+ {
1613
+ "epoch": 0.04,
1614
+ "learning_rate": 0.0007772203389830508,
1615
+ "loss": 2.2615,
1616
+ "step": 268
1617
+ },
1618
+ {
1619
+ "epoch": 0.04,
1620
+ "learning_rate": 0.0007770847457627118,
1621
+ "loss": 2.128,
1622
+ "step": 269
1623
+ },
1624
+ {
1625
+ "epoch": 0.04,
1626
+ "learning_rate": 0.0007769491525423729,
1627
+ "loss": 2.2243,
1628
+ "step": 270
1629
+ },
1630
+ {
1631
+ "epoch": 0.05,
1632
+ "learning_rate": 0.0007768135593220339,
1633
+ "loss": 2.1416,
1634
+ "step": 271
1635
+ },
1636
+ {
1637
+ "epoch": 0.05,
1638
+ "learning_rate": 0.000776677966101695,
1639
+ "loss": 2.1921,
1640
+ "step": 272
1641
+ },
1642
+ {
1643
+ "epoch": 0.05,
1644
+ "learning_rate": 0.000776542372881356,
1645
+ "loss": 2.1748,
1646
+ "step": 273
1647
+ },
1648
+ {
1649
+ "epoch": 0.05,
1650
+ "learning_rate": 0.0007764067796610171,
1651
+ "loss": 2.1006,
1652
+ "step": 274
1653
+ },
1654
+ {
1655
+ "epoch": 0.05,
1656
+ "learning_rate": 0.0007762711864406779,
1657
+ "loss": 2.1855,
1658
+ "step": 275
1659
+ },
1660
+ {
1661
+ "epoch": 0.05,
1662
+ "learning_rate": 0.000776135593220339,
1663
+ "loss": 2.0681,
1664
+ "step": 276
1665
+ },
1666
+ {
1667
+ "epoch": 0.05,
1668
+ "learning_rate": 0.000776,
1669
+ "loss": 2.2588,
1670
+ "step": 277
1671
+ },
1672
+ {
1673
+ "epoch": 0.05,
1674
+ "learning_rate": 0.0007758644067796611,
1675
+ "loss": 2.1486,
1676
+ "step": 278
1677
+ },
1678
+ {
1679
+ "epoch": 0.05,
1680
+ "learning_rate": 0.0007757288135593221,
1681
+ "loss": 2.206,
1682
+ "step": 279
1683
+ },
1684
+ {
1685
+ "epoch": 0.05,
1686
+ "learning_rate": 0.0007755932203389832,
1687
+ "loss": 2.0873,
1688
+ "step": 280
1689
+ },
1690
+ {
1691
+ "epoch": 0.05,
1692
+ "learning_rate": 0.0007754576271186441,
1693
+ "loss": 2.149,
1694
+ "step": 281
1695
+ },
1696
+ {
1697
+ "epoch": 0.05,
1698
+ "learning_rate": 0.0007753220338983051,
1699
+ "loss": 2.1717,
1700
+ "step": 282
1701
+ },
1702
+ {
1703
+ "epoch": 0.05,
1704
+ "learning_rate": 0.0007751864406779661,
1705
+ "loss": 2.1944,
1706
+ "step": 283
1707
+ },
1708
+ {
1709
+ "epoch": 0.05,
1710
+ "learning_rate": 0.0007750508474576272,
1711
+ "loss": 2.1323,
1712
+ "step": 284
1713
+ },
1714
+ {
1715
+ "epoch": 0.05,
1716
+ "learning_rate": 0.0007749152542372882,
1717
+ "loss": 2.2031,
1718
+ "step": 285
1719
+ },
1720
+ {
1721
+ "epoch": 0.05,
1722
+ "learning_rate": 0.0007747796610169493,
1723
+ "loss": 2.1541,
1724
+ "step": 286
1725
+ },
1726
+ {
1727
+ "epoch": 0.05,
1728
+ "learning_rate": 0.0007746440677966102,
1729
+ "loss": 2.1012,
1730
+ "step": 287
1731
+ },
1732
+ {
1733
+ "epoch": 0.05,
1734
+ "learning_rate": 0.0007745084745762712,
1735
+ "loss": 2.1111,
1736
+ "step": 288
1737
+ },
1738
+ {
1739
+ "epoch": 0.05,
1740
+ "learning_rate": 0.0007743728813559322,
1741
+ "loss": 2.1989,
1742
+ "step": 289
1743
+ },
1744
+ {
1745
+ "epoch": 0.05,
1746
+ "learning_rate": 0.0007742372881355932,
1747
+ "loss": 2.1702,
1748
+ "step": 290
1749
+ },
1750
+ {
1751
+ "epoch": 0.05,
1752
+ "learning_rate": 0.0007741016949152543,
1753
+ "loss": 2.095,
1754
+ "step": 291
1755
+ },
1756
+ {
1757
+ "epoch": 0.05,
1758
+ "learning_rate": 0.0007739661016949152,
1759
+ "loss": 2.2087,
1760
+ "step": 292
1761
+ },
1762
+ {
1763
+ "epoch": 0.05,
1764
+ "learning_rate": 0.0007738305084745763,
1765
+ "loss": 2.1621,
1766
+ "step": 293
1767
+ },
1768
+ {
1769
+ "epoch": 0.05,
1770
+ "learning_rate": 0.0007736949152542373,
1771
+ "loss": 2.1102,
1772
+ "step": 294
1773
+ },
1774
+ {
1775
+ "epoch": 0.05,
1776
+ "learning_rate": 0.0007735593220338984,
1777
+ "loss": 2.1092,
1778
+ "step": 295
1779
+ },
1780
+ {
1781
+ "epoch": 0.05,
1782
+ "learning_rate": 0.0007734237288135594,
1783
+ "loss": 2.0995,
1784
+ "step": 296
1785
+ },
1786
+ {
1787
+ "epoch": 0.05,
1788
+ "learning_rate": 0.0007732881355932204,
1789
+ "loss": 2.0985,
1790
+ "step": 297
1791
+ },
1792
+ {
1793
+ "epoch": 0.05,
1794
+ "learning_rate": 0.0007731525423728813,
1795
+ "loss": 2.1749,
1796
+ "step": 298
1797
+ },
1798
+ {
1799
+ "epoch": 0.05,
1800
+ "learning_rate": 0.0007730169491525424,
1801
+ "loss": 2.1833,
1802
+ "step": 299
1803
+ },
1804
+ {
1805
+ "epoch": 0.05,
1806
+ "learning_rate": 0.0007728813559322034,
1807
+ "loss": 2.1927,
1808
+ "step": 300
1809
+ },
1810
+ {
1811
+ "epoch": 0.05,
1812
+ "learning_rate": 0.0007727457627118645,
1813
+ "loss": 2.0672,
1814
+ "step": 301
1815
+ },
1816
+ {
1817
+ "epoch": 0.05,
1818
+ "learning_rate": 0.0007726101694915255,
1819
+ "loss": 2.2456,
1820
+ "step": 302
1821
+ },
1822
+ {
1823
+ "epoch": 0.05,
1824
+ "learning_rate": 0.0007724745762711865,
1825
+ "loss": 2.1152,
1826
+ "step": 303
1827
+ },
1828
+ {
1829
+ "epoch": 0.05,
1830
+ "learning_rate": 0.0007723389830508474,
1831
+ "loss": 2.1347,
1832
+ "step": 304
1833
+ },
1834
+ {
1835
+ "epoch": 0.05,
1836
+ "learning_rate": 0.0007722033898305085,
1837
+ "loss": 2.113,
1838
+ "step": 305
1839
+ },
1840
+ {
1841
+ "epoch": 0.05,
1842
+ "learning_rate": 0.0007720677966101695,
1843
+ "loss": 2.1097,
1844
+ "step": 306
1845
+ },
1846
+ {
1847
+ "epoch": 0.05,
1848
+ "learning_rate": 0.0007719322033898306,
1849
+ "loss": 2.1361,
1850
+ "step": 307
1851
+ },
1852
+ {
1853
+ "epoch": 0.05,
1854
+ "learning_rate": 0.0007717966101694916,
1855
+ "loss": 2.1015,
1856
+ "step": 308
1857
+ },
1858
+ {
1859
+ "epoch": 0.05,
1860
+ "learning_rate": 0.0007716610169491526,
1861
+ "loss": 2.1516,
1862
+ "step": 309
1863
+ },
1864
+ {
1865
+ "epoch": 0.05,
