harish commited on
Commit
a5a0849
1 Parent(s): 7681550

added model

Browse files
Files changed (43) hide show
  1. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/config.json +60 -0
  2. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/optimizer.pt +3 -0
  3. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/pytorch_model.bin +3 -0
  4. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/rng_state.pth +3 -0
  5. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/scheduler.pt +3 -0
  6. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/special_tokens_map.json +107 -0
  7. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/spiece.model +3 -0
  8. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/tokenizer.json +0 -0
  9. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/tokenizer_config.json +112 -0
  10. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/trainer_state.json +1098 -0
  11. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/training_args.bin +3 -0
  12. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/config.json +60 -0
  13. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/optimizer.pt +3 -0
  14. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/pytorch_model.bin +3 -0
  15. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/rng_state.pth +3 -0
  16. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/scheduler.pt +3 -0
  17. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/special_tokens_map.json +107 -0
  18. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/spiece.model +3 -0
  19. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/tokenizer.json +0 -0
  20. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/tokenizer_config.json +112 -0
  21. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/trainer_state.json +1248 -0
  22. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/training_args.bin +3 -0
  23. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/config.json +60 -0
  24. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/rng_state.pth +3 -0
  25. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/scheduler.pt +3 -0
  26. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/special_tokens_map.json +107 -0
  27. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/spiece.model +3 -0
  28. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/tokenizer.json +0 -0
  29. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/tokenizer_config.json +112 -0
  30. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/trainer_state.json +1406 -0
  31. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/training_args.bin +3 -0
  32. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/config.json +60 -0
  33. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs-500Steps.json +0 -0
  34. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs.json +0 -0
  35. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs1250Steps.json +0 -0
  36. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs1750Steps.json +0 -0
  37. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs2500Steps.json +0 -0
  38. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/pytorch_model.bin +3 -0
  39. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/special_tokens_map.json +107 -0
  40. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/spiece.model +3 -0
  41. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/tokenizer.json +0 -0
  42. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/tokenizer_config.json +112 -0
  43. t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/training_args.bin +3 -0
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.21.1",
58
+ "use_cache": true,
59
+ "vocab_size": 32128
60
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:deeb313bf3ddc395360f01ba519b52592ec6865636cd970993fa353badc7e763
3
+ size 1783379133
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a5ceb9e45de01b474db666348438259f41088fbc901a3896433aca22ea1590f
3
+ size 891700799
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ec08f7c8006113b1a3cfff394ab26b2537112941e13a3692aeea177f161c92b
3
+ size 14503
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4d22b99df672a009780d88c7966595b0c30f6d75c13c124599808328a71eadb
3
+ size 623
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/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
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/tokenizer_config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "extra_ids": 100,
106
+ "model_max_length": 512,
107
+ "name_or_path": "t5-base",
108
+ "pad_token": "<pad>",
109
+ "special_tokens_map_file": null,
110
+ "tokenizer_class": "T5Tokenizer",
111
+ "unk_token": "<unk>"
112
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/trainer_state.json ADDED
@@ -0,0 +1,1098 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.127358490566038,
5
+ "global_step": 1750,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 1e-05,
13
+ "loss": 0.002,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 2e-05,
19
+ "loss": 0.002,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.07,
24
+ "learning_rate": 3e-05,
25
+ "loss": 0.002,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.09,
30
+ "learning_rate": 4e-05,
31
+ "loss": 0.0023,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.12,
36
+ "learning_rate": 5e-05,
37
+ "loss": 0.0021,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.14,
42
+ "learning_rate": 6e-05,
43
+ "loss": 0.0022,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "learning_rate": 7e-05,
49
+ "loss": 0.0037,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.19,
54
+ "learning_rate": 8e-05,
55
+ "loss": 0.0054,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.21,
60
+ "learning_rate": 9e-05,
61
+ "loss": 0.0153,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.24,
66
+ "learning_rate": 0.0001,
67
+ "loss": 0.0338,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.26,
72
+ "learning_rate": 9.97584541062802e-05,
73
+ "loss": 0.0546,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.28,
78
+ "learning_rate": 9.951690821256039e-05,
79
+ "loss": 0.0706,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.31,
84
+ "learning_rate": 9.927536231884058e-05,
85
+ "loss": 0.0603,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.33,
90
+ "learning_rate": 9.903381642512077e-05,
91
+ "loss": 0.0698,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.35,
96
+ "learning_rate": 9.879227053140096e-05,
97
+ "loss": 0.0808,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.38,
102
+ "learning_rate": 9.855072463768117e-05,
103
+ "loss": 0.0792,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.4,
108
+ "learning_rate": 9.830917874396136e-05,
109
+ "loss": 0.0656,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.42,
114
+ "learning_rate": 9.806763285024155e-05,
115
+ "loss": 0.0673,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.45,
120
+ "learning_rate": 9.782608695652174e-05,
121
+ "loss": 0.0686,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.47,
126
+ "learning_rate": 9.758454106280194e-05,
127
+ "loss": 0.0717,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.5,
132
+ "learning_rate": 9.734299516908213e-05,
133
+ "loss": 0.066,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.52,
138
+ "learning_rate": 9.710144927536232e-05,
139
+ "loss": 0.08,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.54,
144
+ "learning_rate": 9.685990338164251e-05,
145
+ "loss": 0.084,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.57,
150
+ "learning_rate": 9.66183574879227e-05,
151
+ "loss": 0.0703,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.59,
156
+ "learning_rate": 9.63768115942029e-05,
157
+ "loss": 0.0589,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.61,
162
+ "learning_rate": 9.61352657004831e-05,
163
+ "loss": 0.0708,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.64,
168
+ "learning_rate": 9.589371980676329e-05,
169
+ "loss": 0.0661,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.66,
174
+ "learning_rate": 9.565217391304348e-05,
175
+ "loss": 0.0649,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.68,
180
+ "learning_rate": 9.541062801932368e-05,
181
+ "loss": 0.063,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.71,
186
+ "learning_rate": 9.516908212560387e-05,
187
+ "loss": 0.0628,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.73,
192
+ "learning_rate": 9.492753623188406e-05,
193
+ "loss": 0.0699,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.75,
198
+ "learning_rate": 9.468599033816425e-05,
199
+ "loss": 0.0716,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.78,
204
+ "learning_rate": 9.444444444444444e-05,
205
+ "loss": 0.0623,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.8,
210
+ "learning_rate": 9.420289855072463e-05,
211
+ "loss": 0.0661,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.83,
216
+ "learning_rate": 9.396135265700483e-05,
217
+ "loss": 0.0603,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.85,
222
+ "learning_rate": 9.371980676328503e-05,
223
+ "loss": 0.0591,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.87,
228
+ "learning_rate": 9.347826086956522e-05,
229
+ "loss": 0.054,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.9,
234
+ "learning_rate": 9.323671497584541e-05,
235
+ "loss": 0.0582,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.92,
240
+ "learning_rate": 9.29951690821256e-05,
241
+ "loss": 0.0617,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.94,
246
+ "learning_rate": 9.27536231884058e-05,
247
+ "loss": 0.0615,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.97,
252
+ "learning_rate": 9.251207729468599e-05,
253
+ "loss": 0.0648,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.99,
258
+ "learning_rate": 9.227053140096618e-05,
259
+ "loss": 0.0607,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 1.0,
264
+ "eval_loss": 1.95200514793396,
265
+ "eval_runtime": 26.5048,
266
+ "eval_samples_per_second": 28.41,
267
+ "eval_steps_per_second": 1.811,
268
+ "step": 424
269
+ },
270
+ {
271
+ "epoch": 1.01,
272
+ "learning_rate": 9.202898550724637e-05,
273
+ "loss": 0.0516,
274
+ "step": 430
275
+ },
276
+ {
277
+ "epoch": 1.04,
278
+ "learning_rate": 9.178743961352657e-05,
279
+ "loss": 0.0446,
280
+ "step": 440
281
+ },
282
+ {
283
+ "epoch": 1.06,
284
+ "learning_rate": 9.154589371980676e-05,
285
+ "loss": 0.0452,
286
+ "step": 450
287
+ },
288
+ {
289
+ "epoch": 1.08,
290
+ "learning_rate": 9.130434782608696e-05,
291
+ "loss": 0.0456,
292
+ "step": 460
293
+ },
294
+ {
295
+ "epoch": 1.11,
296
+ "learning_rate": 9.106280193236715e-05,
297
+ "loss": 0.0483,
298
+ "step": 470
299
+ },
300
+ {
301
+ "epoch": 1.13,
302
+ "learning_rate": 9.082125603864735e-05,
