hgroener commited on
Commit
e794a5f
1 Parent(s): 1ca3110

Upload folder using huggingface_hub

Browse files
added_tokens.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</quatrain>": 386,
3
+ "<alexandrine>": 408,
4
+ "<amphibrach>": 407,
5
+ "<anapaest>": 404,
6
+ "<dactyl>": 405,
7
+ "<iambus>": 402,
8
+ "<other>": 406,
9
+ "<quatrain>": 385,
10
+ "<trochee>": 403,
11
+ "[AAAA]": 400,
12
+ "[AAAB]": 398,
13
+ "[AABA]": 396,
14
+ "[AABB]": 397,
15
+ "[AABC]": 388,
16
+ "[ABAA]": 401,
17
+ "[ABAB]": 392,
18
+ "[ABAC]": 389,
19
+ "[ABBA]": 391,
20
+ "[ABBB]": 395,
21
+ "[ABBC]": 394,
22
+ "[ABCA]": 399,
23
+ "[ABCB]": 393,
24
+ "[ABCC]": 390,
25
+ "[ABCD]": 387,
26
+ "[PAD]": 384
27
+ }
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nllg/bygpt5-medium-en",
3
+ "architectures": [
4
+ "ByGPT5LMHeadModel"
5
+ ],
6
+ "d_ff": 3840,
7
+ "d_kv": 64,
8
+ "d_model": 1536,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "gradient_checkpointing": false,
15
+ "initializer_factor": 1.0,
16
+ "is_decoder": true,
17
+ "is_encoder_decoder": false,
18
+ "is_gated_act": true,
19
+ "layer_norm_epsilon": 1e-06,
20
+ "model_type": "bygpt5",
21
+ "num_decoder_layers": 12,
22
+ "num_heads": 16,
23
+ "num_layers": 12,
24
+ "output_past": true,
25
+ "pad_token_id": 0,
26
+ "relative_attention_max_distance": 128,
27
+ "relative_attention_num_buckets": 32,
28
+ "tie_word_embeddings": false,
29
+ "tokenizer_class": "ByT5Tokenizer",
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.20.0",
32
+ "use_cache": true,
33
+ "vocab_size": 409
34
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20d347cf214eaf2e6983fd62d287521ef0ba03969233ca3a9ab87c74289c47f8
3
+ size 2313102842
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81b6de0d2aac6870a31542c82030721dfb3bc692c50a097e3187fb8ecdaad52c
3
+ size 1156556746
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5426cf5999c4e67fe7d7d57f6acc4edcf8de651272ac9370fea96b120cfd6c4b
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd3f26e719d9290a594112b7dccf56a05bf2815c45bd29d2e78307c6bfbef33b
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "[ABCD]",
4
+ "[AABC]",
5
+ "[ABAC]",
6
+ "[ABCC]",
7
+ "[ABBA]",
8
+ "[ABAB]",
9
+ "[ABCB]",
10
+ "[ABBC]",
11
+ "[ABBB]",
12
+ "[AABA]",
13
+ "[AABB]",
14
+ "[AAAB]",
15
+ "[ABCA]",
16
+ "[AAAA]",
17
+ "[ABAA]",
18
+ "<iambus>",
19
+ "<trochee>",
20
+ "<anapaest>",
21
+ "<dactyl>",
22
+ "<other>",
23
+ "<amphibrach>",
24
+ "<alexandrine>"
25
+ ],
26
+ "bos_token": "<quatrain>",
27
+ "eos_token": "</quatrain>",
28
+ "pad_token": "[PAD]",
29
+ "unk_token": {
30
+ "content": "<unk>",
31
+ "lstrip": false,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ }
36
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,160 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "additional_special_tokens": [
6
+ "<extra_id_0>",
7
+ "<extra_id_1>",
8
+ "<extra_id_2>",
9
+ "<extra_id_3>",
10
+ "<extra_id_4>",
11
+ "<extra_id_5>",
12
+ "<extra_id_6>",
13
+ "<extra_id_7>",
14
+ "<extra_id_8>",
15
+ "<extra_id_9>",
16
+ "<extra_id_10>",
17
+ "<extra_id_11>",
18
+ "<extra_id_12>",
19
+ "<extra_id_13>",
20
+ "<extra_id_14>",
21
+ "<extra_id_15>",
22
+ "<extra_id_16>",
23
+ "<extra_id_17>",
24