1866
+ "learning_rate": 0.0007715254237288135,
1867
+ "loss": 2.1099,
1868
+ "step": 310
1869
+ },
1870
+ {
1871
+ "epoch": 0.05,
1872
+ "learning_rate": 0.0007713898305084745,
1873
+ "loss": 2.1754,
1874
+ "step": 311
1875
+ },
1876
+ {
1877
+ "epoch": 0.05,
1878
+ "learning_rate": 0.0007712542372881356,
1879
+ "loss": 2.1101,
1880
+ "step": 312
1881
+ },
1882
+ {
1883
+ "epoch": 0.05,
1884
+ "learning_rate": 0.0007711186440677966,
1885
+ "loss": 2.1652,
1886
+ "step": 313
1887
+ },
1888
+ {
1889
+ "epoch": 0.05,
1890
+ "learning_rate": 0.0007709830508474577,
1891
+ "loss": 2.0852,
1892
+ "step": 314
1893
+ },
1894
+ {
1895
+ "epoch": 0.05,
1896
+ "learning_rate": 0.0007708474576271187,
1897
+ "loss": 2.0926,
1898
+ "step": 315
1899
+ },
1900
+ {
1901
+ "epoch": 0.05,
1902
+ "learning_rate": 0.0007707118644067798,
1903
+ "loss": 2.1592,
1904
+ "step": 316
1905
+ },
1906
+ {
1907
+ "epoch": 0.05,
1908
+ "learning_rate": 0.0007705762711864407,
1909
+ "loss": 2.2111,
1910
+ "step": 317
1911
+ },
1912
+ {
1913
+ "epoch": 0.05,
1914
+ "learning_rate": 0.0007704406779661017,
1915
+ "loss": 2.08,
1916
+ "step": 318
1917
+ },
1918
+ {
1919
+ "epoch": 0.05,
1920
+ "learning_rate": 0.0007703050847457627,
1921
+ "loss": 2.1532,
1922
+ "step": 319
1923
+ },
1924
+ {
1925
+ "epoch": 0.05,
1926
+ "learning_rate": 0.0007701694915254238,
1927
+ "loss": 2.1141,
1928
+ "step": 320
1929
+ },
1930
+ {
1931
+ "epoch": 0.05,
1932
+ "learning_rate": 0.0007700338983050848,
1933
+ "loss": 2.0578,
1934
+ "step": 321
1935
+ },
1936
+ {
1937
+ "epoch": 0.05,
1938
+ "learning_rate": 0.0007698983050847459,
1939
+ "loss": 2.1405,
1940
+ "step": 322
1941
+ },
1942
+ {
1943
+ "epoch": 0.05,
1944
+ "learning_rate": 0.0007697627118644068,
1945
+ "loss": 2.0921,
1946
+ "step": 323
1947
+ },
1948
+ {
1949
+ "epoch": 0.05,
1950
+ "learning_rate": 0.0007696271186440678,
1951
+ "loss": 2.1161,
1952
+ "step": 324
1953
+ },
1954
+ {
1955
+ "epoch": 0.05,
1956
+ "learning_rate": 0.0007694915254237288,
1957
+ "loss": 2.118,
1958
+ "step": 325
1959
+ },
1960
+ {
1961
+ "epoch": 0.05,
1962
+ "learning_rate": 0.0007693559322033899,
1963
+ "loss": 2.054,
1964
+ "step": 326
1965
+ },
1966
+ {
1967
+ "epoch": 0.05,
1968
+ "learning_rate": 0.0007692203389830509,
1969
+ "loss": 2.0837,
1970
+ "step": 327
1971
+ },
1972
+ {
1973
+ "epoch": 0.05,
1974
+ "learning_rate": 0.000769084745762712,
1975
+ "loss": 2.1788,
1976
+ "step": 328
1977
+ },
1978
+ {
1979
+ "epoch": 0.05,
1980
+ "learning_rate": 0.0007689491525423729,
1981
+ "loss": 2.1688,
1982
+ "step": 329
1983
+ },
1984
+ {
1985
+ "epoch": 0.06,
1986
+ "learning_rate": 0.0007688135593220339,
1987
+ "loss": 2.1397,
1988
+ "step": 330
1989
+ },
1990
+ {
1991
+ "epoch": 0.06,
1992
+ "learning_rate": 0.000768677966101695,
1993
+ "loss": 2.0973,
1994
+ "step": 331
1995
+ },
1996
+ {
1997
+ "epoch": 0.06,
1998
+ "learning_rate": 0.0007685423728813559,
1999
+ "loss": 2.1558,
2000
+ "step": 332
2001
+ },
2002
+ {
2003
+ "epoch": 0.06,
2004
+ "learning_rate": 0.000768406779661017,
2005
+ "loss": 2.0784,
2006
+ "step": 333
2007
+ },
2008
+ {
2009
+ "epoch": 0.06,
2010
+ "learning_rate": 0.000768271186440678,
2011
+ "loss": 2.0901,
2012
+ "step": 334
2013
+ },
2014
+ {
2015
+ "epoch": 0.06,
2016
+ "learning_rate": 0.000768135593220339,
2017
+ "loss": 2.083,
2018
+ "step": 335
2019
+ },
2020
+ {
2021
+ "epoch": 0.06,
2022
+ "learning_rate": 0.000768,
2023
+ "loss": 2.0883,
2024
+ "step": 336
2025
+ },
2026
+ {
2027
+ "epoch": 0.06,
2028
+ "learning_rate": 0.0007678644067796611,
2029
+ "loss": 2.1707,
2030
+ "step": 337
2031
+ },
2032
+ {
2033
+ "epoch": 0.06,
2034
+ "learning_rate": 0.0007677288135593221,
2035
+ "loss": 2.1467,
2036
+ "step": 338
2037
+ },
2038
+ {
2039
+ "epoch": 0.06,
2040
+ "learning_rate": 0.0007675932203389831,
2041
+ "loss": 2.0778,
2042
+ "step": 339
2043
+ },
2044
+ {
2045
+ "epoch": 0.06,
2046
+ "learning_rate": 0.000767457627118644,
2047
+ "loss": 2.158,
2048
+ "step": 340
2049
+ },
2050
+ {
2051
+ "epoch": 0.06,
2052
+ "learning_rate": 0.0007673220338983051,
2053
+ "loss": 2.1849,
2054
+ "step": 341
2055
+ },
2056
+ {
2057
+ "epoch": 0.06,
2058
+ "learning_rate": 0.0007671864406779661,
2059
+ "loss": 2.132,
2060
+ "step": 342
2061
+ },
2062
+ {
2063
+ "epoch": 0.06,
2064
+ "learning_rate": 0.0007670508474576272,
2065
+ "loss": 2.1017,
2066
+ "step": 343
2067
+ },
2068
+ {
2069
+ "epoch": 0.06,
2070
+ "learning_rate": 0.0007669152542372882,
2071
+ "loss": 2.1146,
2072
+ "step": 344
2073
+ },
2074
+ {
2075
+ "epoch": 0.06,
2076
+ "learning_rate": 0.0007667796610169493,
2077
+ "loss": 2.0731,
2078
+ "step": 345
2079
+ },
2080
+ {
2081
+ "epoch": 0.06,
2082
+ "learning_rate": 0.0007666440677966102,
2083
+ "loss": 2.1342,
2084
+ "step": 346
2085
+ },
2086
+ {
2087
+ "epoch": 0.06,
2088
+ "learning_rate": 0.0007665084745762712,
2089
+ "loss": 2.1694,
2090
+ "step": 347
2091
+ },
2092
+ {
2093
+ "epoch": 0.06,
2094
+ "learning_rate": 0.0007663728813559322,
2095
+ "loss": 2.0779,
2096
+ "step": 348
2097
+ },
2098
+ {
2099
+ "epoch": 0.06,
2100
+ "learning_rate": 0.0007662372881355933,
2101
+ "loss": 2.151,
2102
+ "step": 349
2103
+ },
2104
+ {
2105
+ "epoch": 0.06,
2106
+ "learning_rate": 0.0007661016949152543,
2107
+ "loss": 2.121,
2108
+ "step": 350
2109
+ },
2110
+ {
2111
+ "epoch": 0.06,
2112
+ "learning_rate": 0.0007659661016949153,
2113
+ "loss": 2.0767,
2114
+ "step": 351
2115
+ },
2116
+ {
2117
+ "epoch": 0.06,
2118
+ "learning_rate": 0.0007658305084745764,