303
+ "loss": 0.0575,
304
+ "step": 480
305
+ },
306
+ {
307
+ "epoch": 1.16,
308
+ "learning_rate": 9.057971014492754e-05,
309
+ "loss": 0.0485,
310
+ "step": 490
311
+ },
312
+ {
313
+ "epoch": 1.18,
314
+ "learning_rate": 9.033816425120773e-05,
315
+ "loss": 0.0471,
316
+ "step": 500
317
+ },
318
+ {
319
+ "epoch": 1.2,
320
+ "learning_rate": 9.009661835748792e-05,
321
+ "loss": 0.0517,
322
+ "step": 510
323
+ },
324
+ {
325
+ "epoch": 1.23,
326
+ "learning_rate": 8.985507246376813e-05,
327
+ "loss": 0.0482,
328
+ "step": 520
329
+ },
330
+ {
331
+ "epoch": 1.25,
332
+ "learning_rate": 8.961352657004832e-05,
333
+ "loss": 0.0449,
334
+ "step": 530
335
+ },
336
+ {
337
+ "epoch": 1.27,
338
+ "learning_rate": 8.937198067632851e-05,
339
+ "loss": 0.0598,
340
+ "step": 540
341
+ },
342
+ {
343
+ "epoch": 1.3,
344
+ "learning_rate": 8.91304347826087e-05,
345
+ "loss": 0.0539,
346
+ "step": 550
347
+ },
348
+ {
349
+ "epoch": 1.32,
350
+ "learning_rate": 8.888888888888889e-05,
351
+ "loss": 0.0481,
352
+ "step": 560
353
+ },
354
+ {
355
+ "epoch": 1.34,
356
+ "learning_rate": 8.864734299516908e-05,
357
+ "loss": 0.0553,
358
+ "step": 570
359
+ },
360
+ {
361
+ "epoch": 1.37,
362
+ "learning_rate": 8.840579710144929e-05,
363
+ "loss": 0.0474,
364
+ "step": 580
365
+ },
366
+ {
367
+ "epoch": 1.39,
368
+ "learning_rate": 8.816425120772948e-05,
369
+ "loss": 0.0536,
370
+ "step": 590
371
+ },
372
+ {
373
+ "epoch": 1.42,
374
+ "learning_rate": 8.792270531400967e-05,
375
+ "loss": 0.0552,
376
+ "step": 600
377
+ },
378
+ {
379
+ "epoch": 1.44,
380
+ "learning_rate": 8.768115942028987e-05,
381
+ "loss": 0.0494,
382
+ "step": 610
383
+ },
384
+ {
385
+ "epoch": 1.46,
386
+ "learning_rate": 8.743961352657006e-05,
387
+ "loss": 0.0605,
388
+ "step": 620
389
+ },
390
+ {
391
+ "epoch": 1.49,
392
+ "learning_rate": 8.719806763285025e-05,
393
+ "loss": 0.0571,
394
+ "step": 630
395
+ },
396
+ {
397
+ "epoch": 1.51,
398
+ "learning_rate": 8.695652173913044e-05,
399
+ "loss": 0.0468,
400
+ "step": 640
401
+ },
402
+ {
403
+ "epoch": 1.53,
404
+ "learning_rate": 8.671497584541063e-05,
405
+ "loss": 0.058,
406
+ "step": 650
407
+ },
408
+ {
409
+ "epoch": 1.56,
410
+ "learning_rate": 8.647342995169082e-05,
411
+ "loss": 0.0523,
412
+ "step": 660
413
+ },
414
+ {
415
+ "epoch": 1.58,
416
+ "learning_rate": 8.623188405797103e-05,
417
+ "loss": 0.0391,
418
+ "step": 670
419
+ },
420
+ {
421
+ "epoch": 1.6,
422
+ "learning_rate": 8.599033816425122e-05,
423
+ "loss": 0.0488,
424
+ "step": 680
425
+ },
426
+ {
427
+ "epoch": 1.63,
428
+ "learning_rate": 8.574879227053141e-05,
429
+ "loss": 0.056,
430
+ "step": 690
431
+ },
432
+ {
433
+ "epoch": 1.65,
434
+ "learning_rate": 8.55072463768116e-05,
435
+ "loss": 0.0599,
436
+ "step": 700
437
+ },
438
+ {
439
+ "epoch": 1.67,
440
+ "learning_rate": 8.52657004830918e-05,
441
+ "loss": 0.0536,
442
+ "step": 710
443
+ },
444
+ {
445
+ "epoch": 1.7,
446
+ "learning_rate": 8.502415458937199e-05,
447
+ "loss": 0.0534,
448
+ "step": 720
449
+ },
450
+ {
451
+ "epoch": 1.72,
452
+ "learning_rate": 8.478260869565218e-05,
453
+ "loss": 0.0484,
454
+ "step": 730
455
+ },
456
+ {
457
+ "epoch": 1.75,
458
+ "learning_rate": 8.454106280193237e-05,
459
+ "loss": 0.0476,
460
+ "step": 740
461
+ },
462
+ {
463
+ "epoch": 1.77,
464
+ "learning_rate": 8.429951690821256e-05,
465
+ "loss": 0.0426,
466
+ "step": 750
467
+ },
468
+ {
469
+ "epoch": 1.79,
470
+ "learning_rate": 8.405797101449276e-05,
471
+ "loss": 0.0486,
472
+ "step": 760
473
+ },
474
+ {
475
+ "epoch": 1.82,
476
+ "learning_rate": 8.381642512077295e-05,
477
+ "loss": 0.0598,
478
+ "step": 770
479
+ },
480
+ {
481
+ "epoch": 1.84,
482
+ "learning_rate": 8.357487922705315e-05,
483
+ "loss": 0.0567,
484
+ "step": 780
485
+ },
486
+ {
487
+ "epoch": 1.86,
488
+ "learning_rate": 8.333333333333334e-05,
489
+ "loss": 0.0535,
490
+ "step": 790
491
+ },
492
+ {
493
+ "epoch": 1.89,
494
+ "learning_rate": 8.309178743961354e-05,
495
+ "loss": 0.0453,
496
+ "step": 800
497
+ },
498
+ {
499
+ "epoch": 1.91,
500
+ "learning_rate": 8.285024154589373e-05,
501
+ "loss": 0.0437,
502
+ "step": 810
503
+ },
504
+ {
505
+ "epoch": 1.93,
506
+ "learning_rate": 8.260869565217392e-05,
507
+ "loss": 0.0549,
508
+ "step": 820
509
+ },
510
+ {
511
+ "epoch": 1.96,
512
+ "learning_rate": 8.236714975845411e-05,
513
+ "loss": 0.0522,
514
+ "step": 830
515
+ },
516
+ {
517
+ "epoch": 1.98,
518
+ "learning_rate": 8.21256038647343e-05,
519
+ "loss": 0.049,
520
+ "step": 840
521
+ },
522
+ {
523
+ "epoch": 2.0,
524
+ "eval_loss": 1.9635112285614014,
525
+ "eval_runtime": 26.4849,
526
+ "eval_samples_per_second": 28.431,
527
+ "eval_steps_per_second": 1.812,
528
+ "step": 848
529
+ },
530
+ {
531
+ "epoch": 2.0,
532
+ "learning_rate": 8.18840579710145e-05,
533
+ "loss": 0.0426,
534
+ "step": 850
535
+ },
536
+ {
537
+ "epoch": 2.03,
538
+ "learning_rate": 8.164251207729469e-05,
539
+ "loss": 0.0259,
540
+ "step": 860
541
+ },
542
+ {
543
+ "epoch": 2.05,
544
+ "learning_rate": 8.140096618357488e-05,
545
+ "loss": 0.0255,
546
+ "step": 870
547
+ },
548
+ {
549
+ "epoch": 2.08,
550
+ "learning_rate": 8.115942028985508e-05,
551
+ "loss": 0.0209,
552
+ "step": 880
553
+ },
554
+ {
555
+ "epoch": 2.1,
556
+ "learning_rate": 8.091787439613528e-05,
557
+ "loss": 0.0244,
558
+ "step": 890
559
+ },
560
+ {
561
+ "epoch": 2.12,
562
+ "learning_rate": 8.067632850241547e-05,
563
+ "loss": 0.0319,
564
+ "step": 900
565
+ },
566
+ {
567
+ "epoch": 2.15,
568
+ "learning_rate": 8.043478260869566e-05,
569
+ "loss": 0.0257,
570
+ "step": 910
571
+ },
572
+ {
573
+ "epoch": 2.17,
574
+ "learning_rate": 8.019323671497585e-05,
575
+ "loss": 0.0286,
576
+ "step": 920
577
+ },
578
+ {
579
+ "epoch": 2.19,
580
+ "learning_rate": 7.995169082125604e-05,
581
+ "loss": 0.0265,
582
+ "step": 930
583
+ },
584
+ {
585
+ "epoch": 2.22,
586
+ "learning_rate": 7.971014492753623e-05,
587
+ "loss": 0.0229,
588
+ "step": 940
589
+ },
590
+ {
591
+ "epoch": 2.24,
592
+ "learning_rate": 7.946859903381643e-05,
593
+ "loss": 0.0237,
594
+ "step": 950
595
+ },
596
+ {
597
+ "epoch": 2.26,
598
+ "learning_rate": 7.922705314009662e-05,
599
+ "loss": 0.0313,
600
+ "step": 960
601
+ },
602
+ {
603
+ "epoch": 2.29,
604
+ "learning_rate": 7.898550724637681e-05,
605
+ "loss": 0.035,
606
+ "step": 970
607
+ },
608
+ {
609
+ "epoch": 2.31,
610
+ "learning_rate": 7.874396135265701e-05,
611
+ "loss": 0.0324,
612
+ "step": 980
613
+ },
614
+ {
615
+ "epoch": 2.33,
616
+ "learning_rate": 7.85024154589372e-05,
617
+ "loss": 0.029,
618
+ "step": 990
619
+ },
620
+ {
621
+ "epoch": 2.36,
622
+ "learning_rate": 7.82608695652174e-05,
623
+ "loss": 0.0297,
624
+ "step": 1000
625
+ },
626
+ {
627
+ "epoch": 2.38,
628
+ "learning_rate": 7.801932367149759e-05,
629
+ "loss": 0.029,
630
+ "step": 1010
631
+ },
632
+ {
633
+ "epoch": 2.41,
634
+ "learning_rate": 7.777777777777778e-05,
635
+ "loss": 0.0289,
636
+ "step": 1020
637
+ },
638
+ {
639
+ "epoch": 2.43,
640
+ "learning_rate": 7.753623188405797e-05,
641
+ "loss": 0.0267,
642
+ "step": 1030
643
+ },
644
+ {
645
+ "epoch": 2.45,
646
+ "learning_rate": 7.729468599033817e-05,
647
+ "loss": 0.0373,
648
+ "step": 1040
649
+ },
650
+ {
651
+ "epoch": 2.48,
652
+ "learning_rate": 7.705314009661836e-05,
653
+ "loss": 0.0269,
654
+ "step": 1050
655
+ },
656
+ {
657
+ "epoch": 2.5,
658
+ "learning_rate": 7.681159420289855e-05,
659
+ "loss": 0.0237,
660
+ "step": 1060
661
+ },
662
+ {
663
+ "epoch": 2.52,
664
+ "learning_rate": 7.657004830917874e-05,
665
+ "loss": 0.0222,
666
+ "step": 1070
667
+ },
668
+ {
669
+ "epoch": 2.55,
670
+ "learning_rate": 7.632850241545893e-05,
671
+ "loss": 0.0378,
672
+ "step": 1080
673
+ },
674
+ {
675
+ "epoch": 2.57,
676
+ "learning_rate": 7.608695652173914e-05,
677
+ "loss": 0.0311,
678
+ "step": 1090
679
+ },
680
+ {
681
+ "epoch": 2.59,
682
+ "learning_rate": 7.584541062801933e-05,
683
+ "loss": 0.0243,
684
+ "step": 1100
685
+ },
686
+ {
687
+ "epoch": 2.62,
688
+ "learning_rate": 7.560386473429952e-05,
689
+ "loss": 0.0295,
690
+ "step": 1110
691
+ },
692
+ {
693
+ "epoch": 2.64,
694
+ "learning_rate": 7.536231884057971e-05,
695
+ "loss": 0.0293,
696
+ "step": 1120
697
+ },
698
+ {
699
+ "epoch": 2.67,
700
+ "learning_rate": 7.51207729468599e-05,
701
+ "loss": 0.029,
702
+ "step": 1130
703
+ },
704
+ {
705
+ "epoch": 2.69,
706
+ "learning_rate": 7.48792270531401e-05,
707
+ "loss": 0.0289,
708
+ "step": 1140
709
+ },
710
+ {
711
+ "epoch": 2.71,
712
+ "learning_rate": 7.463768115942029e-05,
713
+ "loss": 0.0394,
714
+ "step": 1150
715
+ },
716
+ {
717
+ "epoch": 2.74,
718
+ "learning_rate": 7.439613526570048e-05,
719
+ "loss": 0.0355,
720
+ "step": 1160
721
+ },
722
+ {
723
+ "epoch": 2.76,
724
+ "learning_rate": 7.415458937198067e-05,
725
+ "loss": 0.0273,
726
+ "step": 1170
727
+ },
728
+ {
729
+ "epoch": 2.78,
730
+ "learning_rate": 7.391304347826086e-05,
731
+ "loss": 0.0312,
732
+ "step": 1180
733
+ },
734
+ {
735
+ "epoch": 2.81,
736
+ "learning_rate": 7.367149758454107e-05,
737
+ "loss": 0.0284,
738
+ "step": 1190
739
+ },
740
+ {
741
+ "epoch": 2.83,
742
+ "learning_rate": 7.342995169082126e-05,
743
+ "loss": 0.0363,
744
+ "step": 1200
745
+ },
746
+ {
747
+ "epoch": 2.85,
748
+ "learning_rate": 7.318840579710145e-05,
749
+ "loss": 0.0307,
750
+ "step": 1210
751
+ },
752
+ {
753
+ "epoch": 2.88,
754
+ "learning_rate": 7.294685990338164e-05,
755
+ "loss": 0.0357,
756
+ "step": 1220
757
+ },
758
+ {
759
+ "epoch": 2.9,
760
+ "learning_rate": 7.270531400966184e-05,
761
+ "loss": 0.0332,
762
+ "step": 1230
763
+ },
764
+ {
765
+ "epoch": 2.92,
766
+ "learning_rate": 7.246376811594203e-05,
767
+ "loss": 0.0361,
768
+ "step": 1240
769
+ },
770
+ {
771
+ "epoch": 2.95,
772
+ "learning_rate": 7.222222222222222e-05,
773
+ "loss": 0.0318,
774
+ "step": 1250
775
+ },
776
+ {
777
+ "epoch": 2.97,
778
+ "learning_rate": 7.198067632850241e-05,
779
+ "loss": 0.0306,
780
+ "step": 1260
781
+ },
782
+ {
783
+ "epoch": 3.0,
784
+ "learning_rate": 7.17391304347826e-05,
785
+ "loss": 0.0281,
786
+ "step": 1270
787
+ },
788
+ {
789
+ "epoch": 3.0,
790
+ "eval_loss": 1.9812633991241455,
791
+ "eval_runtime": 27.1954,
792
+ "eval_samples_per_second": 27.689,
793
+ "eval_steps_per_second": 1.765,
794
+ "step": 1272
795
+ },
796
+ {
797
+ "epoch": 3.02,
798
+ "learning_rate": 7.14975845410628e-05,
799
+ "loss": 0.0175,
800
+ "step": 1280
801
+ },
802
+ {
803
+ "epoch": 3.04,
804
+ "learning_rate": 7.1256038647343e-05,
805
+ "loss": 0.0208,
806
+ "step": 1290
807
+ },
808
+ {
809
+ "epoch": 3.07,
810
+ "learning_rate": 7.101449275362319e-05,
811
+ "loss": 0.0153,
812
+ "step": 1300
813
+ },
814
+ {
815
+ "epoch": 3.09,
816
+ "learning_rate": 7.077294685990338e-05,
817
+ "loss": 0.0147,
818
+ "step": 1310
819
+ },
820
+ {
821
+ "epoch": 3.11,
822
+ "learning_rate": 7.053140096618357e-05,
