yesj1234 commited on
Commit
141dd3c
1 Parent(s): 4e16e26

Upload folder using huggingface_hub

Browse files
Files changed (43) hide show
  1. README.md +77 -0
  2. added_tokens.json +6 -0
  3. all_results.json +14 -0
  4. checkpoint-22000/added_tokens.json +6 -0
  5. checkpoint-22000/config.json +117 -0
  6. checkpoint-22000/optimizer.pt +3 -0
  7. checkpoint-22000/preprocessor_config.json +10 -0
  8. checkpoint-22000/pytorch_model.bin +3 -0
  9. checkpoint-22000/rng_state_0.pth +3 -0
  10. checkpoint-22000/rng_state_1.pth +3 -0
  11. checkpoint-22000/rng_state_2.pth +3 -0
  12. checkpoint-22000/rng_state_3.pth +3 -0
  13. checkpoint-22000/scheduler.pt +3 -0
  14. checkpoint-22000/special_tokens_map.json +10 -0
  15. checkpoint-22000/tokenizer_config.json +56 -0
  16. checkpoint-22000/trainer_state.json +778 -0
  17. checkpoint-22000/training_args.bin +3 -0
  18. checkpoint-22000/vocab.json +1744 -0
  19. checkpoint-24000/added_tokens.json +6 -0
  20. checkpoint-24000/config.json +117 -0
  21. checkpoint-24000/optimizer.pt +3 -0
  22. checkpoint-24000/preprocessor_config.json +10 -0
  23. checkpoint-24000/pytorch_model.bin +3 -0
  24. checkpoint-24000/rng_state_0.pth +3 -0
  25. checkpoint-24000/rng_state_1.pth +3 -0
  26. checkpoint-24000/rng_state_2.pth +3 -0
  27. checkpoint-24000/rng_state_3.pth +3 -0
  28. checkpoint-24000/scheduler.pt +3 -0
  29. checkpoint-24000/special_tokens_map.json +10 -0
  30. checkpoint-24000/tokenizer_config.json +56 -0
  31. checkpoint-24000/trainer_state.json +847 -0
  32. checkpoint-24000/training_args.bin +3 -0
  33. checkpoint-24000/vocab.json +1744 -0
  34. config.json +117 -0
  35. eval_results.json +9 -0
  36. preprocessor_config.json +10 -0
  37. pytorch_model.bin +3 -0
  38. special_tokens_map.json +10 -0
  39. tokenizer_config.json +56 -0
  40. train_results.json +8 -0
  41. trainer_state.json +868 -0
  42. training_args.bin +3 -0
  43. vocab.json +1744 -0
README.md ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-large-xlsr-53
4
+ tags:
5
+ - automatic-speech-recognition
6
+ - ./sample_speech.py
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: en-xlsr
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # en-xlsr
17
+
18
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the ./SAMPLE_SPEECH.PY - NA dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.3889
21
+ - Cer: 0.1082
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0003
41
+ - train_batch_size: 4
42
+ - eval_batch_size: 4
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 4
46
+ - gradient_accumulation_steps: 4
47
+ - total_train_batch_size: 64
48
+ - total_eval_batch_size: 16
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - lr_scheduler_warmup_steps: 1500
52
+ - num_epochs: 15
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
57
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
58
+ | 1.4503 | 1.22 | 2000 | 1.0610 | 0.2687 |
59
+ | 1.0239 | 2.45 | 4000 | 0.6962 | 0.1904 |
60
+ | 0.8977 | 3.67 | 6000 | 0.5945 | 0.1687 |
61
+ | 0.804 | 4.9 | 8000 | 0.5328 | 0.1492 |
62
+ | 0.698 | 6.12 | 10000 | 0.5014 | 0.1365 |
63
+ | 0.6426 | 7.35 | 12000 | 0.4715 | 0.1322 |
64
+ | 0.61 | 8.57 | 14000 | 0.4530 | 0.1258 |
65
+ | 0.5709 | 9.79 | 16000 | 0.4300 | 0.1201 |
66
+ | 0.5235 | 11.02 | 18000 | 0.4168 | 0.1166 |
67
+ | 0.4778 | 12.24 | 20000 | 0.4057 | 0.1129 |
68
+ | 0.4571 | 13.47 | 22000 | 0.3945 | 0.1100 |
69
+ | 0.4388 | 14.69 | 24000 | 0.3891 | 0.1081 |
70
+
71
+
72
+ ### Framework versions
73
+
74
+ - Transformers 4.34.0
75
+ - Pytorch 2.1.0+cu121
76
+ - Datasets 2.14.5
77
+ - Tokenizers 0.14.1
added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 1743,
3
+ "<s>": 1742,
4
+ "[PAD]": 1741,
5
+ "[UNK]": 1740
6
+ }
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 14.99,
3
+ "eval_cer": 0.1081634945549379,
4
+ "eval_loss": 0.38891980051994324,
5
+ "eval_runtime": 1601.2936,
6
+ "eval_samples": 13093,
7
+ "eval_samples_per_second": 8.177,
8
+ "eval_steps_per_second": 0.511,
9
+ "train_loss": 1.1506779817299688,
10
+ "train_runtime": 54167.2991,
11
+ "train_samples": 104547,
12
+ "train_samples_per_second": 28.951,
13
+ "train_steps_per_second": 0.452
14
+ }
checkpoint-22000/added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 1743,
3
+ "<s>": 1742,
4
+ "[PAD]": 1741,
5
+ "[UNK]": 1740
6
+ }
checkpoint-22000/config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_attn_dim": null,
5
+ "adapter_kernel_size": 3,
6
+ "adapter_stride": 2,
7
+ "add_adapter": false,
8
+ "apply_spec_augment": true,
9
+ "architectures": [
10
+ "Wav2Vec2ForCTC"
11
+ ],
12
+ "attention_dropout": 0.1,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_dim": [
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512
26
+ ],
27
+ "conv_kernel": [
28
+ 10,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 2,
34
+ 2
35
+ ],
36
+ "conv_stride": [
37
+ 5,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2
44
+ ],
45
+ "ctc_loss_reduction": "mean",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": true,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_dropout": 0.0,
52
+ "feat_extract_norm": "layer",
53
+ "feat_proj_dropout": 0.05,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.05,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.05,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.05,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "num_adapter_layers": 3,
80
+ "num_attention_heads": 16,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_feat_extract_layers": 7,
86
+ "num_hidden_layers": 24,
87
+ "num_negatives": 100,
88
+ "output_hidden_size": 1024,
89
+ "pad_token_id": 1741,
90
+ "proj_codevector_dim": 768,
91
+ "tdnn_dilation": [
92
+ 1,
93
+ 2,
94
+ 3,
95
+ 1,
96
+ 1
97
+ ],
98
+ "tdnn_dim": [
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 512,
103
+ 1500
104
+ ],
105
+ "tdnn_kernel": [
106
+ 5,
107
+ 3,
108
+ 3,
109
+ 1,
110
+ 1
111
+ ],
112
+ "torch_dtype": "float32",
113
+ "transformers_version": "4.34.0",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 1744,
116
+ "xvector_output_dim": 512
117
+ }
checkpoint-22000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25fcb59c6cb08679bba4f9e47b0c19bf38e5056d067d48b817a13d88e4308bc6
3
+ size 2504460598
checkpoint-22000/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
checkpoint-22000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59e5060840889b0b4ae7240b94a66edd8b7283f576ae25cbff4e9ab254f7839b
3
+ size 1269052778
checkpoint-22000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33627f5779e5df94aeb61bc78e06ad279190985f8c3abd7806960fd59b56fe43
3
+ size 15024
checkpoint-22000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3221e336a42d20a3cd1fcbfcc5fb855a0f6eff64d1e726316d1139b310402b4e
3
+ size 15024
checkpoint-22000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52cec5e130a71aa404d53bb4652add23d7a0d2f752a3752ae7f0731fcbcddbf4
3
+ size 15088
checkpoint-22000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d27b6a69ddcac4e34afac2ab9690a00bfca4ec0b3ebe4bfa5c32b693201827e
3
+ size 14960
checkpoint-22000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067844c63df10c9eeb76811daadba0a6033e0fed0ccde8c648bc780a57419ddf
3
+ size 1064
checkpoint-22000/special_tokens_map.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>",
4
+ "</s>"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "eos_token": "</s>",
8
+ "pad_token": "[PAD]",
9
+ "unk_token": "[UNK]"
10
+ }
checkpoint-22000/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "1740": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1741": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "1742": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "1743": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "additional_special_tokens": [
37
+ "<s>",
38
+ "</s>"
39
+ ],
40
+ "bos_token": "<s>",
41
+ "clean_up_tokenization_spaces": true,
42
+ "config": null,
43
+ "do_lower_case": false,
44
+ "eos_token": "</s>",
45
+ "model_max_length": 1000000000000000019884624838656,
46
+ "pad_token": "[PAD]",
47
+ "processor_class": "Wav2Vec2Processor",
48
+ "replace_word_delimiter_char": " ",
49
+ "target_lang": null,
50
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
51
+ "tokenizer_file": null,
52
+ "tokenizer_type": "wav2vec2",
53
+ "trust_remote_code": false,
54
+ "unk_token": "[UNK]",
55
+ "word_delimiter_token": "|"
56
+ }
checkpoint-22000/trainer_state.json ADDED
@@ -0,0 +1,778 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 13.465952563121652,
5
+ "eval_steps": 2000,
6
+ "global_step": 22000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12,
13
+ "learning_rate": 3.9999999999999996e-05,
14
+ "loss": 33.0181,
15
+ "step": 200
16
+ },
17
+ {
18
+ "epoch": 0.24,
19
+ "learning_rate": 7.999999999999999e-05,
20
+ "loss": 8.7646,
21
+ "step": 400
22
+ },
23
+ {
24
+ "epoch": 0.37,
25
+ "learning_rate": 0.00011999999999999999,
26
+ "loss": 4.6933,
27
+ "step": 600
28
+ },
29
+ {
30
+ "epoch": 0.49,
31
+ "learning_rate": 0.00015999999999999999,
32
+ "loss": 4.5796,
33
+ "step": 800
34
+ },
35
+ {
36
+ "epoch": 0.61,
37
+ "learning_rate": 0.00019999999999999998,
38
+ "loss": 4.1047,
39
+ "step": 1000
40
+ },
41
+ {
42
+ "epoch": 0.73,
43
+ "learning_rate": 0.00023999999999999998,
44
+ "loss": 2.6327,
45
+ "step": 1200
46
+ },
47
+ {
48
+ "epoch": 0.86,
49
+ "learning_rate": 0.00028,
50
+ "loss": 1.9893,
51
+ "step": 1400
52
+ },
53
+ {
54
+ "epoch": 0.98,
55
+ "learning_rate": 0.00029869536855838224,
56
+ "loss": 1.7241,
57
+ "step": 1600
58
+ },
59
+ {
60
+ "epoch": 1.1,
61
+ "learning_rate": 0.00029608610567514673,
62
+ "loss": 1.5607,
63
+ "step": 1800
64
+ },
65
+ {
66
+ "epoch": 1.22,
67
+ "learning_rate": 0.00029347684279191127,
68
+ "loss": 1.4503,
69
+ "step": 2000
70
+ },
71
+ {
72
+ "epoch": 1.22,
73
+ "eval_cer": 0.268688843618788,
74
+ "eval_loss": 1.0610458850860596,
75
+ "eval_runtime": 1586.9419,
76
+ "eval_samples_per_second": 8.25,
77
+ "eval_steps_per_second": 0.516,
78
+ "step": 2000
79
+ },
80
+ {
81
+ "epoch": 1.35,
82
+ "learning_rate": 0.00029086757990867576,
83
+ "loss": 1.3782,
84
+ "step": 2200
85
+ },
86
+ {
87
+ "epoch": 1.47,
88
+ "learning_rate": 0.0002882583170254403,
89
+ "loss": 1.3244,
90
+ "step": 2400
91
+ },
92
+ {
93
+ "epoch": 1.59,
94
+ "learning_rate": 0.0002856490541422048,
95
+ "loss": 1.2553,
96
+ "step": 2600
97
+ },
98
+ {
99
+ "epoch": 1.71,
100
+ "learning_rate": 0.00028303979125896933,
101
+ "loss": 1.2077,
102
+ "step": 2800
103
+ },
104
+ {
105
+ "epoch": 1.84,
106
+ "learning_rate": 0.0002804305283757338,
107
+ "loss": 1.1868,
108
+ "step": 3000
109
+ },
110
+ {
111
+ "epoch": 1.96,
112
+ "learning_rate": 0.00027782126549249836,
113
+ "loss": 1.1624,
114
+ "step": 3200
115
+ },
116
+ {
117
+ "epoch": 2.08,
118
+ "learning_rate": 0.00027521200260926284,
119
+ "loss": 1.1136,
120
+ "step": 3400
121
+ },
122
+ {
123
+ "epoch": 2.2,
124
+ "learning_rate": 0.0002726027397260274,
125
+ "loss": 1.077,
126
+ "step": 3600
127
+ },
128
+ {
129
+ "epoch": 2.33,
130
+ "learning_rate": 0.00026999347684279187,
131
+ "loss": 1.053,
132
+ "step": 3800
133
+ },
134
+ {
135
+ "epoch": 2.45,
136
+ "learning_rate": 0.0002673842139595564,
137
+ "loss": 1.0239,
138
+ "step": 4000
139
+ },
140
+ {
141
+ "epoch": 2.45,
142
+ "eval_cer": 0.19039349567825709,
143
+ "eval_loss": 0.6961866617202759,
144
+ "eval_runtime": 1607.479,
145
+ "eval_samples_per_second": 8.145,
146
+ "eval_steps_per_second": 0.509,
147
+ "step": 4000
148
+ },
149
+ {
150
+ "epoch": 2.57,
151
+ "learning_rate": 0.0002647749510763209,
152
+ "loss": 1.0206,
153
+ "step": 4200
154
+ },
155
+ {
156
+ "epoch": 2.69,
157
+ "learning_rate": 0.00026216568819308544,
158
+ "loss": 1.0045,
159
+ "step": 4400
160
+ },
161
+ {
162
+ "epoch": 2.82,
163
+ "learning_rate": 0.0002595564253098499,
164
+ "loss": 0.9802,
165
+ "step": 4600
166
+ },
167
+ {
168
+ "epoch": 2.94,
169
+ "learning_rate": 0.00025694716242661447,
170
+ "loss": 0.9839,
171
+ "step": 4800
172
+ },
173
+ {
174
+ "epoch": 3.06,
175
+ "learning_rate": 0.00025433789954337895,
176
+ "loss": 0.9243,
177
+ "step": 5000
178
+ },
179
+ {
180
+ "epoch": 3.18,
181
+ "learning_rate": 0.0002517286366601435,
182
+ "loss": 0.9082,
183
+ "step": 5200
184
+ },
185
+ {
186
+ "epoch": 3.31,
187
+ "learning_rate": 0.000249119373776908,
188
+ "loss": 0.9017,
189
+ "step": 5400
190
+ },
191
+ {
192
+ "epoch": 3.43,
193
+ "learning_rate": 0.0002465101108936725,
194
+ "loss": 0.8871,
195
+ "step": 5600
196
+ },
197
+ {
198
+ "epoch": 3.55,
199
+ "learning_rate": 0.00024390084801043704,
200
+ "loss": 0.9036,
201
+ "step": 5800
202
+ },
203
+ {
204
+ "epoch": 3.67,
205
+ "learning_rate": 0.00024129158512720155,
206
+ "loss": 0.8977,
207
+ "step": 6000
208
+ },
209
+ {
210
+ "epoch": 3.67,
211
+ "eval_cer": 0.16872254319465907,
212
+ "eval_loss": 0.594495415687561,
213
+ "eval_runtime": 1591.6049,
214
+ "eval_samples_per_second": 8.226,
215
+ "eval_steps_per_second": 0.515,
216
+ "step": 6000
217
+ },
218
+ {
219
+ "epoch": 3.79,
220
+ "learning_rate": 0.00023868232224396607,
221
+ "loss": 0.8843,
222
+ "step": 6200
223
+ },
224
+ {
225
+ "epoch": 3.92,
226
+ "learning_rate": 0.00023607305936073058,
227
+ "loss": 0.8757,
228
+ "step": 6400
229
+ },
230
+ {
231
+ "epoch": 4.04,
232
+ "learning_rate": 0.0002334637964774951,
233
+ "loss": 0.8573,
234
+ "step": 6600
235
+ },
236
+ {
237
+ "epoch": 4.16,
238
+ "learning_rate": 0.0002308545335942596,
239
+ "loss": 0.8126,
240
+ "step": 6800
241
+ },
242
+ {
243
+ "epoch": 4.28,
244
+ "learning_rate": 0.00022824527071102412,
245
+ "loss": 0.8192,
246
+ "step": 7000
247
+ },
248
+ {
249
+ "epoch": 4.41,
250
+ "learning_rate": 0.00022563600782778863,
251
+ "loss": 0.8061,
252
+ "step": 7200
253
+ },
254
+ {
255
+ "epoch": 4.53,
256
+ "learning_rate": 0.00022302674494455315,
257
+ "loss": 0.8123,
258
+ "step": 7400
259
+ },
260
+ {
261
+ "epoch": 4.65,
262
+ "learning_rate": 0.00022041748206131766,
263
+ "loss": 0.8046,
264
+ "step": 7600
265
+ },
266
+ {
267
+ "epoch": 4.77,
268
+ "learning_rate": 0.00021780821917808218,
269
+ "loss": 0.7979,
270
+ "step": 7800
271
+ },
272
+ {
273
+ "epoch": 4.9,
274
+ "learning_rate": 0.0002151989562948467,
275
+ "loss": 0.804,
276
+ "step": 8000
277
+ },
278
+ {
279
+ "epoch": 4.9,
280
+ "eval_cer": 0.14924661713942214,
281
+ "eval_loss": 0.5327703952789307,
282
+ "eval_runtime": 1595.6324,
283
+ "eval_samples_per_second": 8.206,
284
+ "eval_steps_per_second": 0.513,
285
+ "step": 8000
286
+ },
287
+ {
288
+ "epoch": 5.02,
289
+ "learning_rate": 0.0002125896934116112,
290
+ "loss": 0.7867,
291
+ "step": 8200
292
+ },
293
+ {
294
+ "epoch": 5.14,
295
+ "learning_rate": 0.00020998043052837572,
296
+ "loss": 0.7557,
297
+ "step": 8400
298
+ },
299
+ {
300
+ "epoch": 5.26,
301
+ "learning_rate": 0.00020737116764514023,
302
+ "loss": 0.7478,
303
+ "step": 8600
304
+ },
305
+ {
306
+ "epoch": 5.39,
307
+ "learning_rate": 0.00020476190476190475,
308
+ "loss": 0.7398,
309
+ "step": 8800
310
+ },
311
+ {
312
+ "epoch": 5.51,
313
+ "learning_rate": 0.00020215264187866926,
314
+ "loss": 0.7408,
315
+ "step": 9000
316
+ },
317
+ {
318
+ "epoch": 5.63,
319
+ "learning_rate": 0.00019954337899543377,
320
+ "loss": 0.75,
321
+ "step": 9200
322
+ },
323
+ {
324
+ "epoch": 5.75,
325
+ "learning_rate": 0.0001969341161121983,
326
+ "loss": 0.7344,
327
+ "step": 9400
328
+ },
329
+ {
330
+ "epoch": 5.88,
331
+ "learning_rate": 0.0001943248532289628,
332
+ "loss": 0.738,
333
+ "step": 9600
334
+ },
335
+ {
336
+ "epoch": 6.0,
337
+ "learning_rate": 0.00019171559034572732,
338
+ "loss": 0.7373,
339
+ "step": 9800
340
+ },
341
+ {
342
+ "epoch": 6.12,
343
+ "learning_rate": 0.00018910632746249183,
344
+ "loss": 0.698,
345
+ "step": 10000
346
+ },
347
+ {
348
+ "epoch": 6.12,
349
+ "eval_cer": 0.13653489424101573,
350
+ "eval_loss": 0.5013594031333923,
351
+ "eval_runtime": 1605.1161,
352
+ "eval_samples_per_second": 8.157,
353
+ "eval_steps_per_second": 0.51,
354
+ "step": 10000
355
+ },
356
+ {
357
+ "epoch": 6.24,
358
+ "learning_rate": 0.00018649706457925634,
359
+ "loss": 0.6943,
360
+ "step": 10200
361
+ },
362
+ {
363
+ "epoch": 6.37,
364
+ "learning_rate": 0.00018388780169602086,
365
+ "loss": 0.6997,
366
+ "step": 10400
367
+ },
368
+ {
369
+ "epoch": 6.49,
370
+ "learning_rate": 0.00018127853881278537,
371
+ "loss": 0.6929,
372
+ "step": 10600
373
+ },
374
+ {
375
+ "epoch": 6.61,
376
+ "learning_rate": 0.00017866927592954989,
377
+ "loss": 0.7003,
378
+ "step": 10800
379
+ },
380
+ {
381
+ "epoch": 6.73,
382
+ "learning_rate": 0.0001760600130463144,
383
+ "loss": 0.6863,
384
+ "step": 11000
385
+ },
386
+ {
387
+ "epoch": 6.86,
388
+ "learning_rate": 0.00017345075016307891,
389
+ "loss": 0.6883,
390
+ "step": 11200
391
+ },
392
+ {
393
+ "epoch": 6.98,
394
+ "learning_rate": 0.00017084148727984343,
395
+ "loss": 0.6787,
396
+ "step": 11400
397
+ },
398
+ {
399
+ "epoch": 7.1,
400
+ "learning_rate": 0.00016823222439660794,
401
+ "loss": 0.6518,
402
+ "step": 11600
403
+ },
404
+ {
405
+ "epoch": 7.22,
406
+ "learning_rate": 0.00016562296151337246,
407
+ "loss": 0.6494,