2119
+ "loss": 2.1442,
2120
+ "step": 352
2121
+ },
2122
+ {
2123
+ "epoch": 0.06,
2124
+ "learning_rate": 0.0007656949152542373,
2125
+ "loss": 2.1636,
2126
+ "step": 353
2127
+ },
2128
+ {
2129
+ "epoch": 0.06,
2130
+ "learning_rate": 0.0007655593220338983,
2131
+ "loss": 2.168,
2132
+ "step": 354
2133
+ },
2134
+ {
2135
+ "epoch": 0.06,
2136
+ "learning_rate": 0.0007654237288135593,
2137
+ "loss": 2.1254,
2138
+ "step": 355
2139
+ },
2140
+ {
2141
+ "epoch": 0.06,
2142
+ "learning_rate": 0.0007652881355932204,
2143
+ "loss": 2.1253,
2144
+ "step": 356
2145
+ },
2146
+ {
2147
+ "epoch": 0.06,
2148
+ "learning_rate": 0.0007651525423728814,
2149
+ "loss": 2.0905,
2150
+ "step": 357
2151
+ },
2152
+ {
2153
+ "epoch": 0.06,
2154
+ "learning_rate": 0.0007650169491525425,
2155
+ "loss": 2.0981,
2156
+ "step": 358
2157
+ },
2158
+ {
2159
+ "epoch": 0.06,
2160
+ "learning_rate": 0.0007648813559322034,
2161
+ "loss": 2.1296,
2162
+ "step": 359
2163
+ },
2164
+ {
2165
+ "epoch": 0.06,
2166
+ "learning_rate": 0.0007647457627118644,
2167
+ "loss": 2.1614,
2168
+ "step": 360
2169
+ },
2170
+ {
2171
+ "epoch": 0.06,
2172
+ "learning_rate": 0.0007646101694915254,
2173
+ "loss": 2.0396,
2174
+ "step": 361
2175
+ },
2176
+ {
2177
+ "epoch": 0.06,
2178
+ "learning_rate": 0.0007644745762711865,
2179
+ "loss": 2.1202,
2180
+ "step": 362
2181
+ },
2182
+ {
2183
+ "epoch": 0.06,
2184
+ "learning_rate": 0.0007643389830508475,
2185
+ "loss": 2.083,
2186
+ "step": 363
2187
+ },
2188
+ {
2189
+ "epoch": 0.06,
2190
+ "learning_rate": 0.0007642033898305086,
2191
+ "loss": 2.1423,
2192
+ "step": 364
2193
+ },
2194
+ {
2195
+ "epoch": 0.06,
2196
+ "learning_rate": 0.0007640677966101695,
2197
+ "loss": 2.2048,
2198
+ "step": 365
2199
+ },
2200
+ {
2201
+ "epoch": 0.06,
2202
+ "learning_rate": 0.0007639322033898306,
2203
+ "loss": 2.1045,
2204
+ "step": 366
2205
+ },
2206
+ {
2207
+ "epoch": 0.06,
2208
+ "learning_rate": 0.0007637966101694915,
2209
+ "loss": 2.0007,
2210
+ "step": 367
2211
+ },
2212
+ {
2213
+ "epoch": 0.06,
2214
+ "learning_rate": 0.0007636610169491526,
2215
+ "loss": 2.1807,
2216
+ "step": 368
2217
+ },
2218
+ {
2219
+ "epoch": 0.06,
2220
+ "learning_rate": 0.0007635254237288136,
2221
+ "loss": 2.1266,
2222
+ "step": 369
2223
+ },
2224
+ {
2225
+ "epoch": 0.06,
2226
+ "learning_rate": 0.0007633898305084747,
2227
+ "loss": 2.1566,
2228
+ "step": 370
2229
+ },
2230
+ {
2231
+ "epoch": 0.06,
2232
+ "learning_rate": 0.0007632542372881356,
2233
+ "loss": 1.9788,
2234
+ "step": 371
2235
+ },
2236
+ {
2237
+ "epoch": 0.06,
2238
+ "learning_rate": 0.0007631186440677966,
2239
+ "loss": 2.1346,
2240
+ "step": 372
2241
+ },
2242
+ {
2243
+ "epoch": 0.06,
2244
+ "learning_rate": 0.0007629830508474577,
2245
+ "loss": 2.1011,
2246
+ "step": 373
2247
+ },
2248
+ {
2249
+ "epoch": 0.06,
2250
+ "learning_rate": 0.0007628474576271187,
2251
+ "loss": 2.1045,
2252
+ "step": 374
2253
+ },
2254
+ {
2255
+ "epoch": 0.06,
2256
+ "learning_rate": 0.0007627118644067797,
2257
+ "loss": 2.0789,
2258
+ "step": 375
2259
+ },
2260
+ {
2261
+ "epoch": 0.06,
2262
+ "learning_rate": 0.0007625762711864407,
2263
+ "loss": 2.1627,
2264
+ "step": 376
2265
+ },
2266
+ {
2267
+ "epoch": 0.06,
2268
+ "learning_rate": 0.0007624406779661017,
2269
+ "loss": 2.1348,
2270
+ "step": 377
2271
+ },
2272
+ {
2273
+ "epoch": 0.06,
2274
+ "learning_rate": 0.0007623050847457627,
2275
+ "loss": 2.0705,
2276
+ "step": 378
2277
+ },
2278
+ {
2279
+ "epoch": 0.06,
2280
+ "learning_rate": 0.0007621694915254238,
2281
+ "loss": 2.1322,
2282
+ "step": 379
2283
+ },
2284
+ {
2285
+ "epoch": 0.06,
2286
+ "learning_rate": 0.0007620338983050848,
2287
+ "loss": 2.0468,
2288
+ "step": 380
2289
+ },
2290
+ {
2291
+ "epoch": 0.06,
2292
+ "learning_rate": 0.0007618983050847458,
2293
+ "loss": 2.0707,
2294
+ "step": 381
2295
+ },
2296
+ {
2297
+ "epoch": 0.06,
2298
+ "learning_rate": 0.0007617627118644068,
2299
+ "loss": 2.1316,
2300
+ "step": 382
2301
+ },
2302
+ {
2303
+ "epoch": 0.06,
2304
+ "learning_rate": 0.0007616271186440678,
2305
+ "loss": 2.115,
2306
+ "step": 383
2307
+ },
2308
+ {
2309
+ "epoch": 0.06,
2310
+ "learning_rate": 0.0007614915254237288,
2311
+ "loss": 2.1173,
2312
+ "step": 384
2313
+ },
2314
+ {
2315
+ "epoch": 0.06,
2316
+ "learning_rate": 0.0007613559322033899,
2317
+ "loss": 2.1069,
2318
+ "step": 385
2319
+ },
2320
+ {
2321
+ "epoch": 0.06,
2322
+ "learning_rate": 0.0007612203389830509,
2323
+ "loss": 2.1525,
2324
+ "step": 386
2325
+ },
2326
+ {
2327
+ "epoch": 0.06,
2328
+ "learning_rate": 0.000761084745762712,
2329
+ "loss": 2.1556,
2330
+ "step": 387
2331
+ },
2332
+ {
2333
+ "epoch": 0.06,
2334
+ "learning_rate": 0.000760949152542373,
2335
+ "loss": 2.1647,
2336
+ "step": 388
2337
+ },
2338
+ {
2339
+ "epoch": 0.06,
2340
+ "learning_rate": 0.0007608135593220339,
2341
+ "loss": 2.0778,
2342
+ "step": 389
2343
+ },
2344
+ {
2345
+ "epoch": 0.07,
2346
+ "learning_rate": 0.0007606779661016949,
2347
+ "loss": 2.1567,
2348
+ "step": 390
2349
+ },
2350
+ {
2351
+ "epoch": 0.07,
2352
+ "learning_rate": 0.0007605423728813559,
2353
+ "loss": 2.0568,
2354
+ "step": 391
2355
+ },
2356
+ {
2357
+ "epoch": 0.07,
2358
+ "learning_rate": 0.000760406779661017,
2359
+ "loss": 2.0702,
2360
+ "step": 392
2361
+ },
2362
+ {
2363
+ "epoch": 0.07,
2364
+ "learning_rate": 0.000760271186440678,
2365
+ "loss": 2.1035,
2366
+ "step": 393
2367
+ },
2368
+ {
2369
+ "epoch": 0.07,
2370
+ "learning_rate": 0.0007601355932203391,
