pszemraj commited on
Commit
4fdd44f
1 Parent(s): 62f7489

update model with additional 1.8ish epochs training

Browse files
.gitattributes CHANGED
@@ -1,7 +1,6 @@
1
  *.7z filter=lfs diff=lfs merge=lfs -text
2
  *.arrow filter=lfs diff=lfs merge=lfs -text
3
  *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bin.* filter=lfs diff=lfs merge=lfs -text
5
  *.bz2 filter=lfs diff=lfs merge=lfs -text
6
  *.ftz filter=lfs diff=lfs merge=lfs -text
7
  *.gz filter=lfs diff=lfs merge=lfs -text
@@ -17,10 +16,11 @@
17
  *.pt filter=lfs diff=lfs merge=lfs -text
18
  *.pth filter=lfs diff=lfs merge=lfs -text
19
  *.rar filter=lfs diff=lfs merge=lfs -text
20
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
21
  *.tar.* filter=lfs diff=lfs merge=lfs -text
22
  *.tflite filter=lfs diff=lfs merge=lfs -text
23
  *.tgz filter=lfs diff=lfs merge=lfs -text
 
24
  *.xz filter=lfs diff=lfs merge=lfs -text
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
1
  *.7z filter=lfs diff=lfs merge=lfs -text
2
  *.arrow filter=lfs diff=lfs merge=lfs -text
3
  *.bin filter=lfs diff=lfs merge=lfs -text
 
4
  *.bz2 filter=lfs diff=lfs merge=lfs -text
5
  *.ftz filter=lfs diff=lfs merge=lfs -text
6
  *.gz filter=lfs diff=lfs merge=lfs -text
16
  *.pt filter=lfs diff=lfs merge=lfs -text
17
  *.pth filter=lfs diff=lfs merge=lfs -text
18
  *.rar filter=lfs diff=lfs merge=lfs -text
19
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
20
  *.tar.* filter=lfs diff=lfs merge=lfs -text
21
  *.tflite filter=lfs diff=lfs merge=lfs -text
22
  *.tgz filter=lfs diff=lfs merge=lfs -text
23
+ *.wasm filter=lfs diff=lfs merge=lfs -text
24
  *.xz filter=lfs diff=lfs merge=lfs -text
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
.gitignore ADDED
@@ -0,0 +1 @@
 
1
+ checkpoint-*/
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "allenai/led-large-16384",
3
  "_num_labels": 3,
4
  "activation_dropout": 0.0,
5
  "activation_function": "gelu",
@@ -52,7 +52,7 @@
52
  "length_penalty": 0.8,
53
  "max_decoder_position_embeddings": 1024,
54
  "max_encoder_position_embeddings": 16384,
55
- "max_length": 1024,
56
  "min_length": 32,
57
  "model_type": "led",
58
  "no_repeat_ngram_size": 3,
@@ -62,7 +62,7 @@
62
  "pad_token_id": 1,
63
  "prefix": " ",
64
  "torch_dtype": "float32",
65
- "transformers_version": "4.16.2",
66
  "use_cache": false,
67
  "vocab_size": 50265
68
  }
1
  {
2
+ "_name_or_path": "pszemraj/led-large-book-summary-11E",
3
  "_num_labels": 3,
4
  "activation_dropout": 0.0,
5
  "activation_function": "gelu",
52
  "length_penalty": 0.8,
53
  "max_decoder_position_embeddings": 1024,
54
  "max_encoder_position_embeddings": 16384,
55
+ "max_length": 512,
56
  "min_length": 32,
57
  "model_type": "led",
58
  "no_repeat_ngram_size": 3,
62
  "pad_token_id": 1,
63
  "prefix": " ",
64
  "torch_dtype": "float32",
65
+ "transformers_version": "4.19.2",
66
  "use_cache": false,
67
  "vocab_size": 50265
68
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6fd9ddcf9d526f237707eb0e06287ebc4b8b122e3624971dedca457623c67c26
3
  size 1839482407
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cae26c66ffc396d3f1b515359fef3cc2b4a080043dbfa56f7a968450c8ad1b27
3
  size 1839482407
tokenizer.json CHANGED
@@ -1,59 +1,74 @@
1
  {
2
  "version": "1.0",
3
- "truncation": null,
4
- "padding": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  "added_tokens": [
6
  {
7
  "id": 0,
8
- "special": true,
9
  "content": "<s>",
10
  "single_word": false,
11
  "lstrip": false,
12
  "rstrip": false,
13
- "normalized": true
 
14
  },
15
  {
16
  "id": 1,
17
- "special": true,
18
  "content": "<pad>",
19
  "single_word": false,
20
  "lstrip": false,
21
  "rstrip": false,
22
- "normalized": true
 
23
  },
24
  {
25
  "id": 2,
26
- "special": true,
27
  "content": "</s>",
28
  "single_word": false,
29
  "lstrip": false,
30
  "rstrip": false,
31
- "normalized": true
 
32
  },
33
  {
34
  "id": 3,
35
- "special": true,
36
  "content": "<unk>",
37
  "single_word": false,
38
  "lstrip": false,
39
  "rstrip": false,
40
- "normalized": true
 
41
  },
42
  {
43
  "id": 50264,
44
- "special": true,
45
  "content": "<mask>",
46
  "single_word": false,
47
  "lstrip": true,
48
  "rstrip": false,
49
- "normalized": true
 
50
  }
51
  ],
52
  "normalizer": null,
53
  "pre_tokenizer": {
54
  "type": "ByteLevel",
55
  "add_prefix_space": false,
56
- "trim_offsets": true
 
57
  },
58
  "post_processor": {
59
  "type": "RobertaProcessing",
@@ -71,7 +86,8 @@
71
  "decoder": {
72
  "type": "ByteLevel",
73
  "add_prefix_space": true,
74
- "trim_offsets": true
 
75
  },
76
  "model": {
77
  "type": "BPE",
1
  {
2
  "version": "1.0",
3
+ "truncation": {
4
+ "direction": "Right",
5
+ "max_length": 1024,
6
+ "strategy": "LongestFirst",
7
+ "stride": 0
8
+ },
9
+ "padding": {
10
+ "strategy": {
11
+ "Fixed": 1024
12
+ },
13
+ "direction": "Right",
14
+ "pad_to_multiple_of": null,
15
+ "pad_id": 1,
16
+ "pad_type_id": 0,
17
+ "pad_token": "<pad>"
18
+ },
19
  "added_tokens": [
20
  {
21
  "id": 0,
 
22
  "content": "<s>",
23
  "single_word": false,
24
  "lstrip": false,
25
  "rstrip": false,
26
+ "normalized": true,
27
+ "special": true
28
  },
29
  {
30
  "id": 1,
 
31
  "content": "<pad>",
32
  "single_word": false,
33
  "lstrip": false,
34
  "rstrip": false,
35
+ "normalized": true,
36
+ "special": true
37
  },
38
  {
39
  "id": 2,
 