408
+ "step": 11800
409
+ },
410
+ {
411
+ "epoch": 7.35,
412
+ "learning_rate": 0.00016301369863013697,
413
+ "loss": 0.6426,
414
+ "step": 12000
415
+ },
416
+ {
417
+ "epoch": 7.35,
418
+ "eval_cer": 0.13216305737125092,
419
+ "eval_loss": 0.47150149941444397,
420
+ "eval_runtime": 1597.7342,
421
+ "eval_samples_per_second": 8.195,
422
+ "eval_steps_per_second": 0.513,
423
+ "step": 12000
424
+ },
425
+ {
426
+ "epoch": 7.47,
427
+ "learning_rate": 0.00016040443574690148,
428
+ "loss": 0.6457,
429
+ "step": 12200
430
+ },
431
+ {
432
+ "epoch": 7.59,
433
+ "learning_rate": 0.000157795172863666,
434
+ "loss": 0.6429,
435
+ "step": 12400
436
+ },
437
+ {
438
+ "epoch": 7.71,
439
+ "learning_rate": 0.0001551859099804305,
440
+ "loss": 0.6512,
441
+ "step": 12600
442
+ },
443
+ {
444
+ "epoch": 7.83,
445
+ "learning_rate": 0.00015257664709719503,
446
+ "loss": 0.6458,
447
+ "step": 12800
448
+ },
449
+ {
450
+ "epoch": 7.96,
451
+ "learning_rate": 0.00014996738421395954,
452
+ "loss": 0.6501,
453
+ "step": 13000
454
+ },
455
+ {
456
+ "epoch": 8.08,
457
+ "learning_rate": 0.00014735812133072405,
458
+ "loss": 0.6095,
459
+ "step": 13200
460
+ },
461
+ {
462
+ "epoch": 8.2,
463
+ "learning_rate": 0.00014474885844748857,
464
+ "loss": 0.6132,
465
+ "step": 13400
466
+ },
467
+ {
468
+ "epoch": 8.32,
469
+ "learning_rate": 0.00014213959556425308,
470
+ "loss": 0.6102,
471
+ "step": 13600
472
+ },
473
+ {
474
+ "epoch": 8.45,
475
+ "learning_rate": 0.0001395303326810176,
476
+ "loss": 0.6125,
477
+ "step": 13800
478
+ },
479
+ {
480
+ "epoch": 8.57,
481
+ "learning_rate": 0.0001369210697977821,
482
+ "loss": 0.61,
483
+ "step": 14000
484
+ },
485
+ {
486
+ "epoch": 8.57,
487
+ "eval_cer": 0.1257692459492199,
488
+ "eval_loss": 0.45295360684394836,
489
+ "eval_runtime": 1603.0798,
490
+ "eval_samples_per_second": 8.167,
491
+ "eval_steps_per_second": 0.511,
492
+ "step": 14000
493
+ },
494
+ {
495
+ "epoch": 8.69,
496
+ "learning_rate": 0.00013431180691454662,
497
+ "loss": 0.606,
498
+ "step": 14200
499
+ },
500
+ {
501
+ "epoch": 8.81,
502
+ "learning_rate": 0.00013170254403131114,
503
+ "loss": 0.5957,
504
+ "step": 14400
505
+ },
506
+ {
507
+ "epoch": 8.94,
508
+ "learning_rate": 0.00012909328114807565,
509
+ "loss": 0.5992,
510
+ "step": 14600
511
+ },
512
+ {
513
+ "epoch": 9.06,
514
+ "learning_rate": 0.00012648401826484017,
515
+ "loss": 0.5752,
516
+ "step": 14800
517
+ },
518
+ {
519
+ "epoch": 9.18,
520
+ "learning_rate": 0.00012387475538160468,
521
+ "loss": 0.5654,
522
+ "step": 15000
523
+ },
524
+ {
525
+ "epoch": 9.3,
526
+ "learning_rate": 0.00012126549249836919,
527
+ "loss": 0.5725,
528
+ "step": 15200
529
+ },
530
+ {
531
+ "epoch": 9.43,
532
+ "learning_rate": 0.00011865622961513371,
533
+ "loss": 0.5713,
534
+ "step": 15400
535
+ },
536
+ {
537
+ "epoch": 9.55,
538
+ "learning_rate": 0.00011604696673189822,
539
+ "loss": 0.5649,
540
+ "step": 15600
541
+ },
542
+ {
543
+ "epoch": 9.67,
544
+ "learning_rate": 0.00011343770384866273,
545
+ "loss": 0.5643,
546
+ "step": 15800
547
+ },
548
+ {
549
+ "epoch": 9.79,
550
+ "learning_rate": 0.00011082844096542725,
551
+ "loss": 0.5709,
552
+ "step": 16000
553
+ },
554
+ {
555
+ "epoch": 9.79,
556
+ "eval_cer": 0.1200554980402634,
557
+ "eval_loss": 0.4299587607383728,
558
+ "eval_runtime": 1609.5227,
559
+ "eval_samples_per_second": 8.135,
560
+ "eval_steps_per_second": 0.509,
561
+ "step": 16000
562
+ },
563
+ {
564
+ "epoch": 9.92,
565
+ "learning_rate": 0.00010821917808219176,
566
+ "loss": 0.5666,
567
+ "step": 16200
568
+ },
569
+ {
570
+ "epoch": 10.04,
571
+ "learning_rate": 0.00010560991519895628,
572
+ "loss": 0.5531,
573
+ "step": 16400
574
+ },
575
+ {
576
+ "epoch": 10.16,
577
+ "learning_rate": 0.00010300065231572079,
578
+ "loss": 0.5389,
579
+ "step": 16600
580
+ },
581
+ {
582
+ "epoch": 10.28,
583
+ "learning_rate": 0.0001003913894324853,
584
+ "loss": 0.5456,
585
+ "step": 16800
586
+ },
587
+ {
588
+ "epoch": 10.41,
589
+ "learning_rate": 9.778212654924982e-05,
590
+ "loss": 0.5353,
591
+ "step": 17000
592
+ },
593
+ {
594
+ "epoch": 10.53,
595
+ "learning_rate": 9.517286366601433e-05,
596
+ "loss": 0.5337,
597
+ "step": 17200
598
+ },
599
+ {
600
+ "epoch": 10.65,
601
+ "learning_rate": 9.256360078277885e-05,
602
+ "loss": 0.5296,
603
+ "step": 17400
604
+ },
605
+ {
606
+ "epoch": 10.77,
607
+ "learning_rate": 8.995433789954336e-05,
608
+ "loss": 0.5372,
609
+ "step": 17600
610
+ },
611
+ {
612
+ "epoch": 10.9,
613
+ "learning_rate": 8.734507501630787e-05,
614
+ "loss": 0.5388,
615
+ "step": 17800
616
+ },
617
+ {
618
+ "epoch": 11.02,
619
+ "learning_rate": 8.473581213307239e-05,
620
+ "loss": 0.5235,
621
+ "step": 18000
622
+ },
623
+ {
624
+ "epoch": 11.02,
625
+ "eval_cer": 0.11664607248141211,
626
+ "eval_loss": 0.4167773723602295,
627
+ "eval_runtime": 1608.2913,
628
+ "eval_samples_per_second": 8.141,
629
+ "eval_steps_per_second": 0.509,
630
+ "step": 18000
631
+ },
632
+ {
633
+ "epoch": 11.14,
634
+ "learning_rate": 8.212654924983692e-05,
635
+ "loss": 0.509,
636
+ "step": 18200
637
+ },
638
+ {
639
+ "epoch": 11.26,
640
+ "learning_rate": 7.951728636660143e-05,
641
+ "loss": 0.5116,
642
+ "step": 18400
643
+ },
644
+ {
645
+ "epoch": 11.38,
646
+ "learning_rate": 7.690802348336594e-05,
647
+ "loss": 0.4967,
648
+ "step": 18600
649
+ },
650
+ {
651
+ "epoch": 11.51,
652
+ "learning_rate": 7.429876060013046e-05,
653
+ "loss": 0.511,
654
+ "step": 18800
655
+ },
656
+ {
657
+ "epoch": 11.63,
658
+ "learning_rate": 7.168949771689497e-05,
659
+ "loss": 0.5056,
660
+ "step": 19000
661
+ },
662
+ {
663
+ "epoch": 11.75,
664
+ "learning_rate": 6.908023483365949e-05,
665
+ "loss": 0.5073,
666
+ "step": 19200
667
+ },
668
+ {
669
+ "epoch": 11.87,
670
+ "learning_rate": 6.6470971950424e-05,
671
+ "loss": 0.4968,
672
+ "step": 19400
673
+ },
674
+ {
675
+ "epoch": 12.0,
676
+ "learning_rate": 6.386170906718851e-05,
677
+ "loss": 0.5009,
678
+ "step": 19600
679
+ },
680
+ {
681
+ "epoch": 12.12,
682
+ "learning_rate": 6.125244618395303e-05,
683
+ "loss": 0.4832,
684
+ "step": 19800
685
+ },
686
+ {
687
+ "epoch": 12.24,
688
+ "learning_rate": 5.864318330071754e-05,
689
+ "loss": 0.4778,
690
+ "step": 20000
691
+ },
692
+ {
693
+ "epoch": 12.24,
694
+ "eval_cer": 0.11294822712906938,
695
+ "eval_loss": 0.40570223331451416,
696
+ "eval_runtime": 1612.661,
697
+ "eval_samples_per_second": 8.119,
698
+ "eval_steps_per_second": 0.508,
699
+ "step": 20000
700
+ },
701
+ {
702
+ "epoch": 12.36,
703
+ "learning_rate": 5.6033920417482055e-05,
704
+ "loss": 0.4775,
705
+ "step": 20200
706
+ },
707
+ {
708
+ "epoch": 12.49,
709
+ "learning_rate": 5.342465753424657e-05,
710
+ "loss": 0.4855,
711
+ "step": 20400
712
+ },
713
+ {
714
+ "epoch": 12.61,
715
+ "learning_rate": 5.081539465101108e-05,
716
+ "loss": 0.4773,
717
+ "step": 20600
718
+ },
719
+ {
720
+ "epoch": 12.73,
721
+ "learning_rate": 4.82061317677756e-05,
722
+ "loss": 0.4745,
723
+ "step": 20800
724
+ },
725
+ {
726
+ "epoch": 12.85,
727
+ "learning_rate": 4.559686888454011e-05,
728
+ "loss": 0.48,
729
+ "step": 21000
730
+ },
731
+ {
732
+ "epoch": 12.98,
733
+ "learning_rate": 4.2987606001304625e-05,
734
+ "loss": 0.463,
735
+ "step": 21200
736
+ },
737
+ {
738
+ "epoch": 13.1,
739
+ "learning_rate": 4.037834311806914e-05,
740
+ "loss": 0.4643,
741
+ "step": 21400
742
+ },
743
+ {
744
+ "epoch": 13.22,
745
+ "learning_rate": 3.776908023483365e-05,
746
+ "loss": 0.449,
747
+ "step": 21600
748
+ },
749
+ {
750
+ "epoch": 13.34,
751
+ "learning_rate": 3.515981735159817e-05,
752
+ "loss": 0.4604,
753
+ "step": 21800
754
+ },
755
+ {
756
+ "epoch": 13.47,
757
+ "learning_rate": 3.255055446836268e-05,
758
+ "loss": 0.4571,
759
+ "step": 22000
760
+ },
761
+ {
762
+ "epoch": 13.47,
763
+ "eval_cer": 0.10995473327241098,
764
+ "eval_loss": 0.3945465385913849,
765
+ "eval_runtime": 1610.0901,
766
+ "eval_samples_per_second": 8.132,
767
+ "eval_steps_per_second": 0.509,
768
+ "step": 22000
769
+ }
770
+ ],
771
+ "logging_steps": 200,
772
+ "max_steps": 24495,
773
+ "num_train_epochs": 15,
774
+ "save_steps": 2000,
775
+ "total_flos": 1.5466215358461693e+20,
776
+ "trial_name": null,
777
+ "trial_params": null
778
+ }
checkpoint-22000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2a53791a3905b5426f4f4a28dc53b7bf0cdde7382afe6d18a7f69f8c493b179
3
+ size 4472
checkpoint-22000/vocab.json ADDED
@@ -0,0 +1,1744 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "&": 1,
3
+ "0": 2,
4
+ "1": 3,
5
+ "2": 4,
6
+ "3": 5,
7
+ "4": 6,
8
+ "5": 7,
9
+ "6": 8,
10
+ "7": 9,
11
+ "8": 10,
12
+ "9": 11,
13
+ "[PAD]": 1741,
14
+ "[UNK]": 1740,
15
+ "\\": 12,
16
+ "a": 13,
17
+ "b": 14,
18
+ "c": 15,
19
+ "d": 16,
20
+ "e": 17,
21
+ "f": 18,
22
+ "g": 19,
23
+ "h": 20,
24
+ "i": 21,
25
+ "j": 22,
26
+ "k": 23,
27
+ "l": 24,
28
+ "m": 25,
29
+ "n": 26,
30
+ "o": 27,
31
+ "p": 28,
32
+ "q": 29,
33
+ "r": 30,
34
+ "s": 31,
35
+ "t": 32,
36
+ "u": 33,
37
+ "v": 34,
38
+ "w": 35,
39
+ "x": 36,
40
+ "y": 37,
41
+ "z": 38,
42
+ "|": 0,
43
+ "°": 39,
44
+ "μ": 40,
45
+ "ᆫ": 41,
46
+ "ㄱ": 42,
47
+ "ㄴ": 43,
48
+ "ㄷ": 44,
49
+ "ㄹ": 45,
50
+ "ㅁ": 46,
51
+ "ㅂ": 47,
52
+ "ㅅ": 48,
53
+ "ㅇ": 49,
54
+ "ㅈ": 50,
55
+ "ㅜ": 51,
56
+ "ㅠ": 52,
57
+ "ㅡ": 53,
58
+ "가": 54,
59
+ "각": 55,
60
+ "간": 56,
61
+ "갇": 57,
62
+ "갈": 58,
63
+ "감": 59,
64
+ "갑": 60,
65
+ "값": 61,
66
+ "갓": 62,
67
+ "갔": 63,
68
+ "강": 64,
69
+ "갖": 65,
70
+ "같": 66,
71
+ "갚": 67,
72
+ "갛": 68,
73
+ "개": 69,
74
+ "객": 70,
75
+ "갠": 71,
76
+ "갤": 72,
77
+ "갬": 73,
78
+ "갭": 74,
79
+ "갯": 75,
80
+ "갰": 76,
81
+ "갱": 77,
82
+ "갸": 78,
83
+ "걀": 79,
84
+ "걍": 80,
85
+ "걔": 81,
86
+ "걘": 82,
87
+ "거": 83,
88
+ "걱": 84,
89
+ "건": 85,
90
+ "걷": 86,
91
+ "걸": 87,
92
+ "검": 88,
93
+ "겁": 89,
94
+ "것": 90,
95
+ "겄": 91,
96
+ "겉": 92,
97
+ "게": 93,
98
+ "겐": 94,
99
+ "겔": 95,
100
+ "겜": 96,
101
+ "겟": 97,
102
+ "겠": 98,
103
+ "겨": 99,
104
+ "격": 100,
105
+ "겪": 101,
106
+ "견": 102,
107
+ "결": 103,
108
+ "겸": 104,
109
+ "겹": 105,
110
+ "겼": 106,
111
+ "경": 107,
112
+ "곁": 108,
113
+ "계": 109,
114
+ "곗": 110,
115
+ "고": 111,
116
+ "곡": 112,
117
+ "곤": 113,
118
+ "곧": 114,
119
+ "골": 115,
120
+ "곪": 116,
121
+ "곰": 117,
122
+ "곱": 118,
123
+ "곳": 119,
124
+ "공": 120,
125
+ "곶": 121,
126
+ "과": 122,
127
+ "곽": 123,
128
+ "관": 124,
129
+ "괄": 125,
130
+ "괍": 126,
131
+ "광": 127,
132
+ "괘": 128,
133
+ "괜": 129,
134
+ "괭": 130,
135
+ "괴": 131,
136
+ "굉": 132,
137
+ "교": 133,
138
+ "굣": 134,
139
+ "구": 135,
140
+ "국": 136,
141
+ "군": 137,
142
+ "굳": 138,
143
+ "굴": 139,
144
+ "굵": 140,
145
+ "굶": 141,
146
+ "굽": 142,
147
+ "굿": 143,
148
+ "궁": 144,
149
+ "궂": 145,
150
+ "궈": 146,
151
+ "권": 147,
152
+ "궐": 148,
153
+ "궜": 149,
154
+ "귀": 150,
155
+ "귄": 151,
156
+ "귈": 152,
157
+ "귓": 153,
158
+ "규": 154,
159
+ "균": 155,
160
+ "귤": 156,
161
+ "그": 157,
162
+ "극": 158,
163
+ "근": 159,
164
+ "귿": 160,
165
+ "글": 161,
166
+ "긁": 162,
167
+ "금": 163,
168
+ "급": 164,
169
+ "긋": 165,
170
+ "긍": 166,
171
+ "긑": 167,
172
+ "기": 168,
173
+ "긴": 169,
174
+ "길": 170,
175
+ "김": 171,
176
+ "깁": 172,
177
+ "깃": 173,
178
+ "깄": 174,
179
+ "깅": 175,
180
+ "깊": 176,
181
+ "까": 177,
182
+ "깍": 178,
183
+ "깎": 179,
184
+ "깐": 180,
185
+ "깔": 181,
186
+ "깜": 182,
187
+ "깝": 183,
188
+ "깟": 184,
189
+ "깠": 185,
190
+ "깡": 186,
191
+ "깥": 187,
192
+ "깨": 188,
193
+ "깬": 189,
194
+ "깰": 190,
195
+ "깻": 191,
196
+ "깼": 192,
197
+ "깽": 193,
198
+ "꺼": 194,
199
+ "꺽": 195,
200
+ "꺾": 196,
201
+ "껀": 197,
202
+ "껄": 198,
203
+ "껌": 199,
204
+ "껍": 200,
205
+ "껏": 201,
206
+ "껐": 202,
207
+ "껑": 203,
208
+ "께": 204,
209
+ "껩": 205,
210
+ "껴": 206,
211
+ "꼈": 207,
212
+ "꼐": 208,
213
+ "꼬": 209,
214
+ "꼭": 210,
215
+ "꼰": 211,
216
+ "꼴": 212,
217
+ "꼼": 213,
218
+ "꼽": 214,
219
+ "꽁": 215,
220
+ "꽂": 216,
221
+ "꽃": 217,
222
+ "꽈": 218,
223
+ "꽉": 219,
224
+ "꽌": 220,
225
+ "꽜": 221,
226
+ "꽝": 222,
227
+ "꽤": 223,
228
+ "꽥": 224,
229
+ "꾀": 225,
230
+ "꾸": 226,
231
+ "꾹": 227,
232
+ "꾼": 228,
233
+ "꿀": 229,
234
+ "꿇": 230,
235
+ "꿈": 231,
236
+ "꿉": 232,
237
+ "꿍": 233,
238
+ "꿔": 234,
239
+ "꿨": 235,
240
+ "꿰": 236,
241
+ "뀌": 237,
242
+ "뀐": 238,
243
+ "뀔": 239,
244
+ "뀝": 240,
245
+ "뀰": 241,
246
+ "끄": 242,
247
+ "끅": 243,
248
+ "끈": 244,
249
+ "끊": 245,
250
+ "끌": 246,
251
+ "끓": 247,
252
+ "끔": 248,
253
+ "끕": 249,
254
+ "끗": 250,
255
+ "끙": 251,
256
+ "끝": 252,
257
+ "끼": 253,
258
+ "끽": 254,
259
+ "낀": 255,
260
+ "낄": 256,
261
+ "낌": 257,
262
+ "낍": 258,
263
+ "낑": 259,
264
+ "나": 260,
265
+ "낙": 261,
266
+ "낚": 262,
267
+ "난": 263,
268
+ "날": 264,
269
+ "낡": 265,
270
+ "남": 266,
271
+ "납": 267,
272
+ "낫": 268,
273
+ "났": 269,
274
+ "낭": 270,
275
+ "낮": 271,
276
+ "낯": 272,
277
+ "낱": 273,
278
+ "낳": 274,
279
+ "내": 275,
280
+ "낵": 276,
281
+ "낸": 277,
282
+ "낼": 278,
283
+ "냄": 279,
284
+ "냅": 280,
285
+ "냇": 281,
286
+ "냈": 282,
287
+ "냉": 283,
288
+ "냐": 284,
289
+ "냑": 285,
290
+ "냠": 286,
291
+ "냥": 287,
292
+ "너": 288,
293
+ "넉": 289,
294
+ "넌": 290,
295
+ "널": 291,
296
+ "넓": 292,
297
+ "넘": 293,
298
+ "넛": 294,
299
+ "넣": 295,
300
+ "네": 296,
301
+ "넥": 297,
302
+ "넬": 298,
303
+ "넵": 299,
304
+ "넷": 300,
305
+ "넹": 301,
306
+ "녀": 302,
307
+ "녁": 303,
308
+ "년": 304,
309
+ "념": 305,
310
+ "녔": 306,
311
+ "녕": 307,
312
+ "녜": 308,
313
+ "노": 309,
314
+ "녹": 310,
315
+ "논": 311,
316
+ "놀": 312,
317
+ "놈": 313,
318
+ "놉": 314,
319
+ "농": 315,
320
+ "높": 316,
321
+ "놓": 317,
322
+ "놔": 318,
323
+ "놨": 319,
324
+ "뇌": 320,
325
+ "뇨": 321,
326
+ "누": 322,
327
+ "눅": 323,
328
+ "눈": 324,
329
+ "눌": 325,
330
+ "눔": 326,
331
+ "눕": 327,
332
+ "눗": 328,
333
+ "눠": 329,
334
+ "눴": 330,
335
+ "뉘": 331,
336
+ "뉜": 332,
337
+ "뉠": 333,
338
+ "뉩": 334,
339
+ "뉴": 335,
340
+ "늄": 336,
341
+ "느": 337,
342
+ "늑": 338,
343
+ "는": 339,
344
+ "늘": 340,
345
+ "늙": 341,
346
+ "늠": 342,
347
+ "늡": 343,
348
+ "능": 344,
349
+ "늦": 345,
350
+ "늬": 346,
351
+ "니": 347,
352
+ "닉": 348,
353
+ "닌": 349,
354
+ "닐": 350,
355
+ "님": 351,
356
+ "닙": 352,
357
+ "닛": 353,
358
+ "닝": 354,
359
+ "다": 355,
360
+ "닥": 356,
361
+ "닦": 357,
362
+ "단": 358,
363
+ "닫": 359,
364
+ "달": 360,
365
+ "닭": 361,
366
+ "닮": 362,
367
+ "닳": 363,
368
+ "담": 364,
369
+ "답": 365,
370
+ "닷": 366,
371
+ "당": 367,
372
+ "닿": 368,
373
+ "대": 369,
374
+ "댁": 370,
375
+ "댄": 371,
376
+ "댈": 372,
377
+ "댐": 373,
378
+ "댑": 374,
379
+ "댓": 375,
380
+ "댔": 376,
381
+ "댕": 377,
382
+ "댜": 378,
383
+ "더": 379,
384
+ "덕": 380,
385
+ "던": 381,
386
+ "덜": 382,
387
+ "덟": 383,
388
+ "덤": 384,
389
+ "덥": 385,
390
+ "덧": 386,
391
+ "덩": 387,
392
+ "덮": 388,
393
+ "데": 389,
394
+ "덱": 390,
395
+ "덴": 391,
396
+ "델": 392,
397
+ "뎅": 393,
398
+ "뎌": 394,
399
+ "뎠": 395,
400
+ "도": 396,
401
+ "독": 397,
402
+ "돈": 398,
403
+ "돋": 399,
404
+ "돌": 400,
405
+ "돔": 401,
406
+ "돕": 402,
407
+ "돗": 403,
408
+ "동": 404,
409
+ "돼": 405,
410
+ "됍": 406,
411
+ "됐": 407,
412
+ "되": 408,
413
+ "된": 409,
414
+ "될": 410,
415
+ "됨": 411,
416
+ "됩": 412,
417
+ "됬": 413,
418
+ "두": 414,
419
+ "둑": 415,
420
+ "둔": 416,
421
+ "둘": 417,
422
+ "둠": 418,
423
+ "둡": 419,
424
+ "둣": 420,
425
+ "둥": 421,
426
+ "둬": 422,
427
+ "뒀": 423,
428
+ "뒈": 424,
429
+ "뒤": 425,
430
+ "뒨": 426,
431
+ "뒷": 427,
432
+ "뒹": 428,
433
+ "듀": 429,
434
+ "드": 430,
435
+ "득": 431,
436
+ "든": 432,
437
+ "듣": 433,
438
+ "들": 434,
439
+ "듦": 435,
440
+ "듬": 436,
441
+ "듭": 437,
442
+ "듯": 438,
443
+ "등": 439,
444
+ "디": 440,
445
+ "딕": 441,
446
+ "딘": 442,
447
+ "딜": 443,
448
+ "딤": 444,
449
+ "딥": 445,
450
+ "딧": 446,
451
+ "딨": 447,
452
+ "딩": 448,
453
+ "딪": 449,
454
+ "따": 450,
455
+ "딱": 451,
456
+ "딴": 452,
457
+ "딸": 453,
458
+ "땀": 454,
459
+ "땁": 455,
460
+ "땃": 456,
461
+ "땄": 457,
462
+ "땅": 458,
463
+ "땋": 459,
464
+ "때": 460,
465
+ "땍": 461,
466
+ "땐": 462,
467
+ "땔": 463,
468
+ "땜": 464,
469
+ "땟": 465,
470
+ "땠": 466,
471
+ "땡": 467,
472
+ "떄": 468,
473
+ "떠": 469,
474
+ "떡": 470,
475
+ "떤": 471,
476
+ "떨": 472,
477
+ "떰": 473,
478
+ "떱": 474,
479
+ "떳": 475,
480
+ "떴": 476,
481
+ "떻": 477,
482
+ "떼": 478,
483
+ "떽": 479,
484
+ "뗀": 480,
485
+ "뗄": 481,
486
+ "뗌": 482,
487
+ "뗍": 483,
488
+ "뗐": 484,
489
+ "뗴": 485,
490
+ "또": 486,
491
+ "똑": 487,
492
+ "똘": 488,
493
+ "똣": 489,
494
+ "똥": 490,
495
+ "뚜": 491,
496
+ "뚝": 492,
497
+ "뚤": 493,
498
+ "뚫": 494,
499
+ "뚱": 495,
500
+ "뛌": 496,
501
+ "뛰": 497,
502
+ "뛴": 498,
503
+ "뛸": 499,
504
+ "뜄": 500,
505
+ "뜨": 501,
506
+ "뜩": 502,
507
+ "뜬": 503,
508
+ "뜯": 504,
509
+ "뜰": 505,
510
+ "뜸": 506,
511
+ "뜹": 507,
512
+ "뜻": 508,
513
+ "띄": 509,
514
+ "띈": 510,
515
+ "띌": 511,
516
+ "띔": 512,
517
+ "띕": 513,
518
+ "띠": 514,
519
+ "띡": 515,
520
+ "띨": 516,
521
+ "띵": 517,
522
+ "라": 518,
523
+ "락": 519,
524
+ "란": 520,
525
+ "랄": 521,
526
+ "람": 522,
527
+ "랍": 523,
528
+ "랏": 524,
529
+ "랐": 525,
530
+ "랑": 526,
531
+ "랖": 527,
532
+ "랗": 528,
533
+ "래": 529,
534
+ "랙": 530,
535
+ "랜": 531,
536
+ "랠": 532,
537
+ "램": 533,
538
+ "랩": 534,
539
+ "랫": 535,
540
+ "랬": 536,
541
+ "랭": 537,
542
+ "랴": 538,
543
+ "략": 539,
544
+ "량": 540,
545
+ "러": 541,
546
+ "럭": 542,
547
+ "런": 543,
548
+ "럴": 544,
549
+ "럼": 545,
550
+ "럽": 546,
551
+ "럿": 547,
552
+ "렀": 548,
553
+ "렁": 549,