2371
+ "loss": 2.1456,
2372
+ "step": 394
2373
+ },
2374
+ {
2375
+ "epoch": 0.07,
2376
+ "learning_rate": 0.00076,
2377
+ "loss": 2.0825,
2378
+ "step": 395
2379
+ },
2380
+ {
2381
+ "epoch": 0.07,
2382
+ "learning_rate": 0.000759864406779661,
2383
+ "loss": 2.1152,
2384
+ "step": 396
2385
+ },
2386
+ {
2387
+ "epoch": 0.07,
2388
+ "learning_rate": 0.000759728813559322,
2389
+ "loss": 2.0893,
2390
+ "step": 397
2391
+ },
2392
+ {
2393
+ "epoch": 0.07,
2394
+ "learning_rate": 0.0007595932203389831,
2395
+ "loss": 2.1312,
2396
+ "step": 398
2397
+ },
2398
+ {
2399
+ "epoch": 0.07,
2400
+ "learning_rate": 0.0007594576271186441,
2401
+ "loss": 2.1767,
2402
+ "step": 399
2403
+ },
2404
+ {
2405
+ "epoch": 0.07,
2406
+ "learning_rate": 0.0007593220338983052,
2407
+ "loss": 2.1467,
2408
+ "step": 400
2409
+ },
2410
+ {
2411
+ "epoch": 0.07,
2412
+ "learning_rate": 0.0007591864406779661,
2413
+ "loss": 2.0759,
2414
+ "step": 401
2415
+ },
2416
+ {
2417
+ "epoch": 0.07,
2418
+ "learning_rate": 0.0007590508474576272,
2419
+ "loss": 2.0779,
2420
+ "step": 402
2421
+ },
2422
+ {
2423
+ "epoch": 0.07,
2424
+ "learning_rate": 0.0007589152542372881,
2425
+ "loss": 2.048,
2426
+ "step": 403
2427
+ },
2428
+ {
2429
+ "epoch": 0.07,
2430
+ "learning_rate": 0.0007587796610169492,
2431
+ "loss": 2.1718,
2432
+ "step": 404
2433
+ },
2434
+ {
2435
+ "epoch": 0.07,
2436
+ "learning_rate": 0.0007586440677966102,
2437
+ "loss": 2.1782,
2438
+ "step": 405
2439
+ },
2440
+ {
2441
+ "epoch": 0.07,
2442
+ "learning_rate": 0.0007585084745762713,
2443
+ "loss": 2.0557,
2444
+ "step": 406
2445
+ },
2446
+ {
2447
+ "epoch": 0.07,
2448
+ "learning_rate": 0.0007583728813559322,
2449
+ "loss": 2.091,
2450
+ "step": 407
2451
+ },
2452
+ {
2453
+ "epoch": 0.07,
2454
+ "learning_rate": 0.0007582372881355933,
2455
+ "loss": 2.1526,
2456
+ "step": 408
2457
+ },
2458
+ {
2459
+ "epoch": 0.07,
2460
+ "learning_rate": 0.0007581016949152543,
2461
+ "loss": 2.1478,
2462
+ "step": 409
2463
+ },
2464
+ {
2465
+ "epoch": 0.07,
2466
+ "learning_rate": 0.0007579661016949153,
2467
+ "loss": 2.0608,
2468
+ "step": 410
2469
+ },
2470
+ {
2471
+ "epoch": 0.07,
2472
+ "learning_rate": 0.0007578305084745763,
2473
+ "loss": 2.1078,
2474
+ "step": 411
2475
+ },
2476
+ {
2477
+ "epoch": 0.07,
2478
+ "learning_rate": 0.0007576949152542373,
2479
+ "loss": 2.1001,
2480
+ "step": 412
2481
+ },
2482
+ {
2483
+ "epoch": 0.07,
2484
+ "learning_rate": 0.0007575593220338983,
2485
+ "loss": 2.1838,
2486
+ "step": 413
2487
+ },
2488
+ {
2489
+ "epoch": 0.07,
2490
+ "learning_rate": 0.0007574237288135593,
2491
+ "loss": 2.0487,
2492
+ "step": 414
2493
+ },
2494
+ {
2495
+ "epoch": 0.07,
2496
+ "learning_rate": 0.0007572881355932204,
2497
+ "loss": 2.0995,
2498
+ "step": 415
2499
+ },
2500
+ {
2501
+ "epoch": 0.07,
2502
+ "learning_rate": 0.0007571525423728814,
2503
+ "loss": 2.0434,
2504
+ "step": 416
2505
+ },
2506
+ {
2507
+ "epoch": 0.07,
2508
+ "learning_rate": 0.0007570169491525424,
2509
+ "loss": 2.0978,
2510
+ "step": 417
2511
+ },
2512
+ {
2513
+ "epoch": 0.07,
2514
+ "learning_rate": 0.0007568813559322034,
2515
+ "loss": 2.1123,
2516
+ "step": 418
2517
+ },
2518
+ {
2519
+ "epoch": 0.07,
2520
+ "learning_rate": 0.0007567457627118644,
2521
+ "loss": 2.1078,
2522
+ "step": 419
2523
+ },
2524
+ {
2525
+ "epoch": 0.07,
2526
+ "learning_rate": 0.0007566101694915254,
2527
+ "loss": 2.0992,
2528
+ "step": 420
2529
+ },
2530
+ {
2531
+ "epoch": 0.07,
2532
+ "learning_rate": 0.0007564745762711865,
2533
+ "loss": 2.1488,
2534
+ "step": 421
2535
+ },
2536
+ {
2537
+ "epoch": 0.07,
2538
+ "learning_rate": 0.0007563389830508475,
2539
+ "loss": 2.0831,
2540
+ "step": 422
2541
+ },
2542
+ {
2543
+ "epoch": 0.07,
2544
+ "learning_rate": 0.0007562033898305086,
2545
+ "loss": 2.061,
2546
+ "step": 423
2547
+ },
2548
+ {
2549
+ "epoch": 0.07,
2550
+ "learning_rate": 0.0007560677966101696,
2551
+ "loss": 2.0328,
2552
+ "step": 424
2553
+ },
2554
+ {
2555
+ "epoch": 0.07,
2556
+ "learning_rate": 0.0007559322033898305,
2557
+ "loss": 2.0889,
2558
+ "step": 425
2559
+ },
2560
+ {
2561
+ "epoch": 0.07,
2562
+ "learning_rate": 0.0007557966101694915,
2563
+ "loss": 2.023,
2564
+ "step": 426
2565
+ },
2566
+ {
2567
+ "epoch": 0.07,
2568
+ "learning_rate": 0.0007556610169491526,
2569
+ "loss": 2.0881,
2570
+ "step": 427
2571
+ },
2572
+ {
2573
+ "epoch": 0.07,
2574
+ "learning_rate": 0.0007555254237288136,
2575
+ "loss": 2.0711,
2576
+ "step": 428
2577
+ },
2578
+ {
2579
+ "epoch": 0.07,
2580
+ "learning_rate": 0.0007553898305084747,
2581
+ "loss": 2.0338,
2582
+ "step": 429
2583
+ },
2584
+ {
2585
+ "epoch": 0.07,
2586
+ "learning_rate": 0.0007552542372881357,
2587
+ "loss": 2.0204,
2588
+ "step": 430
2589
+ },
2590
+ {
2591
+ "epoch": 0.07,
2592
+ "learning_rate": 0.0007551186440677966,
2593
+ "loss": 2.1019,
2594
+ "step": 431
2595
+ },
2596
+ {
2597
+ "epoch": 0.07,
2598
+ "learning_rate": 0.0007549830508474576,
2599
+ "loss": 2.1264,
2600
+ "step": 432
2601
+ },
2602
+ {
2603
+ "epoch": 0.07,
2604
+ "learning_rate": 0.0007548474576271186,
2605
+ "loss": 2.069,
2606
+ "step": 433
2607
+ },
2608
+ {
2609
+ "epoch": 0.07,
2610
+ "learning_rate": 0.0007547118644067797,
2611
+ "loss": 2.0708,
2612
+ "step": 434
2613
+ },
2614
+ {
2615
+ "epoch": 0.07,
2616
+ "learning_rate": 0.0007545762711864407,
2617
+ "loss": 2.0963,
2618
+ "step": 435
2619
+ },
2620
+ {
2621
+ "epoch": 0.07,
2622
+ "learning_rate": 0.0007544406779661018,
2623
+ "loss": 2.0615,