40
  "content": "</s>",
41
  "single_word": false,
42
  "lstrip": false,
43
  "rstrip": false,
44
+ "normalized": true,
45
+ "special": true
46
  },
47
  {
48
  "id": 3,
 
49
  "content": "<unk>",
50
  "single_word": false,
51
  "lstrip": false,
52
  "rstrip": false,
53
+ "normalized": true,
54
+ "special": true
55
  },
56
  {
57
  "id": 50264,
 
58
  "content": "<mask>",
59
  "single_word": false,
60
  "lstrip": true,
61
  "rstrip": false,
62
+ "normalized": true,
63
+ "special": true
64
  }
65
  ],
66
  "normalizer": null,
67
  "pre_tokenizer": {
68
  "type": "ByteLevel",
69
  "add_prefix_space": false,
70
+ "trim_offsets": true,
71
+ "use_regex": true
72
  },
73
  "post_processor": {
74
  "type": "RobertaProcessing",
86
  "decoder": {
87
  "type": "ByteLevel",
88
  "add_prefix_space": true,
89
+ "trim_offsets": true,
90
+ "use_regex": true
91
  },
92
  "model": {
93
  "type": "BPE",
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "trim_offsets": true, "model_max_length": 16384, "special_tokens_map_file": "/root/.cache/huggingface/transformers/2ad921573d53ebf0c0450d63a211e61d8e328324e84830c365abff01f2d115f1.cb2244924ab24d706b02fd7fcedaea4531566537687a539ebb94db511fd122a0", "name_or_path": "allenai/led-large-16384", "tokenizer_class": "LEDTokenizer"}
1
+ {"errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "trim_offsets": true, "model_max_length": 16384, "special_tokens_map_file": "/root/.cache/huggingface/transformers/2ad921573d53ebf0c0450d63a211e61d8e328324e84830c365abff01f2d115f1.cb2244924ab24d706b02fd7fcedaea4531566537687a539ebb94db511fd122a0", "name_or_path": "pszemraj/led-large-book-summary-11E", "tokenizer_class": "LEDTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,1537 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9994069974303224,
5
+ "global_step": 1264,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 2.631578947368421e-06,
13
+ "loss": 1.0727,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 5.263157894736842e-06,
19
+ "loss": 0.8573,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 7.894736842105265e-06,
25
+ "loss": 1.0991,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 1.0526315789473684e-05,
31
+ "loss": 0.8893,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.04,
36
+ "learning_rate": 1.3157894736842108e-05,
37
+ "loss": 0.9549,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "learning_rate": 1.578947368421053e-05,
43
+ "loss": 0.9953,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.06,
48
+ "learning_rate": 1.8421052631578947e-05,
49
+ "loss": 1.0163,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.06,
54
+ "learning_rate": 1.9999868674866755e-05,
55
+ "loss": 0.9562,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.07,
60
+ "learning_rate": 1.9998391306730024e-05,
61
+ "loss": 1.0517,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.08,
66
+ "learning_rate": 1.9995272657365104e-05,
67
+ "loss": 0.9458,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.09,
72
+ "learning_rate": 1.9990513238712407e-05,
73
+ "loss": 0.9163,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.09,
78
+ "learning_rate": 1.998411383205207e-05,
79
+ "loss": 0.8653,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.1,
84
+ "learning_rate": 1.9976075487875692e-05,
85
+ "loss": 1.063,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.11,
90
+ "learning_rate": 1.9966399525713882e-05,
91
+ "loss": 0.8901,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.12,
96
+ "learning_rate": 1.9955087533919662e-05,
97
+ "loss": 1.0556,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.13,
102
+ "learning_rate": 1.994214136940773e-05,
103
+ "loss": 0.9434,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 1.9927563157349646e-05,
109
+ "loss": 0.9989,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.14,
114
+ "learning_rate": 1.9911355290824955e-05,
115
+ "loss": 0.9829,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.15,
120
+ "learning_rate": 1.9893520430428378e-05,
121
+ "loss": 0.9988,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.16,
126
+ "learning_rate": 1.987406150383304e-05,
127
+ "loss": 1.0073,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.17,
132
+ "learning_rate": 1.98529817053099e-05,
133
+ "loss": 0.8863,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.17,
138
+ "learning_rate": 1.983028449520338e-05,
139
+ "loss": 0.8015,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.18,
144
+ "learning_rate": 1.980597359936335e-05,
145
+ "loss": 1.0477,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.19,
150
+ "learning_rate": 1.9780053008533486e-05,
151
+ "loss": 0.961,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.2,
156
+ "learning_rate": 1.97525269776962e-05,
157
+ "loss": 0.9476,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.21,
162
+ "learning_rate": 1.9723400025374168e-05,
163