554
+ "렇": 550,
555
+ "레": 551,
556
+ "렉": 552,
557
+ "렌": 553,
558
+ "렐": 554,
559
+ "렘": 555,
560
+ "렙": 556,
561
+ "렛": 557,
562
+ "렜": 558,
563
+ "려": 559,
564
+ "력": 560,
565
+ "련": 561,
566
+ "렬": 562,
567
+ "렴": 563,
568
+ "렵": 564,
569
+ "렷": 565,
570
+ "렸": 566,
571
+ "령": 567,
572
+ "렿": 568,
573
+ "례": 569,
574
+ "로": 570,
575
+ "록": 571,
576
+ "론": 572,
577
+ "롤": 573,
578
+ "롬": 574,
579
+ "롭": 575,
580
+ "롯": 576,
581
+ "롱": 577,
582
+ "롷": 578,
583
+ "롸": 579,
584
+ "뢰": 580,
585
+ "료": 581,
586
+ "룡": 582,
587
+ "루": 583,
588
+ "룩": 584,
589
+ "룬": 585,
590
+ "룰": 586,
591
+ "룸": 587,
592
+ "룹": 588,
593
+ "룻": 589,
594
+ "룽": 590,
595
+ "뤄": 591,
596
+ "뤘": 592,
597
+ "류": 593,
598
+ "륙": 594,
599
+ "륜": 595,
600
+ "률": 596,
601
+ "륨": 597,
602
+ "륭": 598,
603
+ "르": 599,
604
+ "륵": 600,
605
+ "른": 601,
606
+ "를": 602,
607
+ "름": 603,
608
+ "릅": 604,
609
+ "릇": 605,
610
+ "릉": 606,
611
+ "릎": 607,
612
+ "리": 608,
613
+ "릭": 609,
614
+ "린": 610,
615
+ "릴": 611,
616
+ "림": 612,
617
+ "립": 613,
618
+ "릿": 614,
619
+ "맀": 615,
620
+ "링": 616,
621
+ "마": 617,
622
+ "막": 618,
623
+ "만": 619,
624
+ "많": 620,
625
+ "맏": 621,
626
+ "말": 622,
627
+ "맑": 623,
628
+ "맘": 624,
629
+ "맙": 625,
630
+ "맛": 626,
631
+ "맜": 627,
632
+ "망": 628,
633
+ "맞": 629,
634
+ "맡": 630,
635
+ "맣": 631,
636
+ "매": 632,
637
+ "맥": 633,
638
+ "맨": 634,
639
+ "맴": 635,
640
+ "맵": 636,
641
+ "맷": 637,
642
+ "맸": 638,
643
+ "맹": 639,
644
+ "맺": 640,
645
+ "맻": 641,
646
+ "머": 642,
647
+ "먹": 643,
648
+ "먼": 644,
649
+ "멀": 645,
650
+ "멈": 646,
651
+ "멉": 647,
652
+ "멋": 648,
653
+ "멍": 649,
654
+ "멓": 650,
655
+ "메": 651,
656
+ "멕": 652,
657
+ "멘": 653,
658
+ "멜": 654,
659
+ "멤": 655,
660
+ "멧": 656,
661
+ "멨": 657,
662
+ "멩": 658,
663
+ "며": 659,
664
+ "멱": 660,
665
+ "면": 661,
666
+ "멸": 662,
667
+ "몄": 663,
668
+ "명": 664,
669
+ "몇": 665,
670
+ "모": 666,
671
+ "목": 667,
672
+ "몫": 668,
673
+ "몬": 669,
674
+ "몰": 670,
675
+ "몸": 671,
676
+ "몹": 672,
677
+ "못": 673,
678
+ "몽": 674,
679
+ "묘": 675,
680
+ "무": 676,
681
+ "묵": 677,
682
+ "묶": 678,
683
+ "문": 679,
684
+ "묻": 680,
685
+ "물": 681,
686
+ "묽": 682,
687
+ "뭅": 683,
688
+ "뭇": 684,
689
+ "뭉": 685,
690
+ "뭏": 686,
691
+ "뭐": 687,
692
+ "뭔": 688,
693
+ "뭘": 689,
694
+ "뭡": 690,
695
+ "��": 691,
696
+ "뮤": 692,
697
+ "뮨": 693,
698
+ "므": 694,
699
+ "믄": 695,
700
+ "믈": 696,
701
+ "미": 697,
702
+ "믹": 698,
703
+ "민": 699,
704
+ "믿": 700,
705
+ "밀": 701,
706
+ "밉": 702,
707
+ "밋": 703,
708
+ "밌": 704,
709
+ "밍": 705,
710
+ "및": 706,
711
+ "밑": 707,
712
+ "바": 708,
713
+ "박": 709,
714
+ "밖": 710,
715
+ "반": 711,
716
+ "받": 712,
717
+ "발": 713,
718
+ "밝": 714,
719
+ "밟": 715,
720
+ "밤": 716,
721
+ "밥": 717,
722
+ "밧": 718,
723
+ "방": 719,
724
+ "밭": 720,
725
+ "배": 721,
726
+ "백": 722,
727
+ "밴": 723,
728
+ "밸": 724,
729
+ "뱀": 725,
730
+ "뱁": 726,
731
+ "뱃": 727,
732
+ "뱄": 728,
733
+ "뱅": 729,
734
+ "뱉": 730,
735
+ "버": 731,
736
+ "벅": 732,
737
+ "번": 733,
738
+ "벋": 734,
739
+ "벌": 735,
740
+ "범": 736,
741
+ "법": 737,
742
+ "벗": 738,
743
+ "벙": 739,
744
+ "벚": 740,
745
+ "베": 741,
746
+ "벡": 742,
747
+ "벤": 743,
748
+ "벨": 744,
749
+ "벰": 745,
750
+ "벳": 746,
751
+ "벼": 747,
752
+ "벽": 748,
753
+ "변": 749,
754
+ "별": 750,
755
+ "볌": 751,
756
+ "볍": 752,
757
+ "볐": 753,
758
+ "병": 754,
759
+ "볕": 755,
760
+ "보": 756,
761
+ "복": 757,
762
+ "볶": 758,
763
+ "본": 759,
764
+ "볼": 760,
765
+ "봄": 761,
766
+ "봅": 762,
767
+ "봇": 763,
768
+ "봉": 764,
769
+ "봐": 765,
770
+ "봤": 766,
771
+ "봬": 767,
772
+ "뵀": 768,
773
+ "뵈": 769,
774
+ "뵌": 770,
775
+ "뵐": 771,
776
+ "뵙": 772,
777
+ "부": 773,
778
+ "북": 774,
779
+ "분": 775,
780
+ "붇": 776,
781
+ "불": 777,
782
+ "붉": 778,
783
+ "붐": 779,
784
+ "붑": 780,
785
+ "붓": 781,
786
+ "붕": 782,
787
+ "붙": 783,
788
+ "뷔": 784,
789
+ "뷰": 785,
790
+ "뷸": 786,
791
+ "브": 787,
792
+ "븐": 788,
793
+ "블": 789,
794
+ "비": 790,
795
+ "빅": 791,
796
+ "빈": 792,
797
+ "빌": 793,
798
+ "빔": 794,
799
+ "빕": 795,
800
+ "빗": 796,
801
+ "빙": 797,
802
+ "빚": 798,
803
+ "빛": 799,
804
+ "빠": 800,
805
+ "빡": 801,
806
+ "빤": 802,
807
+ "빨": 803,
808
+ "빳": 804,
809
+ "빴": 805,
810
+ "빵": 806,
811
+ "빻": 807,
812
+ "빼": 808,
813
+ "빽": 809,
814
+ "뺀": 810,
815
+ "뺄": 811,
816
+ "뺌": 812,
817
+ "뺍": 813,
818
+ "뺏": 814,
819
+ "뺐": 815,
820
+ "뺑": 816,
821
+ "뺘": 817,
822
+ "뺨": 818,
823
+ "뺼": 819,
824
+ "뻐": 820,
825
+ "뻑": 821,
826
+ "뻔": 822,
827
+ "뻗": 823,
828
+ "뻘": 824,
829
+ "뻣": 825,
830
+ "뻤": 826,
831
+ "뻥": 827,
832
+ "뻬": 828,
833
+ "뼀": 829,
834
+ "뼈": 830,
835
+ "뼘": 831,
836
+ "뼛": 832,
837
+ "뽀": 833,
838
+ "뽁": 834,
839
+ "뽂": 835,
840
+ "뽄": 836,
841
+ "뽈": 837,
842
+ "뽐": 838,
843
+ "뽑": 839,
844
+ "뽕": 840,
845
+ "뽜": 841,
846
+ "뽝": 842,
847
+ "뾰": 843,
848
+ "뿅": 844,
849
+ "뿌": 845,
850
+ "뿍": 846,
851
+ "뿐": 847,
852
+ "뿔": 848,
853
+ "뿜": 849,
854
+ "뿝": 850,
855
+ "뿟": 851,
856
+ "뿡": 852,
857
+ "뿨": 853,
858
+ "쁘": 854,
859
+ "쁜": 855,
860
+ "쁠": 856,
861
+ "쁨": 857,
862
+ "쁩": 858,
863
+ "삐": 859,
864
+ "삑": 860,
865
+ "삔": 861,
866
+ "삘": 862,
867
+ "삥": 863,
868
+ "사": 864,
869
+ "삭": 865,
870
+ "산": 866,
871
+ "살": 867,
872
+ "삶": 868,
873
+ "삼": 869,
874
+ "삽": 870,
875
+ "삿": 871,
876
+ "샀": 872,
877
+ "상": 873,
878
+ "새": 874,
879
+ "색": 875,
880
+ "샌": 876,
881
+ "샐": 877,
882
+ "샘": 878,
883
+ "샙": 879,
884
+ "샜": 880,
885
+ "생": 881,
886
+ "샤": 882,
887
+ "샥": 883,
888
+ "샴": 884,
889
+ "샵": 885,
890
+ "샷": 886,
891
+ "샹": 887,
892
+ "섀": 888,
893
+ "서": 889,
894
+ "석": 890,
895
+ "섞": 891,
896
+ "선": 892,
897
+ "설": 893,
898
+ "섬": 894,
899
+ "섭": 895,
900
+ "섯": 896,
901
+ "섰": 897,
902
+ "성": 898,
903
+ "세": 899,
904
+ "섹": 900,
905
+ "센": 901,
906
+ "셀": 902,
907
+ "셈": 903,
908
+ "셉": 904,
909
+ "셋": 905,
910
+ "셔": 906,
911
+ "션": 907,
912
+ "셜": 908,
913
+ "셧": 909,
914
+ "셨": 910,
915
+ "셰": 911,
916
+ "솁": 912,
917
+ "소": 913,
918
+ "속": 914,
919
+ "손": 915,
920
+ "솔": 916,
921
+ "솜": 917,
922
+ "솝": 918,
923
+ "솟": 919,
924
+ "송": 920,
925
+ "솥": 921,
926
+ "솨": 922,
927
+ "쇄": 923,
928
+ "쇠": 924,
929
+ "쇼": 925,
930
+ "쇽": 926,
931
+ "숄": 927,
932
+ "숍": 928,
933
+ "숏": 929,
934
+ "숑": 930,
935
+ "수": 931,
936
+ "숙": 932,
937
+ "순": 933,
938
+ "숟": 934,
939
+ "술": 935,
940
+ "숨": 936,
941
+ "숩": 937,
942
+ "숫": 938,
943
+ "숭": 939,
944
+ "숯": 940,
945
+ "숱": 941,
946
+ "숲": 942,
947
+ "숴": 943,
948
+ "쉈": 944,
949
+ "쉐": 945,
950
+ "쉘": 946,
951
+ "쉣": 947,
952
+ "쉬": 948,
953
+ "쉰": 949,
954
+ "쉴": 950,
955
+ "쉼": 951,
956
+ "쉽": 952,
957
+ "슈": 953,
958
+ "슉": 954,
959
+ "슐": 955,
960
+ "슘": 956,
961
+ "슛": 957,
962
+ "슝": 958,
963
+ "스": 959,
964
+ "슥": 960,
965
+ "슨": 961,
966
+ "슬": 962,
967
+ "슴": 963,
968
+ "습": 964,
969
+ "슷": 965,
970
+ "승": 966,
971
+ "시": 967,
972
+ "식": 968,
973
+ "신": 969,
974
+ "싣": 970,
975
+ "실": 971,
976
+ "싫": 972,
977
+ "심": 973,
978
+ "십": 974,
979
+ "싯": 975,
980
+ "싰": 976,
981
+ "싱": 977,
982
+ "싶": 978,
983
+ "싸": 979,
984
+ "싹": 980,
985
+ "싼": 981,
986
+ "쌀": 982,
987
+ "쌈": 983,
988
+ "쌉": 984,
989
+ "쌌": 985,
990
+ "쌍": 986,
991
+ "쌓": 987,
992
+ "쌔": 988,
993
+ "쌜": 989,
994
+ "쌤": 990,
995
+ "쌩": 991,
996
+ "쌰": 992,
997
+ "써": 993,
998
+ "썩": 994,
999
+ "썪": 995,
1000
+ "썬": 996,
1001
+ "썰": 997,
1002
+ "썸": 998,
1003
+ "썹": 999,
1004
+ "썻": 1000,
1005
+ "썼": 1001,
1006
+ "썽": 1002,
1007
+ "쎄": 1003,
1008
+ "쎅": 1004,
1009
+ "쎈": 1005,
1010
+ "쎕": 1006,
1011
+ "쎼": 1007,
1012
+ "쏘": 1008,
1013
+ "쏙": 1009,
1014
+ "쏜": 1010,
1015
+ "쏟": 1011,
1016
+ "쏠": 1012,
1017
+ "쏩": 1013,
1018
+ "쏭": 1014,
1019
+ "쏴": 1015,
1020
+ "쏵": 1016,
1021
+ "쐈": 1017,
1022
+ "쐬": 1018,
1023
+ "쐴": 1019,
1024
+ "쑈": 1020,
1025
+ "쑉": 1021,
1026
+ "쑐": 1022,
1027
+ "쑤": 1023,
1028
+ "쑥": 1024,
1029
+ "쓰": 1025,
1030
+ "쓱": 1026,
1031
+ "쓴": 1027,
1032
+ "쓸": 1028,
1033
+ "쓽": 1029,
1034
+ "씀": 1030,
1035
+ "씁": 1031,
1036
+ "씌": 1032,
1037
+ "씐": 1033,
1038
+ "씨": 1034,
1039
+ "씩": 1035,
1040
+ "씬": 1036,
1041
+ "씰": 1037,
1042
+ "씸": 1038,
1043
+ "씹": 1039,
1044
+ "씻": 1040,
1045
+ "씼": 1041,
1046
+ "씽": 1042,
1047
+ "아": 1043,
1048
+ "악": 1044,
1049
+ "안": 1045,
1050
+ "앉": 1046,
1051
+ "않": 1047,
1052
+ "알": 1048,
1053
+ "앎": 1049,
1054
+ "앓": 1050,
1055
+ "암": 1051,
1056
+ "압": 1052,
1057
+ "앗": 1053,
1058
+ "았": 1054,
1059
+ "앙": 1055,
1060
+ "앞": 1056,
1061
+ "애": 1057,
1062
+ "액": 1058,
1063
+ "앤": 1059,
1064
+ "앨": 1060,
1065
+ "앰": 1061,
1066
+ "앱": 1062,
1067
+ "앴": 1063,
1068
+ "앵": 1064,
1069
+ "야": 1065,
1070
+ "약": 1066,
1071
+ "얀": 1067,
1072
+ "얄": 1068,
1073
+ "얇": 1069,
1074
+ "얌": 1070,
1075
+ "얍": 1071,
1076
+ "얏": 1072,
1077
+ "양": 1073,
1078
+ "얕": 1074,
1079
+ "얗": 1075,
1080
+ "얘": 1076,
1081
+ "얜": 1077,
1082
+ "얬": 1078,
1083
+ "얳": 1079,
1084
+ "어": 1080,
1085
+ "억": 1081,
1086
+ "언": 1082,
1087
+ "얹": 1083,
1088
+ "얻": 1084,
1089
+ "얼": 1085,
1090
+ "엄": 1086,
1091
+ "업": 1087,
1092
+ "없": 1088,
1093
+ "엇": 1089,
1094
+ "었": 1090,
1095
+ "엉": 1091,
1096
+ "엊": 1092,
1097
+ "엌": 1093,
1098
+ "엍": 1094,
1099
+ "엎": 1095,
1100
+ "에": 1096,
1101
+ "엑": 1097,
1102
+ "엔": 1098,
1103
+ "엘": 1099,
1104
+ "엠": 1100,
1105
+ "엣": 1101,
1106
+ "엥": 1102,
1107
+ "여": 1103,
1108
+ "역": 1104,
1109
+ "엮": 1105,
1110
+ "연": 1106,
1111
+ "열": 1107,
1112
+ "염": 1108,
1113
+ "엽": 1109,
1114
+ "엿": 1110,
1115
+ "였": 1111,
1116
+ "영": 1112,
1117
+ "옅": 1113,
1118
+ "옆": 1114,
1119
+ "옇": 1115,
1120
+ "예": 1116,
1121
+ "옐": 1117,
1122
+ "옘": 1118,
1123
+ "옙": 1119,
1124
+ "옛": 1120,
1125
+ "옜": 1121,
1126
+ "오": 1122,
1127
+ "옥": 1123,
1128
+ "온": 1124,
1129
+ "올": 1125,
1130
+ "옮": 1126,
1131
+ "옳": 1127,
1132
+ "옴": 1128,
1133
+ "옵": 1129,
1134
+ "옷": 1130,
1135
+ "옹": 1131,
1136
+ "옽": 1132,
1137
+ "와": 1133,
1138
+ "왁": 1134,
1139
+ "완": 1135,
1140
+ "왈": 1136,
1141
+ "왓": 1137,
1142
+ "왔": 1138,
1143
+ "왕": 1139,
1144
+ "왜": 1140,
1145
+ "왠": 1141,
1146
+ "왯": 1142,
1147
+ "외": 1143,
1148
+ "왼": 1144,
1149
+ "요": 1145,
1150
+ "욕": 1146,
1151
+ "욘": 1147,
1152
+ "욜": 1148,
1153
+ "욤": 1149,
1154
+ "용": 1150,
1155
+ "우": 1151,
1156
+ "욱": 1152,
1157
+ "운": 1153,
1158
+ "울": 1154,
1159
+ "움": 1155,
1160
+ "웁": 1156,
1161
+ "웃": 1157,
1162
+ "웅": 1158,
1163
+ "워": 1159,
1164
+ "웍": 1160,
1165
+ "원": 1161,
1166
+ "월": 1162,
1167
+ "웠": 1163,
1168
+ "웡": 1164,
1169
+ "웨": 1165,
1170
+ "웬": 1166,
1171
+ "웰": 1167,
1172
+ "웸": 1168,
1173
+ "웹": 1169,
1174
+ "웻": 1170,
1175
+ "위": 1171,
1176
+ "윅": 1172,
1177
+ "윈": 1173,
1178
+ "윌": 1174,
1179
+ "윕": 1175,
1180
+ "윗": 1176,
1181
+ "윙": 1177,
1182
+ "유": 1178,
1183
+ "육": 1179,
1184
+ "윤": 1180,
1185
+ "율": 1181,
1186
+ "융": 1182,
1187
+ "윷": 1183,
1188
+ "으": 1184,
1189
+ "윽": 1185,
1190
+ "은": 1186,
1191
+ "을": 1187,
1192
+ "읊": 1188,
1193
+ "음": 1189,
1194
+ "읍": 1190,
1195
+ "읎": 1191,
1196
+ "읏": 1192,
1197
+ "응": 1193,
1198
+ "읓": 1194,
1199
+ "의": 1195,
1200
+ "이": 1196,
1201
+ "익": 1197,
1202
+ "인": 1198,
1203
+ "일": 1199,
1204
+ "읽": 1200,
1205
+ "잃": 1201,
1206
+ "임": 1202,
1207
+ "입": 1203,
1208
+ "잇": 1204,
1209
+ "있": 1205,
1210
+ "잉": 1206,
1211
+ "잊": 1207,
1212
+ "잌": 1208,
1213
+ "잎": 1209,
1214
+ "자": 1210,
1215
+ "작": 1211,
1216
+ "잔": 1212,
1217
+ "잖": 1213,
1218
+ "잘": 1214,
1219
+ "잠": 1215,
1220
+ "잡": 1216,
1221
+ "잣": 1217,
1222
+ "잤": 1218,
1223
+ "장": 1219,
1224
+ "잦": 1220,
1225
+ "재": 1221,
1226
+ "잭": 1222,
1227
+ "잰": 1223,
1228
+ "잴": 1224,
1229
+ "잼": 1225,
1230
+ "잽": 1226,
1231
+ "잿": 1227,
1232
+ "쟀": 1228,
1233
+ "쟁": 1229,
1234
+ "쟈": 1230,
1235
+ "쟤": 1231,
1236
+ "쟨": 1232,
1237
+ "저": 1233,
1238
+ "적": 1234,
1239
+ "전": 1235,
1240
+ "절": 1236,
1241
+ "젊": 1237,
1242
+ "점": 1238,
1243
+ "접": 1239,
1244
+ "젓": 1240,
1245
+ "정": 1241,
1246
+ "젖": 1242,
1247
+ "제": 1243,
1248
+ "젝": 1244,
1249
+ "젠": 1245,
1250
+ "젤": 1246,
1251
+ "젭": 1247,
1252
+ "젯": 1248,
1253
+ "져": 1249,
1254
+ "젼": 1250,
1255
+ "졌": 1251,
1256
+ "조": 1252,
1257
+ "족": 1253,
1258
+ "존": 1254,
1259
+ "졸": 1255,
1260
+ "좀": 1256,
1261
+ "좁": 1257,
1262
+ "좃": 1258,
1263
+ "종": 1259,
1264
+ "좆": 1260,
1265
+ "좋": 1261,
1266
+ "좌": 1262,
1267
+ "좔": 1263,
1268
+ "죄": 1264,
1269
+ "죈": 1265,
1270
+ "죙": 1266,
1271
+ "죠": 1267,
1272
+ "죵": 1268,
1273
+ "주": 1269,
1274
+ "죽": 1270,
1275
+ "준": 1271,
1276
+ "줄": 1272,
1277
+ "줌": 1273,
1278
+ "줍": 1274,
1279
+ "줏": 1275,
1280
+ "중": 1276,
1281
+ "줘": 1277,
1282
+ "줬": 1278,
1283
+ "쥐": 1279,
1284
+ "쥑": 1280,
1285
+ "쥬": 1281,
1286
+ "즈": 1282,
1287
+ "즉": 1283,
1288
+ "즌": 1284,
1289
+ "즐": 1285,
1290
+ "즘": 1286,
1291
+ "즙": 1287,
1292
+ "증": 1288,
1293
+ "지": 1289,
1294
+ "직": 1290,
1295
+ "진": 1291,
1296
+ "질": 1292,
1297
+ "짐": 1293,
1298
+ "집": 1294,
1299
+ "짓": 1295,
1300
+ "징": 1296,
1301
+ "짖": 1297,
1302
+ "짙": 1298,
1303
+ "짚": 1299,
1304
+ "짜": 1300,
1305
+ "짝": 1301,
1306
+ "짠": 1302,
1307
+ "짤": 1303,
1308
+ "짧": 1304,
1309
+ "짬": 1305,
1310
+ "짭": 1306,
1311
+ "짰": 1307,
1312
+ "짱": 1308,
1313
+ "째": 1309,
1314
+ "짼": 1310,
1315
+ "쨋": 1311,
1316
+ "쨌": 1312,
1317
+ "쨍": 1313,
1318
+ "쨔": 1314,
1319
+ "쨰": 1315,
1320
+ "쩌": 1316,
1321
+ "쩍": 1317,
1322
+ "쩐": 1318,
1323
+ "쩔": 1319,
1324
+ "쩜": 1320,
1325
+ "쩝": 1321,
1326
+ "쩠": 1322,
1327
+ "쩡": 1323,
1328
+ "쩨": 1324,
1329
+ "쩬": 1325,
1330
+ "쪄": 1326,
1331
+ "쪘": 1327,
1332
+ "쪠": 1328,
1333
+ "쪼": 1329,
1334
+ "쪽": 1330,
1335
+ "쫀": 1331,
1336
+ "쫄": 1332,
1337
+ "쫌": 1333,
1338
+ "쫍": 1334,
1339
+ "쫑": 1335,
1340
+ "쫒": 1336,
1341
+ "쫓": 1337,
1342
+ "쫘": 1338,
1343
+ "쫙": 1339,
1344
+ "쬐": 1340,
1345
+ "쭈": 1341,
1346
+ "쭉": 1342,
1347
+ "쭐": 1343,
1348
+ "쭘": 1344,
1349
+ "쭝": 1345,
1350
+ "쭤": 1346,
1351
+ "쮸": 1347,
1352
+ "쯔": 1348,
1353
+ "쯤": 1349,
1354
+ "쯥": 1350,
1355
+ "찌": 1351,
1356
+ "찍": 1352,
1357
+ "찐": 1353,
1358
+ "찔": 1354,
1359
+ "찜": 1355,
1360
+ "찝": 1356,
1361
+ "찡": 1357,
1362
+ "찢": 1358,
1363
+ "차": 1359,
1364
+ "착": 1360,
1365
+ "찬": 1361,
1366
+ "찮": 1362,
1367
+ "찰": 1363,
1368
+ "참": 1364,
1369
+ "찹": 1365,
1370
+ "찼": 1366,
1371
+ "창": 1367,
1372
+ "찾": 1368,
1373
+ "채": 1369,
1374
+ "책": 1370,
1375
+ "챌": 1371,
1376
+ "챔": 1372,
1377
+ "챕": 1373,
1378
+ "챗": 1374,
1379
+ "챘": 1375,
1380
+ "챙": 1376,
1381
+ "처": 1377,
1382
+ "척": 1378,
1383
+ "천": 1379,
1384
+ "철": 1380,
1385
+ "첨": 1381,
1386
+ "첩": 1382,
1387
+ "첫": 1383,
1388
+ "청": 1384,
1389
+ "체": 1385,
1390
+ "첵": 1386,
1391
+ "첸": 1387,
1392
+ "첼": 1388,
1393
+ "쳇": 1389,
1394
+ "쳐": 1390,
1395
+ "쳤": 1391,
1396
+ "초": 1392,
1397
+ "촉": 1393,
1398
+ "촌": 1394,
1399
+ "촐": 1395,
1400
+ "촛": 1396,
1401
+ "총": 1397,
1402
+ "촤": 1398,
1403
+ "촥": 1399,
1404
+ "촬": 1400,
1405
+ "최": 1401,
1406
+ "쵸": 1402,
1407
+ "추": 1403,
1408
+ "축": 1404,
1409
+ "춘": 1405,
1410
+ "출": 1406,
1411
+ "춤": 1407,
1412
+ "춥": 1408,
1413
+ "춧": 1409,
1414
+ "충": 1410,
1415
+ "춰": 1411,
1416
+ "췄": 1412,
1417
+ "췌": 1413,
1418
+ "취": 1414,
1419
+ "츄": 1415,
1420
+ "츠": 1416,
1421
+ "측": 1417,
1422
+ "층": 1418,
1423
+ "치": 1419,
1424
+ "칙": 1420,
1425
+ "친": 1421,
1426
+ "칠": 1422,
1427
+ "침": 1423,
1428
+ "칩": 1424,
1429
+ "칫": 1425,
1430
+ "칬": 1426,
1431
+ "칭": 1427,
1432
+ "카": 1428,
1433
+ "칵": 1429,
1434
+ "칸": 1430,
1435
+ "칼": 1431,
1436
+ "캅": 1432,
1437
+ "캉": 1433,
1438
+ "캐": 1434,
1439
+ "캔": 1435,
1440
+ "캠": 1436,
1441
+ "캡": 1437,
1442
+ "캣": 1438,
1443
+ "캬": 1439,
1444
+ "커": 1440,
1445
+ "컥": 1441,
1446
+ "컨": 1442,
1447
+ "컬": 1443,
1448
+ "컴": 1444,
1449
+ "컵": 1445,
1450
+ "컷": 1446,
1451
+ "컸": 1447,
1452
+ "컹": 1448,
1453
+ "컽": 1449,
1454
+ "케": 1450,
1455
+ "켄": 1451,
1456
+ "켈": 1452,
1457
+ "켓": 1453,
1458
+ "켜": 1454,
1459
+ "켠": 1455,
1460
+ "켰": 1456,
1461
+ "코": 1457,
1462
+ "콕": 1458,
1463
+ "콘": 1459,
1464
+ "콜": 1460,
1465
+ "콤": 1461,
1466
+ "콧": 1462,
1467
+ "콩": 1463,
1468
+ "콸": 1464,
1469
+ "쾅": 1465,
1470
+ "쾌": 1466,
1471
+ "쿄": 1467,
1472
+ "쿠": 1468,
1473
+ "쿡": 1469,
1474
+ "쿤": 1470,
1475
+ "쿨": 1471,
1476
+ "쿰": 1472,
1477
+ "쿱": 1473,
1478
+ "쿵": 1474,
1479
+ "퀄": 1475,
1480
+ "퀘": 1476,
1481
+ "퀴": 1477,
1482
+ "퀵": 1478,
1483
+ "퀸": 1479,
1484
+ "퀼": 1480,
1485
+ "큐": 1481,
1486
+ "크": 1482,
1487
+ "큰": 1483,
1488
+ "클": 1484,
1489
+ "큼": 1485,
1490
+ "큽": 1486,
1491
+ "킁": 1487,
1492
+ "키": 1488,
1493
+ "킥": 1489,
1494
+ "킨": 1490,
1495
+ "킬": 1491,
1496
+ "킴": 1492,
1497
+ "킵": 1493,
1498
+ "킷": 1494,
1499
+ "킹": 1495,
1500
+ "타": 1496,
1501
+ "탁": 1497,
1502
+ "탄": 1498,
1503
+ "탈": 1499,
1504
+ "탐": 1500,
1505
+ "탑": 1501,
1506
+ "탓": 1502,
1507
+ "탔": 1503,
1508
+ "탕": 1504,
1509
+ "태": 1505,
1510
+ "택": 1506,
1511
+ "탠": 1507,
1512
+ "탬": 1508,
1513
+ "탭": 1509,
1514
+ "탯": 1510,
1515
+ "탱": 1511,
1516
+ "터": 1512,
1517
+ "턱": 1513,
1518
+ "턴": 1514,
1519
+ "털": 1515,
1520
+ "텀": 1516,
1521
+ "텁": 1517,
1522
+ "텃": 1518,
1523
+ "텄": 1519,
1524
+ "텅": 1520,
1525
+ "테": 1521,
1526
+ "텍": 1522,
1527
+ "텐": 1523,
1528
+ "텔": 1524,
1529
+ "템": 1525,
1530
+ "텝": 1526,
1531
+ "텨": 1527,
1532
+ "텼": 1528,
1533
+ "토": 1529,
1534
+ "톡": 1530,
1535
+ "톤": 1531,
1536
+ "톨": 1532,
1537
+ "톰": 1533,
1538
+ "톱": 1534,
1539
+ "통": 1535,
1540
+ "퇴": 1536,
1541
+ "투": 1537,
1542
+ "툭": 1538,
1543
+ "툰": 1539,
1544
+ "툴": 1540,
1545
+ "툼": 1541,
1546
+ "퉁": 1542,
1547
+ "퉤": 1543,
1548
+ "튀": 1544,
1549
+ "튄": 1545,
1550
+ "튈": 1546,
1551
+ "튑": 1547,