+ "<extra_id_18>",
25
+ "<extra_id_19>",
26
+ "<extra_id_20>",
27
+ "<extra_id_21>",
28
+ "<extra_id_22>",
29
+ "<extra_id_23>",
30
+ "<extra_id_24>",
31
+ "<extra_id_25>",
32
+ "<extra_id_26>",
33
+ "<extra_id_27>",
34
+ "<extra_id_28>",
35
+ "<extra_id_29>",
36
+ "<extra_id_30>",
37
+ "<extra_id_31>",
38
+ "<extra_id_32>",
39
+ "<extra_id_33>",
40
+ "<extra_id_34>",
41
+ "<extra_id_35>",
42
+ "<extra_id_36>",
43
+ "<extra_id_37>",
44
+ "<extra_id_38>",
45
+ "<extra_id_39>",
46
+ "<extra_id_40>",
47
+ "<extra_id_41>",
48
+ "<extra_id_42>",
49
+ "<extra_id_43>",
50
+ "<extra_id_44>",
51
+ "<extra_id_45>",
52
+ "<extra_id_46>",
53
+ "<extra_id_47>",
54
+ "<extra_id_48>",
55
+ "<extra_id_49>",
56
+ "<extra_id_50>",
57
+ "<extra_id_51>",
58
+ "<extra_id_52>",
59
+ "<extra_id_53>",
60
+ "<extra_id_54>",
61
+ "<extra_id_55>",
62
+ "<extra_id_56>",
63
+ "<extra_id_57>",
64
+ "<extra_id_58>",
65
+ "<extra_id_59>",
66
+ "<extra_id_60>",
67
+ "<extra_id_61>",
68
+ "<extra_id_62>",
69
+ "<extra_id_63>",
70
+ "<extra_id_64>",
71
+ "<extra_id_65>",
72
+ "<extra_id_66>",
73
+ "<extra_id_67>",
74
+ "<extra_id_68>",
75
+ "<extra_id_69>",
76
+ "<extra_id_70>",
77
+ "<extra_id_71>",
78
+ "<extra_id_72>",
79
+ "<extra_id_73>",
80
+ "<extra_id_74>",
81
+ "<extra_id_75>",
82
+ "<extra_id_76>",
83
+ "<extra_id_77>",
84
+ "<extra_id_78>",
85
+ "<extra_id_79>",
86
+ "<extra_id_80>",
87
+ "<extra_id_81>",
88
+ "<extra_id_82>",
89
+ "<extra_id_83>",
90
+ "<extra_id_84>",
91
+ "<extra_id_85>",
92
+ "<extra_id_86>",
93
+ "<extra_id_87>",
94
+ "<extra_id_88>",
95
+ "<extra_id_89>",
96
+ "<extra_id_90>",
97
+ "<extra_id_91>",
98
+ "<extra_id_92>",
99
+ "<extra_id_93>",
100
+ "<extra_id_94>",
101
+ "<extra_id_95>",
102
+ "<extra_id_96>",
103
+ "<extra_id_97>",
104
+ "<extra_id_98>",
105
+ "<extra_id_99>",
106
+ "<extra_id_100>",
107
+ "<extra_id_101>",
108
+ "<extra_id_102>",
109
+ "<extra_id_103>",
110
+ "<extra_id_104>",
111
+ "<extra_id_105>",
112
+ "<extra_id_106>",
113
+ "<extra_id_107>",
114
+ "<extra_id_108>",
115
+ "<extra_id_109>",
116
+ "<extra_id_110>",
117
+ "<extra_id_111>",
118
+ "<extra_id_112>",
119
+ "<extra_id_113>",
120
+ "<extra_id_114>",
121
+ "<extra_id_115>",
122
+ "<extra_id_116>",
123
+ "<extra_id_117>",
124
+ "<extra_id_118>",
125
+ "<extra_id_119>",
126
+ "<extra_id_120>",
127
+ "<extra_id_121>",
128
+ "<extra_id_122>",
129
+ "<extra_id_123>",
130
+ "<extra_id_124>"
131
+ ],
132
+ "eos_token": {
133
+ "__type": "AddedToken",
134
+ "content": "</s>",
135
+ "lstrip": false,
136
+ "normalized": true,
137
+ "rstrip": false,
138
+ "single_word": false
139
+ },
140
+ "extra_ids": 125,
141
+ "name_or_path": "nllg/bygpt5-medium-en",
142
+ "pad_token": {
143
+ "__type": "AddedToken",
144
+ "content": "<pad>",
145
+ "lstrip": false,
146
+ "normalized": true,
147
+ "rstrip": false,
148
+ "single_word": false
149
+ },
150
+ "special_tokens_map_file": "/storage/ukp/work/jbelouadi/.cache/huggingface/transformers/1d7794e7e6dcb9aff7f1b0f72b85d665041ce4cfc7c59cf78a0e83302f9ab8e3.063895353d5ef9e19a25220cb616c43abc5e84a2f11b1ffb71c29e097572a109",
151
+ "tokenizer_class": "ByGPT5Tokenizer",
152
+ "unk_token": {
153
+ "__type": "AddedToken",
154
+ "content": "<unk>",
155
+ "lstrip": false,