+ "loss": 1.1,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.21,
168
+ "learning_rate": 1.969267693288855e-05,
169
+ "loss": 0.946,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.22,
174
+ "learning_rate": 1.9660362743574163e-05,
175
+ "loss": 0.946,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.23,
180
+ "learning_rate": 1.9626462761951583e-05,
181
+ "loss": 0.9368,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.24,
186
+ "learning_rate": 1.959098255285636e-05,
187
+ "loss": 1.0338,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.25,
192
+ "learning_rate": 1.9553927940525557e-05,
193
+ "loss": 1.0359,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.25,
198
+ "learning_rate": 1.9515305007641653e-05,
199
+ "loss": 1.0189,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.26,
204
+ "learning_rate": 1.9475120094334046e-05,
205
+ "loss": 1.0388,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.27,
210
+ "learning_rate": 1.9433379797138314e-05,
211
+ "loss": 1.0733,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.28,
216
+ "learning_rate": 1.939009096791333e-05,
217
+ "loss": 0.9984,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.28,
222
+ "learning_rate": 1.9345260712716517e-05,
223
+ "loss": 1.0221,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.29,
228
+ "learning_rate": 1.9298896390637363e-05,
229
+ "loss": 1.2205,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 0.3,
234
+ "learning_rate": 1.9251005612589382e-05,
235
+ "loss": 0.9938,
236
+ "step": 190
237
+ },
238
+ {
239
+ "epoch": 0.31,
240
+ "learning_rate": 1.9201596240060737e-05,
241
+ "loss": 1.0059,
242
+ "step": 195
243
+ },
244
+ {
245
+ "epoch": 0.32,
246
+ "learning_rate": 1.9150676383823775e-05,
247
+ "loss": 1.0072,
248
+ "step": 200
249
+ },
250
+ {
251
+ "epoch": 0.32,
252
+ "learning_rate": 1.9098254402603573e-05,
253
+ "loss": 0.8886,
254
+ "step": 205
255
+ },
256
+ {
257
+ "epoch": 0.33,
258
+ "learning_rate": 1.904433890170583e-05,
259
+ "loss": 1.1496,
260
+ "step": 210
261
+ },
262
+ {
263
+ "epoch": 0.34,
264
+ "learning_rate": 1.898893873160428e-05,
265
+ "loss": 0.8883,
266
+ "step": 215
267
+ },
268
+ {
269
+ "epoch": 0.35,
270
+ "learning_rate": 1.893206298648781e-05,
271
+ "loss": 0.9564,
272
+ "step": 220
273
+ },
274
+ {
275
+ "epoch": 0.36,
276
+ "learning_rate": 1.887372100276764e-05,
277
+ "loss": 0.9264,
278
+ "step": 225
279
+ },
280
+ {
281
+ "epoch": 0.36,
282
+ "learning_rate": 1.8813922357544713e-05,
283
+ "loss": 1.1487,
284
+ "step": 230
285
+ },
286
+ {
287
+ "epoch": 0.37,
288
+ "learning_rate": 1.875267686703754e-05,
289
+ "loss": 1.0543,
290
+ "step": 235
291
+ },
292
+ {
293
+ "epoch": 0.38,
294
+ "learning_rate": 1.8689994584970843e-05,
295
+ "loss": 1.0098,
296
+ "step": 240
297
+ },
298
+ {
299
+ "epoch": 0.39,
300
+ "learning_rate": 1.8625885800925193e-05,
301
+ "loss": 0.915,
302
+ "step": 245
303
+ },
304
+ {
305
+ "epoch": 0.4,
306
+ "learning_rate": 1.8560361038647917e-05,
307
+ "loss": 1.0092,
308
+ "step": 250
309
+ },
310
+ {
311
+ "epoch": 0.4,
312
+ "learning_rate": 1.8493431054325583e-05,
313
+ "loss": 1.0312,
314
+ "step": 255
315
+ },
316
+ {
317
+ "epoch": 0.41,
318
+ "learning_rate": 1.8425106834818336e-05,
319
+ "loss": 1.0963,
320
+ "step": 260
321
+ },
322
+ {
323
+ "epoch": 0.42,
324
+ "learning_rate": 1.8355399595856326e-05,
325
+ "loss": 0.8076,
326
+ "step": 265
327
+ },
328
+ {
329
+ "epoch": 0.43,
330
+ "learning_rate": 1.8284320780198624e-05,
331
+ "loss": 0.8218,
332
+ "step": 270
333
+ },
334
+ {
335
+ "epoch": 0.43,
336
+ "learning_rate": 1.8211882055754824e-05,
337
+ "loss": 0.9424,
338
+ "step": 275
339
+ },
340
+ {
341
+ "epoch": 0.44,
342
+ "learning_rate": 1.8138095313669705e-05,
343
+ "loss": 1.0962,
344
+ "step": 280
345
+ },
346
+ {
347
+ "epoch": 0.45,
348
+ "learning_rate": 1.8062972666371258e-05,
349
+ "loss": 1.0918,
350
+ "step": 285
351
+ },
352
+ {
353
+ "epoch": 0.46,
354
+ "learning_rate": 1.798652644558236e-05,
355
+ "loss": 1.065,
356
+ "step": 290
357
+ },
358
+ {
359
+ "epoch": 0.47,
360
+ "learning_rate": 1.790876920029647e-05,
361
+ "loss": 0.9168,
362
+ "step": 295
363
+ },
364
+ {
365
+ "epoch": 0.47,
366
+ "learning_rate": 1.7829713694717665e-05,
367
+ "loss": 1.0295,
368
+ "step": 300
369
+ },
370
+ {
371
+ "epoch": 0.48,
372
+ "learning_rate": 1.774937290616533e-05,
373
+ "loss": 0.9455,
374
+ "step": 305
375
+ },
376
+ {
377
+ "epoch": 0.49,
378
+ "learning_rate": 1.7667760022943864e-05,
379
+ "loss": 1.0272,
380
+ "step": 310
381
+ },
382
+ {
383
+ "epoch": 0.5,
384
+ "learning_rate": 1.7584888442177774e-05,
385
+ "loss": 1.127,
386
+ "step": 315
387
+ },
388
+ {
389
+ "epoch": 0.51,
390
+ "learning_rate": 1.7500771767612473e-05,
391
+ "loss": 1.0332,
392
+ "step": 320
393
+ },
394
+ {
395
+ "epoch": 0.51,
396
+ "learning_rate": 1.7415423807381162e-05,
397
+ "loss": 1.0603,
398
+ "step": 325
399
+ },