2624
+ "step": 436
2625
+ },
2626
+ {
2627
+ "epoch": 0.07,
2628
+ "learning_rate": 0.0007543050847457627,
2629
+ "loss": 2.0938,
2630
+ "step": 437
2631
+ },
2632
+ {
2633
+ "epoch": 0.07,
2634
+ "learning_rate": 0.0007541694915254237,
2635
+ "loss": 2.1352,
2636
+ "step": 438
2637
+ },
2638
+ {
2639
+ "epoch": 0.07,
2640
+ "learning_rate": 0.0007540338983050847,
2641
+ "loss": 2.0963,
2642
+ "step": 439
2643
+ },
2644
+ {
2645
+ "epoch": 0.07,
2646
+ "learning_rate": 0.0007538983050847458,
2647
+ "loss": 2.0754,
2648
+ "step": 440
2649
+ },
2650
+ {
2651
+ "epoch": 0.07,
2652
+ "learning_rate": 0.0007537627118644068,
2653
+ "loss": 2.1233,
2654
+ "step": 441
2655
+ },
2656
+ {
2657
+ "epoch": 0.07,
2658
+ "learning_rate": 0.0007536271186440679,
2659
+ "loss": 2.1221,
2660
+ "step": 442
2661
+ },
2662
+ {
2663
+ "epoch": 0.07,
2664
+ "learning_rate": 0.0007534915254237288,
2665
+ "loss": 1.9535,
2666
+ "step": 443
2667
+ },
2668
+ {
2669
+ "epoch": 0.07,
2670
+ "learning_rate": 0.0007533559322033899,
2671
+ "loss": 2.1664,
2672
+ "step": 444
2673
+ },
2674
+ {
2675
+ "epoch": 0.07,
2676
+ "learning_rate": 0.0007532203389830509,
2677
+ "loss": 2.0706,
2678
+ "step": 445
2679
+ },
2680
+ {
2681
+ "epoch": 0.07,
2682
+ "learning_rate": 0.0007530847457627119,
2683
+ "loss": 2.0818,
2684
+ "step": 446
2685
+ },
2686
+ {
2687
+ "epoch": 0.07,
2688
+ "learning_rate": 0.0007529491525423729,
2689
+ "loss": 2.1169,
2690
+ "step": 447
2691
+ },
2692
+ {
2693
+ "epoch": 0.07,
2694
+ "learning_rate": 0.000752813559322034,
2695
+ "loss": 2.0383,
2696
+ "step": 448
2697
+ },
2698
+ {
2699
+ "epoch": 0.07,
2700
+ "learning_rate": 0.0007526779661016949,
2701
+ "loss": 2.0979,
2702
+ "step": 449
2703
+ },
2704
+ {
2705
+ "epoch": 0.07,
2706
+ "learning_rate": 0.000752542372881356,
2707
+ "loss": 2.1208,
2708
+ "step": 450
2709
+ },
2710
+ {
2711
+ "epoch": 0.08,
2712
+ "learning_rate": 0.000752406779661017,
2713
+ "loss": 2.0399,
2714
+ "step": 451
2715
+ },
2716
+ {
2717
+ "epoch": 0.08,
2718
+ "learning_rate": 0.000752271186440678,
2719
+ "loss": 1.978,
2720
+ "step": 452
2721
+ },
2722
+ {
2723
+ "epoch": 0.08,
2724
+ "learning_rate": 0.000752135593220339,
2725
+ "loss": 2.0162,
2726
+ "step": 453
2727
+ },
2728
+ {
2729
+ "epoch": 0.08,
2730
+ "learning_rate": 0.000752,
2731
+ "loss": 2.162,
2732
+ "step": 454
2733
+ },
2734
+ {
2735
+ "epoch": 0.08,
2736
+ "learning_rate": 0.000751864406779661,
2737
+ "loss": 2.0827,
2738
+ "step": 455
2739
+ },
2740
+ {
2741
+ "epoch": 0.08,
2742
+ "learning_rate": 0.000751728813559322,
2743
+ "loss": 2.1191,
2744
+ "step": 456
2745
+ },
2746
+ {
2747
+ "epoch": 0.08,
2748
+ "learning_rate": 0.0007515932203389831,
2749
+ "loss": 2.0478,
2750
+ "step": 457
2751
+ },
2752
+ {
2753
+ "epoch": 0.08,
2754
+ "learning_rate": 0.0007514576271186441,
2755
+ "loss": 2.083,
2756
+ "step": 458
2757
+ },
2758
+ {
2759
+ "epoch": 0.08,
2760
+ "learning_rate": 0.0007513220338983052,
2761
+ "loss": 2.0972,
2762
+ "step": 459
2763
+ },
2764
+ {
2765
+ "epoch": 0.08,
2766
+ "learning_rate": 0.0007511864406779661,
2767
+ "loss": 2.0498,
2768
+ "step": 460
2769
+ },
2770
+ {
2771
+ "epoch": 0.08,
2772
+ "learning_rate": 0.0007510508474576271,
2773
+ "loss": 2.0427,
2774
+ "step": 461
2775
+ },
2776
+ {
2777
+ "epoch": 0.08,
2778
+ "learning_rate": 0.0007509152542372881,
2779
+ "loss": 2.0778,
2780
+ "step": 462
2781
+ },
2782
+ {
2783
+ "epoch": 0.08,
2784
+ "learning_rate": 0.0007507796610169492,
2785
+ "loss": 2.035,
2786
+ "step": 463
2787
+ },
2788
+ {
2789
+ "epoch": 0.08,
2790
+ "learning_rate": 0.0007506440677966102,
2791
+ "loss": 2.0465,
2792
+ "step": 464
2793
+ },
2794
+ {
2795
+ "epoch": 0.08,
2796
+ "learning_rate": 0.0007505084745762713,
2797
+ "loss": 2.0468,
2798
+ "step": 465
2799
+ },
2800
+ {
2801
+ "epoch": 0.08,
2802
+ "learning_rate": 0.0007503728813559323,
2803
+ "loss": 2.027,
2804
+ "step": 466
2805
+ },
2806
+ {
2807
+ "epoch": 0.08,
2808
+ "learning_rate": 0.0007502372881355932,
2809
+ "loss": 2.0677,
2810
+ "step": 467
2811
+ },
2812
+ {
2813
+ "epoch": 0.08,
2814
+ "learning_rate": 0.0007501016949152542,
2815
+ "loss": 2.1495,
2816
+ "step": 468
2817
+ },
2818
+ {
2819
+ "epoch": 0.08,
2820
+ "learning_rate": 0.0007499661016949153,
2821
+ "loss": 2.0571,
2822
+ "step": 469
2823
+ },
2824
+ {
2825
+ "epoch": 0.08,
2826
+ "learning_rate": 0.0007498305084745763,
2827
+ "loss": 2.1215,
2828
+ "step": 470
2829
+ },
2830
+ {
2831
+ "epoch": 0.08,
2832
+ "learning_rate": 0.0007496949152542374,
2833
+ "loss": 2.0422,
2834
+ "step": 471
2835
+ },
2836
+ {
2837
+ "epoch": 0.08,
2838
+ "learning_rate": 0.0007495593220338984,
2839
+ "loss": 2.1475,
2840
+ "step": 472
2841
+ },
2842
+ {
2843
+ "epoch": 0.08,
2844
+ "learning_rate": 0.0007494237288135595,
2845
+ "loss": 2.0874,
2846
+ "step": 473
2847
+ },
2848
+ {
2849
+ "epoch": 0.08,
2850
+ "learning_rate": 0.0007492881355932203,
2851
+ "loss": 2.0303,
2852
+ "step": 474
2853
+ },
2854
+ {
2855
+ "epoch": 0.08,
2856
+ "learning_rate": 0.0007491525423728813,
2857
+ "loss": 2.0438,
2858
+ "step": 475
2859
+ },
2860
+ {
2861
+ "epoch": 0.08,
2862
+ "learning_rate": 0.0007490169491525424,
2863
+ "loss": 2.0452,
2864
+ "step": 476
2865
+ },
2866
+ {
2867
+ "epoch": 0.08,
2868
+ "learning_rate": 0.0007488813559322034,
2869
+ "loss": 2.0535,
2870
+ "step": 477
2871
+ },
2872
+ {
2873
+ "epoch": 0.08,
2874
+ "learning_rate": 0.0007487457627118645,
2875
+ "loss": 2.1007,
2876
+ "step": 478
2877
+ },