1552
+ "튕": 1548,
1553
+ "튜": 1549,
1554
+ "트": 1550,
1555
+ "특": 1551,
1556
+ "튼": 1552,
1557
+ "틀": 1553,
1558
+ "틈": 1554,
1559
+ "틉": 1555,
1560
+ "틋": 1556,
1561
+ "틑": 1557,
1562
+ "티": 1558,
1563
+ "틱": 1559,
1564
+ "틴": 1560,
1565
+ "틸": 1561,
1566
+ "팀": 1562,
1567
+ "팁": 1563,
1568
+ "팅": 1564,
1569
+ "파": 1565,
1570
+ "팍": 1566,
1571
+ "팎": 1567,
1572
+ "판": 1568,
1573
+ "팔": 1569,
1574
+ "팜": 1570,
1575
+ "팝": 1571,
1576
+ "팟": 1572,
1577
+ "팠": 1573,
1578
+ "팡": 1574,
1579
+ "팥": 1575,
1580
+ "패": 1576,
1581
+ "팩": 1577,
1582
+ "팬": 1578,
1583
+ "팸": 1579,
1584
+ "팹": 1580,
1585
+ "팻": 1581,
1586
+ "팽": 1582,
1587
+ "퍼": 1583,
1588
+ "퍽": 1584,
1589
+ "펀": 1585,
1590
+ "펄": 1586,
1591
+ "펌": 1587,
1592
+ "펍": 1588,
1593
+ "펐": 1589,
1594
+ "펑": 1590,
1595
+ "페": 1591,
1596
+ "펙": 1592,
1597
+ "펜": 1593,
1598
+ "펠": 1594,
1599
+ "펩": 1595,
1600
+ "펫": 1596,
1601
+ "펭": 1597,
1602
+ "펴": 1598,
1603
+ "편": 1599,
1604
+ "펼": 1600,
1605
+ "폄": 1601,
1606
+ "폈": 1602,
1607
+ "평": 1603,
1608
+ "폐": 1604,
1609
+ "포": 1605,
1610
+ "폭": 1606,
1611
+ "폰": 1607,
1612
+ "폴": 1608,
1613
+ "폼": 1609,
1614
+ "폽": 1610,
1615
+ "폿": 1611,
1616
+ "퐁": 1612,
1617
+ "표": 1613,
1618
+ "푠": 1614,
1619
+ "푸": 1615,
1620
+ "푹": 1616,
1621
+ "푼": 1617,
1622
+ "풀": 1618,
1623
+ "풂": 1619,
1624
+ "품": 1620,
1625
+ "풉": 1621,
1626
+ "풋": 1622,
1627
+ "풍": 1623,
1628
+ "퓨": 1624,
1629
+ "프": 1625,
1630
+ "픈": 1626,
1631
+ "플": 1627,
1632
+ "픔": 1628,
1633
+ "픕": 1629,
1634
+ "픗": 1630,
1635
+ "피": 1631,
1636
+ "픽": 1632,
1637
+ "핀": 1633,
1638
+ "필": 1634,
1639
+ "핌": 1635,
1640
+ "핍": 1636,
1641
+ "핏": 1637,
1642
+ "핑": 1638,
1643
+ "하": 1639,
1644
+ "학": 1640,
1645
+ "한": 1641,
1646
+ "할": 1642,
1647
+ "핥": 1643,
1648
+ "함": 1644,
1649
+ "합": 1645,
1650
+ "핫": 1646,
1651
+ "핬": 1647,
1652
+ "항": 1648,
1653
+ "해": 1649,
1654
+ "핵": 1650,
1655
+ "핸": 1651,
1656
+ "핼": 1652,
1657
+ "햄": 1653,
1658
+ "햅": 1654,
1659
+ "햇": 1655,
1660
+ "했": 1656,
1661
+ "행": 1657,
1662
+ "햐": 1658,
1663
+ "향": 1659,
1664
+ "허": 1660,
1665
+ "헉": 1661,
1666
+ "헌": 1662,
1667
+ "헐": 1663,
1668
+ "험": 1664,
1669
+ "헙": 1665,
1670
+ "헛": 1666,
1671
+ "헝": 1667,
1672
+ "헤": 1668,
1673
+ "헥": 1669,
1674
+ "헨": 1670,
1675
+ "헬": 1671,
1676
+ "헵": 1672,
1677
+ "헷": 1673,
1678
+ "헸": 1674,
1679
+ "헹": 1675,
1680
+ "혀": 1676,
1681
+ "혁": 1677,
1682
+ "현": 1678,
1683
+ "혈": 1679,
1684
+ "혐": 1680,
1685
+ "협": 1681,
1686
+ "혓": 1682,
1687
+ "혔": 1683,
1688
+ "형": 1684,
1689
+ "혜": 1685,
1690
+ "호": 1686,
1691
+ "혹": 1687,
1692
+ "혼": 1688,
1693
+ "홀": 1689,
1694
+ "홈": 1690,
1695
+ "홉": 1691,
1696
+ "홋": 1692,
1697
+ "홍": 1693,
1698
+ "화": 1694,
1699
+ "확": 1695,
1700
+ "환": 1696,
1701
+ "활": 1697,
1702
+ "황": 1698,
1703
+ "횃": 1699,
1704
+ "회": 1700,
1705
+ "획": 1701,
1706
+ "횝": 1702,
1707
+ "횟": 1703,
1708
+ "횡": 1704,
1709
+ "효": 1705,
1710
+ "후": 1706,
1711
+ "훅": 1707,
1712
+ "훈": 1708,
1713
+ "훌": 1709,
1714
+ "훑": 1710,
1715
+ "훔": 1711,
1716
+ "훤": 1712,
1717
+ "훨": 1713,
1718
+ "훼": 1714,
1719
+ "휑": 1715,
1720
+ "휘": 1716,
1721
+ "휙": 1717,
1722
+ "휜": 1718,
1723
+ "휠": 1719,
1724
+ "휩": 1720,
1725
+ "휴": 1721,
1726
+ "흉": 1722,
1727
+ "흐": 1723,
1728
+ "흑": 1724,
1729
+ "흔": 1725,
1730
+ "흘": 1726,
1731
+ "흙": 1727,
1732
+ "흠": 1728,
1733
+ "흡": 1729,
1734
+ "흥": 1730,
1735
+ "흩": 1731,
1736
+ "희": 1732,
1737
+ "흰": 1733,
1738
+ "히": 1734,
1739
+ "힌": 1735,
1740
+ "힐": 1736,
1741
+ "힘": 1737,
1742
+ "힙": 1738,
1743
+ "힝": 1739
1744
+ }
checkpoint-24000/added_tokens.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 1743,
3
+ "<s>": 1742,
4
+ "[PAD]": 1741,
5
+ "[UNK]": 1740
6
+ }
checkpoint-24000/config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_attn_dim": null,
5
+ "adapter_kernel_size": 3,
6
+ "adapter_stride": 2,
7
+ "add_adapter": false,
8
+ "apply_spec_augment": true,
9
+ "architectures": [
10
+ "Wav2Vec2ForCTC"
11
+ ],
12
+ "attention_dropout": 0.1,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_dim": [
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512
26
+ ],
27
+ "conv_kernel": [
28
+ 10,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 2,
34
+ 2
35
+ ],
36
+ "conv_stride": [
37
+ 5,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2
44
+ ],
45
+ "ctc_loss_reduction": "mean",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": true,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_dropout": 0.0,
52
+ "feat_extract_norm": "layer",
53
+ "feat_proj_dropout": 0.05,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.05,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.05,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.05,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "num_adapter_layers": 3,
80
+ "num_attention_heads": 16,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_feat_extract_layers": 7,
86
+ "num_hidden_layers": 24,
87
+ "num_negatives": 100,
88
+ "output_hidden_size": 1024,
89
+ "pad_token_id": 1741,
90
+ "proj_codevector_dim": 768,
91
+ "tdnn_dilation": [
92
+ 1,
93
+ 2,
94
+ 3,
95
+ 1,
96
+ 1
97
+ ],
98
+ "tdnn_dim": [
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 512,
103
+ 1500
104
+ ],
105
+ "tdnn_kernel": [
106
+ 5,
107
+ 3,
108
+ 3,
109
+ 1,
110
+ 1
111
+ ],
112
+ "torch_dtype": "float32",
113
+ "transformers_version": "4.34.0",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 1744,
116
+ "xvector_output_dim": 512
117
+ }
checkpoint-24000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:756b003e4f65921b07ae78325ce08c74266a23e234cc2b7713dda58240611c26
3
+ size 2504460598
checkpoint-24000/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
checkpoint-24000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87d9df2753130a58027ac5e2e647da14f39930736a484cb24abc6260772151d1
3
+ size 1269052778
checkpoint-24000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00fbe61cdb90aa1d37ebc1c2073166464bd3daa8ce51ed87283325e9a89a0661
3
+ size 15024
checkpoint-24000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f9261332cc258dc752bf8e69305493715c37b4a9987f40d34664dc938338648
3
+ size 15024
checkpoint-24000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18e0f490263da9fc10c021fa57e33acf59c8a510565abdbe5b5fff69c81038a1
3
+ size 15088
checkpoint-24000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6657a4c685d306b58f7ef32a42c1530b2923b179c7e575999b358923c913f673
3
+ size 14960
checkpoint-24000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4a978ff70b57f1d4e9e9efe608034b2f0bc18729f1cd9d5bddb316e2e4d984d
3
+ size 1064
checkpoint-24000/special_tokens_map.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>",
4
+ "</s>"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "eos_token": "</s>",
8
+ "pad_token": "[PAD]",
9
+ "unk_token": "[UNK]"
10
+ }
checkpoint-24000/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "1740": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1741": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "1742": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "1743": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "additional_special_tokens": [
37
+ "<s>",
38
+ "</s>"
39
+ ],
40
+ "bos_token": "<s>",
41
+ "clean_up_tokenization_spaces": true,
42
+ "config": null,
43
+ "do_lower_case": false,
44
+ "eos_token": "</s>",
45
+ "model_max_length": 1000000000000000019884624838656,
46
+ "pad_token": "[PAD]",
47
+ "processor_class": "Wav2Vec2Processor",
48
+ "replace_word_delimiter_char": " ",
49
+ "target_lang": null,
50
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
51
+ "tokenizer_file": null,
52
+ "tokenizer_type": "wav2vec2",
53
+ "trust_remote_code": false,
54
+ "unk_token": "[UNK]",
55
+ "word_delimiter_token": "|"
56
+ }
checkpoint-24000/trainer_state.json ADDED
@@ -0,0 +1,847 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 14.690130068859984,
5
+ "eval_steps": 2000,
6
+ "global_step": 24000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12,
13
+ "learning_rate": 3.9999999999999996e-05,
14
+ "loss": 33.0181,
15
+ "step": 200
16
+ },
17
+ {
18
+ "epoch": 0.24,
19
+ "learning_rate": 7.999999999999999e-05,
20
+ "loss": 8.7646,
21
+ "step": 400
22
+ },
23
+ {
24
+ "epoch": 0.37,
25
+ "learning_rate": 0.00011999999999999999,
26
+ "loss": 4.6933,
27
+ "step": 600
28
+ },
29
+ {
30
+ "epoch": 0.49,
31
+ "learning_rate": 0.00015999999999999999,
32
+ "loss": 4.5796,
33
+ "step": 800
34
+ },
35
+ {
36
+ "epoch": 0.61,
37
+ "learning_rate": 0.00019999999999999998,
38
+ "loss": 4.1047,
39
+ "step": 1000
40
+ },
41
+ {
42
+ "epoch": 0.73,
43
+ "learning_rate": 0.00023999999999999998,
44
+ "loss": 2.6327,
45
+ "step": 1200
46
+ },
47
+ {
48
+ "epoch": 0.86,
49
+ "learning_rate": 0.00028,
50
+ "loss": 1.9893,
51
+ "step": 1400
52
+ },
53
+ {
54
+ "epoch": 0.98,
55
+ "learning_rate": 0.00029869536855838224,
56
+ "loss": 1.7241,
57
+ "step": 1600
58
+ },
59
+ {
60
+ "epoch": 1.1,
61
+ "learning_rate": 0.00029608610567514673,
62
+ "loss": 1.5607,
63
+ "step": 1800
64
+ },
65
+ {
66
+ "epoch": 1.22,
67
+ "learning_rate": 0.00029347684279191127,
68
+ "loss": 1.4503,
69
+ "step": 2000
70
+ },
71
+ {
72
+ "epoch": 1.22,
73
+ "eval_cer": 0.268688843618788,
74
+ "eval_loss": 1.0610458850860596,
75
+ "eval_runtime": 1586.9419,
76
+ "eval_samples_per_second": 8.25,
77
+ "eval_steps_per_second": 0.516,
78
+ "step": 2000
79
+ },
80
+ {
81
+ "epoch": 1.35,
82
+ "learning_rate": 0.00029086757990867576,
83
+ "loss": 1.3782,
84
+ "step": 2200
85
+ },
86
+ {
87
+ "epoch": 1.47,
88
+ "learning_rate": 0.0002882583170254403,
89
+ "loss": 1.3244,
90
+ "step": 2400
91
+ },
92
+ {
93
+ "epoch": 1.59,
94
+ "learning_rate": 0.0002856490541422048,
95
+ "loss": 1.2553,
96
+ "step": 2600
97
+ },
98
+ {
99
+ "epoch": 1.71,
100
+ "learning_rate": 0.00028303979125896933,
101
+ "loss": 1.2077,
102
+ "step": 2800
103
+ },
104
+ {
105
+ "epoch": 1.84,
106
+ "learning_rate": 0.0002804305283757338,
107
+ "loss": 1.1868,
108
+ "step": 3000
109
+ },
110
+ {
111
+ "epoch": 1.96,
112
+ "learning_rate": 0.00027782126549249836,
113
+ "loss": 1.1624,
114
+ "step": 3200
115
+ },
116
+ {
117
+ "epoch": 2.08,
118
+ "learning_rate": 0.00027521200260926284,
119
+ "loss": 1.1136,
120
+ "step": 3400
121
+ },
122
+ {
123
+ "epoch": 2.2,
124
+ "learning_rate": 0.0002726027397260274,
125
+ "loss": 1.077,
126
+ "step": 3600
127
+ },
128
+ {
129
+ "epoch": 2.33,
130
+ "learning_rate": 0.00026999347684279187,
131
+ "loss": 1.053,
132
+ "step": 3800
133
+ },
134
+ {
135
+ "epoch": 2.45,
136
+ "learning_rate": 0.0002673842139595564,
137
+ "loss": 1.0239,
138
+ "step": 4000
139
+ },
140
+ {
141
+ "epoch": 2.45,
142
+ "eval_cer": 0.19039349567825709,
143
+ "eval_loss": 0.6961866617202759,
144
+ "eval_runtime": 1607.479,
145
+ "eval_samples_per_second": 8.145,
146
+ "eval_steps_per_second": 0.509,
147
+ "step": 4000
148
+ },
149
+ {
150
+ "epoch": 2.57,
151
+ "learning_rate": 0.0002647749510763209,
152
+ "loss": 1.0206,
153
+ "step": 4200
154
+ },
155
+ {
156
+ "epoch": 2.69,
157
+ "learning_rate": 0.00026216568819308544,
158
+ "loss": 1.0045,
159
+ "step": 4400
160
+ },
161
+ {
162
+ "epoch": 2.82,
163
+ "learning_rate": 0.0002595564253098499,
164
+ "loss": 0.9802,
165
+ "step": 4600
166
+ },
167
+ {
168
+ "epoch": 2.94,
169
+ "learning_rate": 0.00025694716242661447,
170
+ "loss": 0.9839,
171
+ "step": 4800
172
+ },
173
+ {
174
+ "epoch": 3.06,
175
+ "learning_rate": 0.00025433789954337895,
176
+ "loss": 0.9243,
177
+ "step": 5000
178
+ },
179
+ {
180
+ "epoch": 3.18,
181
+ "learning_rate": 0.0002517286366601435,
182
+ "loss": 0.9082,
183
+ "step": 5200
184
+ },
185
+ {
186
+ "epoch": 3.31,
187
+ "learning_rate": 0.000249119373776908,
188
+ "loss": 0.9017,
189
+ "step": 5400
190
+ },
191
+ {
192
+ "epoch": 3.43,
193
+ "learning_rate": 0.0002465101108936725,
194
+ "loss": 0.8871,
195
+ "step": 5600
196
+ },
197
+ {
198
+ "epoch": 3.55,
199
+ "learning_rate": 0.00024390084801043704,
200
+ "loss": 0.9036,
201
+ "step": 5800
202
+ },
203
+ {
204
+ "epoch": 3.67,
205
+ "learning_rate": 0.00024129158512720155,
206
+ "loss": 0.8977,
207
+ "step": 6000
208
+ },
209
+ {
210
+ "epoch": 3.67,
211
+ "eval_cer": 0.16872254319465907,
212
+ "eval_loss": 0.594495415687561,
213
+ "eval_runtime": 1591.6049,
214
+ "eval_samples_per_second": 8.226,
215
+ "eval_steps_per_second": 0.515,
216
+ "step": 6000
217
+ },
218
+ {
219
+ "epoch": 3.79,
220
+ "learning_rate": 0.00023868232224396607,
221
+ "loss": 0.8843,
222
+ "step": 6200
223
+ },
224
+ {
225
+ "epoch": 3.92,
226
+ "learning_rate": 0.00023607305936073058,
227
+ "loss": 0.8757,
228
+ "step": 6400
229
+ },
230
+ {
231
+ "epoch": 4.04,
232
+ "learning_rate": 0.0002334637964774951,
233
+ "loss": 0.8573,
234
+ "step": 6600
235
+ },
236
+ {
237
+ "epoch": 4.16,
238
+ "learning_rate": 0.0002308545335942596,
239
+ "loss": 0.8126,
240
+ "step": 6800
241
+ },
242
+ {
243
+ "epoch": 4.28,
244
+ "learning_rate": 0.00022824527071102412,
245
+ "loss": 0.8192,
246
+ "step": 7000
247
+ },
248
+ {
249
+ "epoch": 4.41,
250
+ "learning_rate": 0.00022563600782778863,
251
+ "loss": 0.8061,
252
+ "step": 7200
253
+ },
254
+ {
255
+ "epoch": 4.53,
256
+ "learning_rate": 0.00022302674494455315,
257
+ "loss": 0.8123,
258
+ "step": 7400
259
+ },
260
+ {
261
+ "epoch": 4.65,
262
+ "learning_rate": 0.00022041748206131766,
263
+ "loss": 0.8046,
264
+ "step": 7600
265
+ },
266
+ {
267
+ "epoch": 4.77,
268
+ "learning_rate": 0.00021780821917808218,
269
+ "loss": 0.7979,
270
+ "step": 7800
271
+ },
272
+ {
273
+ "epoch": 4.9,
274
+ "learning_rate": 0.0002151989562948467,
275
+ "loss": 0.804,
276
+ "step": 8000
277
+ },
278
+ {
279
+ "epoch": 4.9,
280
+ "eval_cer": 0.14924661713942214,
281
+ "eval_loss": 0.5327703952789307,
282
+ "eval_runtime": 1595.6324,
283
+ "eval_samples_per_second": 8.206,
284
+ "eval_steps_per_second": 0.513,
285
+ "step": 8000
286
+ },
287
+ {
288
+ "epoch": 5.02,
289
+ "learning_rate": 0.0002125896934116112,
290
+ "loss": 0.7867,
291
+ "step": 8200
292
+ },
293
+ {
294
+ "epoch": 5.14,
295
+ "learning_rate": 0.00020998043052837572,
296
+ "loss": 0.7557,
297
+ "step": 8400
298
+ },
299
+ {
300
+ "epoch": 5.26,
301
+ "learning_rate": 0.00020737116764514023,
302
+ "loss": 0.7478,
303
+ "step": 8600
304
+ },
305
+ {
306
+ "epoch": 5.39,
307
+ "learning_rate": 0.00020476190476190475,
308
+ "loss": 0.7398,
309
+ "step": 8800
310
+ },
311
+ {
312
+ "epoch": 5.51,
313
+ "learning_rate": 0.00020215264187866926,
314
+ "loss": 0.7408,
315
+ "step": 9000
316
+ },
317
+ {
318
+ "epoch": 5.63,
319
+ "learning_rate": 0.00019954337899543377,
320
+ "loss": 0.75,
321
+ "step": 9200
322
+ },
323
+ {
324
+ "epoch": 5.75,
325
+ "learning_rate": 0.0001969341161121983,
326
+ "loss": 0.7344,
327
+ "step": 9400
328
+ },
329
+ {
330
+ "epoch": 5.88,
331
+ "learning_rate": 0.0001943248532289628,
332
+ "loss": 0.738,
333
+ "step": 9600
334
+ },
335
+ {
336
+ "epoch": 6.0,
337
+ "learning_rate": 0.00019171559034572732,
338
+ "loss": 0.7373,
339
+ "step": 9800
340
+ },
341
+ {
342
+ "epoch": 6.12,
343
+ "learning_rate": 0.00018910632746249183,
344
+ "loss": 0.698,
345
+ "step": 10000
346
+ },
347
+ {
348
+ "epoch": 6.12,
349
+ "eval_cer": 0.13653489424101573,
350
+ "eval_loss": 0.5013594031333923,
351
+ "eval_runtime": 1605.1161,
352
+ "eval_samples_per_second": 8.157,
353
+ "eval_steps_per_second": 0.51,
354
+ "step": 10000
355
+ },
356
+ {
357
+ "epoch": 6.24,
358
+ "learning_rate": 0.00018649706457925634,
359
+ "loss": 0.6943,
360
+ "step": 10200
361
+ },
362
+ {
363
+ "epoch": 6.37,
364
+ "learning_rate": 0.00018388780169602086,
365
+ "loss": 0.6997,
366
+ "step": 10400
367
+ },
368
+ {
369
+ "epoch": 6.49,
370
+ "learning_rate": 0.00018127853881278537,
371
+ "loss": 0.6929,
372
+ "step": 10600
373
+ },
374
+ {
375
+ "epoch": 6.61,
376
+ "learning_rate": 0.00017866927592954989,
377
+ "loss": 0.7003,
378
+ "step": 10800
379
+ },
380
+ {
381
+ "epoch": 6.73,
382
+ "learning_rate": 0.0001760600130463144,
383
+ "loss": 0.6863,
384
+ "step": 11000
385
+ },
386
+ {
387
+ "epoch": 6.86,
388
+ "learning_rate": 0.00017345075016307891,
389
+ "loss": 0.6883,
390
+ "step": 11200
391
+ },
392
+ {
393
+ "epoch": 6.98,
394
+ "learning_rate": 0.00017084148727984343,
395
+ "loss": 0.6787,
396
+ "step": 11400
397
+ },
398
+ {
399
+ "epoch": 7.1,
400
+ "learning_rate": 0.00016823222439660794,
401
+ "loss": 0.6518,
402
+ "step": 11600
403
+ },
404
+ {
405
+ "epoch": 7.22,
406
+ "learning_rate": 0.00016562296151337246,
407
+ "loss": 0.6494,
408
+ "step": 11800
409
+ },
410
+ {
411
+ "epoch": 7.35,
412
+ "learning_rate": 0.00016301369863013697,
413
+ "loss": 0.6426,
414
+ "step": 12000
415
+ },
416
+ {
417
+ "epoch": 7.35,
418
+ "eval_cer": 0.13216305737125092,
419
+ "eval_loss": 0.47150149941444397,
420
+ "eval_runtime": 1597.7342,
421
+ "eval_samples_per_second": 8.195,
422
+ "eval_steps_per_second": 0.513,
423
+ "step": 12000
424
+ },
425
+ {
426
+ "epoch": 7.47,
427
+ "learning_rate": 0.00016040443574690148,
428
+ "loss": 0.6457,
429
+ "step": 12200
430
+ },
431
+ {
432
+ "epoch": 7.59,
433
+ "learning_rate": 0.000157795172863666,
434
+ "loss": 0.6429,
435
+ "step": 12400
436
+ },
437
+ {
438
+ "epoch": 7.71,
439