400
+ {
401
+ "epoch": 0.52,
402
+ "learning_rate": 1.7328858571738157e-05,
403
+ "loss": 1.1075,
404
+ "step": 330
405
+ },
406
+ {
407
+ "epoch": 0.53,
408
+ "learning_rate": 1.7241090270759055e-05,
409
+ "loss": 1.1279,
410
+ "step": 335
411
+ },
412
+ {
413
+ "epoch": 0.54,
414
+ "learning_rate": 1.715213331200807e-05,
415
+ "loss": 1.1037,
416
+ "step": 340
417
+ },
418
+ {
419
+ "epoch": 0.55,
420
+ "learning_rate": 1.7062002298172984e-05,
421
+ "loss": 0.9818,
422
+ "step": 345
423
+ },
424
+ {
425
+ "epoch": 0.55,
426
+ "learning_rate": 1.697071202466803e-05,
427
+ "loss": 0.9672,
428
+ "step": 350
429
+ },
430
+ {
431
+ "epoch": 0.56,
432
+ "learning_rate": 1.687827747720517e-05,
433
+ "loss": 0.8006,
434
+ "step": 355
435
+ },
436
+ {
437
+ "epoch": 0.57,
438
+ "learning_rate": 1.6784713829334124e-05,
439
+ "loss": 0.9823,
440
+ "step": 360
441
+ },
442
+ {
443
+ "epoch": 0.58,
444
+ "learning_rate": 1.6690036439951552e-05,
445
+ "loss": 1.0844,
446
+ "step": 365
447
+ },
448
+ {
449
+ "epoch": 0.59,
450
+ "learning_rate": 1.6594260850779837e-05,
451
+ "loss": 1.0858,
452
+ "step": 370
453
+ },
454
+ {
455
+ "epoch": 0.59,
456
+ "learning_rate": 1.6497402783815834e-05,
457
+ "loss": 0.9117,
458
+ "step": 375
459
+ },
460
+ {
461
+ "epoch": 0.6,
462
+ "learning_rate": 1.6399478138750015e-05,
463
+ "loss": 0.9313,
464
+ "step": 380
465
+ },
466
+ {
467
+ "epoch": 0.61,
468
+ "learning_rate": 1.63005029903565e-05,
469
+ "loss": 0.9776,
470
+ "step": 385
471
+ },
472
+ {
473
+ "epoch": 0.62,
474
+ "learning_rate": 1.620049358585427e-05,
475
+ "loss": 1.1445,
476
+ "step": 390
477
+ },
478
+ {
479
+ "epoch": 0.62,
480
+ "learning_rate": 1.609946634224015e-05,
481
+ "loss": 1.0567,
482
+ "step": 395
483
+ },
484
+ {
485
+ "epoch": 0.63,
486
+ "learning_rate": 1.5997437843593856e-05,
487
+ "loss": 0.9138,
488
+ "step": 400
489
+ },
490
+ {
491
+ "epoch": 0.64,
492
+ "learning_rate": 1.5894424838355654e-05,
493
+ "loss": 1.1003,
494
+ "step": 405
495
+ },
496
+ {
497
+ "epoch": 0.65,
498
+ "learning_rate": 1.5790444236577028e-05,
499
+ "loss": 1.0624,
500
+ "step": 410
501
+ },
502
+ {
503
+ "epoch": 0.66,
504
+ "learning_rate": 1.568551310714482e-05,
505
+ "loss": 0.8902,
506
+ "step": 415
507
+ },
508
+ {
509
+ "epoch": 0.66,
510
+ "learning_rate": 1.557964867497929e-05,
511
+ "loss": 1.0329,
512
+ "step": 420
513
+ },
514
+ {
515
+ "epoch": 0.67,
516
+ "learning_rate": 1.5472868318206566e-05,
517
+ "loss": 1.0023,
518
+ "step": 425
519
+ },
520
+ {
521
+ "epoch": 0.68,
522
+ "learning_rate": 1.5365189565305957e-05,
523
+ "loss": 0.9393,
524
+ "step": 430
525
+ },
526
+ {
527
+ "epoch": 0.69,
528
+ "learning_rate": 1.5256630092232567e-05,
529
+ "loss": 0.9747,
530
+ "step": 435
531
+ },
532
+ {
533
+ "epoch": 0.7,
534
+ "learning_rate": 1.5147207719515692e-05,
535
+ "loss": 1.0078,
536
+ "step": 440
537
+ },
538
+ {
539
+ "epoch": 0.7,
540
+ "learning_rate": 1.5036940409333533e-05,
541
+ "loss": 0.9576,
542
+ "step": 445
543
+ },
544
+ {
545
+ "epoch": 0.71,
546
+ "learning_rate": 1.4925846262564592e-05,
547
+ "loss": 0.992,
548
+ "step": 450
549
+ },
550
+ {
551
+ "epoch": 0.72,
552
+ "learning_rate": 1.4813943515816344e-05,
553
+ "loss": 1.0192,
554
+ "step": 455
555
+ },
556
+ {
557
+ "epoch": 0.73,
558
+ "learning_rate": 1.4701250538431617e-05,
559
+ "loss": 1.0154,
560
+ "step": 460
561
+ },
562
+ {
563
+ "epoch": 0.74,
564
+ "learning_rate": 1.4587785829473173e-05,
565
+ "loss": 1.1043,
566
+ "step": 465
567
+ },
568
+ {
569
+ "epoch": 0.74,
570
+ "learning_rate": 1.4473568014687018e-05,
571
+ "loss": 0.8351,
572
+ "step": 470
573
+ },
574
+ {
575
+ "epoch": 0.75,
576
+ "learning_rate": 1.4358615843444876e-05,
577
+ "loss": 1.0276,
578
+ "step": 475
579
+ },
580
+ {
581
+ "epoch": 0.76,
582
+ "learning_rate": 1.4242948185666419e-05,
583
+ "loss": 0.9423,
584
+ "step": 480
585
+ },
586
+ {
587
+ "epoch": 0.77,
588
+ "learning_rate": 1.4126584028721677e-05,
589
+ "loss": 0.9598,
590
+ "step": 485
591
+ },
592
+ {
593
+ "epoch": 0.77,
594
+ "learning_rate": 1.4009542474314173e-05,
595
+ "loss": 1.1755,
596
+ "step": 490
597
+ },
598
+ {
599
+ "epoch": 0.78,
600
+ "learning_rate": 1.3891842735345285e-05,
601
+ "loss": 1.1018,
602
+ "step": 495
603
+ },
604
+ {
605
+ "epoch": 0.79,
606
+ "learning_rate": 1.3773504132760379e-05,
607
+ "loss": 1.0445,
608
+ "step": 500
609
+ },
610
+ {
611
+ "epoch": 0.8,
612
+ "learning_rate": 1.3654546092377166e-05,
613
+ "loss": 0.9674,
614
+ "step": 505
615
+ },
616
+ {
617
+ "epoch": 0.81,
618
+ "learning_rate": 1.3534988141696891e-05,
619
+ "loss": 1.0473,
620
+ "step": 510
621
+ },
622
+ {
623
+ "epoch": 0.81,
624
+ "learning_rate": 1.3414849906698788e-05,
625
+ "loss": 0.9346,
626
+ "step": 515
627
+ },
628
+ {
629
+ "epoch": 0.82,
630
+ "learning_rate": 1.3294151108618379e-05,
631
+ "loss": 0.9643,
632
+ "step": 520
633
+ },
634
+ {
635
+ "epoch": 0.83,
636