823
+ "loss": 0.0193,
824
+ "step": 1320
825
+ },
826
+ {
827
+ "epoch": 3.14,
828
+ "learning_rate": 7.028985507246377e-05,
829
+ "loss": 0.0173,
830
+ "step": 1330
831
+ },
832
+ {
833
+ "epoch": 3.16,
834
+ "learning_rate": 7.004830917874396e-05,
835
+ "loss": 0.0291,
836
+ "step": 1340
837
+ },
838
+ {
839
+ "epoch": 3.18,
840
+ "learning_rate": 6.980676328502416e-05,
841
+ "loss": 0.019,
842
+ "step": 1350
843
+ },
844
+ {
845
+ "epoch": 3.21,
846
+ "learning_rate": 6.956521739130436e-05,
847
+ "loss": 0.015,
848
+ "step": 1360
849
+ },
850
+ {
851
+ "epoch": 3.23,
852
+ "learning_rate": 6.932367149758455e-05,
853
+ "loss": 0.0208,
854
+ "step": 1370
855
+ },
856
+ {
857
+ "epoch": 3.25,
858
+ "learning_rate": 6.908212560386474e-05,
859
+ "loss": 0.0185,
860
+ "step": 1380
861
+ },
862
+ {
863
+ "epoch": 3.28,
864
+ "learning_rate": 6.884057971014493e-05,
865
+ "loss": 0.0162,
866
+ "step": 1390
867
+ },
868
+ {
869
+ "epoch": 3.3,
870
+ "learning_rate": 6.859903381642512e-05,
871
+ "loss": 0.0166,
872
+ "step": 1400
873
+ },
874
+ {
875
+ "epoch": 3.33,
876
+ "learning_rate": 6.835748792270533e-05,
877
+ "loss": 0.0178,
878
+ "step": 1410
879
+ },
880
+ {
881
+ "epoch": 3.35,
882
+ "learning_rate": 6.811594202898552e-05,
883
+ "loss": 0.0163,
884
+ "step": 1420
885
+ },
886
+ {
887
+ "epoch": 3.37,
888
+ "learning_rate": 6.787439613526571e-05,
889
+ "loss": 0.0227,
890
+ "step": 1430
891
+ },
892
+ {
893
+ "epoch": 3.4,
894
+ "learning_rate": 6.76328502415459e-05,
895
+ "loss": 0.0144,
896
+ "step": 1440
897
+ },
898
+ {
899
+ "epoch": 3.42,
900
+ "learning_rate": 6.73913043478261e-05,
901
+ "loss": 0.0156,
902
+ "step": 1450
903
+ },
904
+ {
905
+ "epoch": 3.44,
906
+ "learning_rate": 6.714975845410629e-05,
907
+ "loss": 0.014,
908
+ "step": 1460
909
+ },
910
+ {
911
+ "epoch": 3.47,
912
+ "learning_rate": 6.690821256038648e-05,
913
+ "loss": 0.0187,
914
+ "step": 1470
915
+ },
916
+ {
917
+ "epoch": 3.49,
918
+ "learning_rate": 6.666666666666667e-05,
919
+ "loss": 0.0198,
920
+ "step": 1480
921
+ },
922
+ {
923
+ "epoch": 3.51,
924
+ "learning_rate": 6.642512077294686e-05,
925
+ "loss": 0.0127,
926
+ "step": 1490
927
+ },
928
+ {
929
+ "epoch": 3.54,
930
+ "learning_rate": 6.618357487922707e-05,
931
+ "loss": 0.0129,
932
+ "step": 1500
933
+ },
934
+ {
935
+ "epoch": 3.56,
936
+ "learning_rate": 6.594202898550726e-05,
937
+ "loss": 0.0174,
938
+ "step": 1510
939
+ },
940
+ {
941
+ "epoch": 3.58,
942
+ "learning_rate": 6.570048309178745e-05,
943
+ "loss": 0.0147,
944
+ "step": 1520
945
+ },
946
+ {
947
+ "epoch": 3.61,
948
+ "learning_rate": 6.545893719806764e-05,
949
+ "loss": 0.0188,
950
+ "step": 1530
951
+ },
952
+ {
953
+ "epoch": 3.63,
954
+ "learning_rate": 6.521739130434783e-05,
955
+ "loss": 0.0149,
956
+ "step": 1540
957
+ },
958
+ {
959
+ "epoch": 3.66,
960
+ "learning_rate": 6.497584541062803e-05,
961
+ "loss": 0.0255,
962
+ "step": 1550
963
+ },
964
+ {
965
+ "epoch": 3.68,
966
+ "learning_rate": 6.473429951690822e-05,
967
+ "loss": 0.0198,
968
+ "step": 1560
969
+ },
970
+ {
971
+ "epoch": 3.7,
972
+ "learning_rate": 6.449275362318841e-05,
973
+ "loss": 0.0219,
974
+ "step": 1570
975
+ },
976
+ {
977
+ "epoch": 3.73,
978
+ "learning_rate": 6.42512077294686e-05,
979
+ "loss": 0.0213,
980
+ "step": 1580
981
+ },
982
+ {
983
+ "epoch": 3.75,
984
+ "learning_rate": 6.400966183574879e-05,
985
+ "loss": 0.0214,
986
+ "step": 1590
987
+ },
988
+ {
989
+ "epoch": 3.77,
990
+ "learning_rate": 6.376811594202898e-05,
991
+ "loss": 0.0185,
992
+ "step": 1600
993
+ },
994
+ {
995
+ "epoch": 3.8,
996
+ "learning_rate": 6.352657004830919e-05,
997
+ "loss": 0.0191,
998
+ "step": 1610
999
+ },
1000
+ {
1001
+ "epoch": 3.82,
1002
+ "learning_rate": 6.328502415458938e-05,
1003
+ "loss": 0.0165,
1004
+ "step": 1620
1005
+ },
1006
+ {
1007
+ "epoch": 3.84,
1008
+ "learning_rate": 6.304347826086957e-05,
1009
+ "loss": 0.0205,
1010
+ "step": 1630
1011
+ },
1012
+ {
1013
+ "epoch": 3.87,
1014
+ "learning_rate": 6.280193236714976e-05,
1015
+ "loss": 0.017,
1016
+ "step": 1640
1017
+ },
1018
+ {
1019
+ "epoch": 3.89,
1020
+ "learning_rate": 6.256038647342996e-05,
1021
+ "loss": 0.0198,
1022
+ "step": 1650
1023
+ },
1024
+ {
1025
+ "epoch": 3.92,
1026
+ "learning_rate": 6.231884057971015e-05,
1027
+ "loss": 0.0171,
1028
+ "step": 1660
1029
+ },
1030
+ {
1031
+ "epoch": 3.94,
1032
+ "learning_rate": 6.207729468599034e-05,
1033
+ "loss": 0.0201,
1034
+ "step": 1670
1035
+ },
1036
+ {
1037
+ "epoch": 3.96,
1038
+ "learning_rate": 6.183574879227053e-05,
1039
+ "loss": 0.0253,
1040
+ "step": 1680
1041
+ },
1042
+ {
1043
+ "epoch": 3.99,
1044
+ "learning_rate": 6.159420289855072e-05,
1045
+ "loss": 0.0204,
1046
+ "step": 1690
1047
+ },
1048
+ {
1049
+ "epoch": 4.0,
1050
+ "eval_loss": 2.0079989433288574,
1051
+ "eval_runtime": 26.4901,
1052
+ "eval_samples_per_second": 28.426,
1053
+ "eval_steps_per_second": 1.812,
1054
+ "step": 1696
1055
+ },
1056
+ {
1057
+ "epoch": 4.01,
1058
+ "learning_rate": 6.135265700483092e-05,
1059
+ "loss": 0.0136,
1060
+ "step": 1700
1061
+ },
1062
+ {
1063
+ "epoch": 4.03,
1064
+ "learning_rate": 6.111111111111112e-05,
1065
+ "loss": 0.0131,
1066
+ "step": 1710
1067
+ },
1068
+ {
1069
+ "epoch": 4.06,
1070
+ "learning_rate": 6.086956521739131e-05,
1071
+ "loss": 0.013,
1072
+ "step": 1720
1073
+ },
1074
+ {
1075
+ "epoch": 4.08,
1076
+ "learning_rate": 6.0628019323671504e-05,
1077
+ "loss": 0.009,
1078
+ "step": 1730
1079
+ },
1080
+ {
1081
+ "epoch": 4.1,
1082
+ "learning_rate": 6.0386473429951696e-05,
1083
+ "loss": 0.0075,
1084
+ "step": 1740
1085
+ },
1086
+ {
1087
+ "epoch": 4.13,
1088
+ "learning_rate": 6.014492753623189e-05,
1089
+ "loss": 0.0095,
1090
+ "step": 1750
1091
+ }
1092
+ ],
1093
+ "max_steps": 4240,
1094
+ "num_train_epochs": 10,
1095
+ "total_flos": 0.0,
1096
+ "trial_name": null,
1097
+ "trial_params": null
1098
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-1750/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d19fa419d807bcbfef5a86725e8eca48bdf21b5f0b9a046119a2780ac6238051
3
+ size 3695
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.21.1",
58
+ "use_cache": true,
59
+ "vocab_size": 32128
60
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c34a3fb6cca8019e3d12449c816a7085a03e89a285e404ebd6d102675de64dfc
3
+ size 1783379133
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:459b0708eee808c1f6a79cc21a41e1bd4c48e2794b24c5b20a1cacb46cb32903
3
+ size 891700799
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ec08f7c8006113b1a3cfff394ab26b2537112941e13a3692aeea177f161c92b
3
+ size 14503
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdd5f6f0a347b761b8f29f28cbcb12a3d5ce941889398b3e473630766e4c5c94
3
+ size 623
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/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
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "extra_ids": 100,
106
+ "model_max_length": 512,
107
+ "name_or_path": "t5-base",
108
+ "pad_token": "<pad>",
109
+ "special_tokens_map_file": null,
110
+ "tokenizer_class": "T5Tokenizer",
111
+ "unk_token": "<unk>"
112
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.716981132075472,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 1e-05,
13
+ "loss": 0.002,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 2e-05,
19
+ "loss": 0.002,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.07,
24
+ "learning_rate": 3e-05,
25
+ "loss": 0.002,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.09,
30
+ "learning_rate": 4e-05,
31
+ "loss": 0.0023,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.12,
36
+ "learning_rate": 5e-05,
37
+ "loss": 0.0021,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.14,
42
+ "learning_rate": 6e-05,
43
+ "loss": 0.0022,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "learning_rate": 7e-05,
49
+ "loss": 0.0037,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.19,
54
+ "learning_rate": 8e-05,
55
+ "loss": 0.0054,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.21,
60
+ "learning_rate": 9e-05,
61
+ "loss": 0.0153,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.24,
66
+ "learning_rate": 0.0001,
67
+ "loss": 0.0338,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.26,
72
+ "learning_rate": 9.97584541062802e-05,
73
+ "loss": 0.0546,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.28,
78
+ "learning_rate": 9.951690821256039e-05,
79
+ "loss": 0.0706,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.31,
84
+ "learning_rate": 9.927536231884058e-05,
85
+ "loss": 0.0603,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.33,
90
+ "learning_rate": 9.903381642512077e-05,
91
+ "loss": 0.0698,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.35,
96
+ "learning_rate": 9.879227053140096e-05,
97
+ "loss": 0.0808,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.38,
102
+ "learning_rate": 9.855072463768117e-05,
103
+ "loss": 0.0792,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.4,
108
+ "learning_rate": 9.830917874396136e-05,
109
+ "loss": 0.0656,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.42,
114
+ "learning_rate": 9.806763285024155e-05,
115
+ "loss": 0.0673,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.45,
120
+ "learning_rate": 9.782608695652174e-05,
121
+ "loss": 0.0686,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.47,
126
+ "learning_rate": 9.758454106280194e-05,
127
+ "loss": 0.0717,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.5,
132
+ "learning_rate": 9.734299516908213e-05,
133
+ "loss": 0.066,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.52,
138
+ "learning_rate": 9.710144927536232e-05,
139
+ "loss": 0.08,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.54,
144
+ "learning_rate": 9.685990338164251e-05,
145
+ "loss": 0.084,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.57,
150
+ "learning_rate": 9.66183574879227e-05,
151
+ "loss": 0.0703,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.59,
156
+ "learning_rate": 9.63768115942029e-05,
157
+ "loss": 0.0589,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.61,
162
+ "learning_rate": 9.61352657004831e-05,
163
+ "loss": 0.0708,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.64,
168
+ "learning_rate": 9.589371980676329e-05,
169
+ "loss": 0.0661,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.66,
174
+ "learning_rate": 9.565217391304348e-05,
175
+ "loss": 0.0649,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.68,
180
+ "learning_rate": 9.541062801932368e-05,
181
+ "loss": 0.063,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.71,
186
+ "learning_rate": 9.516908212560387e-05,
187
+ "loss": 0.0628,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.73,
192
+ "learning_rate": 9.492753623188406e-05,
193
+ "loss": 0.0699,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.75,