+ "learning_rate": 0.0001551859099804305,
440
+ "loss": 0.6512,
441
+ "step": 12600
442
+ },
443
+ {
444
+ "epoch": 7.83,
445
+ "learning_rate": 0.00015257664709719503,
446
+ "loss": 0.6458,
447
+ "step": 12800
448
+ },
449
+ {
450
+ "epoch": 7.96,
451
+ "learning_rate": 0.00014996738421395954,
452
+ "loss": 0.6501,
453
+ "step": 13000
454
+ },
455
+ {
456
+ "epoch": 8.08,
457
+ "learning_rate": 0.00014735812133072405,
458
+ "loss": 0.6095,
459
+ "step": 13200
460
+ },
461
+ {
462
+ "epoch": 8.2,
463
+ "learning_rate": 0.00014474885844748857,
464
+ "loss": 0.6132,
465
+ "step": 13400
466
+ },
467
+ {
468
+ "epoch": 8.32,
469
+ "learning_rate": 0.00014213959556425308,
470
+ "loss": 0.6102,
471
+ "step": 13600
472
+ },
473
+ {
474
+ "epoch": 8.45,
475
+ "learning_rate": 0.0001395303326810176,
476
+ "loss": 0.6125,
477
+ "step": 13800
478
+ },
479
+ {
480
+ "epoch": 8.57,
481
+ "learning_rate": 0.0001369210697977821,
482
+ "loss": 0.61,
483
+ "step": 14000
484
+ },
485
+ {
486
+ "epoch": 8.57,
487
+ "eval_cer": 0.1257692459492199,
488
+ "eval_loss": 0.45295360684394836,
489
+ "eval_runtime": 1603.0798,
490
+ "eval_samples_per_second": 8.167,
491
+ "eval_steps_per_second": 0.511,
492
+ "step": 14000
493
+ },
494
+ {
495
+ "epoch": 8.69,
496
+ "learning_rate": 0.00013431180691454662,
497
+ "loss": 0.606,
498
+ "step": 14200
499
+ },
500
+ {
501
+ "epoch": 8.81,
502
+ "learning_rate": 0.00013170254403131114,
503
+ "loss": 0.5957,
504
+ "step": 14400
505
+ },
506
+ {
507
+ "epoch": 8.94,
508
+ "learning_rate": 0.00012909328114807565,
509
+ "loss": 0.5992,
510
+ "step": 14600
511
+ },
512
+ {
513
+ "epoch": 9.06,
514
+ "learning_rate": 0.00012648401826484017,
515
+ "loss": 0.5752,
516
+ "step": 14800
517
+ },
518
+ {
519
+ "epoch": 9.18,
520
+ "learning_rate": 0.00012387475538160468,
521
+ "loss": 0.5654,
522
+ "step": 15000
523
+ },
524
+ {
525
+ "epoch": 9.3,
526
+ "learning_rate": 0.00012126549249836919,
527
+ "loss": 0.5725,
528
+ "step": 15200
529
+ },
530
+ {
531
+ "epoch": 9.43,
532
+ "learning_rate": 0.00011865622961513371,
533
+ "loss": 0.5713,
534
+ "step": 15400
535
+ },
536
+ {
537
+ "epoch": 9.55,
538
+ "learning_rate": 0.00011604696673189822,
539
+ "loss": 0.5649,
540
+ "step": 15600
541
+ },
542
+ {
543
+ "epoch": 9.67,
544
+ "learning_rate": 0.00011343770384866273,
545
+ "loss": 0.5643,
546
+ "step": 15800
547
+ },
548
+ {
549
+ "epoch": 9.79,
550
+ "learning_rate": 0.00011082844096542725,
551
+ "loss": 0.5709,
552
+ "step": 16000
553
+ },
554
+ {
555
+ "epoch": 9.79,
556
+ "eval_cer": 0.1200554980402634,
557
+ "eval_loss": 0.4299587607383728,
558
+ "eval_runtime": 1609.5227,
559
+ "eval_samples_per_second": 8.135,
560
+ "eval_steps_per_second": 0.509,
561
+ "step": 16000
562
+ },
563
+ {
564
+ "epoch": 9.92,
565
+ "learning_rate": 0.00010821917808219176,
566
+ "loss": 0.5666,
567
+ "step": 16200
568
+ },
569
+ {
570
+ "epoch": 10.04,
571
+ "learning_rate": 0.00010560991519895628,
572
+ "loss": 0.5531,
573
+ "step": 16400
574
+ },
575
+ {
576
+ "epoch": 10.16,
577
+ "learning_rate": 0.00010300065231572079,
578
+ "loss": 0.5389,
579
+ "step": 16600
580
+ },
581
+ {
582
+ "epoch": 10.28,
583
+ "learning_rate": 0.0001003913894324853,
584
+ "loss": 0.5456,
585
+ "step": 16800
586
+ },
587
+ {
588
+ "epoch": 10.41,
589
+ "learning_rate": 9.778212654924982e-05,
590
+ "loss": 0.5353,
591
+ "step": 17000
592
+ },
593
+ {
594
+ "epoch": 10.53,
595
+ "learning_rate": 9.517286366601433e-05,
596
+ "loss": 0.5337,
597
+ "step": 17200
598
+ },
599
+ {
600
+ "epoch": 10.65,
601
+ "learning_rate": 9.256360078277885e-05,
602
+ "loss": 0.5296,
603
+ "step": 17400
604
+ },
605
+ {
606
+ "epoch": 10.77,
607
+ "learning_rate": 8.995433789954336e-05,
608
+ "loss": 0.5372,
609
+ "step": 17600
610
+ },
611
+ {
612
+ "epoch": 10.9,
613
+ "learning_rate": 8.734507501630787e-05,
614
+ "loss": 0.5388,
615
+ "step": 17800
616
+ },
617
+ {
618
+ "epoch": 11.02,
619
+ "learning_rate": 8.473581213307239e-05,
620
+ "loss": 0.5235,
621
+ "step": 18000
622
+ },
623
+ {
624
+ "epoch": 11.02,
625
+ "eval_cer": 0.11664607248141211,
626
+ "eval_loss": 0.4167773723602295,
627
+ "eval_runtime": 1608.2913,
628
+ "eval_samples_per_second": 8.141,
629
+ "eval_steps_per_second": 0.509,
630
+ "step": 18000
631
+ },
632
+ {
633
+ "epoch": 11.14,
634
+ "learning_rate": 8.212654924983692e-05,
635
+ "loss": 0.509,
636
+ "step": 18200
637
+ },
638
+ {
639
+ "epoch": 11.26,
640
+ "learning_rate": 7.951728636660143e-05,
641
+ "loss": 0.5116,
642
+ "step": 18400
643
+ },
644
+ {
645
+ "epoch": 11.38,
646
+ "learning_rate": 7.690802348336594e-05,
647
+ "loss": 0.4967,
648
+ "step": 18600
649
+ },
650
+ {
651
+ "epoch": 11.51,
652
+ "learning_rate": 7.429876060013046e-05,
653
+ "loss": 0.511,
654
+ "step": 18800
655
+ },
656
+ {
657
+ "epoch": 11.63,
658
+ "learning_rate": 7.168949771689497e-05,
659
+ "loss": 0.5056,
660
+ "step": 19000
661
+ },
662
+ {
663
+ "epoch": 11.75,
664
+ "learning_rate": 6.908023483365949e-05,
665
+ "loss": 0.5073,
666
+ "step": 19200
667
+ },
668
+ {
669
+ "epoch": 11.87,
670
+ "learning_rate": 6.6470971950424e-05,
671
+ "loss": 0.4968,
672
+ "step": 19400
673
+ },
674
+ {
675
+ "epoch": 12.0,
676
+ "learning_rate": 6.386170906718851e-05,
677
+ "loss": 0.5009,
678
+ "step": 19600
679
+ },
680
+ {
681
+ "epoch": 12.12,
682
+ "learning_rate": 6.125244618395303e-05,
683
+ "loss": 0.4832,
684
+ "step": 19800
685
+ },
686
+ {
687
+ "epoch": 12.24,
688
+ "learning_rate": 5.864318330071754e-05,
689
+ "loss": 0.4778,
690
+ "step": 20000
691
+ },
692
+ {
693
+ "epoch": 12.24,
694
+ "eval_cer": 0.11294822712906938,
695
+ "eval_loss": 0.40570223331451416,
696
+ "eval_runtime": 1612.661,
697
+ "eval_samples_per_second": 8.119,
698
+ "eval_steps_per_second": 0.508,
699
+ "step": 20000
700
+ },
701
+ {
702
+ "epoch": 12.36,
703
+ "learning_rate": 5.6033920417482055e-05,
704
+ "loss": 0.4775,
705
+ "step": 20200
706
+ },
707
+ {
708
+ "epoch": 12.49,
709
+ "learning_rate": 5.342465753424657e-05,
710
+ "loss": 0.4855,
711
+ "step": 20400
712
+ },
713
+ {
714
+ "epoch": 12.61,
715
+ "learning_rate": 5.081539465101108e-05,
716
+ "loss": 0.4773,
717
+ "step": 20600
718
+ },
719
+ {
720
+ "epoch": 12.73,
721
+ "learning_rate": 4.82061317677756e-05,
722
+ "loss": 0.4745,
723
+ "step": 20800
724
+ },
725
+ {
726
+ "epoch": 12.85,
727
+ "learning_rate": 4.559686888454011e-05,
728
+ "loss": 0.48,
729
+ "step": 21000
730
+ },
731
+ {
732
+ "epoch": 12.98,
733
+ "learning_rate": 4.2987606001304625e-05,
734
+ "loss": 0.463,
735
+ "step": 21200
736
+ },
737
+ {
738
+ "epoch": 13.1,
739
+ "learning_rate": 4.037834311806914e-05,
740
+ "loss": 0.4643,
741
+ "step": 21400
742
+ },
743
+ {
744
+ "epoch": 13.22,
745
+ "learning_rate": 3.776908023483365e-05,
746
+ "loss": 0.449,
747
+ "step": 21600
748
+ },
749
+ {
750
+ "epoch": 13.34,
751
+ "learning_rate": 3.515981735159817e-05,
752
+ "loss": 0.4604,
753
+ "step": 21800
754
+ },
755
+ {
756
+ "epoch": 13.47,
757
+ "learning_rate": 3.255055446836268e-05,
758
+ "loss": 0.4571,
759
+ "step": 22000
760
+ },
761
+ {
762
+ "epoch": 13.47,
763
+ "eval_cer": 0.10995473327241098,
764
+ "eval_loss": 0.3945465385913849,
765
+ "eval_runtime": 1610.0901,
766
+ "eval_samples_per_second": 8.132,
767
+ "eval_steps_per_second": 0.509,
768
+ "step": 22000
769
+ },
770
+ {
771
+ "epoch": 13.59,
772
+ "learning_rate": 2.99412915851272e-05,
773
+ "loss": 0.4539,
774
+ "step": 22200
775
+ },
776
+ {
777
+ "epoch": 13.71,
778
+ "learning_rate": 2.7332028701891712e-05,
779
+ "loss": 0.4569,
780
+ "step": 22400
781
+ },
782
+ {
783
+ "epoch": 13.83,
784
+ "learning_rate": 2.4722765818656226e-05,
785
+ "loss": 0.4533,
786
+ "step": 22600
787
+ },
788
+ {
789
+ "epoch": 13.96,
790
+ "learning_rate": 2.211350293542074e-05,
791
+ "loss": 0.4545,
792
+ "step": 22800
793
+ },
794
+ {
795
+ "epoch": 14.08,
796
+ "learning_rate": 1.9504240052185254e-05,
797
+ "loss": 0.447,
798
+ "step": 23000
799
+ },
800
+ {
801
+ "epoch": 14.2,
802
+ "learning_rate": 1.6894977168949768e-05,
803
+ "loss": 0.4443,
804
+ "step": 23200
805
+ },
806
+ {
807
+ "epoch": 14.32,
808
+ "learning_rate": 1.4285714285714284e-05,
809
+ "loss": 0.4423,
810
+ "step": 23400
811
+ },
812
+ {
813
+ "epoch": 14.45,
814
+ "learning_rate": 1.1676451402478798e-05,
815
+ "loss": 0.4434,
816
+ "step": 23600
817
+ },
818
+ {
819
+ "epoch": 14.57,
820
+ "learning_rate": 9.067188519243312e-06,
821
+ "loss": 0.438,
822
+ "step": 23800
823
+ },
824
+ {
825
+ "epoch": 14.69,
826
+ "learning_rate": 6.4579256360078264e-06,
827
+ "loss": 0.4388,
828
+ "step": 24000
829
+ },
830
+ {
831
+ "epoch": 14.69,
832
+ "eval_cer": 0.10809973860058716,
833
+ "eval_loss": 0.38906005024909973,
834
+ "eval_runtime": 1607.4288,
835
+ "eval_samples_per_second": 8.145,
836
+ "eval_steps_per_second": 0.51,
837
+ "step": 24000
838
+ }
839
+ ],
840
+ "logging_steps": 200,
841
+ "max_steps": 24495,
842
+ "num_train_epochs": 15,
843
+ "save_steps": 2000,
844
+ "total_flos": 1.6873620478584737e+20,
845
+ "trial_name": null,
846
+ "trial_params": null
847
+ }
checkpoint-24000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2a53791a3905b5426f4f4a28dc53b7bf0cdde7382afe6d18a7f69f8c493b179
3
+ size 4472
checkpoint-24000/vocab.json ADDED
@@ -0,0 +1,1744 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "&": 1,
3
+ "0": 2,
4
+ "1": 3,
5
+ "2": 4,
6
+ "3": 5,
7
+ "4": 6,
8
+ "5": 7,
9
+ "6": 8,
10
+ "7": 9,
11
+ "8": 10,
12
+ "9": 11,
13
+ "[PAD]": 1741,
14
+ "[UNK]": 1740,
15
+ "\\": 12,
16
+ "a": 13,
17
+ "b": 14,
18
+ "c": 15,
19
+ "d": 16,
20
+ "e": 17,
21
+ "f": 18,
22
+ "g": 19,
23
+ "h": 20,
24
+ "i": 21,
25
+ "j": 22,
26
+ "k": 23,
27
+ "l": 24,
28
+ "m": 25,
29
+ "n": 26,
30
+ "o": 27,
31
+ "p": 28,
32
+ "q": 29,
33
+ "r": 30,
34
+ "s": 31,
35
+ "t": 32,
36
+ "u": 33,
37
+ "v": 34,
38
+ "w": 35,
39
+ "x": 36,
40
+ "y": 37,
41
+ "z": 38,
42
+ "|": 0,
43
+ "°": 39,
44
+ "μ": 40,
45
+ "ᆫ": 41,
46
+ "ㄱ": 42,
47
+ "ㄴ": 43,
48
+ "ㄷ": 44,
49
+ "ㄹ": 45,
50
+ "ㅁ": 46,
51
+ "ㅂ": 47,
52
+ "ㅅ": 48,
53
+ "ㅇ": 49,
54
+ "ㅈ": 50,
55
+ "ㅜ": 51,
56
+ "ㅠ": 52,
57
+ "ㅡ": 53,
58
+ "가": 54,
59
+ "각": 55,
60
+ "간": 56,
61
+ "갇": 57,
62
+ "갈": 58,
63
+ "감": 59,
64
+ "갑": 60,
65
+ "값": 61,
66
+ "갓": 62,
67
+ "갔": 63,
68
+ "강": 64,
69
+ "갖": 65,
70
+ "같": 66,
71
+ "갚": 67,
72
+ "갛": 68,
73
+ "개": 69,
74
+ "객": 70,
75
+ "갠": 71,
76
+ "갤": 72,
77
+ "갬": 73,
78
+ "갭": 74,
79
+ "갯": 75,
80
+ "갰": 76,
81
+ "갱": 77,
82
+ "갸": 78,
83
+ "걀": 79,
84
+ "걍": 80,
85
+ "걔": 81,
86
+ "걘": 82,
87
+ "거": 83,
88
+ "걱": 84,
89
+ "건": 85,
90
+ "걷": 86,
91
+ "걸": 87,
92
+ "검": 88,
93
+ "겁": 89,
94
+ "것": 90,
95
+ "겄": 91,
96
+ "겉": 92,
97
+ "게": 93,
98
+ "겐": 94,
99
+ "겔": 95,
100
+ "겜": 96,
101
+ "겟": 97,
102
+ "겠": 98,
103
+ "겨": 99,
104
+ "격": 100,
105
+ "겪": 101,
106
+ "견": 102,
107
+ "결": 103,
108
+ "겸": 104,
109
+ "겹": 105,
110
+ "겼": 106,
111
+ "경": 107,
112
+ "곁": 108,
113
+ "계": 109,
114
+ "곗": 110,
115
+ "고": 111,
116
+ "곡": 112,
117
+ "곤": 113,
118
+ "곧": 114,
119
+ "골": 115,
120
+ "곪": 116,
121
+ "곰": 117,
122
+ "곱": 118,
123
+ "곳": 119,
124
+ "공": 120,
125
+ "곶": 121,
126
+ "과": 122,
127
+ "곽": 123,
128
+ "관": 124,
129
+ "괄": 125,
130
+ "괍": 126,
131
+ "광": 127,
132
+ "괘": 128,
133
+ "괜": 129,
134
+ "괭": 130,
135
+ "괴": 131,
136
+ "굉": 132,
137
+ "교": 133,
138
+ "굣": 134,
139
+ "구": 135,
140
+ "국": 136,
141
+ "군": 137,
142
+ "굳": 138,
143
+ "굴": 139,
144
+ "굵": 140,
145
+ "굶": 141,
146
+ "굽": 142,
147
+ "굿": 143,
148
+ "궁": 144,
149
+ "궂": 145,
150
+ "궈": 146,
151
+ "권": 147,
152
+ "궐": 148,
153
+ "궜": 149,
154
+ "귀": 150,
155
+ "귄": 151,
156
+ "귈": 152,
157
+ "귓": 153,
158
+ "규": 154,
159
+ "균": 155,
160
+ "귤": 156,
161
+ "그": 157,
162
+ "극": 158,
163
+ "근": 159,
164
+ "귿": 160,
165
+ "글": 161,
166
+ "긁": 162,
167
+ "금": 163,
168
+ "급": 164,
169
+ "긋": 165,
170
+ "긍": 166,
171
+ "긑": 167,
172
+ "기": 168,
173
+ "긴": 169,
174
+ "길": 170,
175
+ "김": 171,
176
+ "깁": 172,
177
+ "깃": 173,
178
+ "깄": 174,
179
+ "깅": 175,
180
+ "깊": 176,
181
+ "까": 177,
182
+ "깍": 178,
183
+ "깎": 179,
184
+ "깐": 180,
185
+ "깔": 181,
186
+ "깜": 182,
187
+ "깝": 183,
188
+ "깟": 184,
189
+ "깠": 185,
190
+ "깡": 186,
191
+ "깥": 187,
192
+ "깨": 188,
193
+ "깬": 189,
194
+ "깰": 190,
195
+ "깻": 191,
196
+ "깼": 192,
197
+ "깽": 193,
198
+ "꺼": 194,
199
+ "꺽": 195,
200
+ "꺾": 196,
201
+ "껀": 197,
202
+ "껄": 198,
203
+ "껌": 199,
204
+ "껍": 200,
205
+ "껏": 201,
206
+ "껐": 202,
207
+ "껑": 203,
208
+ "께": 204,
209
+ "껩": 205,
210
+ "껴": 206,
211
+ "꼈": 207,
212
+ "꼐": 208,
213
+ "꼬": 209,
214
+ "꼭": 210,
215
+ "꼰": 211,
216
+ "꼴": 212,
217
+ "꼼": 213,
218
+ "꼽": 214,
219
+ "꽁": 215,
220
+ "꽂": 216,
221
+ "꽃": 217,
222
+ "꽈": 218,
223
+ "꽉": 219,
224
+ "꽌": 220,
225
+ "꽜": 221,
226
+ "꽝": 222,
227
+ "꽤": 223,
228
+ "꽥": 224,
229
+ "꾀": 225,
230
+ "꾸": 226,
231
+ "꾹": 227,
232
+ "꾼": 228,
233
+ "꿀": 229,
234
+ "꿇": 230,
235
+ "꿈": 231,
236
+ "꿉": 232,
237
+ "꿍": 233,
238
+ "꿔": 234,
239
+ "꿨": 235,
240
+ "꿰": 236,
241
+ "뀌": 237,
242
+ "뀐": 238,
243
+ "뀔": 239,
244
+ "뀝": 240,
245
+ "뀰": 241,
246
+ "끄": 242,
247
+ "끅": 243,
248
+ "끈": 244,
249
+ "끊": 245,
250
+ "끌": 246,
251
+ "끓": 247,
252
+ "끔": 248,
253
+ "끕": 249,
254
+ "끗": 250,
255
+ "끙": 251,
256
+ "끝": 252,
257
+ "끼": 253,
258
+ "끽": 254,
259
+ "낀": 255,
260
+ "낄": 256,
261
+ "낌": 257,
262
+ "낍": 258,
263
+ "낑": 259,
264
+ "나": 260,
265
+ "낙": 261,
266
+ "낚": 262,
267
+ "난": 263,
268
+ "날": 264,
269
+ "낡": 265,
270
+ "남": 266,
271
+ "납": 267,
272
+ "낫": 268,
273
+ "났": 269,
274
+ "낭": 270,
275
+ "낮": 271,
276
+ "낯": 272,
277
+ "낱": 273,
278
+ "낳": 274,
279
+ "내": 275,
280
+ "낵": 276,
281
+ "낸": 277,
282
+ "낼": 278,
283
+ "냄": 279,
284
+ "냅": 280,
285
+ "냇": 281,
286
+ "냈": 282,
287
+ "냉": 283,
288
+ "냐": 284,
289
+ "냑": 285,
290
+ "냠": 286,
291
+ "냥": 287,
292
+ "너": 288,
293
+ "넉": 289,
294
+ "넌": 290,
295
+ "널": 291,
296
+ "넓": 292,
297
+ "넘": 293,
298
+ "넛": 294,
299
+ "넣": 295,
300
+ "네": 296,
301
+ "넥": 297,
302
+ "넬": 298,
303
+ "넵": 299,
304
+ "넷": 300,
305
+ "넹": 301,
306
+ "녀": 302,
307
+ "녁": 303,
308
+ "년": 304,
309
+ "념": 305,
310
+ "녔": 306,
311
+ "녕": 307,
312
+ "녜": 308,
313
+ "노": 309,
314
+ "녹": 310,
315
+ "논": 311,
316
+ "놀": 312,
317
+ "놈": 313,
318
+ "놉": 314,
319
+ "농": 315,
320
+ "높": 316,
321
+ "놓": 317,
322
+ "놔": 318,
323
+ "놨": 319,
324
+ "뇌": 320,
325
+ "뇨": 321,
326
+ "누": 322,
327
+ "눅": 323,
328
+ "눈": 324,
329
+ "눌": 325,
330
+ "눔": 326,
331
+ "눕": 327,
332
+ "눗": 328,
333
+ "눠": 329,
334
+ "눴": 330,
335
+ "뉘": 331,
336
+ "뉜": 332,
337
+ "뉠": 333,
338
+ "뉩": 334,
339
+ "뉴": 335,
340
+ "늄": 336,
341
+ "느": 337,
342
+ "늑": 338,
343
+ "는": 339,
344
+ "늘": 340,
345
+ "늙": 341,
346
+ "늠": 342,
347
+ "늡": 343,
348
+ "능": 344,
349
+ "늦": 345,
350
+ "늬": 346,
351
+ "니": 347,
352
+ "닉": 348,
353
+ "닌": 349,
354
+ "닐": 350,
355
+ "님": 351,
356
+ "닙": 352,
357
+ "닛": 353,
358
+ "닝": 354,
359
+ "다": 355,
360
+ "닥": 356,
361
+ "닦": 357,
362
+ "단": 358,
363
+ "닫": 359,
364
+ "달": 360,
365
+ "닭": 361,
366
+ "닮": 362,
367
+ "닳": 363,
368
+ "담": 364,
369
+ "답": 365,
370
+ "닷": 366,
371
+ "당": 367,
372
+ "닿": 368,
373
+ "대": 369,
374
+ "댁": 370,
375
+ "댄": 371,
376
+ "댈": 372,
377
+ "댐": 373,
378
+ "댑": 374,
379
+ "댓": 375,
380
+ "댔": 376,
381
+ "댕": 377,
382
+ "댜": 378,
383
+ "더": 379,
384
+ "덕": 380,
385
+ "던": 381,
386
+ "덜": 382,
387
+ "덟": 383,
388
+ "덤": 384,
389
+ "덥": 385,
390
+ "덧": 386,
391
+ "덩": 387,
392
+ "덮": 388,
393
+ "데": 389,
394
+ "덱": 390,
395
+ "덴": 391,
396
+ "델": 392,
397
+ "뎅": 393,
398
+ "뎌": 394,
399
+ "뎠": 395,
400
+ "도": 396,
401
+ "독": 397,
402
+ "돈": 398,
403
+ "돋": 399,
404
+ "돌": 400,
405
+ "돔": 401,
406
+ "돕": 402,
407
+ "돗": 403,
408
+ "동": 404,
409
+ "돼": 405,
410
+ "됍": 406,
411
+ "됐": 407,
412
+ "되": 408,
413
+ "된": 409,
414
+ "될": 410,
415
+ "됨": 411,
416
+ "됩": 412,
417
+ "됬": 413,
418
+ "두": 414,
419
+ "둑": 415,
420
+ "둔": 416,
421
+ "둘": 417,
422
+ "둠": 418,
423
+ "둡": 419,
424
+ "둣": 420,
425
+ "둥": 421,
426
+ "둬": 422,
427
+ "뒀": 423,
428
+ "뒈": 424,
429
+ "뒤": 425,
430
+ "뒨": 426,
431
+ "뒷": 427,
432
+ "뒹": 428,
433
+ "듀": 429,
434
+ "드": 430,
435
+ "득": 431,
436
+ "든": 432,
437
+ "듣": 433,
438
+ "들": 434,
439
+ "듦": 435,
440
+ "듬": 436,
441
+ "듭": 437,
442
+ "듯": 438,
443
+ "등": 439,
444
+ "디": 440,
445
+ "딕": 441,
446
+ "딘": 442,
447
+ "딜": 443,
448
+ "딤": 444,
449
+ "딥": 445,
450
+ "딧": 446,
451
+ "딨": 447,
452
+ "딩": 448,
453
+ "딪": 449,
454
+ "따": 450,
455
+ "딱": 451,
456
+ "딴": 452,
457
+ "딸": 453,
458
+ "땀": 454,
459
+ "땁": 455,
460
+ "땃": 456,
461
+ "땄": 457,
462
+ "땅": 458,
463
+ "땋": 459,
464
+ "때": 460,
465
+ "땍": 461,
466
+ "땐": 462,
467
+ "땔": 463,
468
+ "땜": 464,
469
+ "땟": 465,
470
+ "땠": 466,
471
+ "땡": 467,
472
+ "떄": 468,
473
+ "떠": 469,
474
+ "떡": 470,
475
+ "떤": 471,
476
+ "떨": 472,
477
+ "떰": 473,
478
+ "떱": 474,
479
+ "떳": 475,
480
+ "떴": 476,
481
+ "떻": 477,
482
+ "떼": 478,
483
+ "떽": 479,
484
+ "뗀": 480,
485
+ "뗄": 481,
486
+ "뗌": 482,
487
+ "뗍": 483,
488
+ "뗐": 484,
489
+ "뗴": 485,
490
+ "또": 486,
491
+ "똑": 487,
492
+ "똘": 488,
493
+ "똣": 489,
494
+ "똥": 490,
495
+ "뚜": 491,
496
+ "뚝": 492,
497
+ "뚤": 493,
498
+ "뚫": 494,
499
+ "뚱": 495,
500
+ "뛌": 496,
501
+ "뛰": 497,
502