+ "learning_rate": 1.3172911560710167e-05,
637
+ "loss": 1.113,
638
+ "step": 525
639
+ },
640
+ {
641
+ "epoch": 0.84,
642
+ "learning_rate": 1.3051151164995188e-05,
643
+ "loss": 1.0155,
644
+ "step": 530
645
+ },
646
+ {
647
+ "epoch": 0.85,
648
+ "learning_rate": 1.2928889908994003e-05,
649
+ "loss": 0.9675,
650
+ "step": 535
651
+ },
652
+ {
653
+ "epoch": 0.85,
654
+ "learning_rate": 1.280614786244566e-05,
655
+ "loss": 0.8424,
656
+ "step": 540
657
+ },
658
+ {
659
+ "epoch": 0.86,
660
+ "learning_rate": 1.2682945174013148e-05,
661
+ "loss": 1.1247,
662
+ "step": 545
663
+ },
664
+ {
665
+ "epoch": 0.87,
666
+ "learning_rate": 1.2559302067975914e-05,
667
+ "loss": 0.9641,
668
+ "step": 550
669
+ },
670
+ {
671
+ "epoch": 0.88,
672
+ "learning_rate": 1.243523884090995e-05,
673
+ "loss": 1.0329,
674
+ "step": 555
675
+ },
676
+ {
677
+ "epoch": 0.89,
678
+ "learning_rate": 1.2310775858356017e-05,
679
+ "loss": 1.0533,
680
+ "step": 560
681
+ },
682
+ {
683
+ "epoch": 0.89,
684
+ "learning_rate": 1.2185933551476545e-05,
685
+ "loss": 1.0216,
686
+ "step": 565
687
+ },
688
+ {
689
+ "epoch": 0.9,
690
+ "learning_rate": 1.2060732413701773e-05,
691
+ "loss": 1.0346,
692
+ "step": 570
693
+ },
694
+ {
695
+ "epoch": 0.91,
696
+ "learning_rate": 1.1935192997365666e-05,
697
+ "loss": 0.9749,
698
+ "step": 575
699
+ },
700
+ {
701
+ "epoch": 0.92,
702
+ "learning_rate": 1.1809335910332136e-05,
703
+ "loss": 1.0901,
704
+ "step": 580
705
+ },
706
+ {
707
+ "epoch": 0.93,
708
+ "learning_rate": 1.1683181812612186e-05,
709
+ "loss": 1.0043,
710
+ "step": 585
711
+ },
712
+ {
713
+ "epoch": 0.93,
714
+ "learning_rate": 1.1556751412972462e-05,
715
+ "loss": 0.9162,
716
+ "step": 590
717
+ },
718
+ {
719
+ "epoch": 0.94,
720
+ "learning_rate": 1.1430065465535827e-05,
721
+ "loss": 0.9228,
722
+ "step": 595
723
+ },
724
+ {
725
+ "epoch": 0.95,
726
+ "learning_rate": 1.1303144766374476e-05,
727
+ "loss": 0.9568,
728
+ "step": 600
729
+ },
730
+ {
731
+ "epoch": 0.96,
732
+ "learning_rate": 1.1176010150096158e-05,
733
+ "loss": 1.063,
734
+ "step": 605
735
+ },
736
+ {
737
+ "epoch": 0.96,
738
+ "learning_rate": 1.104868248642408e-05,
739
+ "loss": 0.9938,
740
+ "step": 610
741
+ },
742
+ {
743
+ "epoch": 0.97,
744
+ "learning_rate": 1.092118267677106e-05,
745
+ "loss": 1.1056,
746
+ "step": 615
747
+ },
748
+ {
749
+ "epoch": 0.98,
750
+ "learning_rate": 1.0793531650808469e-05,
751
+ "loss": 0.9269,
752
+ "step": 620
753
+ },
754
+ {
755
+ "epoch": 0.99,
756
+ "learning_rate": 1.0665750363030498e-05,
757
+ "loss": 1.0452,
758
+ "step": 625
759
+ },
760
+ {
761
+ "epoch": 1.0,
762
+ "learning_rate": 1.0537859789314424e-05,
763
+ "loss": 0.8855,
764
+ "step": 630
765
+ },
766
+ {
767
+ "epoch": 1.0,
768
+ "learning_rate": 1.0409880923477293e-05,
769
+ "loss": 1.0583,
770
+ "step": 635
771
+ },
772
+ {
773
+ "epoch": 1.01,
774
+ "learning_rate": 1.028183477382971e-05,
775
+ "loss": 0.8533,
776
+ "step": 640
777
+ },
778
+ {
779
+ "epoch": 1.02,
780
+ "learning_rate": 1.0153742359727226e-05,
781
+ "loss": 0.942,
782
+ "step": 645
783
+ },
784
+ {
785
+ "epoch": 1.03,
786
+ "learning_rate": 1.0025624708119901e-05,
787
+ "loss": 1.0044,
788
+ "step": 650
789
+ },
790
+ {
791
+ "epoch": 1.04,
792
+ "learning_rate": 9.897502850100648e-06,
793
+ "loss": 0.9652,
794
+ "step": 655
795
+ },
796
+ {
797
+ "epoch": 1.04,
798
+ "learning_rate": 9.76939781745289e-06,
799
+ "loss": 1.0202,
800
+ "step": 660
801
+ },
802
+ {
803
+ "epoch": 1.05,
804
+ "learning_rate": 9.641330639198083e-06,
805
+ "loss": 0.9401,
806
+ "step": 665
807
+ },
808
+ {
809
+ "epoch": 1.06,
810
+ "learning_rate": 9.513322338143714e-06,
811
+ "loss": 0.8343,
812
+ "step": 670
813
+ },
814
+ {
815
+ "epoch": 1.07,
816
+ "learning_rate": 9.385393927432307e-06,
817
+ "loss": 0.8904,
818
+ "step": 675
819
+ },
820
+ {
821
+ "epoch": 1.08,
822
+ "learning_rate": 9.257566407092032e-06,
823
+ "loss": 0.9143,
824
+ "step": 680
825
+ },
826
+ {
827
+ "epoch": 1.08,
828
+ "learning_rate": 9.129860760589441e-06,
829
+ "loss": 0.8408,
830
+ "step": 685
831
+ },
832
+ {
833
+ "epoch": 1.09,
834
+ "learning_rate": 9.002297951384945e-06,
835
+ "loss": 0.9426,
836
+ "step": 690
837
+ },
838
+ {
839
+ "epoch": 1.1,
840
+ "learning_rate": 8.874898919491564e-06,
841
+ "loss": 0.9476,
842
+ "step": 695
843
+ },
844
+ {
845
+ "epoch": 1.11,
846
+ "learning_rate": 8.74768457803754e-06,
847
+ "loss": 0.9992,
848
+ "step": 700
849
+ },
850
+ {
851
+ "epoch": 1.12,
852
+ "learning_rate": 8.62067580983333e-06,
853
+ "loss": 0.8491,
854
+ "step": 705
855
+ },
856
+ {
857
+ "epoch": 1.12,
858
+ "learning_rate": 8.493893463943617e-06,
859
+ "loss": 0.8829,
860
+ "step": 710
861
+ },
862
+ {
863
+ "epoch": 1.13,
864
+ "learning_rate": 8.367358352264834e-06,
865
+ "loss": 0.7933,
866
+ "step": 715
867
+ },
868
+ {
869
+ "epoch": 1.14,
870
+ "learning_rate": 8.241091246108796e-06,
871
+ "loss": 0.9976,