198
+ "learning_rate": 9.468599033816425e-05,
199
+ "loss": 0.0716,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.78,
204
+ "learning_rate": 9.444444444444444e-05,
205
+ "loss": 0.0623,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.8,
210
+ "learning_rate": 9.420289855072463e-05,
211
+ "loss": 0.0661,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.83,
216
+ "learning_rate": 9.396135265700483e-05,
217
+ "loss": 0.0603,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.85,
222
+ "learning_rate": 9.371980676328503e-05,
223
+ "loss": 0.0591,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.87,
228
+ "learning_rate": 9.347826086956522e-05,
229
+ "loss": 0.054,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.9,
234
+ "learning_rate": 9.323671497584541e-05,
235
+ "loss": 0.0582,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.92,
240
+ "learning_rate": 9.29951690821256e-05,
241
+ "loss": 0.0617,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.94,
246
+ "learning_rate": 9.27536231884058e-05,
247
+ "loss": 0.0615,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.97,
252
+ "learning_rate": 9.251207729468599e-05,
253
+ "loss": 0.0648,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.99,
258
+ "learning_rate": 9.227053140096618e-05,
259
+ "loss": 0.0607,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 1.0,
264
+ "eval_loss": 1.95200514793396,
265
+ "eval_runtime": 26.5048,
266
+ "eval_samples_per_second": 28.41,
267
+ "eval_steps_per_second": 1.811,
268
+ "step": 424
269
+ },
270
+ {
271
+ "epoch": 1.01,
272
+ "learning_rate": 9.202898550724637e-05,
273
+ "loss": 0.0516,
274
+ "step": 430
275
+ },
276
+ {
277
+ "epoch": 1.04,
278
+ "learning_rate": 9.178743961352657e-05,
279
+ "loss": 0.0446,
280
+ "step": 440
281
+ },
282
+ {
283
+ "epoch": 1.06,
284
+ "learning_rate": 9.154589371980676e-05,
285
+ "loss": 0.0452,
286
+ "step": 450
287
+ },
288
+ {
289
+ "epoch": 1.08,
290
+ "learning_rate": 9.130434782608696e-05,
291
+ "loss": 0.0456,
292
+ "step": 460
293
+ },
294
+ {
295
+ "epoch": 1.11,
296
+ "learning_rate": 9.106280193236715e-05,
297
+ "loss": 0.0483,
298
+ "step": 470
299
+ },
300
+ {
301
+ "epoch": 1.13,
302
+ "learning_rate": 9.082125603864735e-05,
303
+ "loss": 0.0575,
304
+ "step": 480
305
+ },
306
+ {
307
+ "epoch": 1.16,
308
+ "learning_rate": 9.057971014492754e-05,
309
+ "loss": 0.0485,
310
+ "step": 490
311
+ },
312
+ {
313
+ "epoch": 1.18,
314
+ "learning_rate": 9.033816425120773e-05,
315
+ "loss": 0.0471,
316
+ "step": 500
317
+ },
318
+ {
319
+ "epoch": 1.2,
320
+ "learning_rate": 9.009661835748792e-05,
321
+ "loss": 0.0517,
322
+ "step": 510
323
+ },
324
+ {
325
+ "epoch": 1.23,
326
+ "learning_rate": 8.985507246376813e-05,
327
+ "loss": 0.0482,
328
+ "step": 520
329
+ },
330
+ {
331
+ "epoch": 1.25,
332
+ "learning_rate": 8.961352657004832e-05,
333
+ "loss": 0.0449,
334
+ "step": 530
335
+ },
336
+ {
337
+ "epoch": 1.27,
338
+ "learning_rate": 8.937198067632851e-05,
339
+ "loss": 0.0598,
340
+ "step": 540
341
+ },
342
+ {
343
+ "epoch": 1.3,
344
+ "learning_rate": 8.91304347826087e-05,
345
+ "loss": 0.0539,
346
+ "step": 550
347
+ },
348
+ {
349
+ "epoch": 1.32,
350
+ "learning_rate": 8.888888888888889e-05,
351
+ "loss": 0.0481,
352
+ "step": 560
353
+ },
354
+ {
355
+ "epoch": 1.34,
356
+ "learning_rate": 8.864734299516908e-05,
357
+ "loss": 0.0553,
358
+ "step": 570
359
+ },
360
+ {
361
+ "epoch": 1.37,
362
+ "learning_rate": 8.840579710144929e-05,
363
+ "loss": 0.0474,
364
+ "step": 580
365
+ },
366
+ {
367
+ "epoch": 1.39,
368
+ "learning_rate": 8.816425120772948e-05,
369
+ "loss": 0.0536,
370
+ "step": 590
371
+ },
372
+ {
373
+ "epoch": 1.42,
374
+ "learning_rate": 8.792270531400967e-05,
375
+ "loss": 0.0552,
376
+ "step": 600
377
+ },
378
+ {
379
+ "epoch": 1.44,
380
+ "learning_rate": 8.768115942028987e-05,
381
+ "loss": 0.0494,
382
+ "step": 610
383
+ },
384
+ {
385
+ "epoch": 1.46,
386
+ "learning_rate": 8.743961352657006e-05,
387
+ "loss": 0.0605,
388
+ "step": 620
389
+ },
390
+ {
391
+ "epoch": 1.49,
392
+ "learning_rate": 8.719806763285025e-05,
393
+ "loss": 0.0571,
394
+ "step": 630
395
+ },
396
+ {
397
+ "epoch": 1.51,
398
+ "learning_rate": 8.695652173913044e-05,
399
+ "loss": 0.0468,
400
+ "step": 640
401
+ },
402
+ {
403
+ "epoch": 1.53,
404
+ "learning_rate": 8.671497584541063e-05,
405
+ "loss": 0.058,
406
+ "step": 650
407
+ },
408
+ {
409
+ "epoch": 1.56,
410
+ "learning_rate": 8.647342995169082e-05,
411
+ "loss": 0.0523,
412
+ "step": 660
413
+ },
414
+ {
415
+ "epoch": 1.58,
416
+ "learning_rate": 8.623188405797103e-05,
417
+ "loss": 0.0391,
418
+ "step": 670
419
+ },
420
+ {
421
+ "epoch": 1.6,
422
+ "learning_rate": 8.599033816425122e-05,
423
+ "loss": 0.0488,
424
+ "step": 680
425
+ },
426
+ {
427
+ "epoch": 1.63,
428
+ "learning_rate": 8.574879227053141e-05,
429
+ "loss": 0.056,
430
+ "step": 690
431
+ },
432
+ {
433
+ "epoch": 1.65,
434
+ "learning_rate": 8.55072463768116e-05,
435
+ "loss": 0.0599,
436
+ "step": 700
437
+ },
438
+ {
439
+ "epoch": 1.67,
440
+ "learning_rate": 8.52657004830918e-05,
441
+ "loss": 0.0536,
442
+ "step": 710
443
+ },
444
+ {
445
+ "epoch": 1.7,
446
+ "learning_rate": 8.502415458937199e-05,
447
+ "loss": 0.0534,
448
+ "step": 720
449
+ },
450
+ {
451
+ "epoch": 1.72,
452
+ "learning_rate": 8.478260869565218e-05,
453
+ "loss": 0.0484,
454
+ "step": 730
455
+ },
456
+ {
457
+ "epoch": 1.75,
458
+ "learning_rate": 8.454106280193237e-05,
459
+ "loss": 0.0476,
460
+ "step": 740
461
+ },
462
+ {
463
+ "epoch": 1.77,
464
+ "learning_rate": 8.429951690821256e-05,
465
+ "loss": 0.0426,
466
+ "step": 750
467
+ },
468
+ {
469
+ "epoch": 1.79,
470
+ "learning_rate": 8.405797101449276e-05,
471
+ "loss": 0.0486,
472
+ "step": 760
473
+ },
474
+ {
475
+ "epoch": 1.82,
476
+ "learning_rate": 8.381642512077295e-05,
477
+ "loss": 0.0598,
478
+ "step": 770
479
+ },
480
+ {
481
+ "epoch": 1.84,
482
+ "learning_rate": 8.357487922705315e-05,
483
+ "loss": 0.0567,
484
+ "step": 780
485
+ },
486
+ {
487
+ "epoch": 1.86,
488
+ "learning_rate": 8.333333333333334e-05,
489
+ "loss": 0.0535,
490
+ "step": 790
491
+ },
492
+ {
493
+ "epoch": 1.89,
494
+ "learning_rate": 8.309178743961354e-05,
495
+ "loss": 0.0453,
496
+ "step": 800
497
+ },
498
+ {
499
+ "epoch": 1.91,
500
+ "learning_rate": 8.285024154589373e-05,
501
+ "loss": 0.0437,
502
+ "step": 810
503
+ },
504
+ {
505
+ "epoch": 1.93,
506
+ "learning_rate": 8.260869565217392e-05,
507
+ "loss": 0.0549,
508
+ "step": 820
509
+ },
510
+ {
511
+ "epoch": 1.96,
512
+ "learning_rate": 8.236714975845411e-05,
513
+ "loss": 0.0522,
514
+ "step": 830
515
+ },
516
+ {
517
+ "epoch": 1.98,
518
+ "learning_rate": 8.21256038647343e-05,
519
+ "loss": 0.049,
520
+ "step": 840
521
+ },
522
+ {
523
+ "epoch": 2.0,
524
+ "eval_loss": 1.9635112285614014,
525
+ "eval_runtime": 26.4849,
526
+ "eval_samples_per_second": 28.431,
527
+ "eval_steps_per_second": 1.812,
528
+ "step": 848
529
+ },
530
+ {
531
+ "epoch": 2.0,
532
+ "learning_rate": 8.18840579710145e-05,
533
+ "loss": 0.0426,
534
+ "step": 850
535
+ },
536
+ {
537
+ "epoch": 2.03,
538
+ "learning_rate": 8.164251207729469e-05,
539
+ "loss": 0.0259,
540
+ "step": 860
541
+ },
542
+ {
543
+ "epoch": 2.05,
544
+ "learning_rate": 8.140096618357488e-05,
545
+ "loss": 0.0255,
546
+ "step": 870
547
+ },
548
+ {
549
+ "epoch": 2.08,
550
+ "learning_rate": 8.115942028985508e-05,
551
+ "loss": 0.0209,
552
+ "step": 880
553
+ },
554
+ {
555
+ "epoch": 2.1,
556
+ "learning_rate": 8.091787439613528e-05,
557
+ "loss": 0.0244,
558
+ "step": 890
559
+ },
560
+ {
561
+ "epoch": 2.12,
562
+ "learning_rate": 8.067632850241547e-05,
563
+ "loss": 0.0319,
564
+ "step": 900
565
+ },
566
+ {
567
+ "epoch": 2.15,
568
+ "learning_rate": 8.043478260869566e-05,
569
+ "loss": 0.0257,
570
+ "step": 910
571
+ },
572
+ {
573
+ "epoch": 2.17,
574
+ "learning_rate": 8.019323671497585e-05,
575
+ "loss": 0.0286,
576
+ "step": 920
577
+ },
578
+ {
579
+ "epoch": 2.19,
580
+ "learning_rate": 7.995169082125604e-05,
581
+ "loss": 0.0265,
582
+ "step": 930
583
+ },
584
+ {
585
+ "epoch": 2.22,
586
+ "learning_rate": 7.971014492753623e-05,
587
+ "loss": 0.0229,
588
+ "step": 940
589
+ },
590
+ {
591
+ "epoch": 2.24,
592
+ "learning_rate": 7.946859903381643e-05,
593
+ "loss": 0.0237,
594
+ "step": 950
595
+ },
596
+ {
597
+ "epoch": 2.26,
598
+ "learning_rate": 7.922705314009662e-05,
599
+ "loss": 0.0313,
600
+ "step": 960
601
+ },
602
+ {
603
+ "epoch": 2.29,
604
+ "learning_rate": 7.898550724637681e-05,
605
+ "loss": 0.035,
606
+ "step": 970
607
+ },
608
+ {
609
+ "epoch": 2.31,
610
+ "learning_rate": 7.874396135265701e-05,
611
+ "loss": 0.0324,
612
+ "step": 980
613
+ },
614
+ {
615
+ "epoch": 2.33,
616
+ "learning_rate": 7.85024154589372e-05,
617
+ "loss": 0.029,
618
+ "step": 990
619
+ },
620
+ {
621
+ "epoch": 2.36,
622
+ "learning_rate": 7.82608695652174e-05,
623
+ "loss": 0.0297,
624
+ "step": 1000
625
+ },
626
+ {
627
+ "epoch": 2.38,
628
+ "learning_rate": 7.801932367149759e-05,
629
+ "loss": 0.029,
630
+ "step": 1010
631
+ },
632
+ {
633
+ "epoch": 2.41,
634
+ "learning_rate": 7.777777777777778e-05,
635
+ "loss": 0.0289,
636
+ "step": 1020
637
+ },
638
+ {
639
+ "epoch": 2.43,
640
+ "learning_rate": 7.753623188405797e-05,
641
+ "loss": 0.0267,
642
+ "step": 1030
643
+ },
644
+ {
645
+ "epoch": 2.45,
646
+ "learning_rate": 7.729468599033817e-05,
647
+ "loss": 0.0373,
648
+ "step": 1040
649
+ },
650
+ {
651
+ "epoch": 2.48,
652
+ "learning_rate": 7.705314009661836e-05,
653
+ "loss": 0.0269,
654
+ "step": 1050
655
+ },
656
+ {
657
+ "epoch": 2.5,
658
+ "learning_rate": 7.681159420289855e-05,
659
+ "loss": 0.0237,
660
+ "step": 1060
661
+ },
662
+ {
663
+ "epoch": 2.52,
664
+ "learning_rate": 7.657004830917874e-05,
665
+ "loss": 0.0222,
666
+ "step": 1070
667
+ },
668
+ {
669
+ "epoch": 2.55,
670
+ "learning_rate": 7.632850241545893e-05,
671
+ "loss": 0.0378,
672
+ "step": 1080
673
+ },
674
+ {
675
+ "epoch": 2.57,
676
+ "learning_rate": 7.608695652173914e-05,
677
+ "loss": 0.0311,
678
+ "step": 1090
679
+ },
680
+ {
681
+ "epoch": 2.59,
682
+ "learning_rate": 7.584541062801933e-05,
683
+ "loss": 0.0243,
684
+ "step": 1100
685
+ },
686
+ {
687
+ "epoch": 2.62,
688
+ "learning_rate": 7.560386473429952e-05,
689
+ "loss": 0.0295,
690
+ "step": 1110
691
+ },
692
+ {
693
+ "epoch": 2.64,
694
+ "learning_rate": 7.536231884057971e-05,
695
+ "loss": 0.0293,
696
+ "step": 1120
697
+ },
698
+ {
699
+ "epoch": 2.67,
700
+ "learning_rate": 7.51207729468599e-05,
701
+ "loss": 0.029,
702
+ "step": 1130
703
+ },
704
+ {
705
+ "epoch": 2.69,
706
+ "learning_rate": 7.48792270531401e-05,
707
+ "loss": 0.0289,
708
+ "step": 1140
709
+ },
710
+ {
711
+ "epoch": 2.71,
712
+ "learning_rate": 7.463768115942029e-05,
713
+ "loss": 0.0394,
714
+ "step": 1150
715