2878
+ {
2879
+ "epoch": 0.08,
2880
+ "learning_rate": 0.0007486101694915254,
2881
+ "loss": 1.9706,
2882
+ "step": 479
2883
+ },
2884
+ {
2885
+ "epoch": 0.08,
2886
+ "learning_rate": 0.0007484745762711865,
2887
+ "loss": 2.1412,
2888
+ "step": 480
2889
+ },
2890
+ {
2891
+ "epoch": 0.08,
2892
+ "learning_rate": 0.0007483389830508475,
2893
+ "loss": 2.0067,
2894
+ "step": 481
2895
+ },
2896
+ {
2897
+ "epoch": 0.08,
2898
+ "learning_rate": 0.0007482033898305085,
2899
+ "loss": 2.0692,
2900
+ "step": 482
2901
+ },
2902
+ {
2903
+ "epoch": 0.08,
2904
+ "learning_rate": 0.0007480677966101695,
2905
+ "loss": 2.0327,
2906
+ "step": 483
2907
+ },
2908
+ {
2909
+ "epoch": 0.08,
2910
+ "learning_rate": 0.0007479322033898306,
2911
+ "loss": 2.0774,
2912
+ "step": 484
2913
+ },
2914
+ {
2915
+ "epoch": 0.08,
2916
+ "learning_rate": 0.0007477966101694915,
2917
+ "loss": 2.0983,
2918
+ "step": 485
2919
+ },
2920
+ {
2921
+ "epoch": 0.08,
2922
+ "learning_rate": 0.0007476610169491526,
2923
+ "loss": 2.14,
2924
+ "step": 486
2925
+ },
2926
+ {
2927
+ "epoch": 0.08,
2928
+ "learning_rate": 0.0007475254237288136,
2929
+ "loss": 2.1073,
2930
+ "step": 487
2931
+ },
2932
+ {
2933
+ "epoch": 0.08,
2934
+ "learning_rate": 0.0007473898305084746,
2935
+ "loss": 2.0778,
2936
+ "step": 488
2937
+ },
2938
+ {
2939
+ "epoch": 0.08,
2940
+ "learning_rate": 0.0007472542372881356,
2941
+ "loss": 1.9671,
2942
+ "step": 489
2943
+ },
2944
+ {
2945
+ "epoch": 0.08,
2946
+ "learning_rate": 0.0007471186440677967,
2947
+ "loss": 2.0635,
2948
+ "step": 490
2949
+ },
2950
+ {
2951
+ "epoch": 0.08,
2952
+ "learning_rate": 0.0007469830508474576,
2953
+ "loss": 2.0576,
2954
+ "step": 491
2955
+ },
2956
+ {
2957
+ "epoch": 0.08,
2958
+ "learning_rate": 0.0007468474576271187,
2959
+ "loss": 2.101,
2960
+ "step": 492
2961
+ },
2962
+ {
2963
+ "epoch": 0.08,
2964
+ "learning_rate": 0.0007467118644067797,
2965
+ "loss": 2.1317,
2966
+ "step": 493
2967
+ },
2968
+ {
2969
+ "epoch": 0.08,
2970
+ "learning_rate": 0.0007465762711864408,
2971
+ "loss": 2.0516,
2972
+ "step": 494
2973
+ },
2974
+ {
2975
+ "epoch": 0.08,
2976
+ "learning_rate": 0.0007464406779661017,
2977
+ "loss": 2.1134,
2978
+ "step": 495
2979
+ },
2980
+ {
2981
+ "epoch": 0.08,
2982
+ "learning_rate": 0.0007463050847457627,
2983
+ "loss": 2.0108,
2984
+ "step": 496
2985
+ },
2986
+ {
2987
+ "epoch": 0.08,
2988
+ "learning_rate": 0.0007461694915254237,
2989
+ "loss": 2.0717,
2990
+ "step": 497
2991
+ },
2992
+ {
2993
+ "epoch": 0.08,
2994
+ "learning_rate": 0.0007460338983050847,
2995
+ "loss": 2.062,
2996
+ "step": 498
2997
+ },
2998
+ {
2999
+ "epoch": 0.08,
3000
+ "learning_rate": 0.0007458983050847458,
3001
+ "loss": 2.136,
3002
+ "step": 499
3003
+ },
3004
+ {
3005
+ "epoch": 0.08,
3006
+ "learning_rate": 0.0007457627118644068,
3007
+ "loss": 1.9764,
3008
+ "step": 500
3009
+ },
3010
+ {
3011
+ "epoch": 0.08,
3012
+ "learning_rate": 0.0007456271186440679,
3013
+ "loss": 1.9965,
3014
+ "step": 501
3015
+ },
3016
+ {
3017
+ "epoch": 0.08,
3018
+ "learning_rate": 0.0007454915254237289,
3019
+ "loss": 1.9598,
3020
+ "step": 502
3021
+ },
3022
+ {
3023
+ "epoch": 0.08,
3024
+ "learning_rate": 0.0007453559322033898,
3025
+ "loss": 2.0202,
3026
+ "step": 503
3027
+ },
3028
+ {
3029
+ "epoch": 0.08,
3030
+ "learning_rate": 0.0007452203389830508,
3031
+ "loss": 2.0236,
3032
+ "step": 504
3033
+ },
3034
+ {
3035
+ "epoch": 0.08,
3036
+ "learning_rate": 0.0007450847457627119,
3037
+ "loss": 2.0403,
3038
+ "step": 505
3039
+ },
3040
+ {
3041
+ "epoch": 0.08,
3042
+ "learning_rate": 0.0007449491525423729,
3043
+ "loss": 2.0985,
3044
+ "step": 506
3045
+ },
3046
+ {
3047
+ "epoch": 0.08,
3048
+ "learning_rate": 0.000744813559322034,
3049
+ "loss": 2.0491,
3050
+ "step": 507
3051
+ },
3052
+ {
3053
+ "epoch": 0.08,
3054
+ "learning_rate": 0.000744677966101695,
3055
+ "loss": 2.0475,
3056
+ "step": 508
3057
+ },
3058
+ {
3059
+ "epoch": 0.08,
3060
+ "learning_rate": 0.000744542372881356,
3061
+ "loss": 2.0317,
3062
+ "step": 509
3063
+ },
3064
+ {
3065
+ "epoch": 0.09,
3066
+ "learning_rate": 0.0007444067796610169,
3067
+ "loss": 2.0934,
3068
+ "step": 510
3069
+ },
3070
+ {
3071
+ "epoch": 0.09,
3072
+ "learning_rate": 0.000744271186440678,
3073
+ "loss": 2.0595,
3074
+ "step": 511
3075
+ },
3076
+ {
3077
+ "epoch": 0.09,
3078
+ "learning_rate": 0.000744135593220339,
3079
+ "loss": 2.0955,
3080
+ "step": 512
3081
+ },
3082
+ {
3083
+ "epoch": 0.09,
3084
+ "learning_rate": 0.0007440000000000001,
3085
+ "loss": 2.0815,
3086
+ "step": 513
3087
+ },
3088
+ {
3089
+ "epoch": 0.09,
3090
+ "learning_rate": 0.0007438644067796611,
3091
+ "loss": 1.9808,
3092
+ "step": 514
3093
+ },
3094
+ {
3095
+ "epoch": 0.09,
3096
+ "learning_rate": 0.0007437288135593222,
3097
+ "loss": 2.1262,
3098
+ "step": 515
3099
+ },
3100
+ {
3101
+ "epoch": 0.09,
3102
+ "learning_rate": 0.0007435932203389831,
3103
+ "loss": 2.0687,
3104
+ "step": 516
3105
+ },
3106
+ {
3107
+ "epoch": 0.09,
3108
+ "learning_rate": 0.000743457627118644,
3109
+ "loss": 1.9892,
3110
+ "step": 517
3111
+ },
3112
+ {
3113
+ "epoch": 0.09,
3114
+ "learning_rate": 0.0007433220338983051,
3115
+ "loss": 2.0759,
3116
+ "step": 518
3117
+ },
3118
+ {
3119
+ "epoch": 0.09,
3120
+ "learning_rate": 0.0007431864406779661,
3121
+ "loss": 2.1157,
3122
+ "step": 519
3123
+ },
3124
+ {
3125
+ "epoch": 0.09,
3126
+ "learning_rate": 0.0007430508474576272,
3127
+ "loss": 2.0402,
3128
+ "step": 520
3129
+ },
3130
+ {
3131