+ "뛴": 498,
503
+ "뛸": 499,
504
+ "뜄": 500,
505
+ "뜨": 501,
506
+ "뜩": 502,
507
+ "뜬": 503,
508
+ "뜯": 504,
509
+ "뜰": 505,
510
+ "뜸": 506,
511
+ "뜹": 507,
512
+ "뜻": 508,
513
+ "띄": 509,
514
+ "띈": 510,
515
+ "띌": 511,
516
+ "띔": 512,
517
+ "띕": 513,
518
+ "띠": 514,
519
+ "띡": 515,
520
+ "띨": 516,
521
+ "띵": 517,
522
+ "라": 518,
523
+ "락": 519,
524
+ "란": 520,
525
+ "랄": 521,
526
+ "람": 522,
527
+ "랍": 523,
528
+ "랏": 524,
529
+ "랐": 525,
530
+ "랑": 526,
531
+ "랖": 527,
532
+ "랗": 528,
533
+ "래": 529,
534
+ "랙": 530,
535
+ "랜": 531,
536
+ "랠": 532,
537
+ "램": 533,
538
+ "랩": 534,
539
+ "랫": 535,
540
+ "랬": 536,
541
+ "랭": 537,
542
+ "랴": 538,
543
+ "략": 539,
544
+ "량": 540,
545
+ "러": 541,
546
+ "럭": 542,
547
+ "런": 543,
548
+ "럴": 544,
549
+ "럼": 545,
550
+ "럽": 546,
551
+ "럿": 547,
552
+ "렀": 548,
553
+ "렁": 549,
554
+ "렇": 550,
555
+ "레": 551,
556
+ "렉": 552,
557
+ "렌": 553,
558
+ "렐": 554,
559
+ "렘": 555,
560
+ "렙": 556,
561
+ "렛": 557,
562
+ "렜": 558,
563
+ "려": 559,
564
+ "력": 560,
565
+ "련": 561,
566
+ "렬": 562,
567
+ "렴": 563,
568
+ "렵": 564,
569
+ "렷": 565,
570
+ "렸": 566,
571
+ "령": 567,
572
+ "렿": 568,
573
+ "례": 569,
574
+ "로": 570,
575
+ "록": 571,
576
+ "론": 572,
577
+ "롤": 573,
578
+ "롬": 574,
579
+ "롭": 575,
580
+ "롯": 576,
581
+ "롱": 577,
582
+ "롷": 578,
583
+ "롸": 579,
584
+ "뢰": 580,
585
+ "료": 581,
586
+ "룡": 582,
587
+ "루": 583,
588
+ "룩": 584,
589
+ "룬": 585,
590
+ "룰": 586,
591
+ "룸": 587,
592
+ "룹": 588,
593
+ "룻": 589,
594
+ "룽": 590,
595
+ "뤄": 591,
596
+ "뤘": 592,
597
+ "류": 593,
598
+ "륙": 594,
599
+ "륜": 595,
600
+ "률": 596,
601
+ "륨": 597,
602
+ "륭": 598,
603
+ "르": 599,
604
+ "륵": 600,
605
+ "른": 601,
606
+ "를": 602,
607
+ "름": 603,
608
+ "릅": 604,
609
+ "릇": 605,
610
+ "릉": 606,
611
+ "릎": 607,
612
+ "리": 608,
613
+ "릭": 609,
614
+ "린": 610,
615
+ "릴": 611,
616
+ "림": 612,
617
+ "립": 613,
618
+ "릿": 614,
619
+ "맀": 615,
620
+ "링": 616,
621
+ "마": 617,
622
+ "막": 618,
623
+ "만": 619,
624
+ "많": 620,
625
+ "맏": 621,
626
+ "말": 622,
627
+ "맑": 623,
628
+ "맘": 624,
629
+ "맙": 625,
630
+ "맛": 626,
631
+ "맜": 627,
632
+ "망": 628,
633
+ "맞": 629,
634
+ "맡": 630,
635
+ "맣": 631,
636
+ "매": 632,
637
+ "맥": 633,
638
+ "맨": 634,
639
+ "맴": 635,
640
+ "맵": 636,
641
+ "맷": 637,
642
+ "맸": 638,
643
+ "맹": 639,
644
+ "맺": 640,
645
+ "맻": 641,
646
+ "머": 642,
647
+ "먹": 643,
648
+ "먼": 644,
649
+ "멀": 645,
650
+ "멈": 646,
651
+ "멉": 647,
652
+ "멋": 648,
653
+ "멍": 649,
654
+ "멓": 650,
655
+ "메": 651,
656
+ "멕": 652,
657
+ "멘": 653,
658
+ "멜": 654,
659
+ "멤": 655,
660
+ "멧": 656,
661
+ "멨": 657,
662
+ "멩": 658,
663
+ "며": 659,
664
+ "멱": 660,
665
+ "면": 661,
666
+ "멸": 662,
667
+ "몄": 663,
668
+ "명": 664,
669
+ "몇": 665,
670
+ "모": 666,
671
+ "목": 667,
672
+ "몫": 668,
673
+ "몬": 669,
674
+ "몰": 670,
675
+ "몸": 671,
676
+ "몹": 672,
677
+ "못": 673,
678
+ "몽": 674,
679
+ "묘": 675,
680
+ "무": 676,
681
+ "묵": 677,
682
+ "묶": 678,
683
+ "문": 679,
684
+ "묻": 680,
685
+ "물": 681,
686
+ "묽": 682,
687
+ "뭅": 683,
688
+ "뭇": 684,
689
+ "뭉": 685,
690
+ "뭏": 686,
691
+ "뭐": 687,
692
+ "뭔": 688,
693
+ "뭘": 689,
694
+ "뭡": 690,
695
+ "��": 691,
696
+ "뮤": 692,
697
+ "뮨": 693,
698
+ "므": 694,
699
+ "믄": 695,
700
+ "믈": 696,
701
+ "미": 697,
702
+ "믹": 698,
703
+ "민": 699,
704
+ "믿": 700,
705
+ "밀": 701,
706
+ "밉": 702,
707
+ "밋": 703,
708
+ "밌": 704,
709
+ "밍": 705,
710
+ "및": 706,
711
+ "밑": 707,
712
+ "바": 708,
713
+ "박": 709,
714
+ "밖": 710,
715
+ "반": 711,
716
+ "받": 712,
717
+ "발": 713,
718
+ "밝": 714,
719
+ "밟": 715,
720
+ "밤": 716,
721
+ "밥": 717,
722
+ "밧": 718,
723
+ "방": 719,
724
+ "밭": 720,
725
+ "배": 721,
726
+ "백": 722,
727
+ "밴": 723,
728
+ "밸": 724,
729
+ "뱀": 725,
730
+ "뱁": 726,
731
+ "뱃": 727,
732
+ "뱄": 728,
733
+ "뱅": 729,
734
+ "뱉": 730,
735
+ "버": 731,
736
+ "벅": 732,
737
+ "번": 733,
738
+ "벋": 734,
739
+ "벌": 735,
740
+ "범": 736,
741
+ "법": 737,
742
+ "벗": 738,
743
+ "벙": 739,
744
+ "벚": 740,
745
+ "베": 741,
746
+ "벡": 742,
747
+ "벤": 743,
748
+ "벨": 744,
749
+ "벰": 745,
750
+ "벳": 746,
751
+ "벼": 747,
752
+ "벽": 748,
753
+ "변": 749,
754
+ "별": 750,
755
+ "볌": 751,
756
+ "볍": 752,
757
+ "볐": 753,
758
+ "병": 754,
759
+ "볕": 755,
760
+ "보": 756,
761
+ "복": 757,
762
+ "볶": 758,
763
+ "본": 759,
764
+ "볼": 760,
765
+ "봄": 761,
766
+ "봅": 762,
767
+ "봇": 763,
768
+ "봉": 764,
769
+ "봐": 765,
770
+ "봤": 766,
771
+ "봬": 767,
772
+ "뵀": 768,
773
+ "뵈": 769,
774
+ "뵌": 770,
775
+ "뵐": 771,
776
+ "뵙": 772,
777
+ "부": 773,
778
+ "북": 774,
779
+ "분": 775,
780
+ "붇": 776,
781
+ "불": 777,
782
+ "붉": 778,
783
+ "붐": 779,
784
+ "붑": 780,
785
+ "붓": 781,
786
+ "붕": 782,
787
+ "붙": 783,
788
+ "뷔": 784,
789
+ "뷰": 785,
790
+ "뷸": 786,
791
+ "브": 787,
792
+ "븐": 788,
793
+ "블": 789,
794
+ "비": 790,
795
+ "빅": 791,
796
+ "빈": 792,
797
+ "빌": 793,
798
+ "빔": 794,
799
+ "빕": 795,
800
+ "빗": 796,
801
+ "빙": 797,
802
+ "빚": 798,
803
+ "빛": 799,
804
+ "빠": 800,
805
+ "빡": 801,
806
+ "빤": 802,
807
+ "빨": 803,
808
+ "빳": 804,
809
+ "빴": 805,
810
+ "빵": 806,
811
+ "빻": 807,
812
+ "빼": 808,
813
+ "빽": 809,
814
+ "뺀": 810,
815
+ "뺄": 811,
816
+ "뺌": 812,
817
+ "뺍": 813,
818
+ "뺏": 814,
819
+ "뺐": 815,
820
+ "뺑": 816,
821
+ "뺘": 817,
822
+ "뺨": 818,
823
+ "뺼": 819,
824
+ "뻐": 820,
825
+ "뻑": 821,
826
+ "뻔": 822,
827
+ "뻗": 823,
828
+ "뻘": 824,
829
+ "뻣": 825,
830
+ "뻤": 826,
831
+ "뻥": 827,
832
+ "뻬": 828,
833
+ "뼀": 829,
834
+ "뼈": 830,
835
+ "뼘": 831,
836
+ "뼛": 832,
837
+ "뽀": 833,
838
+ "뽁": 834,
839
+ "뽂": 835,
840
+ "뽄": 836,
841
+ "뽈": 837,
842
+ "뽐": 838,
843
+ "뽑": 839,
844
+ "뽕": 840,
845
+ "뽜": 841,
846
+ "뽝": 842,
847
+ "뾰": 843,
848
+ "뿅": 844,
849
+ "뿌": 845,
850
+ "뿍": 846,
851
+ "뿐": 847,
852
+ "뿔": 848,
853
+ "뿜": 849,
854
+ "뿝": 850,
855
+ "뿟": 851,
856
+ "뿡": 852,
857
+ "뿨": 853,
858
+ "쁘": 854,
859
+ "쁜": 855,
860
+ "쁠": 856,
861
+ "쁨": 857,
862
+ "쁩": 858,
863
+ "삐": 859,
864
+ "삑": 860,
865
+ "삔": 861,
866
+ "삘": 862,
867
+ "삥": 863,
868
+ "사": 864,
869
+ "삭": 865,
870
+ "산": 866,
871
+ "살": 867,
872
+ "삶": 868,
873
+ "삼": 869,
874
+ "삽": 870,
875
+ "삿": 871,
876
+ "샀": 872,
877
+ "상": 873,
878
+ "새": 874,
879
+ "색": 875,
880
+ "샌": 876,
881
+ "샐": 877,
882
+ "샘": 878,
883
+ "샙": 879,
884
+ "샜": 880,
885
+ "생": 881,
886
+ "샤": 882,
887
+ "샥": 883,
888
+ "샴": 884,
889
+ "샵": 885,
890
+ "샷": 886,
891
+ "샹": 887,
892
+ "섀": 888,
893
+ "서": 889,
894
+ "석": 890,
895
+ "섞": 891,
896
+ "선": 892,
897
+ "설": 893,
898
+ "섬": 894,
899
+ "섭": 895,
900
+ "섯": 896,
901
+ "섰": 897,
902
+ "성": 898,
903
+ "세": 899,
904
+ "섹": 900,
905
+ "센": 901,
906
+ "셀": 902,
907
+ "셈": 903,
908
+ "셉": 904,
909
+ "셋": 905,
910
+ "셔": 906,
911
+ "션": 907,
912
+ "셜": 908,
913
+ "셧": 909,
914
+ "셨": 910,
915
+ "셰": 911,
916
+ "솁": 912,
917
+ "소": 913,
918
+ "속": 914,
919
+ "손": 915,
920
+ "솔": 916,
921
+ "솜": 917,
922
+ "솝": 918,
923
+ "솟": 919,
924
+ "송": 920,
925
+ "솥": 921,
926
+ "솨": 922,
927
+ "쇄": 923,
928
+ "쇠": 924,
929
+ "쇼": 925,
930
+ "쇽": 926,
931
+ "숄": 927,
932
+ "숍": 928,
933
+ "숏": 929,
934
+ "숑": 930,
935
+ "수": 931,
936
+ "숙": 932,
937
+ "순": 933,
938
+ "숟": 934,
939
+ "술": 935,
940
+ "숨": 936,
941
+ "숩": 937,
942
+ "숫": 938,
943
+ "숭": 939,
944
+ "숯": 940,
945
+ "숱": 941,
946
+ "숲": 942,
947
+ "숴": 943,
948
+ "쉈": 944,
949
+ "쉐": 945,
950
+ "쉘": 946,
951
+ "쉣": 947,
952
+ "쉬": 948,
953
+ "쉰": 949,
954
+ "쉴": 950,
955
+ "쉼": 951,
956
+ "쉽": 952,
957
+ "슈": 953,
958
+ "슉": 954,
959
+ "슐": 955,
960
+ "슘": 956,
961
+ "슛": 957,
962
+ "슝": 958,
963
+ "스": 959,
964
+ "슥": 960,
965
+ "슨": 961,
966
+ "슬": 962,
967
+ "슴": 963,
968
+ "습": 964,
969
+ "슷": 965,
970
+ "승": 966,
971
+ "시": 967,
972
+ "식": 968,
973
+ "신": 969,
974
+ "싣": 970,
975
+ "실": 971,
976
+ "싫": 972,
977
+ "심": 973,
978
+ "십": 974,
979
+ "싯": 975,
980
+ "싰": 976,
981
+ "싱": 977,
982
+ "싶": 978,
983
+ "싸": 979,
984
+ "싹": 980,
985
+ "싼": 981,
986
+ "쌀": 982,
987
+ "쌈": 983,
988
+ "쌉": 984,
989
+ "쌌": 985,
990
+ "쌍": 986,
991
+ "쌓": 987,
992
+ "쌔": 988,
993
+ "쌜": 989,
994
+ "쌤": 990,
995
+ "쌩": 991,
996
+ "쌰": 992,
997
+ "써": 993,
998
+ "썩": 994,
999
+ "썪": 995,
1000
+ "썬": 996,
1001
+ "썰": 997,
1002
+ "썸": 998,
1003
+ "썹": 999,
1004
+ "썻": 1000,
1005
+ "썼": 1001,
1006
+ "썽": 1002,
1007
+ "쎄": 1003,
1008
+ "쎅": 1004,
1009
+ "쎈": 1005,
1010
+ "쎕": 1006,
1011
+ "쎼": 1007,
1012
+ "쏘": 1008,
1013
+ "쏙": 1009,
1014
+ "쏜": 1010,
1015
+ "쏟": 1011,
1016
+ "쏠": 1012,
1017
+ "쏩": 1013,
1018
+ "쏭": 1014,
1019
+ "쏴": 1015,
1020
+ "쏵": 1016,
1021
+ "쐈": 1017,
1022
+ "쐬": 1018,
1023
+ "쐴": 1019,
1024
+ "쑈": 1020,
1025
+ "쑉": 1021,
1026
+ "쑐": 1022,
1027
+ "쑤": 1023,
1028
+ "쑥": 1024,
1029
+ "쓰": 1025,
1030
+ "쓱": 1026,
1031
+ "쓴": 1027,
1032
+ "쓸": 1028,
1033
+ "쓽": 1029,
1034
+ "씀": 1030,
1035
+ "씁": 1031,
1036
+ "씌": 1032,
1037
+ "씐": 1033,
1038
+ "씨": 1034,
1039
+ "씩": 1035,
1040
+ "씬": 1036,
1041
+ "씰": 1037,
1042
+ "씸": 1038,
1043
+ "씹": 1039,
1044
+ "씻": 1040,
1045
+ "씼": 1041,
1046
+ "씽": 1042,
1047
+ "아": 1043,
1048
+ "악": 1044,
1049
+ "안": 1045,
1050
+ "앉": 1046,
1051
+ "않": 1047,
1052
+ "알": 1048,
1053
+ "앎": 1049,
1054
+ "앓": 1050,
1055
+ "암": 1051,
1056
+ "압": 1052,
1057
+ "앗": 1053,
1058
+ "았": 1054,
1059
+ "앙": 1055,
1060
+ "앞": 1056,
1061
+ "애": 1057,
1062
+ "액": 1058,
1063
+ "앤": 1059,
1064
+ "앨": 1060,
1065
+ "앰": 1061,
1066
+ "앱": 1062,
1067
+ "앴": 1063,
1068
+ "앵": 1064,
1069
+ "야": 1065,
1070
+ "약": 1066,
1071
+ "얀": 1067,
1072
+ "얄": 1068,
1073
+ "얇": 1069,
1074
+ "얌": 1070,
1075
+ "얍": 1071,
1076
+ "얏": 1072,
1077
+ "양": 1073,
1078
+ "얕": 1074,
1079
+ "얗": 1075,
1080
+ "얘": 1076,
1081
+ "얜": 1077,
1082
+ "얬": 1078,
1083
+ "얳": 1079,
1084
+ "어": 1080,
1085
+ "억": 1081,
1086
+ "언": 1082,
1087
+ "얹": 1083,
1088
+ "얻": 1084,
1089
+ "얼": 1085,
1090
+ "엄": 1086,
1091
+ "업": 1087,
1092
+ "없": 1088,
1093
+ "엇": 1089,
1094
+ "었": 1090,
1095
+ "엉": 1091,
1096
+ "엊": 1092,
1097
+ "엌": 1093,
1098
+ "엍": 1094,
1099
+ "엎": 1095,
1100
+ "에": 1096,
1101
+ "엑": 1097,
1102
+ "엔": 1098,
1103
+ "엘": 1099,
1104
+ "엠": 1100,
1105
+ "엣": 1101,
1106
+ "엥": 1102,
1107
+ "여": 1103,
1108
+ "역": 1104,
1109
+ "엮": 1105,
1110
+ "연": 1106,
1111
+ "열": 1107,
1112
+ "염": 1108,
1113
+ "엽": 1109,
1114
+ "엿": 1110,
1115
+ "였": 1111,
1116
+ "영": 1112,
1117
+ "옅": 1113,
1118
+ "옆": 1114,
1119
+ "옇": 1115,
1120
+ "예": 1116,
1121
+ "옐": 1117,
1122
+ "옘": 1118,
1123
+ "옙": 1119,
1124
+ "옛": 1120,
1125
+ "옜": 1121,
1126
+ "오": 1122,
1127
+ "옥": 1123,
1128
+ "온": 1124,
1129
+ "올": 1125,
1130
+ "옮": 1126,
1131
+ "옳": 1127,
1132
+ "옴": 1128,
1133
+ "옵": 1129,
1134
+ "옷": 1130,
1135
+ "옹": 1131,
1136
+ "옽": 1132,
1137
+ "와": 1133,
1138
+ "왁": 1134,
1139
+ "완": 1135,
1140
+ "왈": 1136,
1141
+ "왓": 1137,
1142
+ "왔": 1138,
1143
+ "왕": 1139,
1144
+ "왜": 1140,
1145
+ "왠": 1141,
1146
+ "왯": 1142,
1147
+ "외": 1143,
1148
+ "왼": 1144,
1149
+ "요": 1145,
1150
+ "욕": 1146,
1151
+ "욘": 1147,
1152
+ "욜": 1148,
1153
+ "욤": 1149,
1154
+ "용": 1150,
1155
+ "우": 1151,
1156
+ "욱": 1152,
1157
+ "운": 1153,
1158
+ "울": 1154,
1159
+ "움": 1155,
1160
+ "웁": 1156,
1161
+ "웃": 1157,
1162
+ "웅": 1158,
1163
+ "워": 1159,
1164
+ "웍": 1160,
1165
+ "원": 1161,
1166
+ "월": 1162,
1167
+ "웠": 1163,
1168
+ "웡": 1164,
1169
+ "웨": 1165,
1170
+ "웬": 1166,
1171
+ "웰": 1167,
1172
+ "웸": 1168,
1173
+ "웹": 1169,
1174
+ "웻": 1170,
1175
+ "위": 1171,
1176
+ "윅": 1172,
1177
+ "윈": 1173,
1178
+ "윌": 1174,
1179
+ "윕": 1175,
1180
+ "윗": 1176,
1181
+ "윙": 1177,
1182
+ "유": 1178,
1183
+ "육": 1179,
1184
+ "윤": 1180,
1185
+ "율": 1181,
1186
+ "융": 1182,
1187
+ "윷": 1183,
1188
+ "으": 1184,
1189
+ "윽": 1185,
1190
+ "은": 1186,
1191
+ "을": 1187,
1192
+ "읊": 1188,
1193
+ "음": 1189,
1194
+ "읍": 1190,
1195
+ "읎": 1191,
1196
+ "읏": 1192,
1197
+ "응": 1193,
1198
+ "읓": 1194,
1199
+ "의": 1195,
1200
+ "이": 1196,
1201
+ "익": 1197,
1202
+ "인": 1198,
1203
+ "일": 1199,
1204
+ "읽": 1200,
1205
+ "잃": 1201,
1206
+ "임": 1202,
1207
+ "입": 1203,
1208
+ "잇": 1204,
1209
+ "있": 1205,
1210
+ "잉": 1206,
1211
+ "잊": 1207,
1212
+ "잌": 1208,
1213
+ "잎": 1209,
1214
+ "자": 1210,
1215
+ "작": 1211,
1216
+ "잔": 1212,
1217
+ "잖": 1213,
1218
+ "잘": 1214,
1219
+ "잠": 1215,
1220
+ "잡": 1216,
1221
+ "잣": 1217,
1222
+ "잤": 1218,
1223
+ "장": 1219,
1224
+ "잦": 1220,
1225
+ "재": 1221,
1226
+ "잭": 1222,
1227
+ "잰": 1223,
1228
+ "잴": 1224,
1229
+ "잼": 1225,
1230
+ "잽": 1226,
1231
+ "잿": 1227,
1232
+ "쟀": 1228,
1233
+ "쟁": 1229,
1234
+ "쟈": 1230,
1235
+ "쟤": 1231,
1236
+ "쟨": 1232,
1237
+ "저": 1233,
1238
+ "적": 1234,
1239
+ "전": 1235,
1240
+ "절": 1236,
1241
+ "젊": 1237,
1242
+ "점": 1238,
1243
+ "접": 1239,
1244
+ "젓": 1240,
1245
+ "정": 1241,
1246
+ "젖": 1242,
1247
+ "제": 1243,
1248
+ "젝": 1244,
1249
+ "젠": 1245,
1250
+ "젤": 1246,
1251
+ "젭": 1247,
1252
+ "젯": 1248,
1253
+ "져": 1249,
1254
+ "젼": 1250,
1255
+ "졌": 1251,
1256
+ "조": 1252,
1257
+ "족": 1253,
1258
+ "존": 1254,
1259
+ "졸": 1255,
1260
+ "좀": 1256,
1261
+ "좁": 1257,
1262
+ "좃": 1258,
1263
+ "종": 1259,
1264
+ "좆": 1260,
1265
+ "좋": 1261,
1266
+ "좌": 1262,
1267
+ "좔": 1263,
1268
+ "죄": 1264,
1269
+ "죈": 1265,
1270
+ "죙": 1266,
1271
+ "죠": 1267,
1272
+ "죵": 1268,
1273
+ "주": 1269,
1274
+ "죽": 1270,
1275
+ "준": 1271,
1276
+ "줄": 1272,
1277
+ "줌": 1273,
1278
+ "줍": 1274,
1279
+ "줏": 1275,
1280
+ "중": 1276,
1281
+ "줘": 1277,
1282
+ "줬": 1278,
1283
+ "쥐": 1279,
1284
+ "쥑": 1280,
1285
+ "쥬": 1281,
1286
+ "즈": 1282,
1287
+ "즉": 1283,
1288
+ "즌": 1284,
1289
+ "즐": 1285,
1290
+ "즘": 1286,
1291
+ "즙": 1287,
1292
+ "증": 1288,
1293
+ "지": 1289,
1294
+ "직": 1290,
1295
+ "진": 1291,
1296
+ "질": 1292,
1297
+ "짐": 1293,
1298
+ "집": 1294,
1299
+ "짓": 1295,
1300
+ "징": 1296,
1301
+ "짖": 1297,
1302
+ "짙": 1298,
1303
+ "짚": 1299,
1304
+ "짜": 1300,
1305
+ "짝": 1301,
1306
+ "짠": 1302,
1307
+ "짤": 1303,
1308
+ "짧": 1304,
1309
+ "짬": 1305,
1310
+ "짭": 1306,
1311
+ "짰": 1307,
1312
+ "짱": 1308,
1313
+ "째": 1309,
1314
+ "짼": 1310,
1315
+ "쨋": 1311,
1316
+ "쨌": 1312,
1317
+ "쨍": 1313,
1318
+ "쨔": 1314,
1319
+ "쨰": 1315,
1320
+ "쩌": 1316,
1321
+ "쩍": 1317,
1322
+ "쩐": 1318,
1323
+ "쩔": 1319,
1324
+ "쩜": 1320,
1325
+ "쩝": 1321,
1326
+ "쩠": 1322,
1327
+ "쩡": 1323,
1328
+ "쩨": 1324,
1329
+ "쩬": 1325,
1330
+ "쪄": 1326,
1331
+ "쪘": 1327,
1332
+ "쪠": 1328,
1333
+ "쪼": 1329,
1334
+ "쪽": 1330,
1335
+ "쫀": 1331,
1336
+ "쫄": 1332,
1337
+ "쫌": 1333,
1338
+ "쫍": 1334,
1339
+ "쫑": 1335,
1340
+ "쫒": 1336,
1341
+ "쫓": 1337,
1342
+ "쫘": 1338,
1343
+ "쫙": 1339,
1344
+ "쬐": 1340,
1345
+ "쭈": 1341,
1346
+ "쭉": 1342,
1347
+ "쭐": 1343,
1348
+ "쭘": 1344,
1349
+ "쭝": 1345,
1350
+ "쭤": 1346,
1351
+ "쮸": 1347,
1352
+ "쯔": 1348,
1353
+ "쯤": 1349,
1354
+ "쯥": 1350,
1355
+ "찌": 1351,
1356
+ "찍": 1352,
1357
+ "찐": 1353,
1358
+ "찔": 1354,
1359
+ "찜": 1355,
1360
+ "찝": 1356,
1361
+ "찡": 1357,
1362
+ "찢": 1358,
1363
+ "차": 1359,
1364
+ "착": 1360,
1365
+ "찬": 1361,
1366
+ "찮": 1362,
1367
+ "찰": 1363,
1368
+ "참": 1364,
1369
+ "찹": 1365,
1370
+ "찼": 1366,
1371
+ "창": 1367,
1372
+ "찾": 1368,
1373
+ "채": 1369,
1374
+ "책": 1370,
1375
+ "챌": 1371,
1376
+ "챔": 1372,
1377
+ "챕": 1373,
1378
+ "챗": 1374,
1379
+ "챘": 1375,
1380
+ "챙": 1376,
1381
+ "처": 1377,
1382
+ "척": 1378,
1383
+ "천": 1379,
1384
+ "철": 1380,
1385
+ "첨": 1381,
1386
+ "첩": 1382,
1387
+ "첫": 1383,
1388
+ "청": 1384,
1389
+ "체": 1385,
1390
+ "첵": 1386,
1391
+ "첸": 1387,
1392
+ "첼": 1388,
1393
+ "쳇": 1389,
1394
+ "쳐": 1390,
1395
+ "쳤": 1391,
1396
+ "초": 1392,
1397
+ "촉": 1393,
1398
+ "촌": 1394,
1399
+ "촐": 1395,
1400
+ "촛": 1396,
1401
+ "총": 1397,
1402
+ "촤": 1398,
1403
+ "촥": 1399,
1404
+ "촬": 1400,
1405
+ "최": 1401,
1406
+ "쵸": 1402,
1407
+ "추": 1403,
1408
+ "축": 1404,
1409
+ "춘": 1405,
1410
+ "출": 1406,
1411
+ "춤": 1407,
1412
+ "춥": 1408,
1413
+ "춧": 1409,
1414
+ "충": 1410,
1415
+ "춰": 1411,
1416
+ "췄": 1412,
1417
+ "췌": 1413,
1418
+ "취": 1414,
1419
+ "츄": 1415,
1420
+ "츠": 1416,
1421
+ "측": 1417,
1422
+ "층": 1418,
1423
+ "치": 1419,
1424
+ "칙": 1420,
1425
+ "친": 1421,
1426
+ "칠": 1422,
1427
+ "침": 1423,
1428
+ "칩": 1424,
1429
+ "칫": 1425,
1430
+ "칬": 1426,
1431
+ "칭": 1427,
1432
+ "카": 1428,
1433
+ "칵": 1429,
1434
+ "칸": 1430,
1435
+ "칼": 1431,
1436
+ "캅": 1432,
1437
+ "캉": 1433,
1438
+ "캐": 1434,
1439
+ "캔": 1435,
1440
+ "캠": 1436,
1441
+ "캡": 1437,
1442
+ "캣": 1438,
1443
+ "캬": 1439,
1444
+ "커": 1440,
1445
+ "컥": 1441,
1446
+ "컨": 1442,
1447
+ "컬": 1443,
1448
+ "컴": 1444,
1449
+ "컵": 1445,
1450
+ "컷": 1446,
1451
+ "컸": 1447,
1452
+ "컹": 1448,
1453
+ "컽": 1449,
1454
+ "케": 1450,
1455
+ "켄": 1451,
1456
+ "켈": 1452,
1457
+ "켓": 1453,
1458
+ "켜": 1454,
1459
+ "켠": 1455,
1460
+ "켰": 1456,
1461
+ "코": 1457,
1462
+ "콕": 1458,
1463
+ "콘": 1459,
1464
+ "콜": 1460,
1465
+ "콤": 1461,
1466
+ "콧": 1462,
1467
+ "콩": 1463,
1468
+ "콸": 1464,
1469
+ "쾅": 1465,
1470
+ "쾌": 1466,
1471
+ "쿄": 1467,
1472
+ "쿠": 1468,
1473
+ "쿡": 1469,
1474
+ "쿤": 1470,
1475
+ "쿨": 1471,
1476
+ "쿰": 1472,
1477
+ "쿱": 1473,
1478
+ "쿵": 1474,
1479
+ "퀄": 1475,
1480
+ "퀘": 1476,
1481
+ "퀴": 1477,
1482
+ "퀵": 1478,
1483
+ "퀸": 1479,
1484
+ "퀼": 1480,
1485
+ "큐": 1481,
1486
+ "크": 1482,
1487
+ "큰": 1483,
1488
+ "클": 1484,
1489
+ "큼": 1485,
1490
+ "큽": 1486,
1491
+ "킁": 1487,
1492
+ "키": 1488,
1493
+ "킥": 1489,
1494
+ "킨": 1490,
1495
+ "킬": 1491,
1496
+ "킴": 1492,
1497
+ "킵": 1493,
1498
+ "킷": 1494,
1499
+ "킹": 1495,
1500
+ "타": 1496,
1501
+ "탁": 1497,
1502
+ "탄": 1498,
1503
+ "탈": 1499,
1504
+ "탐": 1500,
1505
+ "탑": 1501,
1506