+ },
716
+ {
717
+ "epoch": 2.74,
718
+ "learning_rate": 7.439613526570048e-05,
719
+ "loss": 0.0355,
720
+ "step": 1160
721
+ },
722
+ {
723
+ "epoch": 2.76,
724
+ "learning_rate": 7.415458937198067e-05,
725
+ "loss": 0.0273,
726
+ "step": 1170
727
+ },
728
+ {
729
+ "epoch": 2.78,
730
+ "learning_rate": 7.391304347826086e-05,
731
+ "loss": 0.0312,
732
+ "step": 1180
733
+ },
734
+ {
735
+ "epoch": 2.81,
736
+ "learning_rate": 7.367149758454107e-05,
737
+ "loss": 0.0284,
738
+ "step": 1190
739
+ },
740
+ {
741
+ "epoch": 2.83,
742
+ "learning_rate": 7.342995169082126e-05,
743
+ "loss": 0.0363,
744
+ "step": 1200
745
+ },
746
+ {
747
+ "epoch": 2.85,
748
+ "learning_rate": 7.318840579710145e-05,
749
+ "loss": 0.0307,
750
+ "step": 1210
751
+ },
752
+ {
753
+ "epoch": 2.88,
754
+ "learning_rate": 7.294685990338164e-05,
755
+ "loss": 0.0357,
756
+ "step": 1220
757
+ },
758
+ {
759
+ "epoch": 2.9,
760
+ "learning_rate": 7.270531400966184e-05,
761
+ "loss": 0.0332,
762
+ "step": 1230
763
+ },
764
+ {
765
+ "epoch": 2.92,
766
+ "learning_rate": 7.246376811594203e-05,
767
+ "loss": 0.0361,
768
+ "step": 1240
769
+ },
770
+ {
771
+ "epoch": 2.95,
772
+ "learning_rate": 7.222222222222222e-05,
773
+ "loss": 0.0318,
774
+ "step": 1250
775
+ },
776
+ {
777
+ "epoch": 2.97,
778
+ "learning_rate": 7.198067632850241e-05,
779
+ "loss": 0.0306,
780
+ "step": 1260
781
+ },
782
+ {
783
+ "epoch": 3.0,
784
+ "learning_rate": 7.17391304347826e-05,
785
+ "loss": 0.0281,
786
+ "step": 1270
787
+ },
788
+ {
789
+ "epoch": 3.0,
790
+ "eval_loss": 1.9812633991241455,
791
+ "eval_runtime": 27.1954,
792
+ "eval_samples_per_second": 27.689,
793
+ "eval_steps_per_second": 1.765,
794
+ "step": 1272
795
+ },
796
+ {
797
+ "epoch": 3.02,
798
+ "learning_rate": 7.14975845410628e-05,
799
+ "loss": 0.0175,
800
+ "step": 1280
801
+ },
802
+ {
803
+ "epoch": 3.04,
804
+ "learning_rate": 7.1256038647343e-05,
805
+ "loss": 0.0208,
806
+ "step": 1290
807
+ },
808
+ {
809
+ "epoch": 3.07,
810
+ "learning_rate": 7.101449275362319e-05,
811
+ "loss": 0.0153,
812
+ "step": 1300
813
+ },
814
+ {
815
+ "epoch": 3.09,
816
+ "learning_rate": 7.077294685990338e-05,
817
+ "loss": 0.0147,
818
+ "step": 1310
819
+ },
820
+ {
821
+ "epoch": 3.11,
822
+ "learning_rate": 7.053140096618357e-05,
823
+ "loss": 0.0193,
824
+ "step": 1320
825
+ },
826
+ {
827
+ "epoch": 3.14,
828
+ "learning_rate": 7.028985507246377e-05,
829
+ "loss": 0.0173,
830
+ "step": 1330
831
+ },
832
+ {
833
+ "epoch": 3.16,
834
+ "learning_rate": 7.004830917874396e-05,
835
+ "loss": 0.0291,
836
+ "step": 1340
837
+ },
838
+ {
839
+ "epoch": 3.18,
840
+ "learning_rate": 6.980676328502416e-05,
841
+ "loss": 0.019,
842
+ "step": 1350
843
+ },
844
+ {
845
+ "epoch": 3.21,
846
+ "learning_rate": 6.956521739130436e-05,
847
+ "loss": 0.015,
848
+ "step": 1360
849
+ },
850
+ {
851
+ "epoch": 3.23,
852
+ "learning_rate": 6.932367149758455e-05,
853
+ "loss": 0.0208,
854
+ "step": 1370
855
+ },
856
+ {
857
+ "epoch": 3.25,
858
+ "learning_rate": 6.908212560386474e-05,
859
+ "loss": 0.0185,
860
+ "step": 1380
861
+ },
862
+ {
863
+ "epoch": 3.28,
864
+ "learning_rate": 6.884057971014493e-05,
865
+ "loss": 0.0162,
866
+ "step": 1390
867
+ },
868
+ {
869
+ "epoch": 3.3,
870
+ "learning_rate": 6.859903381642512e-05,
871
+ "loss": 0.0166,
872
+ "step": 1400
873
+ },
874
+ {
875
+ "epoch": 3.33,
876
+ "learning_rate": 6.835748792270533e-05,
877
+ "loss": 0.0178,
878
+ "step": 1410
879
+ },
880
+ {
881
+ "epoch": 3.35,
882
+ "learning_rate": 6.811594202898552e-05,
883
+ "loss": 0.0163,
884
+ "step": 1420
885
+ },
886
+ {
887
+ "epoch": 3.37,
888
+ "learning_rate": 6.787439613526571e-05,
889
+ "loss": 0.0227,
890
+ "step": 1430
891
+ },
892
+ {
893
+ "epoch": 3.4,
894
+ "learning_rate": 6.76328502415459e-05,
895
+ "loss": 0.0144,
896
+ "step": 1440
897
+ },
898
+ {
899
+ "epoch": 3.42,
900
+ "learning_rate": 6.73913043478261e-05,
901
+ "loss": 0.0156,
902
+ "step": 1450
903
+ },
904
+ {
905
+ "epoch": 3.44,
906
+ "learning_rate": 6.714975845410629e-05,
907
+ "loss": 0.014,
908
+ "step": 1460
909
+ },
910
+ {
911
+ "epoch": 3.47,
912
+ "learning_rate": 6.690821256038648e-05,
913
+ "loss": 0.0187,
914
+ "step": 1470
915
+ },
916
+ {
917
+ "epoch": 3.49,
918
+ "learning_rate": 6.666666666666667e-05,
919
+ "loss": 0.0198,
920
+ "step": 1480
921
+ },
922
+ {
923
+ "epoch": 3.51,
924
+ "learning_rate": 6.642512077294686e-05,
925
+ "loss": 0.0127,
926
+ "step": 1490
927
+ },
928
+ {
929
+ "epoch": 3.54,
930
+ "learning_rate": 6.618357487922707e-05,
931
+ "loss": 0.0129,
932
+ "step": 1500
933
+ },
934
+ {
935
+ "epoch": 3.56,
936
+ "learning_rate": 6.594202898550726e-05,
937
+ "loss": 0.0174,
938
+ "step": 1510
939
+ },
940
+ {
941
+ "epoch": 3.58,
942
+ "learning_rate": 6.570048309178745e-05,
943
+ "loss": 0.0147,
944
+ "step": 1520
945
+ },
946
+ {
947
+ "epoch": 3.61,
948
+ "learning_rate": 6.545893719806764e-05,
949
+ "loss": 0.0188,
950
+ "step": 1530
951
+ },
952
+ {
953
+ "epoch": 3.63,
954
+ "learning_rate": 6.521739130434783e-05,
955
+ "loss": 0.0149,
956
+ "step": 1540
957
+ },
958
+ {
959
+ "epoch": 3.66,
960
+ "learning_rate": 6.497584541062803e-05,
961
+ "loss": 0.0255,
962
+ "step": 1550
963
+ },
964
+ {
965
+ "epoch": 3.68,
966
+ "learning_rate": 6.473429951690822e-05,
967
+ "loss": 0.0198,
968
+ "step": 1560
969
+ },
970
+ {
971
+ "epoch": 3.7,
972
+ "learning_rate": 6.449275362318841e-05,
973
+ "loss": 0.0219,
974
+ "step": 1570
975
+ },
976
+ {
977
+ "epoch": 3.73,
978
+ "learning_rate": 6.42512077294686e-05,
979
+ "loss": 0.0213,
980
+ "step": 1580
981
+ },
982
+ {
983
+ "epoch": 3.75,
984
+ "learning_rate": 6.400966183574879e-05,
985
+ "loss": 0.0214,
986
+ "step": 1590
987
+ },
988
+ {
989
+ "epoch": 3.77,
990
+ "learning_rate": 6.376811594202898e-05,
991
+ "loss": 0.0185,
992
+ "step": 1600
993
+ },
994
+ {
995
+ "epoch": 3.8,
996
+ "learning_rate": 6.352657004830919e-05,
997
+ "loss": 0.0191,
998
+ "step": 1610
999
+ },
1000
+ {
1001
+ "epoch": 3.82,
1002
+ "learning_rate": 6.328502415458938e-05,
1003
+ "loss": 0.0165,
1004
+ "step": 1620
1005
+ },
1006
+ {
1007
+ "epoch": 3.84,
1008
+ "learning_rate": 6.304347826086957e-05,
1009
+ "loss": 0.0205,
1010
+ "step": 1630
1011
+ },
1012
+ {
1013
+ "epoch": 3.87,
1014
+ "learning_rate": 6.280193236714976e-05,
1015
+ "loss": 0.017,
1016
+ "step": 1640
1017
+ },
1018
+ {
1019
+ "epoch": 3.89,
1020
+ "learning_rate": 6.256038647342996e-05,
1021
+ "loss": 0.0198,
1022
+ "step": 1650
1023
+ },
1024
+ {
1025
+ "epoch": 3.92,
1026
+ "learning_rate": 6.231884057971015e-05,
1027
+ "loss": 0.0171,
1028
+ "step": 1660
1029
+ },
1030
+ {
1031
+ "epoch": 3.94,
1032
+ "learning_rate": 6.207729468599034e-05,
1033
+ "loss": 0.0201,
1034
+ "step": 1670
1035
+ },
1036
+ {
1037
+ "epoch": 3.96,
1038
+ "learning_rate": 6.183574879227053e-05,
1039
+ "loss": 0.0253,
1040
+ "step": 1680
1041
+ },
1042
+ {
1043
+ "epoch": 3.99,
1044
+ "learning_rate": 6.159420289855072e-05,
1045
+ "loss": 0.0204,
1046
+ "step": 1690
1047
+ },
1048
+ {
1049
+ "epoch": 4.0,
1050
+ "eval_loss": 2.0079989433288574,
1051
+ "eval_runtime": 26.4901,
1052
+ "eval_samples_per_second": 28.426,
1053
+ "eval_steps_per_second": 1.812,
1054
+ "step": 1696
1055
+ },
1056
+ {
1057
+ "epoch": 4.01,
1058
+ "learning_rate": 6.135265700483092e-05,
1059
+ "loss": 0.0136,
1060
+ "step": 1700
1061
+ },
1062
+ {
1063
+ "epoch": 4.03,
1064
+ "learning_rate": 6.111111111111112e-05,
1065
+ "loss": 0.0131,
1066
+ "step": 1710
1067
+ },
1068
+ {
1069
+ "epoch": 4.06,
1070
+ "learning_rate": 6.086956521739131e-05,
1071
+ "loss": 0.013,
1072
+ "step": 1720
1073
+ },
1074
+ {
1075
+ "epoch": 4.08,
1076
+ "learning_rate": 6.0628019323671504e-05,
1077
+ "loss": 0.009,
1078
+ "step": 1730
1079
+ },
1080
+ {
1081
+ "epoch": 4.1,
1082
+ "learning_rate": 6.0386473429951696e-05,
1083
+ "loss": 0.0075,
1084
+ "step": 1740
1085
+ },
1086
+ {
1087
+ "epoch": 4.13,
1088
+ "learning_rate": 6.014492753623189e-05,
1089
+ "loss": 0.0095,
1090
+ "step": 1750
1091
+ },
1092
+ {
1093
+ "epoch": 4.15,
1094
+ "learning_rate": 5.990338164251208e-05,
1095
+ "loss": 0.0078,
1096
+ "step": 1760
1097
+ },
1098
+ {
1099
+ "epoch": 4.17,
1100
+ "learning_rate": 5.966183574879227e-05,
1101
+ "loss": 0.0122,
1102
+ "step": 1770
1103
+ },
1104
+ {
1105
+ "epoch": 4.2,
1106
+ "learning_rate": 5.942028985507246e-05,
1107
+ "loss": 0.0089,
1108
+ "step": 1780
1109
+ },
1110
+ {
1111
+ "epoch": 4.22,
1112
+ "learning_rate": 5.9178743961352655e-05,
1113
+ "loss": 0.0085,
1114
+ "step": 1790
1115
+ },
1116
+ {
1117
+ "epoch": 4.25,
1118
+ "learning_rate": 5.8937198067632847e-05,
1119
+ "loss": 0.0103,
1120
+ "step": 1800
1121
+ },
1122
+ {
1123
+ "epoch": 4.27,
1124
+ "learning_rate": 5.869565217391305e-05,
1125
+ "loss": 0.0091,
1126
+ "step": 1810
1127
+ },
1128
+ {
1129
+ "epoch": 4.29,
1130
+ "learning_rate": 5.8454106280193244e-05,
1131
+ "loss": 0.0091,
1132
+ "step": 1820
1133
+ },
1134
+ {
1135
+ "epoch": 4.32,
1136
+ "learning_rate": 5.8212560386473435e-05,
1137
+ "loss": 0.0084,
1138
+ "step": 1830
1139
+ },
1140
+ {
1141
+ "epoch": 4.34,
1142
+ "learning_rate": 5.797101449275363e-05,
1143
+ "loss": 0.0095,
1144
+ "step": 1840
1145
+ },
1146
+ {
1147
+ "epoch": 4.36,
1148
+ "learning_rate": 5.772946859903382e-05,
1149
+ "loss": 0.0088,
1150
+ "step": 1850
1151
+ },
1152
+ {
1153
+ "epoch": 4.39,
1154
+ "learning_rate": 5.748792270531401e-05,
1155
+ "loss": 0.0096,
1156
+ "step": 1860
1157
+ },
1158
+ {
1159
+ "epoch": 4.41,
1160
+ "learning_rate": 5.72463768115942e-05,
1161
+ "loss": 0.0094,
1162
+ "step": 1870
1163
+ },
1164
+ {
1165
+ "epoch": 4.43,
1166
+ "learning_rate": 5.7004830917874394e-05,
1167
+ "loss": 0.0111,
1168
+ "step": 1880
1169
+ },
1170
+ {
1171
+ "epoch": 4.46,
1172
+ "learning_rate": 5.6763285024154586e-05,
1173
+ "loss": 0.015,
1174
+ "step": 1890
1175
+ },
1176
+ {
1177
+ "epoch": 4.48,
1178
+ "learning_rate": 5.652173913043478e-05,
1179
+ "loss": 0.0078,
1180
+ "step": 1900
1181
+ },
1182
+ {
1183
+ "epoch": 4.5,
1184
+ "learning_rate": 5.6280193236714976e-05,
1185
+ "loss": 0.0113,
1186
+ "step": 1910
1187
+ },
1188
+ {
1189
+ "epoch": 4.53,
1190
+ "learning_rate": 5.6038647342995175e-05,
1191
+ "loss": 0.0114,
1192
+ "step": 1920
1193
+ },
1194
+ {
1195
+ "epoch": 4.55,
1196
+ "learning_rate": 5.5797101449275366e-05,
1197
+ "loss": 0.0089,
1198
+ "step": 1930
1199
+ },
1200
+ {
1201
+ "epoch": 4.58,
1202
+ "learning_rate": 5.555555555555556e-05,
1203
+ "loss": 0.0137,
1204
+ "step": 1940
1205
+ },
1206
+ {
1207
+ "epoch": 4.6,
1208
+ "learning_rate": 5.531400966183575e-05,
1209
+ "loss": 0.0122,
1210
+ "step": 1950
1211
+ },
1212
+ {
1213
+ "epoch": 4.62,
1214
+ "learning_rate": 5.507246376811594e-05,
1215
+ "loss": 0.0075,
1216
+ "step": 1960
1217
+ },
1218
+ {
1219
+ "epoch": 4.65,
1220
+ "learning_rate": 5.483091787439614e-05,