+ "epoch": 0.09,
3132
+ "learning_rate": 0.0007429152542372881,
3133
+ "loss": 2.1073,
3134
+ "step": 521
3135
+ },
3136
+ {
3137
+ "epoch": 0.09,
3138
+ "learning_rate": 0.0007427796610169492,
3139
+ "loss": 2.0317,
3140
+ "step": 522
3141
+ },
3142
+ {
3143
+ "epoch": 0.09,
3144
+ "learning_rate": 0.0007426440677966102,
3145
+ "loss": 1.9934,
3146
+ "step": 523
3147
+ },
3148
+ {
3149
+ "epoch": 0.09,
3150
+ "learning_rate": 0.0007425084745762712,
3151
+ "loss": 1.9867,
3152
+ "step": 524
3153
+ },
3154
+ {
3155
+ "epoch": 0.09,
3156
+ "learning_rate": 0.0007423728813559322,
3157
+ "loss": 2.0976,
3158
+ "step": 525
3159
+ },
3160
+ {
3161
+ "epoch": 0.09,
3162
+ "learning_rate": 0.0007422372881355933,
3163
+ "loss": 1.9917,
3164
+ "step": 526
3165
+ },
3166
+ {
3167
+ "epoch": 0.09,
3168
+ "learning_rate": 0.0007421016949152542,
3169
+ "loss": 2.0608,
3170
+ "step": 527
3171
+ },
3172
+ {
3173
+ "epoch": 0.09,
3174
+ "learning_rate": 0.0007419661016949153,
3175
+ "loss": 1.9874,
3176
+ "step": 528
3177
+ },
3178
+ {
3179
+ "epoch": 0.09,
3180
+ "learning_rate": 0.0007418305084745763,
3181
+ "loss": 2.0441,
3182
+ "step": 529
3183
+ },
3184
+ {
3185
+ "epoch": 0.09,
3186
+ "learning_rate": 0.0007416949152542374,
3187
+ "loss": 2.0061,
3188
+ "step": 530
3189
+ },
3190
+ {
3191
+ "epoch": 0.09,
3192
+ "learning_rate": 0.0007415593220338983,
3193
+ "loss": 2.0045,
3194
+ "step": 531
3195
+ },
3196
+ {
3197
+ "epoch": 0.09,
3198
+ "learning_rate": 0.0007414237288135594,
3199
+ "loss": 2.0507,
3200
+ "step": 532
3201
+ },
3202
+ {
3203
+ "epoch": 0.09,
3204
+ "learning_rate": 0.0007412881355932203,
3205
+ "loss": 2.0163,
3206
+ "step": 533
3207
+ },
3208
+ {
3209
+ "epoch": 0.09,
3210
+ "learning_rate": 0.0007411525423728814,
3211
+ "loss": 2.1097,
3212
+ "step": 534
3213
+ },
3214
+ {
3215
+ "epoch": 0.09,
3216
+ "learning_rate": 0.0007410169491525424,
3217
+ "loss": 2.0441,
3218
+ "step": 535
3219
+ },
3220
+ {
3221
+ "epoch": 0.09,
3222
+ "learning_rate": 0.0007408813559322035,
3223
+ "loss": 2.0815,
3224
+ "step": 536
3225
+ },
3226
+ {
3227
+ "epoch": 0.09,
3228
+ "learning_rate": 0.0007407457627118645,
3229
+ "loss": 2.122,
3230
+ "step": 537
3231
+ },
3232
+ {
3233
+ "epoch": 0.09,
3234
+ "learning_rate": 0.0007406101694915255,
3235
+ "loss": 2.1065,
3236
+ "step": 538
3237
+ },
3238
+ {
3239
+ "epoch": 0.09,
3240
+ "learning_rate": 0.0007404745762711864,
3241
+ "loss": 1.9847,
3242
+ "step": 539
3243
+ },
3244
+ {
3245
+ "epoch": 0.09,
3246
+ "learning_rate": 0.0007403389830508474,
3247
+ "loss": 2.0269,
3248
+ "step": 540
3249
+ },
3250
+ {
3251
+ "epoch": 0.09,
3252
+ "learning_rate": 0.0007402033898305085,
3253
+ "loss": 2.0521,
3254
+ "step": 541
3255
+ },
3256
+ {
3257
+ "epoch": 0.09,
3258
+ "learning_rate": 0.0007400677966101695,
3259
+ "loss": 1.9796,
3260
+ "step": 542
3261
+ },
3262
+ {
3263
+ "epoch": 0.09,
3264
+ "learning_rate": 0.0007399322033898306,
3265
+ "loss": 2.0629,
3266
+ "step": 543
3267
+ },
3268
+ {
3269
+ "epoch": 0.09,
3270
+ "learning_rate": 0.0007397966101694916,
3271
+ "loss": 2.0144,
3272
+ "step": 544
3273
+ },
3274
+ {
3275
+ "epoch": 0.09,
3276
+ "learning_rate": 0.0007396610169491525,
3277
+ "loss": 1.9843,
3278
+ "step": 545
3279
+ },
3280
+ {
3281
+ "epoch": 0.09,
3282
+ "learning_rate": 0.0007395254237288135,
3283
+ "loss": 2.0864,
3284
+ "step": 546
3285
+ },
3286
+ {
3287
+ "epoch": 0.09,
3288
+ "learning_rate": 0.0007393898305084746,
3289
+ "loss": 2.0842,
3290
+ "step": 547
3291
+ },
3292
+ {
3293
+ "epoch": 0.09,
3294
+ "learning_rate": 0.0007392542372881356,
3295
+ "loss": 1.9326,
3296
+ "step": 548
3297
+ },
3298
+ {
3299
+ "epoch": 0.09,
3300
+ "learning_rate": 0.0007391186440677967,
3301
+ "loss": 2.0396,
3302
+ "step": 549
3303
+ },
3304
+ {
3305
+ "epoch": 0.09,
3306
+ "learning_rate": 0.0007389830508474577,
3307
+ "loss": 2.0398,
3308
+ "step": 550
3309
+ },
3310
+ {
3311
+ "epoch": 0.09,
3312
+ "learning_rate": 0.0007388474576271188,
3313
+ "loss": 2.0183,
3314
+ "step": 551
3315
+ },
3316
+ {
3317
+ "epoch": 0.09,
3318
+ "learning_rate": 0.0007387118644067797,
3319
+ "loss": 2.0774,
3320
+ "step": 552
3321
+ },
3322
+ {
3323
+ "epoch": 0.09,
3324
+ "learning_rate": 0.0007385762711864407,
3325
+ "loss": 2.0977,
3326
+ "step": 553
3327
+ },
3328
+ {
3329
+ "epoch": 0.09,
3330
+ "learning_rate": 0.0007384406779661017,
3331
+ "loss": 2.0751,
3332
+ "step": 554
3333
+ },
3334
+ {
3335
+ "epoch": 0.09,
3336
+ "learning_rate": 0.0007383050847457628,
3337
+ "loss": 2.0928,
3338
+ "step": 555
3339
+ },
3340
+ {
3341
+ "epoch": 0.09,
3342
+ "learning_rate": 0.0007381694915254238,
3343
+ "loss": 2.0599,
3344
+ "step": 556
3345
+ },
3346
+ {
3347
+ "epoch": 0.09,
3348
+ "learning_rate": 0.0007380338983050849,
3349
+ "loss": 2.1053,
3350
+ "step": 557
3351
+ },
3352
+ {
3353
+ "epoch": 0.09,
3354
+ "learning_rate": 0.0007378983050847458,
3355
+ "loss": 2.0203,
3356
+ "step": 558
3357
+ },
3358
+ {
3359
+ "epoch": 0.09,
3360
+ "learning_rate": 0.0007377627118644068,
3361
+ "loss": 1.9643,
3362
+ "step": 559
3363
+ },
3364
+ {
3365
+ "epoch": 0.09,
3366
+ "learning_rate": 0.0007376271186440678,
3367
+ "loss": 2.1094,
3368
+ "step": 560
3369
+ },
3370
+ {
3371
+ "epoch": 0.09,
3372
+ "learning_rate": 0.0007374915254237288,
3373
+ "loss": 2.0148,
3374
+ "step": 561
3375
+ },
3376
+ {
3377
+ "epoch": 0.09,
3378
+ "learning_rate": 0.0007373559322033899,
3379
+ "loss": 2.0172,
3380
+ "step": 562
3381
+ },
3382
+ {
3383
+ "epoch": 0.09,