872
+ "step": 720
873
+ },
874
+ {
875
+ "epoch": 1.15,
876
+ "learning_rate": 8.115112872793006e-06,
877
+ "loss": 0.9947,
878
+ "step": 725
879
+ },
880
+ {
881
+ "epoch": 1.15,
882
+ "learning_rate": 7.989443912238151e-06,
883
+ "loss": 0.9171,
884
+ "step": 730
885
+ },
886
+ {
887
+ "epoch": 1.16,
888
+ "learning_rate": 7.864104993573422e-06,
889
+ "loss": 0.9348,
890
+ "step": 735
891
+ },
892
+ {
893
+ "epoch": 1.17,
894
+ "learning_rate": 7.73911669175013e-06,
895
+ "loss": 0.9357,
896
+ "step": 740
897
+ },
898
+ {
899
+ "epoch": 1.18,
900
+ "learning_rate": 7.614499524164251e-06,
901
+ "loss": 0.9273,
902
+ "step": 745
903
+ },
904
+ {
905
+ "epoch": 1.19,
906
+ "learning_rate": 7.490273947288389e-06,
907
+ "loss": 0.8991,
908
+ "step": 750
909
+ },
910
+ {
911
+ "epoch": 1.19,
912
+ "learning_rate": 7.366460353313762e-06,
913
+ "loss": 0.8373,
914
+ "step": 755
915
+ },
916
+ {
917
+ "epoch": 1.2,
918
+ "learning_rate": 7.2430790668027274e-06,
919
+ "loss": 0.8316,
920
+ "step": 760
921
+ },
922
+ {
923
+ "epoch": 1.21,
924
+ "learning_rate": 7.120150341352413e-06,
925
+ "loss": 0.7532,
926
+ "step": 765
927
+ },
928
+ {
929
+ "epoch": 1.22,
930
+ "learning_rate": 6.99769435627e-06,
931
+ "loss": 0.8851,
932
+ "step": 770
933
+ },
934
+ {
935
+ "epoch": 1.23,
936
+ "learning_rate": 6.875731213260193e-06,
937
+ "loss": 0.9678,
938
+ "step": 775
939
+ },
940
+ {
941
+ "epoch": 1.23,
942
+ "learning_rate": 6.754280933125441e-06,
943
+ "loss": 0.9968,
944
+ "step": 780
945
+ },
946
+ {
947
+ "epoch": 1.24,
948
+ "learning_rate": 6.633363452479431e-06,
949
+ "loss": 0.9126,
950
+ "step": 785
951
+ },
952
+ {
953
+ "epoch": 1.25,
954
+ "learning_rate": 6.512998620474396e-06,
955
+ "loss": 0.8765,
956
+ "step": 790
957
+ },
958
+ {
959
+ "epoch": 1.26,
960
+ "learning_rate": 6.393206195542791e-06,
961
+ "loss": 0.8761,
962
+ "step": 795
963
+ },
964
+ {
965
+ "epoch": 1.27,
966
+ "learning_rate": 6.27400584215386e-06,
967
+ "loss": 0.99,
968
+ "step": 800
969
+ },
970
+ {
971
+ "epoch": 1.27,
972
+ "learning_rate": 6.155417127585617e-06,
973
+ "loss": 0.7691,
974
+ "step": 805
975
+ },
976
+ {
977
+ "epoch": 1.28,
978
+ "learning_rate": 6.037459518712796e-06,
979
+ "loss": 0.7794,
980
+ "step": 810
981
+ },
982
+ {
983
+ "epoch": 1.29,
984
+ "learning_rate": 5.920152378811268e-06,
985
+ "loss": 0.9978,
986
+ "step": 815
987
+ },
988
+ {
989
+ "epoch": 1.3,
990
+ "learning_rate": 5.803514964379482e-06,
991
+ "loss": 0.8165,
992
+ "step": 820
993
+ },
994
+ {
995
+ "epoch": 1.31,
996
+ "learning_rate": 5.68756642197741e-06,
997
+ "loss": 0.8918,
998
+ "step": 825
999
+ },
1000
+ {
1001
+ "epoch": 1.31,
1002
+ "learning_rate": 5.572325785083563e-06,
1003
+ "loss": 0.9572,
1004
+ "step": 830
1005
+ },
1006
+ {
1007
+ "epoch": 1.32,
1008
+ "learning_rate": 5.457811970970564e-06,
1009
+ "loss": 0.9112,
1010
+ "step": 835
1011
+ },
1012
+ {
1013
+ "epoch": 1.33,
1014
+ "learning_rate": 5.3440437775997636e-06,
1015
+ "loss": 0.9659,
1016
+ "step": 840
1017
+ },
1018
+ {
1019
+ "epoch": 1.34,
1020
+ "learning_rate": 5.231039880535511e-06,
1021
+ "loss": 0.9516,
1022
+ "step": 845
1023
+ },
1024
+ {
1025
+ "epoch": 1.34,
1026
+ "learning_rate": 5.118818829879442e-06,
1027
+ "loss": 1.076,
1028
+ "step": 850
1029
+ },
1030
+ {
1031
+ "epoch": 1.35,
1032
+ "learning_rate": 5.0073990472254075e-06,
1033
+ "loss": 0.8136,
1034
+ "step": 855
1035
+ },
1036
+ {
1037
+ "epoch": 1.36,
1038
+ "learning_rate": 4.8967988226354945e-06,
1039
+ "loss": 0.9711,
1040
+ "step": 860
1041
+ },
1042
+ {
1043
+ "epoch": 1.37,
1044
+ "learning_rate": 4.787036311637609e-06,
1045
+ "loss": 0.8879,
1046
+ "step": 865
1047
+ },
1048
+ {
1049
+ "epoch": 1.38,
1050
+ "learning_rate": 4.678129532245189e-06,
1051
+ "loss": 1.0423,
1052
+ "step": 870
1053
+ },
1054
+ {
1055
+ "epoch": 1.38,
1056
+ "learning_rate": 4.570096361999445e-06,
1057
+ "loss": 0.892,
1058
+ "step": 875
1059
+ },
1060
+ {
1061
+ "epoch": 1.39,
1062
+ "learning_rate": 4.462954535034692e-06,
1063
+ "loss": 0.9188,
1064
+ "step": 880
1065
+ },
1066
+ {
1067
+ "epoch": 1.4,
1068
+ "learning_rate": 4.356721639167202e-06,
1069
+ "loss": 0.8706,
1070
+ "step": 885
1071
+ },
1072
+ {
1073
+ "epoch": 1.41,
1074
+ "learning_rate": 4.251415113008096e-06,
1075
+ "loss": 0.8926,
1076
+ "step": 890
1077
+ },
1078
+ {
1079
+ "epoch": 1.42,
1080
+ "learning_rate": 4.147052243100706e-06,
1081
+ "loss": 0.9126,
1082
+ "step": 895
1083
+ },
1084
+ {
1085
+ "epoch": 1.42,
1086
+ "learning_rate": 4.043650161082913e-06,
1087
+ "loss": 0.9201,
1088
+ "step": 900
1089
+ },
1090
+ {
1091
+ "epoch": 1.43,
1092
+ "learning_rate": 3.941225840874925e-06,
1093
+ "loss": 0.8758,
1094
+ "step": 905
1095
+ },
1096
+ {
1097
+ "epoch": 1.44,
1098
+ "learning_rate": 3.839796095892905e-06,
1099
+ "loss": 0.8818,
1100
+ "step": 910
1101
+ },
1102
+ {
1103
+ "epoch": 1.45,
1104
+ "learning_rate": 3.7393775762889963e-06,