1221
+ "loss": 0.011,
1222
+ "step": 1970
1223
+ },
1224
+ {
1225
+ "epoch": 4.67,
1226
+ "learning_rate": 5.458937198067633e-05,
1227
+ "loss": 0.0105,
1228
+ "step": 1980
1229
+ },
1230
+ {
1231
+ "epoch": 4.69,
1232
+ "learning_rate": 5.4347826086956524e-05,
1233
+ "loss": 0.0108,
1234
+ "step": 1990
1235
+ },
1236
+ {
1237
+ "epoch": 4.72,
1238
+ "learning_rate": 5.4106280193236716e-05,
1239
+ "loss": 0.0093,
1240
+ "step": 2000
1241
+ }
1242
+ ],
1243
+ "max_steps": 4240,
1244
+ "num_train_epochs": 10,
1245
+ "total_flos": 0.0,
1246
+ "trial_name": null,
1247
+ "trial_params": null
1248
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d19fa419d807bcbfef5a86725e8eca48bdf21b5f0b9a046119a2780ac6238051
3
+ size 3695
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.21.1",
58
+ "use_cache": true,
59
+ "vocab_size": 32128
60
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa2c64da4ff2e97cada62ba26f541dc4cd118a576d68c6ee3a09a7d8baf43b1e
3
+ size 14503
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38681d5def2caab7afa29ea06d6f3c2a336d06c4f7df84205bf13fdaae6acfa5
3
+ size 623
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/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
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/tokenizer_config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "extra_ids": 100,
106
+ "model_max_length": 512,
107
+ "name_or_path": "t5-base",
108
+ "pad_token": "<pad>",
109
+ "special_tokens_map_file": null,
110
+ "tokenizer_class": "T5Tokenizer",
111
+ "unk_token": "<unk>"
112
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/trainer_state.json ADDED
@@ -0,0 +1,1406 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.306603773584905,
5
+ "global_step": 2250,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 1e-05,
13
+ "loss": 0.002,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 2e-05,
19
+ "loss": 0.002,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.07,
24
+ "learning_rate": 3e-05,
25
+ "loss": 0.002,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.09,
30
+ "learning_rate": 4e-05,
31
+ "loss": 0.0023,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.12,
36
+ "learning_rate": 5e-05,
37
+ "loss": 0.0021,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.14,
42
+ "learning_rate": 6e-05,
43
+ "loss": 0.0022,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "learning_rate": 7e-05,
49
+ "loss": 0.0037,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.19,
54
+ "learning_rate": 8e-05,
55
+ "loss": 0.0054,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.21,
60
+ "learning_rate": 9e-05,
61
+ "loss": 0.0153,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.24,
66
+ "learning_rate": 0.0001,
67
+ "loss": 0.0338,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.26,
72
+ "learning_rate": 9.97584541062802e-05,
73
+ "loss": 0.0546,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.28,
78
+ "learning_rate": 9.951690821256039e-05,
79
+ "loss": 0.0706,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.31,
84
+ "learning_rate": 9.927536231884058e-05,
85
+ "loss": 0.0603,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.33,
90
+ "learning_rate": 9.903381642512077e-05,
91
+ "loss": 0.0698,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.35,
96
+ "learning_rate": 9.879227053140096e-05,
97
+ "loss": 0.0808,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.38,
102
+ "learning_rate": 9.855072463768117e-05,
103
+ "loss": 0.0792,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.4,
108
+ "learning_rate": 9.830917874396136e-05,
109
+ "loss": 0.0656,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.42,
114
+ "learning_rate": 9.806763285024155e-05,
115
+ "loss": 0.0673,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.45,
120
+ "learning_rate": 9.782608695652174e-05,
121
+ "loss": 0.0686,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.47,
126
+ "learning_rate": 9.758454106280194e-05,
127
+ "loss": 0.0717,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.5,
132
+ "learning_rate": 9.734299516908213e-05,
133
+ "loss": 0.066,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.52,
138
+ "learning_rate": 9.710144927536232e-05,
139
+ "loss": 0.08,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.54,
144
+ "learning_rate": 9.685990338164251e-05,
145
+ "loss": 0.084,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.57,
150
+ "learning_rate": 9.66183574879227e-05,
151
+ "loss": 0.0703,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.59,
156
+ "learning_rate": 9.63768115942029e-05,
157
+ "loss": 0.0589,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.61,
162
+ "learning_rate": 9.61352657004831e-05,
163
+ "loss": 0.0708,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.64,
168
+ "learning_rate": 9.589371980676329e-05,
169
+ "loss": 0.0661,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.66,
174
+ "learning_rate": 9.565217391304348e-05,
175
+ "loss": 0.0649,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.68,
180
+ "learning_rate": 9.541062801932368e-05,
181
+ "loss": 0.063,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.71,
186
+ "learning_rate": 9.516908212560387e-05,
187
+ "loss": 0.0628,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.73,
192
+ "learning_rate": 9.492753623188406e-05,
193
+ "loss": 0.0699,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.75,
198
+ "learning_rate": 9.468599033816425e-05,
199
+ "loss": 0.0716,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.78,
204
+ "learning_rate": 9.444444444444444e-05,
205
+ "loss": 0.0623,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.8,
210
+ "learning_rate": 9.420289855072463e-05,
211
+ "loss": 0.0661,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.83,
216
+ "learning_rate": 9.396135265700483e-05,
217
+ "loss": 0.0603,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.85,
222
+ "learning_rate": 9.371980676328503e-05,
223
+ "loss": 0.0591,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.87,
228
+ "learning_rate": 9.347826086956522e-05,
229
+ "loss": 0.054,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.9,
234
+ "learning_rate": 9.323671497584541e-05,
235
+ "loss": 0.0582,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.92,
240
+ "learning_rate": 9.29951690821256e-05,
241
+ "loss": 0.0617,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.94,
246
+ "learning_rate": 9.27536231884058e-05,
247
+ "loss": 0.0615,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.97,
252
+ "learning_rate": 9.251207729468599e-05,
253
+ "loss": 0.0648,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.99,
258
+ "learning_rate": 9.227053140096618e-05,
259
+ "loss": 0.0607,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 1.0,
264
+ "eval_loss": 1.95200514793396,
265
+ "eval_runtime": 26.5048,
266
+ "eval_samples_per_second": 28.41,
267
+ "eval_steps_per_second": 1.811,
268
+ "step": 424
269
+ },
270
+ {
271
+ "epoch": 1.01,
272
+ "learning_rate": 9.202898550724637e-05,
273
+ "loss": 0.0516,
274
+ "step": 430
275
+ },
276
+ {
277
+ "epoch": 1.04,
278
+ "learning_rate": 9.178743961352657e-05,
279
+ "loss": 0.0446,
280
+ "step": 440
281
+ },
282
+ {
283
+ "epoch": 1.06,
284
+ "learning_rate": 9.154589371980676e-05,
285
+ "loss": 0.0452,
286
+ "step": 450
287
+ },
288
+ {
289
+ "epoch": 1.08,
290
+ "learning_rate": 9.130434782608696e-05,
291
+ "loss": 0.0456,
292
+ "step": 460
293
+ },
294
+ {
295
+ "epoch": 1.11,
296
+ "learning_rate": 9.106280193236715e-05,
297
+ "loss": 0.0483,
298
+ "step": 470
299
+ },
300
+ {
301
+ "epoch": 1.13,
302
+ "learning_rate": 9.082125603864735e-05,
303
+ "loss": 0.0575,
304
+ "step": 480
305
+ },
306
+ {
307
+ "epoch": 1.16,
308
+ "learning_rate": 9.057971014492754e-05,
309
+ "loss": 0.0485,
310
+ "step": 490
311
+ },
312
+ {
313
+ "epoch": 1.18,
314
+ "learning_rate": 9.033816425120773e-05,
315
+ "loss": 0.0471,
316
+ "step": 500
317
+ },
318
+ {
319
+ "epoch": 1.2,
320
+ "learning_rate": 9.009661835748792e-05,
321
+ "loss": 0.0517,
322
+ "step": 510
323
+ },
324
+ {
325
+ "epoch": 1.23,
326
+ "learning_rate": 8.985507246376813e-05,
327
+ "loss": 0.0482,
328
+ "step": 520
329
+ },
330
+ {
331
+ "epoch": 1.25,
332
+ "learning_rate": 8.961352657004832e-05,
333
+ "loss": 0.0449,
334
+ "step": 530
335
+ },
336
+ {
337
+ "epoch": 1.27,
338
+ "learning_rate": 8.937198067632851e-05,
339
+ "loss": 0.0598,
340
+ "step": 540
341
+ },
342
+ {
343
+ "epoch": 1.3,
344
+ "learning_rate": 8.91304347826087e-05,
345
+ "loss": 0.0539,
346
+ "step": 550
347
+ },
348
+ {
349
+ "epoch": 1.32,
350
+ "learning_rate": 8.888888888888889e-05,
351
+ "loss": 0.0481,
352
+ "step": 560
353
+ },
354
+ {
355
+ "epoch": 1.34,
356
+ "learning_rate": 8.864734299516908e-05,
357
+ "loss": 0.0553,
358
+ "step": 570
359
+ },
360
+ {
361
+ "epoch": 1.37,
362
+ "learning_rate": 8.840579710144929e-05,
363
+ "loss": 0.0474,
364
+ "step": 580
365
+ },
366
+ {
367
+ "epoch": 1.39,
368
+ "learning_rate": 8.816425120772948e-05,
369
+ "loss": 0.0536,
370
+ "step": 590
371
+ },
372
+ {
373
+ "epoch": 1.42,
374
+ "learning_rate": 8.792270531400967e-05,
375
+ "loss": 0.0552,
376
+ "step": 600
377
+ },
378
+ {
379
+ "epoch": 1.44,
380
+ "learning_rate": 8.768115942028987e-05,
381
+ "loss": 0.0494,
382
+ "step": 610
383
+ },
384
+ {
385
+ "epoch": 1.46,
386
+ "learning_rate": 8.743961352657006e-05,
387
+ "loss": 0.0605,
388
+ "step": 620
389
+ },
390
+ {
391
+ "epoch": 1.49,
392
+ "learning_rate": 8.719806763285025e-05,
393
+ "loss": 0.0571,
394
+ "step": 630
395
+ },
396
+ {
397
+ "epoch": 1.51,
398
+ "learning_rate": 8.695652173913044e-05,
399
+ "loss": 0.0468,
400
+ "step": 640
401
+ },
402
+ {
403
+ "epoch": 1.53,
404
+ "learning_rate": 8.671497584541063e-05,
405
+ "loss": 0.058,
406
+ "step": 650
407
+ },
408
+ {
409
+ "epoch": 1.56,
410
+ "learning_rate": 8.647342995169082e-05,
411
+ "loss": 0.0523,
412
+ "step": 660
413
+ },
414
+ {
415
+ "epoch": 1.58,
416
+ "learning_rate": 8.623188405797103e-05,
417
+ "loss": 0.0391,
418
+ "step": 670
419
+ },
420
+ {
421
+ "epoch": 1.6,
422
+ "learning_rate": 8.599033816425122e-05,
423
+ "loss": 0.0488,
424
+ "step": 680
425
+ },
426
+ {
427
+ "epoch": 1.63,
428
+ "learning_rate": 8.574879227053141e-05,
429
+ "loss": 0.056,
430
+ "step": 690
431
+ },
432
+ {
433
+ "epoch": 1.65,
434
+ "learning_rate": 8.55072463768116e-05,
435
+ "loss": 0.0599,
436
+ "step": 700
437
+ },
438
+ {
439
+ "epoch": 1.67,
440
+ "learning_rate": 8.52657004830918e-05,
441
+ "loss": 0.0536,
442
+ "step": 710
443
+ },
444
+ {
445
+ "epoch": 1.7,
446
+ "learning_rate": 8.502415458937199e-05,
447
+ "loss": 0.0534,
448
+ "step": 720
449
+ },
450
+ {
451
+ "epoch": 1.72,
452
+ "learning_rate": 8.478260869565218e-05,
453
+ "loss": 0.0484,
454
+ "step": 730
455
+ },
456
+ {
457