3384
+ "learning_rate": 0.0007372203389830509,
3385
+ "loss": 1.9883,
3386
+ "step": 563
3387
+ },
3388
+ {
3389
+ "epoch": 0.09,
3390
+ "learning_rate": 0.0007370847457627119,
3391
+ "loss": 2.0124,
3392
+ "step": 564
3393
+ },
3394
+ {
3395
+ "epoch": 0.09,
3396
+ "learning_rate": 0.0007369491525423729,
3397
+ "loss": 2.0097,
3398
+ "step": 565
3399
+ },
3400
+ {
3401
+ "epoch": 0.09,
3402
+ "learning_rate": 0.0007368135593220339,
3403
+ "loss": 2.023,
3404
+ "step": 566
3405
+ },
3406
+ {
3407
+ "epoch": 0.09,
3408
+ "learning_rate": 0.0007366779661016949,
3409
+ "loss": 2.079,
3410
+ "step": 567
3411
+ },
3412
+ {
3413
+ "epoch": 0.09,
3414
+ "learning_rate": 0.000736542372881356,
3415
+ "loss": 2.1124,
3416
+ "step": 568
3417
+ },
3418
+ {
3419
+ "epoch": 0.09,
3420
+ "learning_rate": 0.000736406779661017,
3421
+ "loss": 2.0094,
3422
+ "step": 569
3423
+ },
3424
+ {
3425
+ "epoch": 0.1,
3426
+ "learning_rate": 0.000736271186440678,
3427
+ "loss": 2.1219,
3428
+ "step": 570
3429
+ },
3430
+ {
3431
+ "epoch": 0.1,
3432
+ "learning_rate": 0.000736135593220339,
3433
+ "loss": 1.9903,
3434
+ "step": 571
3435
+ },
3436
+ {
3437
+ "epoch": 0.1,
3438
+ "learning_rate": 0.0007360000000000001,
3439
+ "loss": 1.9831,
3440
+ "step": 572
3441
+ },
3442
+ {
3443
+ "epoch": 0.1,
3444
+ "learning_rate": 0.0007358644067796611,
3445
+ "loss": 2.0807,
3446
+ "step": 573
3447
+ },
3448
+ {
3449
+ "epoch": 0.1,
3450
+ "learning_rate": 0.0007357288135593221,
3451
+ "loss": 2.062,
3452
+ "step": 574
3453
+ },
3454
+ {
3455
+ "epoch": 0.1,
3456
+ "learning_rate": 0.000735593220338983,
3457
+ "loss": 2.1067,
3458
+ "step": 575
3459
+ },
3460
+ {
3461
+ "epoch": 0.1,
3462
+ "learning_rate": 0.0007354576271186441,
3463
+ "loss": 2.0922,
3464
+ "step": 576
3465
+ },
3466
+ {
3467
+ "epoch": 0.1,
3468
+ "learning_rate": 0.0007353220338983051,
3469
+ "loss": 2.0415,
3470
+ "step": 577
3471
+ },
3472
+ {
3473
+ "epoch": 0.1,
3474
+ "learning_rate": 0.0007351864406779661,
3475
+ "loss": 1.9444,
3476
+ "step": 578
3477
+ },
3478
+ {
3479
+ "epoch": 0.1,
3480
+ "learning_rate": 0.0007350508474576272,
3481
+ "loss": 2.048,
3482
+ "step": 579
3483
+ },
3484
+ {
3485
+ "epoch": 0.1,
3486
+ "learning_rate": 0.0007349152542372882,
3487
+ "loss": 1.9961,
3488
+ "step": 580
3489
+ },
3490
+ {
3491
+ "epoch": 0.1,
3492
+ "learning_rate": 0.0007347796610169492,
3493
+ "loss": 2.083,
3494
+ "step": 581
3495
+ },
3496
+ {
3497
+ "epoch": 0.1,
3498
+ "learning_rate": 0.0007346440677966101,
3499
+ "loss": 2.0359,
3500
+ "step": 582
3501
+ },
3502
+ {
3503
+ "epoch": 0.1,
3504
+ "learning_rate": 0.0007345084745762712,
3505
+ "loss": 2.0901,
3506
+ "step": 583
3507
+ },
3508
+ {
3509
+ "epoch": 0.1,
3510
+ "learning_rate": 0.0007343728813559322,
3511
+ "loss": 2.0923,
3512
+ "step": 584
3513
+ },
3514
+ {
3515
+ "epoch": 0.1,
3516
+ "learning_rate": 0.0007342372881355933,
3517
+ "loss": 1.9558,
3518
+ "step": 585
3519
+ },
3520
+ {
3521
+ "epoch": 0.1,
3522
+ "learning_rate": 0.0007341016949152543,
3523
+ "loss": 2.0329,
3524
+ "step": 586
3525
+ },
3526
+ {
3527
+ "epoch": 0.1,
3528
+ "learning_rate": 0.0007339661016949154,
3529
+ "loss": 2.0818,
3530
+ "step": 587
3531
+ },
3532
+ {
3533
+ "epoch": 0.1,
3534
+ "learning_rate": 0.0007338305084745762,
3535
+ "loss": 2.0637,
3536
+ "step": 588
3537
+ },
3538
+ {
3539
+ "epoch": 0.1,
3540
+ "learning_rate": 0.0007336949152542373,
3541
+ "loss": 2.0038,
3542
+ "step": 589
3543
+ },
3544
+ {
3545
+ "epoch": 0.1,
3546
+ "learning_rate": 0.0007335593220338983,
3547
+ "loss": 2.0047,
3548
+ "step": 590
3549
+ },
3550
+ {
3551
+ "epoch": 0.1,
3552
+ "learning_rate": 0.0007334237288135594,
3553
+ "loss": 2.108,
3554
+ "step": 591
3555
+ },
3556
+ {
3557
+ "epoch": 0.1,
3558
+ "learning_rate": 0.0007332881355932204,
3559
+ "loss": 1.9784,
3560
+ "step": 592
3561
+ },
3562
+ {
3563
+ "epoch": 0.1,
3564
+ "learning_rate": 0.0007331525423728815,
3565
+ "loss": 2.032,
3566
+ "step": 593
3567
+ },
3568
+ {
3569
+ "epoch": 0.1,
3570
+ "learning_rate": 0.0007330169491525424,
3571
+ "loss": 2.032,
3572
+ "step": 594
3573
+ },
3574
+ {
3575
+ "epoch": 0.1,
3576
+ "learning_rate": 0.0007328813559322034,
3577
+ "loss": 2.0375,
3578
+ "step": 595
3579
+ },
3580
+ {
3581
+ "epoch": 0.1,
3582
+ "learning_rate": 0.0007327457627118644,
3583
+ "loss": 1.9258,
3584
+ "step": 596
3585
+ },
3586
+ {
3587
+ "epoch": 0.1,
3588
+ "learning_rate": 0.0007326101694915255,
3589
+ "loss": 2.0336,
3590
+ "step": 597
3591
+ },
3592
+ {
3593
+ "epoch": 0.1,
3594
+ "learning_rate": 0.0007324745762711865,
3595
+ "loss": 1.9972,
3596
+ "step": 598
3597
+ },
3598
+ {
3599
+ "epoch": 0.1,
3600
+ "learning_rate": 0.0007323389830508475,
3601
+ "loss": 2.0291,
3602
+ "step": 599
3603
+ },
3604
+ {
3605
+ "epoch": 0.1,
3606
+ "learning_rate": 0.0007322033898305085,
3607
+ "loss": 2.0544,
3608
+ "step": 600
3609
+ }
3610
+ ],
3611
+ "max_steps": 6000,
3612
+ "num_train_epochs": 9223372036854775807,
3613
+ "total_flos": 3.09706257494016e+17,
3614
+ "trial_name": null,
3615
+ "trial_params": null
3616
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67f743bb2a5a33d61bcab37d9a6d28ff91d1afc088eb9159ab7efb7ace667af5
3
+ size 4091
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e91453ea049f2d2cbc0ee4103c112ffa69d976b44b48ebf7ae0a9534698d3ad1
3
  size 990452905
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:874374987dbaa7df5e77d2d330b1bedbad6e692df7e14eef7d62092f8be4bf23
3
  size 990452905