156
+ "normalized": true,
157
+ "rstrip": false,
158
+ "single_word": false
159
+ }
160
+ }
trainer_state.json ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.0628538131713867,
3
+ "best_model_checkpoint": "bygpt5_quatags_con/medium/en/checkpoint-90832",
4
+ "epoch": 7.999955964595535,
5
+ "global_step": 90832,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.0,
12
+ "learning_rate": 3.948830567446131e-05,
13
+ "loss": 1.2434,
14
+ "step": 11354
15
+ },
16
+ {
17
+ "epoch": 1.0,
18
+ "eval_loss": 1.1300886869430542,
19
+ "eval_runtime": 441.3582,
20
+ "eval_samples_per_second": 45.736,
21
+ "eval_steps_per_second": 5.719,
22
+ "step": 11354
23
+ },
24
+ {
25
+ "epoch": 2.0,
26
+ "learning_rate": 3.704453340788571e-05,
27
+ "loss": 1.1308,
28
+ "step": 22708
29
+ },
30
+ {
31
+ "epoch": 2.0,
32
+ "eval_loss": 1.1001691818237305,
33
+ "eval_runtime": 440.6446,
34
+ "eval_samples_per_second": 45.81,
35
+ "eval_steps_per_second": 5.728,
36
+ "step": 22708
37
+ },
38
+ {
39
+ "epoch": 3.0,
40
+ "learning_rate": 3.282841708036157e-05,
41
+ "loss": 1.0962,
42
+ "step": 34062
43
+ },
44
+ {
45
+ "epoch": 3.0,
46
+ "eval_loss": 1.082809329032898,
47
+ "eval_runtime": 440.424,
48
+ "eval_samples_per_second": 45.833,
49
+ "eval_steps_per_second": 5.731,
50
+ "step": 34062
51
+ },
52
+ {
53
+ "epoch": 4.0,
54
+ "learning_rate": 2.727836169568317e-05,
55
+ "loss": 1.0715,
56
+ "step": 45416
57
+ },
58
+ {
59
+ "epoch": 4.0,
60
+ "eval_loss": 1.073394536972046,
61
+ "eval_runtime": 440.7618,
62
+ "eval_samples_per_second": 45.798,
63
+ "eval_steps_per_second": 5.726,
64
+ "step": 45416
65
+ },
66
+ {
67
+ "epoch": 5.0,
68
+ "learning_rate": 2.0971479426039337e-05,
69
+ "loss": 1.0522,
70
+ "step": 56770
71
+ },
72
+ {
73
+ "epoch": 5.0,
74
+ "eval_loss": 1.0667953491210938,
75
+ "eval_runtime": 440.157,
76
+ "eval_samples_per_second": 45.861,
77
+ "eval_steps_per_second": 5.734,
78
+ "step": 56770
79
+ },
80
+ {
81
+ "epoch": 6.0,
82
+ "learning_rate": 1.4563579677660773e-05,
83
+ "loss": 1.0372,
84
+ "step": 68124
85
+ },
86
+ {
87
+ "epoch": 6.0,
88
+ "eval_loss": 1.0641292333602905,
89
+ "eval_runtime": 439.9553,
90
+ "eval_samples_per_second": 45.882,
91
+ "eval_steps_per_second": 5.737,
92
+ "step": 68124
93
+ },
94
+ {
95
+ "epoch": 7.0,
96
+ "learning_rate": 8.720975971159198e-06,
97
+ "loss": 1.0258,
98
+ "step": 79478
99
+ },
100
+ {
101
+ "epoch": 7.0,
102
+ "eval_loss": 1.0637840032577515,
103
+ "eval_runtime": 440.5762,
104
+ "eval_samples_per_second": 45.817,
105
+ "eval_steps_per_second": 5.729,
106
+ "step": 79478
107
+ },
108
+ {
109
+ "epoch": 8.0,
110
+ "learning_rate": 4.051200571096512e-06,
111
+ "loss": 1.0178,
112
+ "step": 90832
113
+ },
114
+ {
115
+ "epoch": 8.0,
116
+ "eval_loss": 1.0628538131713867,
117
+ "eval_runtime": 440.6394,
118
+ "eval_samples_per_second": 45.811,
119
+ "eval_steps_per_second": 5.728,
120
+ "step": 90832
121
+ }
122
+ ],
123
+ "max_steps": 113540,
124
+ "num_train_epochs": 10,
125
+ "total_flos": 5.170056667435008e+17,
126
+ "trial_name": null,
127
+ "trial_params": null
128
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:666e443b505c62b8f5e852eddae329f59fbe49199853678af293e90af2d5ece5
3
+ size 3768