+ "epoch": 1.75,
458
+ "learning_rate": 8.454106280193237e-05,
459
+ "loss": 0.0476,
460
+ "step": 740
461
+ },
462
+ {
463
+ "epoch": 1.77,
464
+ "learning_rate": 8.429951690821256e-05,
465
+ "loss": 0.0426,
466
+ "step": 750
467
+ },
468
+ {
469
+ "epoch": 1.79,
470
+ "learning_rate": 8.405797101449276e-05,
471
+ "loss": 0.0486,
472
+ "step": 760
473
+ },
474
+ {
475
+ "epoch": 1.82,
476
+ "learning_rate": 8.381642512077295e-05,
477
+ "loss": 0.0598,
478
+ "step": 770
479
+ },
480
+ {
481
+ "epoch": 1.84,
482
+ "learning_rate": 8.357487922705315e-05,
483
+ "loss": 0.0567,
484
+ "step": 780
485
+ },
486
+ {
487
+ "epoch": 1.86,
488
+ "learning_rate": 8.333333333333334e-05,
489
+ "loss": 0.0535,
490
+ "step": 790
491
+ },
492
+ {
493
+ "epoch": 1.89,
494
+ "learning_rate": 8.309178743961354e-05,
495
+ "loss": 0.0453,
496
+ "step": 800
497
+ },
498
+ {
499
+ "epoch": 1.91,
500
+ "learning_rate": 8.285024154589373e-05,
501
+ "loss": 0.0437,
502
+ "step": 810
503
+ },
504
+ {
505
+ "epoch": 1.93,
506
+ "learning_rate": 8.260869565217392e-05,
507
+ "loss": 0.0549,
508
+ "step": 820
509
+ },
510
+ {
511
+ "epoch": 1.96,
512
+ "learning_rate": 8.236714975845411e-05,
513
+ "loss": 0.0522,
514
+ "step": 830
515
+ },
516
+ {
517
+ "epoch": 1.98,
518
+ "learning_rate": 8.21256038647343e-05,
519
+ "loss": 0.049,
520
+ "step": 840
521
+ },
522
+ {
523
+ "epoch": 2.0,
524
+ "eval_loss": 1.9635112285614014,
525
+ "eval_runtime": 26.4849,
526
+ "eval_samples_per_second": 28.431,
527
+ "eval_steps_per_second": 1.812,
528
+ "step": 848
529
+ },
530
+ {
531
+ "epoch": 2.0,
532
+ "learning_rate": 8.18840579710145e-05,
533
+ "loss": 0.0426,
534
+ "step": 850
535
+ },
536
+ {
537
+ "epoch": 2.03,
538
+ "learning_rate": 8.164251207729469e-05,
539
+ "loss": 0.0259,
540
+ "step": 860
541
+ },
542
+ {
543
+ "epoch": 2.05,
544
+ "learning_rate": 8.140096618357488e-05,
545
+ "loss": 0.0255,
546
+ "step": 870
547
+ },
548
+ {
549
+ "epoch": 2.08,
550
+ "learning_rate": 8.115942028985508e-05,
551
+ "loss": 0.0209,
552
+ "step": 880
553
+ },
554
+ {
555
+ "epoch": 2.1,
556
+ "learning_rate": 8.091787439613528e-05,
557
+ "loss": 0.0244,
558
+ "step": 890
559
+ },
560
+ {
561
+ "epoch": 2.12,
562
+ "learning_rate": 8.067632850241547e-05,
563
+ "loss": 0.0319,
564
+ "step": 900
565
+ },
566
+ {
567
+ "epoch": 2.15,
568
+ "learning_rate": 8.043478260869566e-05,
569
+ "loss": 0.0257,
570
+ "step": 910
571
+ },
572
+ {
573
+ "epoch": 2.17,
574
+ "learning_rate": 8.019323671497585e-05,
575
+ "loss": 0.0286,
576
+ "step": 920
577
+ },
578
+ {
579
+ "epoch": 2.19,
580
+ "learning_rate": 7.995169082125604e-05,
581
+ "loss": 0.0265,
582
+ "step": 930
583
+ },
584
+ {
585
+ "epoch": 2.22,
586
+ "learning_rate": 7.971014492753623e-05,
587
+ "loss": 0.0229,
588
+ "step": 940
589
+ },
590
+ {
591
+ "epoch": 2.24,
592
+ "learning_rate": 7.946859903381643e-05,
593
+ "loss": 0.0237,
594
+ "step": 950
595
+ },
596
+ {
597
+ "epoch": 2.26,
598
+ "learning_rate": 7.922705314009662e-05,
599
+ "loss": 0.0313,
600
+ "step": 960
601
+ },
602
+ {
603
+ "epoch": 2.29,
604
+ "learning_rate": 7.898550724637681e-05,
605
+ "loss": 0.035,
606
+ "step": 970
607
+ },
608
+ {
609
+ "epoch": 2.31,
610
+ "learning_rate": 7.874396135265701e-05,
611
+ "loss": 0.0324,
612
+ "step": 980
613
+ },
614
+ {
615
+ "epoch": 2.33,
616
+ "learning_rate": 7.85024154589372e-05,
617
+ "loss": 0.029,
618
+ "step": 990
619
+ },
620
+ {
621
+ "epoch": 2.36,
622
+ "learning_rate": 7.82608695652174e-05,
623
+ "loss": 0.0297,
624
+ "step": 1000
625
+ },
626
+ {
627
+ "epoch": 2.38,
628
+ "learning_rate": 7.801932367149759e-05,
629
+ "loss": 0.029,
630
+ "step": 1010
631
+ },
632
+ {
633
+ "epoch": 2.41,
634
+ "learning_rate": 7.777777777777778e-05,
635
+ "loss": 0.0289,
636
+ "step": 1020
637
+ },
638
+ {
639
+ "epoch": 2.43,
640
+ "learning_rate": 7.753623188405797e-05,
641
+ "loss": 0.0267,
642
+ "step": 1030
643
+ },
644
+ {
645
+ "epoch": 2.45,
646
+ "learning_rate": 7.729468599033817e-05,
647
+ "loss": 0.0373,
648
+ "step": 1040
649
+ },
650
+ {
651
+ "epoch": 2.48,
652
+ "learning_rate": 7.705314009661836e-05,
653
+ "loss": 0.0269,
654
+ "step": 1050
655
+ },
656
+ {
657
+ "epoch": 2.5,
658
+ "learning_rate": 7.681159420289855e-05,
659
+ "loss": 0.0237,
660
+ "step": 1060
661
+ },
662
+ {
663
+ "epoch": 2.52,
664
+ "learning_rate": 7.657004830917874e-05,
665
+ "loss": 0.0222,
666
+ "step": 1070
667
+ },
668
+ {
669
+ "epoch": 2.55,
670
+ "learning_rate": 7.632850241545893e-05,
671
+ "loss": 0.0378,
672
+ "step": 1080
673
+ },
674
+ {
675
+ "epoch": 2.57,
676
+ "learning_rate": 7.608695652173914e-05,
677
+ "loss": 0.0311,
678
+ "step": 1090
679
+ },
680
+ {
681
+ "epoch": 2.59,
682
+ "learning_rate": 7.584541062801933e-05,
683
+ "loss": 0.0243,
684
+ "step": 1100
685
+ },
686
+ {
687
+ "epoch": 2.62,
688
+ "learning_rate": 7.560386473429952e-05,
689
+ "loss": 0.0295,
690
+ "step": 1110
691
+ },
692
+ {
693
+ "epoch": 2.64,
694
+ "learning_rate": 7.536231884057971e-05,
695
+ "loss": 0.0293,
696
+ "step": 1120
697
+ },
698
+ {
699
+ "epoch": 2.67,
700
+ "learning_rate": 7.51207729468599e-05,
701
+ "loss": 0.029,
702
+ "step": 1130
703
+ },
704
+ {
705
+ "epoch": 2.69,
706
+ "learning_rate": 7.48792270531401e-05,
707
+ "loss": 0.0289,
708
+ "step": 1140
709
+ },
710
+ {
711
+ "epoch": 2.71,
712
+ "learning_rate": 7.463768115942029e-05,
713
+ "loss": 0.0394,
714
+ "step": 1150
715
+ },
716
+ {
717
+ "epoch": 2.74,
718
+ "learning_rate": 7.439613526570048e-05,
719
+ "loss": 0.0355,
720
+ "step": 1160
721
+ },
722
+ {
723
+ "epoch": 2.76,
724
+ "learning_rate": 7.415458937198067e-05,
725
+ "loss": 0.0273,
726
+ "step": 1170
727
+ },
728
+ {
729
+ "epoch": 2.78,
730
+ "learning_rate": 7.391304347826086e-05,
731
+ "loss": 0.0312,
732
+ "step": 1180
733
+ },
734
+ {
735
+ "epoch": 2.81,
736
+ "learning_rate": 7.367149758454107e-05,
737
+ "loss": 0.0284,
738
+ "step": 1190
739
+ },
740
+ {
741
+ "epoch": 2.83,
742
+ "learning_rate": 7.342995169082126e-05,
743
+ "loss": 0.0363,
744
+ "step": 1200
745
+ },
746
+ {
747
+ "epoch": 2.85,
748
+ "learning_rate": 7.318840579710145e-05,
749
+ "loss": 0.0307,
750
+ "step": 1210
751
+ },
752
+ {
753
+ "epoch": 2.88,
754
+ "learning_rate": 7.294685990338164e-05,
755
+ "loss": 0.0357,
756
+ "step": 1220
757
+ },
758
+ {
759
+ "epoch": 2.9,
760
+ "learning_rate": 7.270531400966184e-05,
761
+ "loss": 0.0332,
762
+ "step": 1230
763
+ },
764
+ {
765
+ "epoch": 2.92,
766
+ "learning_rate": 7.246376811594203e-05,
767
+ "loss": 0.0361,
768
+ "step": 1240
769
+ },
770
+ {
771
+ "epoch": 2.95,
772
+ "learning_rate": 7.222222222222222e-05,
773
+ "loss": 0.0318,
774
+ "step": 1250
775
+ },
776
+ {
777
+ "epoch": 2.97,
778
+ "learning_rate": 7.198067632850241e-05,
779
+ "loss": 0.0306,
780
+ "step": 1260
781
+ },
782
+ {
783
+ "epoch": 3.0,
784
+ "learning_rate": 7.17391304347826e-05,
785
+ "loss": 0.0281,
786
+ "step": 1270
787
+ },
788
+ {
789
+ "epoch": 3.0,
790
+ "eval_loss": 1.9812633991241455,
791
+ "eval_runtime": 27.1954,
792
+ "eval_samples_per_second": 27.689,
793
+ "eval_steps_per_second": 1.765,
794
+ "step": 1272
795
+ },
796
+ {
797
+ "epoch": 3.02,
798
+ "learning_rate": 7.14975845410628e-05,
799
+ "loss": 0.0175,
800
+ "step": 1280
801
+ },
802
+ {
803
+ "epoch": 3.04,
804
+ "learning_rate": 7.1256038647343e-05,
805
+ "loss": 0.0208,
806
+ "step": 1290
807
+ },
808
+ {
809
+ "epoch": 3.07,
810
+ "learning_rate": 7.101449275362319e-05,
811
+ "loss": 0.0153,
812
+ "step": 1300
813
+ },
814
+ {
815
+ "epoch": 3.09,
816
+ "learning_rate": 7.077294685990338e-05,
817
+ "loss": 0.0147,
818
+ "step": 1310
819
+ },
820
+ {
821
+ "epoch": 3.11,
822
+ "learning_rate": 7.053140096618357e-05,
823
+ "loss": 0.0193,
824
+ "step": 1320
825
+ },
826
+ {
827
+ "epoch": 3.14,
828
+ "learning_rate": 7.028985507246377e-05,
829
+ "loss": 0.0173,
830
+ "step": 1330
831
+ },
832
+ {
833
+ "epoch": 3.16,
834
+ "learning_rate": 7.004830917874396e-05,
835
+ "loss": 0.0291,
836
+ "step": 1340
837
+ },
838
+ {
839
+ "epoch": 3.18,
840
+ "learning_rate": 6.980676328502416e-05,
841
+ "loss": 0.019,
842
+ "step": 1350
843
+ },
844
+ {
845
+ "epoch": 3.21,
846
+ "learning_rate": 6.956521739130436e-05,
847
+ "loss": 0.015,
848
+ "step": 1360
849
+ },
850
+ {
851
+ "epoch": 3.23,
852
+ "learning_rate": 6.932367149758455e-05,
853
+ "loss": 0.0208,
854
+ "step": 1370
855
+ },
856
+ {
857
+ "epoch": 3.25,
858
+ "learning_rate": 6.908212560386474e-05,
859
+ "loss": 0.0185,
860
+ "step": 1380
861
+ },
862
+ {
863
+ "epoch": 3.28,
864
+ "learning_rate": 6.884057971014493e-05,
865
+ "loss": 0.0162,
866
+ "step": 1390
867
+ },
868
+ {
869
+ "epoch": 3.3,
870
+ "learning_rate": 6.859903381642512e-05,
871
+ "loss": 0.0166,
872
+ "step": 1400
873
+ },
874
+ {
875
+ "epoch": 3.33,
876
+ "learning_rate": 6.835748792270533e-05,
877
+ "loss": 0.0178,
878
+ "step": 1410
879
+ },
880
+ {
881
+ "epoch": 3.35,
882
+ "learning_rate": 6.811594202898552e-05,
883
+ "loss": 0.0163,
884
+ "step": 1420
885
+ },
886
+ {
887
+ "epoch": 3.37,
888
+ "learning_rate": 6.787439613526571e-05,
889
+ "loss": 0.0227,
890
+ "step": 1430
891
+ },
892
+ {
893
+ "epoch": 3.4,
894
+ "learning_rate": 6.76328502415459e-05,
895
+ "loss": 0.0144,
896
+ "step": 1440
897
+ },
898
+ {
899
+ "epoch": 3.42,
900
+ "learning_rate": 6.73913043478261e-05,
901
+ "loss": 0.0156,
902
+ "step": 1450
903
+ },
904
+ {
905
+ "epoch": 3.44,
906
+ "learning_rate": 6.714975845410629e-05,
907
+ "loss": 0.014,
908
+ "step": 1460
909
+ },
910
+ {
911
+ "epoch": 3.47,
912
+ "learning_rate": 6.690821256038648e-05,
913
+ "loss": 0.0187,
914
+ "step": 1470
915
+ },
916
+ {
917
+ "epoch": 3.49,
918
+ "learning_rate": 6.666666666666667e-05,
919
+ "loss": 0.0198,
920
+ "step": 1480
921
+ },
922
+ {
923
+ "epoch": 3.51,
924
+ "learning_rate": 6.642512077294686e-05,
925
+ "loss": 0.0127,
926
+ "step": 1490
927
+ },
928
+ {
929
+ "epoch": 3.54,
930
+ "learning_rate": 6.618357487922707e-05,
931
+ "loss": 0.0129,
932
+ "step": 1500
933
+ },
934
+ {
935
+ "epoch": 3.56,
936
+ "learning_rate": 6.594202898550726e-05,
937
+ "loss": 0.0174,
938
+ "step": 1510
939
+ },
940
+ {
941
+ "epoch": 3.58,
942
+ "learning_rate": 6.570048309178745e-05,
943
+ "loss": 0.0147,
944
+ "step": 1520
945
+ },
946
+ {
947
+ "epoch": 3.61,
948
+ "learning_rate": 6.545893719806764e-05,
949
+ "loss": 0.0188,
950
+ "step": 1530
951
+ },
952
+ {
953
+ "epoch": 3.63,
954
+ "learning_rate": 6.521739130434783e-05,
955
+ "loss": 0.0149,
956
+ "step": 1540
957
+ },
958
+ {
959
+ "epoch": 3.66,
960
+ "learning_rate": 6.497584541062803e-05,
961
+ "loss": 0.0255,
962
+ "step": 1550
963
+ },
964
+ {
965
+ "epoch": 3.68,
966
+ "learning_rate": 6.473429951690822e-05,
967
+ "loss": 0.0198,
968
+ "step": 1560
969
+ },
970
+ {
971
+ "epoch": 3.7,
972
+ "learning_rate": 6.449275362318841e-05,
973
+ "loss": 0.0219,
974