1105
+ "loss": 0.9018,
1106
+ "step": 915
1107
+ },
1108
+ {
1109
+ "epoch": 1.46,
1110
+ "learning_rate": 3.639986766218112e-06,
1111
+ "loss": 0.8346,
1112
+ "step": 920
1113
+ },
1114
+ {
1115
+ "epoch": 1.46,
1116
+ "learning_rate": 3.541639981131996e-06,
1117
+ "loss": 0.9619,
1118
+ "step": 925
1119
+ },
1120
+ {
1121
+ "epoch": 1.47,
1122
+ "learning_rate": 3.4443533651009474e-06,
1123
+ "loss": 1.0202,
1124
+ "step": 930
1125
+ },
1126
+ {
1127
+ "epoch": 1.48,
1128
+ "learning_rate": 3.348142888163726e-06,
1129
+ "loss": 0.9186,
1130
+ "step": 935
1131
+ },
1132
+ {
1133
+ "epoch": 1.49,
1134
+ "learning_rate": 3.2530243437059773e-06,
1135
+ "loss": 0.9326,
1136
+ "step": 940
1137
+ },
1138
+ {
1139
+ "epoch": 1.49,
1140
+ "learning_rate": 3.1590133458676787e-06,
1141
+ "loss": 0.8347,
1142
+ "step": 945
1143
+ },
1144
+ {
1145
+ "epoch": 1.5,
1146
+ "learning_rate": 3.066125326980027e-06,
1147
+ "loss": 0.8914,
1148
+ "step": 950
1149
+ },
1150
+ {
1151
+ "epoch": 1.51,
1152
+ "learning_rate": 2.9743755350321213e-06,
1153
+ "loss": 0.8869,
1154
+ "step": 955
1155
+ },
1156
+ {
1157
+ "epoch": 1.52,
1158
+ "learning_rate": 2.8837790311679625e-06,
1159
+ "loss": 0.9833,
1160
+ "step": 960
1161
+ },
1162
+ {
1163
+ "epoch": 1.53,
1164
+ "learning_rate": 2.7943506872140844e-06,
1165
+ "loss": 0.8893,
1166
+ "step": 965
1167
+ },
1168
+ {
1169
+ "epoch": 1.53,
1170
+ "learning_rate": 2.7061051832382836e-06,
1171
+ "loss": 0.8616,
1172
+ "step": 970
1173
+ },
1174
+ {
1175
+ "epoch": 1.54,
1176
+ "learning_rate": 2.6190570051398035e-06,
1177
+ "loss": 0.9909,
1178
+ "step": 975
1179
+ },
1180
+ {
1181
+ "epoch": 1.55,
1182
+ "learning_rate": 2.5332204422714368e-06,
1183
+ "loss": 0.9515,
1184
+ "step": 980
1185
+ },
1186
+ {
1187
+ "epoch": 1.56,
1188
+ "learning_rate": 2.4486095850938352e-06,
1189
+ "loss": 1.0862,
1190
+ "step": 985
1191
+ },
1192
+ {
1193
+ "epoch": 1.57,
1194
+ "learning_rate": 2.365238322862511e-06,
1195
+ "loss": 0.9123,
1196
+ "step": 990
1197
+ },
1198
+ {
1199
+ "epoch": 1.57,
1200
+ "learning_rate": 2.2831203413478555e-06,
1201
+ "loss": 0.8965,
1202
+ "step": 995
1203
+ },
1204
+ {
1205
+ "epoch": 1.58,
1206
+ "learning_rate": 2.202269120588546e-06,
1207
+ "loss": 0.9261,
1208
+ "step": 1000
1209
+ },
1210
+ {
1211
+ "epoch": 1.59,
1212
+ "learning_rate": 2.122697932678748e-06,
1213
+ "loss": 0.9044,
1214
+ "step": 1005
1215
+ },
1216
+ {
1217
+ "epoch": 1.6,
1218
+ "learning_rate": 2.0444198395894332e-06,
1219
+ "loss": 0.923,
1220
+ "step": 1010
1221
+ },
1222
+ {
1223
+ "epoch": 1.61,
1224
+ "learning_rate": 1.9674476910242055e-06,
1225
+ "loss": 0.9581,
1226
+ "step": 1015
1227
+ },
1228
+ {
1229
+ "epoch": 1.61,
1230
+ "learning_rate": 1.891794122309949e-06,
1231
+ "loss": 1.0578,
1232
+ "step": 1020
1233
+ },
1234
+ {
1235
+ "epoch": 1.62,
1236
+ "learning_rate": 1.8174715523227017e-06,
1237
+ "loss": 0.886,
1238
+ "step": 1025
1239
+ },
1240
+ {
1241
+ "epoch": 1.63,
1242
+ "learning_rate": 1.7444921814490256e-06,
1243
+ "loss": 0.8365,
1244
+ "step": 1030
1245
+ },
1246
+ {
1247
+ "epoch": 1.64,
1248
+ "learning_rate": 1.6728679895832622e-06,
1249
+ "loss": 0.9913,
1250
+ "step": 1035
1251
+ },
1252
+ {
1253
+ "epoch": 1.65,
1254
+ "learning_rate": 1.6026107341609842e-06,
1255
+ "loss": 0.8921,
1256
+ "step": 1040
1257
+ },
1258
+ {
1259
+ "epoch": 1.65,
1260
+ "learning_rate": 1.5337319482289503e-06,
1261
+ "loss": 0.8323,
1262
+ "step": 1045
1263
+ },
1264
+ {
1265
+ "epoch": 1.66,
1266
+ "learning_rate": 1.4662429385519084e-06,
1267
+ "loss": 0.8869,
1268
+ "step": 1050
1269
+ },
1270
+ {
1271
+ "epoch": 1.67,
1272
+ "learning_rate": 1.400154783756541e-06,
1273
+ "loss": 0.9799,
1274
+ "step": 1055
1275
+ },
1276
+ {
1277
+ "epoch": 1.68,
1278
+ "learning_rate": 1.3354783325128561e-06,
1279
+ "loss": 0.9416,
1280
+ "step": 1060
1281
+ },
1282
+ {
1283
+ "epoch": 1.68,
1284
+ "learning_rate": 1.2722242017533192e-06,
1285
+ "loss": 0.9801,
1286
+ "step": 1065
1287
+ },
1288
+ {
1289
+ "epoch": 1.69,
1290
+ "learning_rate": 1.2104027749300574e-06,
1291
+ "loss": 0.9614,
1292
+ "step": 1070
1293
+ },
1294
+ {
1295
+ "epoch": 1.7,
1296
+ "learning_rate": 1.150024200310348e-06,
1297
+ "loss": 1.0792,
1298
+ "step": 1075
1299
+ },
1300
+ {
1301
+ "epoch": 1.71,
1302
+ "learning_rate": 1.0910983893107419e-06,
1303
+ "loss": 0.8622,
1304
+ "step": 1080
1305
+ },
1306
+ {
1307
+ "epoch": 1.72,
1308
+ "learning_rate": 1.0336350148700668e-06,
1309
+ "loss": 0.8627,
1310
+ "step": 1085
1311
+ },
1312
+ {
1313
+ "epoch": 1.72,
1314
+ "learning_rate": 9.776435098615578e-07,
1315
+ "loss": 0.857,
1316
+ "step": 1090
1317
+ },
1318
+ {
1319
+ "epoch": 1.73,
1320
+ "learning_rate": 9.231330655444193e-07,
1321
+ "loss": 1.0396,
1322
+ "step": 1095
1323
+ },
1324
+ {
1325
+ "epoch": 1.74,
1326
+ "learning_rate": 8.701126300550322e-07,
1327
+ "loss": 1.0174,
1328
+ "step": 1100