+ "탓": 1502,
1507
+ "탔": 1503,
1508
+ "탕": 1504,
1509
+ "태": 1505,
1510
+ "택": 1506,
1511
+ "탠": 1507,
1512
+ "탬": 1508,
1513
+ "탭": 1509,
1514
+ "탯": 1510,
1515
+ "탱": 1511,
1516
+ "터": 1512,
1517
+ "턱": 1513,
1518
+ "턴": 1514,
1519
+ "털": 1515,
1520
+ "텀": 1516,
1521
+ "텁": 1517,
1522
+ "텃": 1518,
1523
+ "텄": 1519,
1524
+ "텅": 1520,
1525
+ "테": 1521,
1526
+ "텍": 1522,
1527
+ "텐": 1523,
1528
+ "텔": 1524,
1529
+ "템": 1525,
1530
+ "텝": 1526,
1531
+ "텨": 1527,
1532
+ "텼": 1528,
1533
+ "토": 1529,
1534
+ "톡": 1530,
1535
+ "톤": 1531,
1536
+ "톨": 1532,
1537
+ "톰": 1533,
1538
+ "톱": 1534,
1539
+ "통": 1535,
1540
+ "퇴": 1536,
1541
+ "투": 1537,
1542
+ "툭": 1538,
1543
+ "툰": 1539,
1544
+ "툴": 1540,
1545
+ "툼": 1541,
1546
+ "퉁": 1542,
1547
+ "퉤": 1543,
1548
+ "튀": 1544,
1549
+ "튄": 1545,
1550
+ "튈": 1546,
1551
+ "튑": 1547,
1552
+ "튕": 1548,
1553
+ "튜": 1549,
1554
+ "트": 1550,
1555
+ "특": 1551,
1556
+ "튼": 1552,
1557
+ "틀": 1553,
1558
+ "틈": 1554,
1559
+ "틉": 1555,
1560
+ "틋": 1556,
1561
+ "틑": 1557,
1562
+ "티": 1558,
1563
+ "틱": 1559,
1564
+ "틴": 1560,
1565
+ "틸": 1561,
1566
+ "팀": 1562,
1567
+ "팁": 1563,
1568
+ "팅": 1564,
1569
+ "파": 1565,
1570
+ "팍": 1566,
1571
+ "팎": 1567,
1572
+ "판": 1568,
1573
+ "팔": 1569,
1574
+ "팜": 1570,
1575
+ "팝": 1571,
1576
+ "팟": 1572,
1577
+ "팠": 1573,
1578
+ "팡": 1574,
1579
+ "팥": 1575,
1580
+ "패": 1576,
1581
+ "팩": 1577,
1582
+ "팬": 1578,
1583
+ "팸": 1579,
1584
+ "팹": 1580,
1585
+ "팻": 1581,
1586
+ "팽": 1582,
1587
+ "퍼": 1583,
1588
+ "퍽": 1584,
1589
+ "펀": 1585,
1590
+ "펄": 1586,
1591
+ "펌": 1587,
1592
+ "펍": 1588,
1593
+ "펐": 1589,
1594
+ "펑": 1590,
1595
+ "페": 1591,
1596
+ "펙": 1592,
1597
+ "펜": 1593,
1598
+ "펠": 1594,
1599
+ "펩": 1595,
1600
+ "펫": 1596,
1601
+ "펭": 1597,
1602
+ "펴": 1598,
1603
+ "편": 1599,
1604
+ "펼": 1600,
1605
+ "폄": 1601,
1606
+ "폈": 1602,
1607
+ "평": 1603,
1608
+ "폐": 1604,
1609
+ "포": 1605,
1610
+ "폭": 1606,
1611
+ "폰": 1607,
1612
+ "폴": 1608,
1613
+ "폼": 1609,
1614
+ "폽": 1610,
1615
+ "폿": 1611,
1616
+ "퐁": 1612,
1617
+ "표": 1613,
1618
+ "푠": 1614,
1619
+ "푸": 1615,
1620
+ "푹": 1616,
1621
+ "푼": 1617,
1622
+ "풀": 1618,
1623
+ "풂": 1619,
1624
+ "품": 1620,
1625
+ "풉": 1621,
1626
+ "풋": 1622,
1627
+ "풍": 1623,
1628
+ "퓨": 1624,
1629
+ "프": 1625,
1630
+ "픈": 1626,
1631
+ "플": 1627,
1632
+ "픔": 1628,
1633
+ "픕": 1629,
1634
+ "픗": 1630,
1635
+ "피": 1631,
1636
+ "픽": 1632,
1637
+ "핀": 1633,
1638
+ "필": 1634,
1639
+ "핌": 1635,
1640
+ "핍": 1636,
1641
+ "핏": 1637,
1642
+ "핑": 1638,
1643
+ "하": 1639,
1644
+ "학": 1640,
1645
+ "한": 1641,
1646
+ "할": 1642,
1647
+ "핥": 1643,
1648
+ "함": 1644,
1649
+ "합": 1645,
1650
+ "핫": 1646,
1651
+ "핬": 1647,
1652
+ "항": 1648,
1653
+ "해": 1649,
1654
+ "핵": 1650,
1655
+ "핸": 1651,
1656
+ "핼": 1652,
1657
+ "햄": 1653,
1658
+ "햅": 1654,
1659
+ "햇": 1655,
1660
+ "했": 1656,
1661
+ "행": 1657,
1662
+ "햐": 1658,
1663
+ "향": 1659,
1664
+ "허": 1660,
1665
+ "헉": 1661,
1666
+ "헌": 1662,
1667
+ "헐": 1663,
1668
+ "험": 1664,
1669
+ "헙": 1665,
1670
+ "헛": 1666,
1671
+ "헝": 1667,
1672
+ "헤": 1668,
1673
+ "헥": 1669,
1674
+ "헨": 1670,
1675
+ "헬": 1671,
1676
+ "헵": 1672,
1677
+ "헷": 1673,
1678
+ "헸": 1674,
1679
+ "헹": 1675,
1680
+ "혀": 1676,
1681
+ "혁": 1677,
1682
+ "현": 1678,
1683
+ "혈": 1679,
1684
+ "혐": 1680,
1685
+ "협": 1681,
1686
+ "혓": 1682,
1687
+ "혔": 1683,
1688
+ "형": 1684,
1689
+ "혜": 1685,
1690
+ "호": 1686,
1691
+ "혹": 1687,
1692
+ "혼": 1688,
1693
+ "홀": 1689,
1694
+ "홈": 1690,
1695
+ "홉": 1691,
1696
+ "홋": 1692,
1697
+ "홍": 1693,
1698
+ "화": 1694,
1699
+ "확": 1695,
1700
+ "환": 1696,
1701
+ "활": 1697,
1702
+ "황": 1698,
1703
+ "횃": 1699,
1704
+ "회": 1700,
1705
+ "획": 1701,
1706
+ "횝": 1702,
1707
+ "횟": 1703,
1708
+ "횡": 1704,
1709
+ "효": 1705,
1710
+ "후": 1706,
1711
+ "훅": 1707,
1712
+ "훈": 1708,
1713
+ "훌": 1709,
1714
+ "훑": 1710,
1715
+ "훔": 1711,
1716
+ "훤": 1712,
1717
+ "훨": 1713,
1718
+ "훼": 1714,
1719
+ "휑": 1715,
1720
+ "휘": 1716,
1721
+ "휙": 1717,
1722
+ "휜": 1718,
1723
+ "휠": 1719,
1724
+ "휩": 1720,
1725
+ "휴": 1721,
1726
+ "흉": 1722,
1727
+ "흐": 1723,
1728
+ "흑": 1724,
1729
+ "흔": 1725,
1730
+ "흘": 1726,
1731
+ "흙": 1727,
1732
+ "흠": 1728,
1733
+ "흡": 1729,
1734
+ "흥": 1730,
1735
+ "흩": 1731,
1736
+ "희": 1732,
1737
+ "흰": 1733,
1738
+ "히": 1734,
1739
+ "힌": 1735,
1740
+ "힐": 1736,
1741
+ "힘": 1737,
1742
+ "힙": 1738,
1743
+ "힝": 1739
1744
+ }
config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_attn_dim": null,
5
+ "adapter_kernel_size": 3,
6
+ "adapter_stride": 2,
7
+ "add_adapter": false,
8
+ "apply_spec_augment": true,
9
+ "architectures": [
10
+ "Wav2Vec2ForCTC"
11
+ ],
12
+ "attention_dropout": 0.1,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_dim": [
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512
26
+ ],
27
+ "conv_kernel": [
28
+ 10,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 2,
34
+ 2
35
+ ],
36
+ "conv_stride": [
37
+ 5,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2
44
+ ],
45
+ "ctc_loss_reduction": "mean",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": true,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_dropout": 0.0,
52
+ "feat_extract_norm": "layer",
53
+ "feat_proj_dropout": 0.05,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.05,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.05,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.05,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "num_adapter_layers": 3,
80
+ "num_attention_heads": 16,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_feat_extract_layers": 7,
86
+ "num_hidden_layers": 24,
87
+ "num_negatives": 100,
88
+ "output_hidden_size": 1024,
89
+ "pad_token_id": 1741,
90
+ "proj_codevector_dim": 768,
91
+ "tdnn_dilation": [
92
+ 1,
93
+ 2,
94
+ 3,
95
+ 1,
96
+ 1
97
+ ],
98
+ "tdnn_dim": [
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 512,
103
+ 1500
104
+ ],
105
+ "tdnn_kernel": [
106
+ 5,
107
+ 3,
108
+ 3,
109
+ 1,
110
+ 1
111
+ ],
112
+ "torch_dtype": "float32",
113
+ "transformers_version": "4.34.0",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 1744,
116
+ "xvector_output_dim": 512
117
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 14.99,
3
+ "eval_cer": 0.1081634945549379,
4
+ "eval_loss": 0.38891980051994324,
5
+ "eval_runtime": 1601.2936,
6
+ "eval_samples": 13093,
7
+ "eval_samples_per_second": 8.177,
8
+ "eval_steps_per_second": 0.511
9
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:416f31c2909f85b637313f51236e7dacc2f9ca4d62360e2b41f00bd10c9bba50
3
+ size 1269052778
special_tokens_map.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>",
4
+ "</s>"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "eos_token": "</s>",
8
+ "pad_token": "[PAD]",
9
+ "unk_token": "[UNK]"
10
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "1740": {
4
+ "content": "[UNK]",
5
+ "lstrip": true,
6
+ "normalized": false,
7
+ "rstrip": true,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1741": {
12
+ "content": "[PAD]",
13
+ "lstrip": true,
14
+ "normalized": false,
15
+ "rstrip": true,
16
+ "single_word": false,
17
+ "special": false
18
+ },
19
+ "1742": {
20
+ "content": "<s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "1743": {
28
+ "content": "</s>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "additional_special_tokens": [
37
+ "<s>",
38
+ "</s>"
39
+ ],
40
+ "bos_token": "<s>",
41
+ "clean_up_tokenization_spaces": true,
42
+ "config": null,
43
+ "do_lower_case": false,
44
+ "eos_token": "</s>",
45
+ "model_max_length": 1000000000000000019884624838656,
46
+ "pad_token": "[PAD]",
47
+ "processor_class": "Wav2Vec2Processor",
48
+ "replace_word_delimiter_char": " ",
49
+ "target_lang": null,
50
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
51
+ "tokenizer_file": null,
52
+ "tokenizer_type": "wav2vec2",
53
+ "trust_remote_code": false,
54
+ "unk_token": "[UNK]",
55
+ "word_delimiter_token": "|"
56
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 14.99,
3
+ "train_loss": 1.1506779817299688,
4
+ "train_runtime": 54167.2991,
5
+ "train_samples": 104547,
6
+ "train_samples_per_second": 28.951,
7
+ "train_steps_per_second": 0.452
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,868 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 14.993114001530222,
5
+ "eval_steps": 2000,
6
+ "global_step": 24495,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12,
13
+ "learning_rate": 3.9999999999999996e-05,
14
+ "loss": 33.0181,
15
+ "step": 200
16
+ },
17
+ {
18
+ "epoch": 0.24,
19
+ "learning_rate": 7.999999999999999e-05,
20
+ "loss": 8.7646,
21
+ "step": 400
22
+ },
23
+ {
24
+ "epoch": 0.37,
25
+ "learning_rate": 0.00011999999999999999,
26
+ "loss": 4.6933,
27
+ "step": 600
28
+ },
29
+ {
30
+ "epoch": 0.49,
31
+ "learning_rate": 0.00015999999999999999,
32
+ "loss": 4.5796,
33
+ "step": 800
34
+ },
35
+ {
36
+ "epoch": 0.61,
37
+ "learning_rate": 0.00019999999999999998,
38
+ "loss": 4.1047,
39
+ "step": 1000
40
+ },
41
+ {
42
+ "epoch": 0.73,
43
+ "learning_rate": 0.00023999999999999998,
44
+ "loss": 2.6327,
45
+ "step": 1200
46
+ },
47
+ {
48
+ "epoch": 0.86,
49
+ "learning_rate": 0.00028,
50
+ "loss": 1.9893,
51
+ "step": 1400
52
+ },
53
+ {
54
+ "epoch": 0.98,
55
+ "learning_rate": 0.00029869536855838224,
56
+ "loss": 1.7241,
57
+ "step": 1600
58
+ },
59
+ {
60
+ "epoch": 1.1,
61
+ "learning_rate": 0.00029608610567514673,
62
+ "loss": 1.5607,
63
+ "step": 1800
64
+ },
65
+ {
66
+ "epoch": 1.22,
67
+ "learning_rate": 0.00029347684279191127,
68
+ "loss": 1.4503,
69
+ "step": 2000
70
+ },
71
+ {
72
+ "epoch": 1.22,
73
+ "eval_cer": 0.268688843618788,
74
+ "eval_loss": 1.0610458850860596,
75
+ "eval_runtime": 1586.9419,
76
+ "eval_samples_per_second": 8.25,
77
+ "eval_steps_per_second": 0.516,
78
+ "step": 2000
79
+ },
80
+ {
81
+ "epoch": 1.35,
82
+ "learning_rate": 0.00029086757990867576,
83
+ "loss": 1.3782,
84
+ "step": 2200
85
+ },
86
+ {
87
+ "epoch": 1.47,
88
+ "learning_rate": 0.0002882583170254403,
89
+ "loss": 1.3244,
90
+ "step": 2400
91
+ },
92
+ {
93
+ "epoch": 1.59,
94
+ "learning_rate": 0.0002856490541422048,
95
+ "loss": 1.2553,
96
+ "step": 2600
97
+ },
98
+ {
99
+ "epoch": 1.71,
100
+ "learning_rate": 0.00028303979125896933,
101
+ "loss": 1.2077,
102
+ "step": 2800
103
+ },
104
+ {
105
+ "epoch": 1.84,
106
+ "learning_rate": 0.0002804305283757338,
107
+ "loss": 1.1868,
108
+ "step": 3000
109
+ },
110
+ {
111
+ "epoch": 1.96,
112
+ "learning_rate": 0.00027782126549249836,
113
+ "loss": 1.1624,
114
+ "step": 3200
115
+ },
116
+ {
117
+ "epoch": 2.08,
118
+ "learning_rate": 0.00027521200260926284,
119
+ "loss": 1.1136,
120
+ "step": 3400
121
+ },
122
+ {
123
+ "epoch": 2.2,
124
+ "learning_rate": 0.0002726027397260274,
125
+ "loss": 1.077,
126
+ "step": 3600
127
+ },
128
+ {
129
+ "epoch": 2.33,
130
+ "learning_rate": 0.00026999347684279187,
131
+ "loss": 1.053,
132
+ "step": 3800
133
+ },
134
+ {
135
+ "epoch": 2.45,
136
+ "learning_rate": 0.0002673842139595564,
137
+ "loss": 1.0239,
138
+ "step": 4000
139
+ },
140
+ {
141
+ "epoch": 2.45,
142
+ "eval_cer": 0.19039349567825709,
143
+ "eval_loss": 0.6961866617202759,
144
+ "eval_runtime": 1607.479,
145
+ "eval_samples_per_second": 8.145,
146
+ "eval_steps_per_second": 0.509,
147
+ "step": 4000
148
+ },
149
+ {
150
+ "epoch": 2.57,
151
+ "learning_rate": 0.0002647749510763209,
152
+ "loss": 1.0206,
153
+ "step": 4200
154
+ },
155
+ {
156
+ "epoch": 2.69,
157
+ "learning_rate": 0.00026216568819308544,
158
+ "loss": 1.0045,
159
+ "step": 4400
160
+ },
161
+ {
162
+ "epoch": 2.82,
163
+ "learning_rate": 0.0002595564253098499,
164
+ "loss": 0.9802,
165
+ "step": 4600
166
+ },
167
+ {
168
+ "epoch": 2.94,
169
+ "learning_rate": 0.00025694716242661447,
170
+ "loss": 0.9839,
171
+ "step": 4800
172
+ },
173
+ {
174
+ "epoch": 3.06,
175
+ "learning_rate": 0.00025433789954337895,
176
+ "loss": 0.9243,
177
+ "step": 5000
178
+ },
179
+ {
180
+ "epoch": 3.18,
181
+ "learning_rate": 0.0002517286366601435,
182
+ "loss": 0.9082,
183
+ "step": 5200
184
+ },
185
+ {
186
+ "epoch": 3.31,
187
+ "learning_rate": 0.000249119373776908,
188
+ "loss": 0.9017,
189
+ "step": 5400
190
+ },
191
+ {
192
+ "epoch": 3.43,
193
+ "learning_rate": 0.0002465101108936725,
194
+ "loss": 0.8871,
195
+ "step": 5600
196
+ },
197
+ {
198
+ "epoch": 3.55,
199
+ "learning_rate": 0.00024390084801043704,
200
+ "loss": 0.9036,
201
+ "step": 5800
202
+ },
203
+ {
204
+ "epoch": 3.67,
205
+ "learning_rate": 0.00024129158512720155,
206
+ "loss": 0.8977,
207
+ "step": 6000
208
+ },
209
+ {
210
+ "epoch": 3.67,
211
+ "eval_cer": 0.16872254319465907,
212
+ "eval_loss": 0.594495415687561,
213
+ "eval_runtime": 1591.6049,
214
+ "eval_samples_per_second": 8.226,
215
+ "eval_steps_per_second": 0.515,
216
+ "step": 6000
217
+ },
218
+ {
219
+ "epoch": 3.79,
220
+ "learning_rate": 0.00023868232224396607,
221
+ "loss": 0.8843,
222
+ "step": 6200
223
+ },
224
+ {
225
+ "epoch": 3.92,
226
+ "learning_rate": 0.00023607305936073058,
227
+ "loss": 0.8757,
228
+ "step": 6400
229
+ },
230
+ {
231
+ "epoch": 4.04,
232
+ "learning_rate": 0.0002334637964774951,
233
+ "loss": 0.8573,
234
+ "step": 6600
235
+ },
236
+ {
237
+ "epoch": 4.16,
238
+ "learning_rate": 0.0002308545335942596,
239
+ "loss": 0.8126,
240
+ "step": 6800
241
+ },
242
+ {
243
+ "epoch": 4.28,
244
+ "learning_rate": 0.00022824527071102412,
245
+ "loss": 0.8192,
246
+ "step": 7000
247
+ },
248
+ {
249
+ "epoch": 4.41,
250
+ "learning_rate": 0.00022563600782778863,
251
+ "loss": 0.8061,
252
+ "step": 7200
253
+ },
254
+ {
255
+ "epoch": 4.53,
256
+ "learning_rate": 0.00022302674494455315,
257
+ "loss": 0.8123,
258
+ "step": 7400
259
+ },
260
+ {
261
+ "epoch": 4.65,
262
+ "learning_rate": 0.00022041748206131766,
263
+ "loss": 0.8046,
264
+ "step": 7600
265
+ },
266
+ {
267
+ "epoch": 4.77,
268
+ "learning_rate": 0.00021780821917808218,
269
+ "loss": 0.7979,
270
+ "step": 7800
271
+ },
272
+ {
273
+ "epoch": 4.9,
274
+ "learning_rate": 0.0002151989562948467,
275
+ "loss": 0.804,
276
+ "step": 8000
277
+ },
278
+ {
279
+ "epoch": 4.9,
280
+ "eval_cer": 0.14924661713942214,
281
+ "eval_loss": 0.5327703952789307,
282
+ "eval_runtime": 1595.6324,
283
+ "eval_samples_per_second": 8.206,
284
+ "eval_steps_per_second": 0.513,
285
+ "step": 8000
286
+ },
287
+ {
288
+ "epoch": 5.02,
289
+ "learning_rate": 0.0002125896934116112,
290
+ "loss": 0.7867,
291
+ "step": 8200
292
+ },
293
+ {
294
+ "epoch": 5.14,
295
+ "learning_rate": 0.00020998043052837572,
296
+ "loss": 0.7557,
297
+ "step": 8400
298
+ },
299
+ {
300
+ "epoch": 5.26,
301
+ "learning_rate": 0.00020737116764514023,
302
+ "loss": 0.7478,
303
+ "step": 8600
304
+ },
305
+ {
306
+ "epoch": 5.39,
307
+ "learning_rate": 0.00020476190476190475,
308
+ "loss": 0.7398,
309
+ "step": 8800
310
+ },
311
+ {
312
+ "epoch": 5.51,
313
+ "learning_rate": 0.00020215264187866926,
314
+ "loss": 0.7408,
315
+ "step": 9000
316
+ },
317
+ {
318
+ "epoch": 5.63,
319
+ "learning_rate": 0.00019954337899543377,
320
+ "loss": 0.75,
321
+ "step": 9200
322
+ },
323
+ {
324
+ "epoch": 5.75,
325
+ "learning_rate": 0.0001969341161121983,
326
+ "loss": 0.7344,
327
+ "step": 9400
328
+ },
329
+ {
330
+ "epoch": 5.88,
331
+ "learning_rate": 0.0001943248532289628,
332
+ "loss": 0.738,
333
+ "step": 9600
334
+ },
335
+ {
336
+ "epoch": 6.0,
337
+ "learning_rate": 0.00019171559034572732,
338
+ "loss": 0.7373,
339
+ "step": 9800
340
+ },
341
+ {
342
+ "epoch": 6.12,
343
+ "learning_rate": 0.00018910632746249183,
344
+ "loss": 0.698,
345
+ "step": 10000
346
+ },
347
+ {
348
+ "epoch": 6.12,
349
+ "eval_cer": 0.13653489424101573,
350
+ "eval_loss": 0.5013594031333923,
351
+ "eval_runtime": 1605.1161,
352
+ "eval_samples_per_second": 8.157,
353
+ "eval_steps_per_second": 0.51,
354
+ "step": 10000
355
+ },
356
+ {
357
+ "epoch": 6.24,
358
+ "learning_rate": 0.00018649706457925634,
359
+ "loss": 0.6943,
360
+ "step": 10200
361
+ },
362
+ {
363
+ "epoch": 6.37,
364
+ "learning_rate": 0.00018388780169602086,
365
+ "loss": 0.6997,
366
+ "step": 10400
367
+ },
368
+ {
369
+ "epoch": 6.49,
370
+ "learning_rate": 0.00018127853881278537,
371
+ "loss": 0.6929,
372
+ "step": 10600
373
+ },
374
+ {
375
+ "epoch": 6.61,
376
+ "learning_rate": 0.00017866927592954989,
377
+ "loss": 0.7003,
378
+ "step": 10800
379
+ },
380
+ {
381
+ "epoch": 6.73,
382
+ "learning_rate": 0.0001760600130463144,
383
+ "loss": 0.6863,
384
+ "step": 11000
385
+ },
386
+ {
387
+ "epoch": 6.86,
388
+ "learning_rate": 0.00017345075016307891,
389
+ "loss": 0.6883,
390
+ "step": 11200
391
+ },
392
+ {
393
+ "epoch": 6.98,
394
+ "learning_rate": 0.00017084148727984343,
395
+ "loss": 0.6787,
396
+ "step": 11400
397
+ },
398
+ {
399
+ "epoch": 7.1,
400
+ "learning_rate": 0.00016823222439660794,
401
+ "loss": 0.6518,
402
+ "step": 11600
403
+ },
404
+ {
405
+ "epoch": 7.22,
406
+ "learning_rate": 0.00016562296151337246,
407
+ "loss": 0.6494,
408
+ "step": 11800
409
+ },
410
+ {
411
+ "epoch": 7.35,
412
+ "learning_rate": 0.00016301369863013697,
413
+ "loss": 0.6426,
414
+ "step": 12000
415
+ },
416
+ {
417
+ "epoch": 7.35,
418
+ "eval_cer": 0.13216305737125092,
419
+ "eval_loss": 0.47150149941444397,
420
+ "eval_runtime": 1597.7342,
421
+ "eval_samples_per_second": 8.195,