+ "step": 1570
975
+ },
976
+ {
977
+ "epoch": 3.73,
978
+ "learning_rate": 6.42512077294686e-05,
979
+ "loss": 0.0213,
980
+ "step": 1580
981
+ },
982
+ {
983
+ "epoch": 3.75,
984
+ "learning_rate": 6.400966183574879e-05,
985
+ "loss": 0.0214,
986
+ "step": 1590
987
+ },
988
+ {
989
+ "epoch": 3.77,
990
+ "learning_rate": 6.376811594202898e-05,
991
+ "loss": 0.0185,
992
+ "step": 1600
993
+ },
994
+ {
995
+ "epoch": 3.8,
996
+ "learning_rate": 6.352657004830919e-05,
997
+ "loss": 0.0191,
998
+ "step": 1610
999
+ },
1000
+ {
1001
+ "epoch": 3.82,
1002
+ "learning_rate": 6.328502415458938e-05,
1003
+ "loss": 0.0165,
1004
+ "step": 1620
1005
+ },
1006
+ {
1007
+ "epoch": 3.84,
1008
+ "learning_rate": 6.304347826086957e-05,
1009
+ "loss": 0.0205,
1010
+ "step": 1630
1011
+ },
1012
+ {
1013
+ "epoch": 3.87,
1014
+ "learning_rate": 6.280193236714976e-05,
1015
+ "loss": 0.017,
1016
+ "step": 1640
1017
+ },
1018
+ {
1019
+ "epoch": 3.89,
1020
+ "learning_rate": 6.256038647342996e-05,
1021
+ "loss": 0.0198,
1022
+ "step": 1650
1023
+ },
1024
+ {
1025
+ "epoch": 3.92,
1026
+ "learning_rate": 6.231884057971015e-05,
1027
+ "loss": 0.0171,
1028
+ "step": 1660
1029
+ },
1030
+ {
1031
+ "epoch": 3.94,
1032
+ "learning_rate": 6.207729468599034e-05,
1033
+ "loss": 0.0201,
1034
+ "step": 1670
1035
+ },
1036
+ {
1037
+ "epoch": 3.96,
1038
+ "learning_rate": 6.183574879227053e-05,
1039
+ "loss": 0.0253,
1040
+ "step": 1680
1041
+ },
1042
+ {
1043
+ "epoch": 3.99,
1044
+ "learning_rate": 6.159420289855072e-05,
1045
+ "loss": 0.0204,
1046
+ "step": 1690
1047
+ },
1048
+ {
1049
+ "epoch": 4.0,
1050
+ "eval_loss": 2.0079989433288574,
1051
+ "eval_runtime": 26.4901,
1052
+ "eval_samples_per_second": 28.426,
1053
+ "eval_steps_per_second": 1.812,
1054
+ "step": 1696
1055
+ },
1056
+ {
1057
+ "epoch": 4.01,
1058
+ "learning_rate": 6.135265700483092e-05,
1059
+ "loss": 0.0136,
1060
+ "step": 1700
1061
+ },
1062
+ {
1063
+ "epoch": 4.03,
1064
+ "learning_rate": 6.111111111111112e-05,
1065
+ "loss": 0.0131,
1066
+ "step": 1710
1067
+ },
1068
+ {
1069
+ "epoch": 4.06,
1070
+ "learning_rate": 6.086956521739131e-05,
1071
+ "loss": 0.013,
1072
+ "step": 1720
1073
+ },
1074
+ {
1075
+ "epoch": 4.08,
1076
+ "learning_rate": 6.0628019323671504e-05,
1077
+ "loss": 0.009,
1078
+ "step": 1730
1079
+ },
1080
+ {
1081
+ "epoch": 4.1,
1082
+ "learning_rate": 6.0386473429951696e-05,
1083
+ "loss": 0.0075,
1084
+ "step": 1740
1085
+ },
1086
+ {
1087
+ "epoch": 4.13,
1088
+ "learning_rate": 6.014492753623189e-05,
1089
+ "loss": 0.0095,
1090
+ "step": 1750
1091
+ },
1092
+ {
1093
+ "epoch": 4.15,
1094
+ "learning_rate": 5.990338164251208e-05,
1095
+ "loss": 0.0078,
1096
+ "step": 1760
1097
+ },
1098
+ {
1099
+ "epoch": 4.17,
1100
+ "learning_rate": 5.966183574879227e-05,
1101
+ "loss": 0.0122,
1102
+ "step": 1770
1103
+ },
1104
+ {
1105
+ "epoch": 4.2,
1106
+ "learning_rate": 5.942028985507246e-05,
1107
+ "loss": 0.0089,
1108
+ "step": 1780
1109
+ },
1110
+ {
1111
+ "epoch": 4.22,
1112
+ "learning_rate": 5.9178743961352655e-05,
1113
+ "loss": 0.0085,
1114
+ "step": 1790
1115
+ },
1116
+ {
1117
+ "epoch": 4.25,
1118
+ "learning_rate": 5.8937198067632847e-05,
1119
+ "loss": 0.0103,
1120
+ "step": 1800
1121
+ },
1122
+ {
1123
+ "epoch": 4.27,
1124
+ "learning_rate": 5.869565217391305e-05,
1125
+ "loss": 0.0091,
1126
+ "step": 1810
1127
+ },
1128
+ {
1129
+ "epoch": 4.29,
1130
+ "learning_rate": 5.8454106280193244e-05,
1131
+ "loss": 0.0091,
1132
+ "step": 1820
1133
+ },
1134
+ {
1135
+ "epoch": 4.32,
1136
+ "learning_rate": 5.8212560386473435e-05,
1137
+ "loss": 0.0084,
1138
+ "step": 1830
1139
+ },
1140
+ {
1141
+ "epoch": 4.34,
1142
+ "learning_rate": 5.797101449275363e-05,
1143
+ "loss": 0.0095,
1144
+ "step": 1840
1145
+ },
1146
+ {
1147
+ "epoch": 4.36,
1148
+ "learning_rate": 5.772946859903382e-05,
1149
+ "loss": 0.0088,
1150
+ "step": 1850
1151
+ },
1152
+ {
1153
+ "epoch": 4.39,
1154
+ "learning_rate": 5.748792270531401e-05,
1155
+ "loss": 0.0096,
1156
+ "step": 1860
1157
+ },
1158
+ {
1159
+ "epoch": 4.41,
1160
+ "learning_rate": 5.72463768115942e-05,
1161
+ "loss": 0.0094,
1162
+ "step": 1870
1163
+ },
1164
+ {
1165
+ "epoch": 4.43,
1166
+ "learning_rate": 5.7004830917874394e-05,
1167
+ "loss": 0.0111,
1168
+ "step": 1880
1169
+ },
1170
+ {
1171
+ "epoch": 4.46,
1172
+ "learning_rate": 5.6763285024154586e-05,
1173
+ "loss": 0.015,
1174
+ "step": 1890
1175
+ },
1176
+ {
1177
+ "epoch": 4.48,
1178
+ "learning_rate": 5.652173913043478e-05,
1179
+ "loss": 0.0078,
1180
+ "step": 1900
1181
+ },
1182
+ {
1183
+ "epoch": 4.5,
1184
+ "learning_rate": 5.6280193236714976e-05,
1185
+ "loss": 0.0113,
1186
+ "step": 1910
1187
+ },
1188
+ {
1189
+ "epoch": 4.53,
1190
+ "learning_rate": 5.6038647342995175e-05,
1191
+ "loss": 0.0114,
1192
+ "step": 1920
1193
+ },
1194
+ {
1195
+ "epoch": 4.55,
1196
+ "learning_rate": 5.5797101449275366e-05,
1197
+ "loss": 0.0089,
1198
+ "step": 1930
1199
+ },
1200
+ {
1201
+ "epoch": 4.58,
1202
+ "learning_rate": 5.555555555555556e-05,
1203
+ "loss": 0.0137,
1204
+ "step": 1940
1205
+ },
1206
+ {
1207
+ "epoch": 4.6,
1208
+ "learning_rate": 5.531400966183575e-05,
1209
+ "loss": 0.0122,
1210
+ "step": 1950
1211
+ },
1212
+ {
1213
+ "epoch": 4.62,
1214
+ "learning_rate": 5.507246376811594e-05,
1215
+ "loss": 0.0075,
1216
+ "step": 1960
1217
+ },
1218
+ {
1219
+ "epoch": 4.65,
1220
+ "learning_rate": 5.483091787439614e-05,
1221
+ "loss": 0.011,
1222
+ "step": 1970
1223
+ },
1224
+ {
1225
+ "epoch": 4.67,
1226
+ "learning_rate": 5.458937198067633e-05,
1227
+ "loss": 0.0105,
1228
+ "step": 1980
1229
+ },
1230
+ {
1231
+ "epoch": 4.69,
1232
+ "learning_rate": 5.4347826086956524e-05,
1233
+ "loss": 0.0108,
1234
+ "step": 1990
1235
+ },
1236
+ {
1237
+ "epoch": 4.72,
1238
+ "learning_rate": 5.4106280193236716e-05,
1239
+ "loss": 0.0093,
1240
+ "step": 2000
1241
+ },
1242
+ {
1243
+ "epoch": 4.74,
1244
+ "learning_rate": 5.386473429951691e-05,
1245
+ "loss": 0.0137,
1246
+ "step": 2010
1247
+ },
1248
+ {
1249
+ "epoch": 4.76,
1250
+ "learning_rate": 5.3623188405797106e-05,
1251
+ "loss": 0.0103,
1252
+ "step": 2020
1253
+ },
1254
+ {
1255
+ "epoch": 4.79,
1256
+ "learning_rate": 5.3381642512077304e-05,
1257
+ "loss": 0.0095,
1258
+ "step": 2030
1259
+ },
1260
+ {
1261
+ "epoch": 4.81,
1262
+ "learning_rate": 5.3140096618357496e-05,
1263
+ "loss": 0.0108,
1264
+ "step": 2040
1265
+ },
1266
+ {
1267
+ "epoch": 4.83,
1268
+ "learning_rate": 5.289855072463769e-05,
1269
+ "loss": 0.011,
1270
+ "step": 2050
1271
+ },
1272
+ {
1273
+ "epoch": 4.86,
1274
+ "learning_rate": 5.265700483091788e-05,
1275
+ "loss": 0.0112,
1276
+ "step": 2060
1277
+ },
1278
+ {
1279
+ "epoch": 4.88,
1280
+ "learning_rate": 5.241545893719807e-05,
1281
+ "loss": 0.0107,
1282
+ "step": 2070
1283
+ },
1284
+ {
1285
+ "epoch": 4.91,
1286
+ "learning_rate": 5.217391304347826e-05,
1287
+ "loss": 0.0114,
1288
+ "step": 2080
1289
+ },
1290
+ {
1291
+ "epoch": 4.93,
1292
+ "learning_rate": 5.1932367149758455e-05,
1293
+ "loss": 0.0094,
1294
+ "step": 2090
1295
+ },
1296
+ {
1297
+ "epoch": 4.95,
1298
+ "learning_rate": 5.1690821256038647e-05,
1299
+ "loss": 0.0096,
1300
+ "step": 2100
1301
+ },
1302
+ {
1303
+ "epoch": 4.98,
1304
+ "learning_rate": 5.144927536231884e-05,
1305
+ "loss": 0.0118,
1306
+ "step": 2110
1307
+ },
1308
+ {
1309
+ "epoch": 5.0,
1310
+ "learning_rate": 5.1207729468599044e-05,
1311
+ "loss": 0.0093,
1312
+ "step": 2120
1313
+ },
1314
+ {
1315
+ "epoch": 5.0,
1316
+ "eval_loss": 2.034162759780884,
1317
+ "eval_runtime": 26.6455,
1318
+ "eval_samples_per_second": 28.26,
1319
+ "eval_steps_per_second": 1.801,
1320
+ "step": 2120
1321
+ },
1322
+ {
1323
+ "epoch": 5.02,
1324
+ "learning_rate": 5.0966183574879235e-05,
1325
+ "loss": 0.0044,
1326
+ "step": 2130
1327
+ },
1328
+ {
1329
+ "epoch": 5.05,
1330
+ "learning_rate": 5.072463768115943e-05,
1331
+ "loss": 0.0046,
1332
+ "step": 2140
1333
+ },
1334
+ {
1335
+ "epoch": 5.07,
1336
+ "learning_rate": 5.048309178743962e-05,
1337
+ "loss": 0.0071,
1338
+ "step": 2150
1339
+ },
1340
+ {
1341
+ "epoch": 5.09,
1342
+ "learning_rate": 5.024154589371981e-05,
1343
+ "loss": 0.005,
1344
+ "step": 2160
1345
+ },
1346
+ {
1347
+ "epoch": 5.12,
1348
+ "learning_rate": 5e-05,
1349
+ "loss": 0.0032,
1350
+ "step": 2170
1351
+ },
1352
+ {
1353
+ "epoch": 5.14,
1354
+ "learning_rate": 4.9758454106280194e-05,
1355
+ "loss": 0.0038,
1356
+ "step": 2180
1357
+ },
1358
+ {
1359
+ "epoch": 5.17,
1360
+ "learning_rate": 4.9516908212560386e-05,
1361
+ "loss": 0.005,
1362
+ "step": 2190
1363
+ },
1364
+ {
1365
+ "epoch": 5.19,
1366
+ "learning_rate": 4.9275362318840584e-05,
1367
+ "loss": 0.0033,
1368
+ "step": 2200
1369
+ },
1370
+ {
1371
+ "epoch": 5.21,
1372
+ "learning_rate": 4.9033816425120776e-05,
1373
+ "loss": 0.0058,
1374
+ "step": 2210
1375
+ },
1376
+ {
1377
+ "epoch": 5.24,
1378
+ "learning_rate": 4.879227053140097e-05,
1379
+ "loss": 0.0074,
1380
+ "step": 2220
1381
+ },
1382
+ {
1383
+ "epoch": 5.26,
1384
+ "learning_rate": 4.855072463768116e-05,
1385
+ "loss": 0.0038,
1386
+ "step": 2230
1387
+ },
1388
+ {
1389
+ "epoch": 5.28,
1390
+ "learning_rate": 4.830917874396135e-05,
1391
+ "loss": 0.0051,
1392
+ "step": 2240
1393
+ },
1394
+ {
1395
+ "epoch": 5.31,
1396
+ "learning_rate": 4.806763285024155e-05,
1397
+ "loss": 0.0042,
1398
+ "step": 2250
1399
+ }
1400
+ ],
1401
+ "max_steps": 4240,
1402
+ "num_train_epochs": 10,
1403
+ "total_flos": 0.0,
1404
+ "trial_name": null,
1405
+ "trial_params": null
1406
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/checkpoint-2250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d19fa419d807bcbfef5a86725e8eca48bdf21b5f0b9a046119a2780ac6238051
3
+ size 3695
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.21.1",
58
+ "use_cache": true,
59
+ "vocab_size": 32128
60
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs-500Steps.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs1250Steps.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs1750Steps.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/outputs2500Steps.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37cd659b3f4a43460cd3776e03eec35d8dc2f89568520fb17ba0aff142a036a7
3
+ size 891700799
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/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
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/tokenizer_config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "extra_ids": 100,
106
+ "model_max_length": 512,
107
+ "name_or_path": "t5-base",
108
+ "pad_token": "<pad>",
109
+ "special_tokens_map_file": null,
110
+ "tokenizer_class": "T5Tokenizer",
111
+ "unk_token": "<unk>"
112
+ }
t5-e2e-10epochs-lr1e4-alpha0-1PLUSalpha0-9-e10/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1e67a880c0e90b8af114008a893453d93e514b1300673ac56565c1387e9829e
3
+ size 3695