1329
+ },
1330
+ {
1331
+ "epoch": 1.75,
1332
+ "learning_rate": 8.185909069380782e-07,
1333
+ "loss": 1.031,
1334
+ "step": 1105
1335
+ },
1336
+ {
1337
+ "epoch": 1.76,
1338
+ "learning_rate": 7.685763537178093e-07,
1339
+ "loss": 0.9254,
1340
+ "step": 1110
1341
+ },
1342
+ {
1343
+ "epoch": 1.76,
1344
+ "learning_rate": 7.200771805097206e-07,
1345
+ "loss": 0.9294,
1346
+ "step": 1115
1347
+ },
1348
+ {
1349
+ "epoch": 1.77,
1350
+ "learning_rate": 6.731013486728044e-07,
1351
+ "loss": 0.8641,
1352
+ "step": 1120
1353
+ },
1354
+ {
1355
+ "epoch": 1.78,
1356
+ "learning_rate": 6.276565695026671e-07,
1357
+ "loss": 0.9669,
1358
+ "step": 1125
1359
+ },
1360
+ {
1361
+ "epoch": 1.79,
1362
+ "learning_rate": 5.837503029656888e-07,
1363
+ "loss": 0.8447,
1364
+ "step": 1130
1365
+ },
1366
+ {
1367
+ "epoch": 1.8,
1368
+ "learning_rate": 5.413897564744253e-07,
1369
+ "loss": 0.9566,
1370
+ "step": 1135
1371
+ },
1372
+ {
1373
+ "epoch": 1.8,
1374
+ "learning_rate": 5.005818837044885e-07,
1375
+ "loss": 0.9457,
1376
+ "step": 1140
1377
+ },
1378
+ {
1379
+ "epoch": 1.81,
1380
+ "learning_rate": 4.613333834530631e-07,
1381
+ "loss": 0.8523,
1382
+ "step": 1145
1383
+ },
1384
+ {
1385
+ "epoch": 1.82,
1386
+ "learning_rate": 4.23650698539273e-07,
1387
+ "loss": 0.9696,
1388
+ "step": 1150
1389
+ },
1390
+ {
1391
+ "epoch": 1.83,
1392
+ "learning_rate": 3.8754001474655354e-07,
1393
+ "loss": 0.7867,
1394
+ "step": 1155
1395
+ },
1396
+ {
1397
+ "epoch": 1.83,
1398
+ "learning_rate": 3.530072598072454e-07,
1399
+ "loss": 0.9534,
1400
+ "step": 1160
1401
+ },
1402
+ {
1403
+ "epoch": 1.84,
1404
+ "learning_rate": 3.200581024295102e-07,
1405
+ "loss": 0.9209,
1406
+ "step": 1165
1407
+ },
1408
+ {
1409
+ "epoch": 1.85,
1410
+ "learning_rate": 2.886979513667998e-07,
1411
+ "loss": 0.8257,
1412
+ "step": 1170
1413
+ },
1414
+ {
1415
+ "epoch": 1.86,
1416
+ "learning_rate": 2.589319545299807e-07,
1417
+ "loss": 0.9424,
1418
+ "step": 1175
1419
+ },
1420
+ {
1421
+ "epoch": 1.87,
1422
+ "learning_rate": 2.3076499814227992e-07,
1423
+ "loss": 0.873,
1424
+ "step": 1180
1425
+ },
1426
+ {
1427
+ "epoch": 1.87,
1428
+ "learning_rate": 2.042017059371948e-07,
1429
+ "loss": 0.843,
1430
+ "step": 1185
1431
+ },
1432
+ {
1433
+ "epoch": 1.88,
1434
+ "learning_rate": 1.7924643839947632e-07,
1435
+ "loss": 0.9136,
1436
+ "step": 1190
1437
+ },
1438
+ {
1439
+ "epoch": 1.89,
1440
+ "learning_rate": 1.559032920493464e-07,
1441
+ "loss": 1.0255,
1442
+ "step": 1195
1443
+ },
1444
+ {
1445
+ "epoch": 1.9,
1446
+ "learning_rate": 1.3417609877002691e-07,
1447
+ "loss": 0.8384,
1448
+ "step": 1200
1449
+ },
1450
+ {
1451
+ "epoch": 1.91,
1452
+ "learning_rate": 1.1406842517872608e-07,
1453
+ "loss": 0.7154,
1454
+ "step": 1205
1455
+ },
1456
+ {
1457
+ "epoch": 1.91,
1458
+ "learning_rate": 9.558357204115464e-08,
1459
+ "loss": 0.8805,
1460
+ "step": 1210
1461
+ },
1462
+ {
1463
+ "epoch": 1.92,
1464
+ "learning_rate": 7.872457372969711e-08,
1465
+ "loss": 0.7834,
1466
+ "step": 1215
1467
+ },
1468
+ {
1469
+ "epoch": 1.93,
1470
+ "learning_rate": 6.34941977253023e-08,
1471
+ "loss": 0.8752,
1472
+ "step": 1220
1473
+ },
1474
+ {
1475
+ "epoch": 1.94,
1476
+ "learning_rate": 4.989494416318685e-08,
1477
+ "loss": 0.9148,
1478
+ "step": 1225
1479
+ },
1480
+ {
1481
+ "epoch": 1.95,
1482
+ "learning_rate": 3.7929045422432364e-08,
1483
+ "loss": 0.8612,
1484
+ "step": 1230
1485
+ },
1486
+ {
1487
+ "epoch": 1.95,
1488
+ "learning_rate": 2.7598465759526294e-08,
1489
+ "loss": 0.9299,
1490
+ "step": 1235
1491
+ },
1492
+ {
1493
+ "epoch": 1.96,
1494
+ "learning_rate": 1.8904900985918796e-08,
1495
+ "loss": 0.9918,
1496
+ "step": 1240
1497
+ },
1498
+ {
1499
+ "epoch": 1.97,
1500
+ "learning_rate": 1.184977818965205e-08,
1501
+ "loss": 0.9008,
1502
+ "step": 1245
1503
+ },
1504
+ {
1505
+ "epoch": 1.98,
1506
+ "learning_rate": 6.434255501095443e-09,
1507
+ "loss": 0.9461,
1508
+ "step": 1250
1509
+ },
1510
+ {
1511
+ "epoch": 1.99,
1512
+ "learning_rate": 2.659221902830966e-09,
1513
+ "loss": 0.8181,
1514
+ "step": 1255
1515
+ },
1516
+ {
1517
+ "epoch": 1.99,
1518
+ "learning_rate": 5.252970837255067e-10,
1519
+ "loss": 0.8643,
1520
+ "step": 1260
1521
+ },
1522
+ {
1523
+ "epoch": 2.0,
1524
+ "step": 1264,
1525
+ "total_flos": 3.882961297263821e+17,
1526
+ "train_loss": 0.9575079066466682,
1527
+ "train_runtime": 77706.2136,
1528
+ "train_samples_per_second": 0.26,
1529
+ "train_steps_per_second": 0.016
1530
+ }
1531
+ ],
1532
+ "max_steps": 1264,
1533
+ "num_train_epochs": 2,
1534
+ "total_flos": 3.882961297263821e+17,
1535
+ "trial_name": null,
1536
+ "trial_params": null
1537
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d2cfacbe013a88ea45672bbf16b7c9b6dedbe580c5f57c175c53b218c0dc6b3f
3
- size 4207
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2c8d84c6ea850e743bbd2858bd2ac8b6913bd2628de22bb48d811f906d778c2
3
+ size 4399