422
+ "eval_steps_per_second": 0.513,
423
+ "step": 12000
424
+ },
425
+ {
426
+ "epoch": 7.47,
427
+ "learning_rate": 0.00016040443574690148,
428
+ "loss": 0.6457,
429
+ "step": 12200
430
+ },
431
+ {
432
+ "epoch": 7.59,
433
+ "learning_rate": 0.000157795172863666,
434
+ "loss": 0.6429,
435
+ "step": 12400
436
+ },
437
+ {
438
+ "epoch": 7.71,
439
+ "learning_rate": 0.0001551859099804305,
440
+ "loss": 0.6512,
441
+ "step": 12600
442
+ },
443
+ {
444
+ "epoch": 7.83,
445
+ "learning_rate": 0.00015257664709719503,
446
+ "loss": 0.6458,
447
+ "step": 12800
448
+ },
449
+ {
450
+ "epoch": 7.96,
451
+ "learning_rate": 0.00014996738421395954,
452
+ "loss": 0.6501,
453
+ "step": 13000
454
+ },
455
+ {
456
+ "epoch": 8.08,
457
+ "learning_rate": 0.00014735812133072405,
458
+ "loss": 0.6095,
459
+ "step": 13200
460
+ },
461
+ {
462
+ "epoch": 8.2,
463
+ "learning_rate": 0.00014474885844748857,
464
+ "loss": 0.6132,
465
+ "step": 13400
466
+ },
467
+ {
468
+ "epoch": 8.32,
469
+ "learning_rate": 0.00014213959556425308,
470
+ "loss": 0.6102,
471
+ "step": 13600
472
+ },
473
+ {
474
+ "epoch": 8.45,
475
+ "learning_rate": 0.0001395303326810176,
476
+ "loss": 0.6125,
477
+ "step": 13800
478
+ },
479
+ {
480
+ "epoch": 8.57,
481
+ "learning_rate": 0.0001369210697977821,
482
+ "loss": 0.61,
483
+ "step": 14000
484
+ },
485
+ {
486
+ "epoch": 8.57,
487
+ "eval_cer": 0.1257692459492199,
488
+ "eval_loss": 0.45295360684394836,
489
+ "eval_runtime": 1603.0798,
490
+ "eval_samples_per_second": 8.167,
491
+ "eval_steps_per_second": 0.511,
492
+ "step": 14000
493
+ },
494
+ {
495
+ "epoch": 8.69,
496
+ "learning_rate": 0.00013431180691454662,
497
+ "loss": 0.606,
498
+ "step": 14200
499
+ },
500
+ {
501
+ "epoch": 8.81,
502
+ "learning_rate": 0.00013170254403131114,
503
+ "loss": 0.5957,
504
+ "step": 14400
505
+ },
506
+ {
507
+ "epoch": 8.94,
508
+ "learning_rate": 0.00012909328114807565,
509
+ "loss": 0.5992,
510
+ "step": 14600
511
+ },
512
+ {
513
+ "epoch": 9.06,
514
+ "learning_rate": 0.00012648401826484017,
515
+ "loss": 0.5752,
516
+ "step": 14800
517
+ },
518
+ {
519
+ "epoch": 9.18,
520
+ "learning_rate": 0.00012387475538160468,
521
+ "loss": 0.5654,
522
+ "step": 15000
523
+ },
524
+ {
525
+ "epoch": 9.3,
526
+ "learning_rate": 0.00012126549249836919,
527
+ "loss": 0.5725,
528
+ "step": 15200
529
+ },
530
+ {
531
+ "epoch": 9.43,
532
+ "learning_rate": 0.00011865622961513371,
533
+ "loss": 0.5713,
534
+ "step": 15400
535
+ },
536
+ {
537
+ "epoch": 9.55,
538
+ "learning_rate": 0.00011604696673189822,
539
+ "loss": 0.5649,
540
+ "step": 15600
541
+ },
542
+ {
543
+ "epoch": 9.67,
544
+ "learning_rate": 0.00011343770384866273,
545
+ "loss": 0.5643,
546
+ "step": 15800
547
+ },
548
+ {
549
+ "epoch": 9.79,
550
+ "learning_rate": 0.00011082844096542725,
551
+ "loss": 0.5709,
552
+ "step": 16000
553
+ },
554
+ {
555
+ "epoch": 9.79,
556
+ "eval_cer": 0.1200554980402634,
557
+ "eval_loss": 0.4299587607383728,
558
+ "eval_runtime": 1609.5227,
559
+ "eval_samples_per_second": 8.135,
560
+ "eval_steps_per_second": 0.509,
561
+ "step": 16000
562
+ },
563
+ {
564
+ "epoch": 9.92,
565
+ "learning_rate": 0.00010821917808219176,
566
+ "loss": 0.5666,
567
+ "step": 16200
568
+ },
569
+ {
570
+ "epoch": 10.04,
571
+ "learning_rate": 0.00010560991519895628,
572
+ "loss": 0.5531,
573
+ "step": 16400
574
+ },
575
+ {
576
+ "epoch": 10.16,
577
+ "learning_rate": 0.00010300065231572079,
578
+ "loss": 0.5389,
579
+ "step": 16600
580
+ },
581
+ {
582
+ "epoch": 10.28,
583
+ "learning_rate": 0.0001003913894324853,
584
+ "loss": 0.5456,
585
+ "step": 16800
586
+ },
587
+ {
588
+ "epoch": 10.41,
589
+ "learning_rate": 9.778212654924982e-05,
590
+ "loss": 0.5353,
591
+ "step": 17000
592
+ },
593
+ {
594
+ "epoch": 10.53,
595
+ "learning_rate": 9.517286366601433e-05,
596
+ "loss": 0.5337,
597
+ "step": 17200
598
+ },
599
+ {
600
+ "epoch": 10.65,
601
+ "learning_rate": 9.256360078277885e-05,
602
+ "loss": 0.5296,
603
+ "step": 17400
604
+ },
605
+ {
606
+ "epoch": 10.77,
607
+ "learning_rate": 8.995433789954336e-05,
608
+ "loss": 0.5372,
609
+ "step": 17600
610
+ },
611
+ {
612
+ "epoch": 10.9,
613
+ "learning_rate": 8.734507501630787e-05,
614
+ "loss": 0.5388,
615
+ "step": 17800
616
+ },
617
+ {
618
+ "epoch": 11.02,
619
+ "learning_rate": 8.473581213307239e-05,
620
+ "loss": 0.5235,
621
+ "step": 18000
622
+ },
623
+ {
624
+ "epoch": 11.02,
625
+ "eval_cer": 0.11664607248141211,
626
+ "eval_loss": 0.4167773723602295,
627
+ "eval_runtime": 1608.2913,
628
+ "eval_samples_per_second": 8.141,
629
+ "eval_steps_per_second": 0.509,
630
+ "step": 18000
631
+ },
632
+ {
633
+ "epoch": 11.14,
634
+ "learning_rate": 8.212654924983692e-05,
635
+ "loss": 0.509,
636
+ "step": 18200
637
+ },
638
+ {
639
+ "epoch": 11.26,
640
+ "learning_rate": 7.951728636660143e-05,
641
+ "loss": 0.5116,
642
+ "step": 18400
643
+ },
644
+ {
645
+ "epoch": 11.38,
646
+ "learning_rate": 7.690802348336594e-05,
647
+ "loss": 0.4967,
648
+ "step": 18600
649
+ },
650
+ {
651
+ "epoch": 11.51,
652
+ "learning_rate": 7.429876060013046e-05,
653
+ "loss": 0.511,
654
+ "step": 18800
655
+ },
656
+ {
657
+ "epoch": 11.63,
658
+ "learning_rate": 7.168949771689497e-05,
659
+ "loss": 0.5056,
660
+ "step": 19000
661
+ },
662
+ {
663
+ "epoch": 11.75,
664
+ "learning_rate": 6.908023483365949e-05,
665
+ "loss": 0.5073,
666
+ "step": 19200
667
+ },
668
+ {
669
+ "epoch": 11.87,
670
+ "learning_rate": 6.6470971950424e-05,
671
+ "loss": 0.4968,
672
+ "step": 19400
673
+ },
674
+ {
675
+ "epoch": 12.0,
676
+ "learning_rate": 6.386170906718851e-05,
677
+ "loss": 0.5009,
678
+ "step": 19600
679
+ },
680
+ {
681
+ "epoch": 12.12,
682
+ "learning_rate": 6.125244618395303e-05,
683
+ "loss": 0.4832,
684
+ "step": 19800
685
+ },
686
+ {
687
+ "epoch": 12.24,
688
+ "learning_rate": 5.864318330071754e-05,
689
+ "loss": 0.4778,
690
+ "step": 20000
691
+ },
692
+ {
693
+ "epoch": 12.24,
694
+ "eval_cer": 0.11294822712906938,
695
+ "eval_loss": 0.40570223331451416,
696
+ "eval_runtime": 1612.661,
697
+ "eval_samples_per_second": 8.119,
698
+ "eval_steps_per_second": 0.508,
699
+ "step": 20000
700
+ },
701
+ {
702
+ "epoch": 12.36,
703
+ "learning_rate": 5.6033920417482055e-05,
704
+ "loss": 0.4775,
705
+ "step": 20200
706
+ },
707
+ {
708
+ "epoch": 12.49,
709
+ "learning_rate": 5.342465753424657e-05,
710
+ "loss": 0.4855,
711
+ "step": 20400
712
+ },
713
+ {
714
+ "epoch": 12.61,
715
+ "learning_rate": 5.081539465101108e-05,
716
+ "loss": 0.4773,
717
+ "step": 20600
718
+ },
719
+ {
720
+ "epoch": 12.73,
721
+ "learning_rate": 4.82061317677756e-05,
722
+ "loss": 0.4745,
723
+ "step": 20800
724
+ },
725
+ {
726
+ "epoch": 12.85,
727
+ "learning_rate": 4.559686888454011e-05,
728
+ "loss": 0.48,
729
+ "step": 21000
730
+ },
731
+ {
732
+ "epoch": 12.98,
733
+ "learning_rate": 4.2987606001304625e-05,
734
+ "loss": 0.463,
735
+ "step": 21200
736
+ },
737
+ {
738
+ "epoch": 13.1,
739
+ "learning_rate": 4.037834311806914e-05,
740
+ "loss": 0.4643,
741
+ "step": 21400
742
+ },
743
+ {
744
+ "epoch": 13.22,
745
+ "learning_rate": 3.776908023483365e-05,
746
+ "loss": 0.449,
747
+ "step": 21600
748
+ },
749
+ {
750
+ "epoch": 13.34,
751
+ "learning_rate": 3.515981735159817e-05,
752
+ "loss": 0.4604,
753
+ "step": 21800
754
+ },
755
+ {
756
+ "epoch": 13.47,
757
+ "learning_rate": 3.255055446836268e-05,
758
+ "loss": 0.4571,
759
+ "step": 22000
760
+ },
761
+ {
762
+ "epoch": 13.47,
763
+ "eval_cer": 0.10995473327241098,
764
+ "eval_loss": 0.3945465385913849,
765
+ "eval_runtime": 1610.0901,
766
+ "eval_samples_per_second": 8.132,
767
+ "eval_steps_per_second": 0.509,
768
+ "step": 22000
769
+ },
770
+ {
771
+ "epoch": 13.59,
772
+ "learning_rate": 2.99412915851272e-05,
773
+ "loss": 0.4539,
774
+ "step": 22200
775
+ },
776
+ {
777
+ "epoch": 13.71,
778
+ "learning_rate": 2.7332028701891712e-05,
779
+ "loss": 0.4569,
780
+ "step": 22400
781
+ },
782
+ {
783
+ "epoch": 13.83,
784
+ "learning_rate": 2.4722765818656226e-05,
785
+ "loss": 0.4533,
786
+ "step": 22600
787
+ },
788
+ {
789
+ "epoch": 13.96,
790
+ "learning_rate": 2.211350293542074e-05,
791
+ "loss": 0.4545,
792
+ "step": 22800
793
+ },
794
+ {
795
+ "epoch": 14.08,
796
+ "learning_rate": 1.9504240052185254e-05,
797
+ "loss": 0.447,
798
+ "step": 23000
799
+ },
800
+ {
801
+ "epoch": 14.2,
802
+ "learning_rate": 1.6894977168949768e-05,
803
+ "loss": 0.4443,
804
+ "step": 23200
805
+ },
806
+ {
807
+ "epoch": 14.32,
808
+ "learning_rate": 1.4285714285714284e-05,
809
+ "loss": 0.4423,
810
+ "step": 23400
811
+ },
812
+ {
813
+ "epoch": 14.45,
814
+ "learning_rate": 1.1676451402478798e-05,
815
+ "loss": 0.4434,
816
+ "step": 23600
817
+ },
818
+ {
819
+ "epoch": 14.57,
820
+ "learning_rate": 9.067188519243312e-06,
821
+ "loss": 0.438,
822
+ "step": 23800
823
+ },
824
+ {
825
+ "epoch": 14.69,
826
+ "learning_rate": 6.4579256360078264e-06,
827
+ "loss": 0.4388,
828
+ "step": 24000
829
+ },
830
+ {
831
+ "epoch": 14.69,
832
+ "eval_cer": 0.10809973860058716,
833
+ "eval_loss": 0.38906005024909973,
834
+ "eval_runtime": 1607.4288,
835
+ "eval_samples_per_second": 8.145,
836
+ "eval_steps_per_second": 0.51,
837
+ "step": 24000
838
+ },
839
+ {
840
+ "epoch": 14.81,
841
+ "learning_rate": 3.848662752772341e-06,
842
+ "loss": 0.4406,
843
+ "step": 24200
844
+ },
845
+ {
846
+ "epoch": 14.93,
847
+ "learning_rate": 1.2393998695368556e-06,
848
+ "loss": 0.449,
849
+ "step": 24400
850
+ },
851
+ {
852
+ "epoch": 14.99,
853
+ "step": 24495,
854
+ "total_flos": 1.7220710227304147e+20,
855
+ "train_loss": 1.1506779817299688,
856
+ "train_runtime": 54167.2991,
857
+ "train_samples_per_second": 28.951,
858
+ "train_steps_per_second": 0.452
859
+ }
860
+ ],
861
+ "logging_steps": 200,
862
+ "max_steps": 24495,
863
+ "num_train_epochs": 15,
864
+ "save_steps": 2000,
865
+ "total_flos": 1.7220710227304147e+20,
866
+ "trial_name": null,
867
+ "trial_params": null
868
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2a53791a3905b5426f4f4a28dc53b7bf0cdde7382afe6d18a7f69f8c493b179
3
+ size 4472
vocab.json ADDED
@@ -0,0 +1,1744 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "&": 1,
3
+ "0": 2,
4
+ "1": 3,
5
+ "2": 4,
6
+ "3": 5,
7
+ "4": 6,
8
+ "5": 7,
9
+ "6": 8,
10
+ "7": 9,
11
+ "8": 10,
12
+ "9": 11,
13
+ "[PAD]": 1741,
14
+ "[UNK]": 1740,
15
+ "\\": 12,
16
+ "a": 13,
17
+ "b": 14,
18
+ "c": 15,
19
+ "d": 16,
20
+ "e": 17,
21
+ "f": 18,
22
+ "g": 19,
23
+ "h": 20,
24
+ "i": 21,
25
+ "j": 22,
26
+ "k": 23,
27
+ "l": 24,
28
+ "m": 25,
29
+ "n": 26,
30
+ "o": 27,
31
+ "p": 28,
32
+ "q": 29,
33
+ "r": 30,
34
+ "s": 31,
35
+ "t": 32,
36
+ "u": 33,
37
+ "v": 34,
38
+ "w": 35,
39
+ "x": 36,
40
+ "y": 37,
41
+ "z": 38,
42
+ "|": 0,
43
+ "°": 39,
44
+ "μ": 40,
45
+ "ᆫ": 41,
46
+ "ㄱ": 42,
47
+ "ㄴ": 43,
48
+ "ㄷ": 44,
49
+ "ㄹ": 45,
50
+ "ㅁ": 46,
51
+ "ㅂ": 47,
52
+ "ㅅ": 48,
53
+ "ㅇ": 49,
54
+ "ㅈ": 50,
55
+ "ㅜ": 51,
56
+ "ㅠ": 52,
57
+ "ㅡ": 53,
58
+ "가": 54,
59
+ "각": 55,
60
+ "간": 56,
61
+ "갇": 57,
62
+ "갈": 58,
63
+ "감": 59,
64
+ "갑": 60,
65
+ "값": 61,
66
+ "갓": 62,
67
+ "갔": 63,
68
+ "강": 64,
69
+ "갖": 65,
70
+ "같": 66,
71
+ "갚": 67,
72
+ "갛": 68,
73
+ "개": 69,
74
+ "객": 70,
75
+ "갠": 71,
76
+ "갤": 72,
77
+ "갬": 73,
78
+ "갭": 74,
79
+ "갯": 75,
80
+ "갰": 76,
81
+ "갱": 77,
82
+ "갸": 78,
83
+ "걀": 79,
84
+ "걍": 80,
85
+ "걔": 81,
86
+ "걘": 82,
87
+ "거": 83,
88
+ "걱": 84,
89
+ "건": 85,
90
+ "걷": 86,
91
+ "걸": 87,
92
+ "검": 88,
93
+ "겁": 89,
94
+ "것": 90,
95
+ "겄": 91,
96
+ "겉": 92,
97
+ "게": 93,
98
+ "겐": 94,
99
+ "겔": 95,
100
+ "겜": 96,
101
+ "겟": 97,
102
+ "겠": 98,
103
+ "겨": 99,
104
+ "격": 100,
105
+ "겪": 101,
106
+ "견": 102,
107
+ "결": 103,
108
+ "겸": 104,
109
+ "겹": 105,
110
+ "겼": 106,
111
+ "경": 107,
112
+ "곁": 108,
113
+ "계": 109,
114
+ "곗": 110,
115
+ "고": 111,
116
+ "곡": 112,
117
+ "곤": 113,
118
+ "곧": 114,
119
+ "골": 115,
120
+ "곪": 116,
121
+ "곰": 117,
122
+ "곱": 118,
123
+ "곳": 119,
124
+ "공": 120,
125
+ "곶": 121,
126
+ "과": 122,
127
+ "곽": 123,
128
+ "관": 124,
129
+ "괄": 125,
130
+ "괍": 126,
131
+ "광": 127,
132
+ "괘": 128,
133
+ "괜": 129,
134
+ "괭": 130,
135
+ "괴": 131,
136
+ "굉": 132,
137
+ "교": 133,
138
+ "굣": 134,
139
+ "구": 135,
140
+ "국": 136,
141
+ "군": 137,
142
+ "굳": 138,
143
+ "굴": 139,
144
+ "굵": 140,
145
+ "굶": 141,
146
+ "굽": 142,
147
+ "굿": 143,
148
+ "궁": 144,
149
+ "궂": 145,
150
+ "궈": 146,
151
+ "권": 147,
152
+ "궐": 148,
153
+ "궜": 149,
154
+ "귀": 150,
155
+ "귄": 151,
156
+ "귈": 152,
157
+ "귓": 153,
158
+ "규": 154,
159
+ "균": 155,
160
+ "귤": 156,
161
+ "그": 157,
162
+ "극": 158,
163
+ "근": 159,
164
+ "귿": 160,
165
+ "글": 161,
166
+ "긁": 162,
167
+ "금": 163,
168
+ "급": 164,
169
+ "긋": 165,
170
+ "긍": 166,
171
+ "긑": 167,
172
+ "기": 168,
173
+ "긴": 169,
174
+ "길": 170,
175
+ "김": 171,
176
+ "깁": 172,
177
+ "깃": 173,
178
+ "깄": 174,
179
+ "깅": 175,
180
+ "깊": 176,
181
+ "까": 177,
182
+ "깍": 178,
183
+ "깎": 179,
184
+ "깐": 180,
185
+ "깔": 181,
186
+ "깜": 182,
187
+ "깝": 183,
188
+ "깟": 184,
189
+ "깠": 185,
190
+ "깡": 186,
191
+ "깥": 187,
192
+ "깨": 188,
193
+ "깬": 189,
194
+ "깰": 190,
195
+ "깻": 191,
196
+ "깼": 192,
197
+ "깽": 193,
198
+ "꺼": 194,
199
+ "꺽": 195,
200
+ "꺾": 196,
201
+ "껀": 197,
202
+ "껄": 198,
203
+ "껌": 199,
204
+ "껍": 200,
205
+ "껏": 201,
206
+ "껐": 202,
207
+ "껑": 203,
208
+ "께": 204,
209
+ "껩": 205,
210
+ "껴": 206,
211
+ "꼈": 207,
212
+ "꼐": 208,
213
+ "꼬": 209,
214
+ "꼭": 210,
215
+ "꼰": 211,
216
+ "꼴": 212,
217
+ "꼼": 213,
218
+ "꼽": 214,
219
+ "꽁": 215,
220
+ "꽂": 216,
221
+ "꽃": 217,
222
+ "꽈": 218,
223
+ "꽉": 219,
224
+ "꽌": 220,
225
+ "꽜": 221,
226
+ "꽝": 222,
227
+ "꽤": 223,
228
+ "꽥": 224,
229
+ "꾀": 225,
230
+ "꾸": 226,
231
+ "꾹": 227,
232
+ "꾼": 228,
233
+ "꿀": 229,
234
+ "꿇": 230,
235
+ "꿈": 231,
236
+ "꿉": 232,
237
+ "꿍": 233,
238
+ "꿔": 234,
239
+ "꿨": 235,
240
+ "꿰": 236,
241
+ "뀌": 237,
242
+ "뀐": 238,
243
+ "뀔": 239,
244
+ "뀝": 240,
245
+ "뀰": 241,
246
+ "끄": 242,
247
+ "끅": 243,
248
+ "끈": 244,
249
+ "끊": 245,
250
+ "끌": 246,
251
+ "끓": 247,
252
+ "끔": 248,
253
+ "끕": 249,
254
+ "끗": 250,
255
+ "끙": 251,
256
+ "끝": 252,
257
+ "끼": 253,
258
+ "끽": 254,
259
+ "낀": 255,
260
+ "낄": 256,
261
+ "낌": 257,
262
+ "낍": 258,
263
+ "낑": 259,
264
+ "나": 260,
265
+ "낙": 261,
266
+ "낚": 262,
267
+ "난": 263,
268
+ "날": 264,
269
+ "낡": 265,
270
+ "남": 266,
271
+ "납": 267,
272
+ "낫": 268,
273
+ "났": 269,
274
+ "낭": 270,
275
+ "낮": 271,
276
+ "낯": 272,
277
+ "낱": 273,
278
+ "낳": 274,
279
+ "내": 275,
280
+ "낵": 276,
281
+ "낸": 277,
282
+ "낼": 278,
283
+ "냄": 279,
284
+ "냅": 280,
285
+ "냇": 281,
286
+ "냈": 282,
287
+ "냉": 283,
288
+ "냐": 284,
289
+ "냑": 285,
290
+ "냠": 286,
291
+ "냥": 287,
292
+ "너": 288,
293
+ "넉": 289,
294
+ "넌": 290,
295
+ "널": 291,
296
+ "넓": 292,
297
+ "넘": 293,
298
+ "넛": 294,
299
+ "넣": 295,
300
+ "네": 296,
301
+ "넥": 297,
302
+ "넬": 298,
303
+ "넵": 299,
304
+ "넷": 300,
305
+ "넹": 301,
306
+ "녀": 302,
307
+ "녁": 303,
308
+ "년": 304,
309
+ "념": 305,
310
+ "녔": 306,
311
+ "녕": 307,
312
+ "녜": 308,
313
+ "노": 309,
314
+ "녹": 310,
315
+ "논": 311,
316
+ "놀": 312,
317
+ "놈": 313,
318
+ "놉": 314,
319
+ "농": 315,
320
+ "높": 316,
321
+ "놓": 317,
322
+ "놔": 318,
323
+ "놨": 319,
324
+ "뇌": 320,
325
+ "뇨": 321,
326
+ "누": 322,
327
+ "눅": 323,
328
+ "눈": 324,
329
+ "눌": 325,
330
+ "눔": 326,
331
+ "눕": 327,
332
+ "눗": 328,
333
+ "눠": 329,
334
+ "눴": 330,
335
+ "뉘": 331,
336
+ "뉜": 332,
337
+ "뉠": 333,
338
+ "뉩": 334,
339
+ "뉴": 335,
340
+ "늄": 336,
341
+ "느": 337,
342
+ "늑": 338,
343
+ "는": 339,
344
+ "늘": 340,
345
+ "늙": 341,
346
+ "늠": 342,
347
+ "늡": 343,
348
+ "능": 344,
349
+ "늦": 345,
350
+ "늬": 346,
351
+ "니": 347,
352
+ "닉": 348,
353
+ "닌": 349,
354
+ "닐": 350,
355
+ "님": 351,
356
+ "닙": 352,
357
+ "닛": 353,
358
+ "닝": 354,
359
+ "다": 355,
360
+ "닥": 356,
361
+ "닦": 357,
362
+ "단": 358,
363
+ "닫": 359,
364
+ "달": 360,
365
+ "닭": 361,
366
+ "닮": 362,
367
+ "닳": 363,
368
+ "담": 364,
369
+ "답": 365,
370
+ "닷": 366,
371
+ "당": 367,
372
+ "닿": 368,
373
+ "대": 369,
374
+ "댁": 370,
375
+ "댄": 371,
376
+ "댈": 372,
377
+ "댐": 373,
378
+ "댑": 374,
379
+ "댓": 375,
380
+ "댔": 376,
381
+ "댕": 377,
382
+ "댜": 378,
383
+ "더": 379,
384
+ "덕": 380,
385
+ "던": 381,
386
+ "덜": 382,
387
+ "덟": 383,
388
+ "덤": 384,
389
+ "덥": 385,
390
+ "덧": 386,
391
+ "덩": 387,
392
+ "덮": 388,
393
+ "데": 389,
394
+ "덱": 390,
395
+ "덴": 391,
396
+ "델": 392,
397
+ "뎅": 393,
398
+ "뎌": 394,
399
+ "뎠": 395,
400
+ "도": 396,
401
+ "독": 397,
402
+ "돈": 398,
403
+ "돋": 399,
404
+ "돌": 400,
405
+ "돔": 401,
406
+ "돕": 402,
407
+ "돗": 403,
408
+ "동": 404,
409
+ "돼": 405,
410
+ "됍": 406,
411
+ "됐": 407,
412
+ "되": 408,
413
+ "된": 409,
414
+ "될": 410,
415
+ "됨": 411,
416
+ "됩": 412,
417
+ "됬": 413,
418
+ "두": 414,
419
+ "둑": 415,
420
+ "둔": 416,
421
+ "둘": 417,
422
+ "둠": 418,
423
+ "둡": 419,
424
+ "둣": 420,
425
+ "둥": 421,
426
+ "둬": 422,
427
+ "뒀": 423,
428
+ "뒈": 424,
429
+ "뒤": 425,
430
+ "뒨": 426,
431
+ "뒷": 427,
432
+ "뒹": 428,
433
+ "듀": 429,
434
+ "드": 430,
435
+ "득": 431,
436
+ "든": 432,
437
+ "듣": 433,
438
+ "들": 434,
439
+ "듦": 435,
440
+ "듬": 436,
441
+ "듭": 437,
442
+ "듯": 438,
443
+ "등": 439,
444
+ "디": 440,
445
+ "딕": 441,
446
+ "딘": 442,
447
+ "딜": 443,
448
+ "딤": 444,
449
+ "딥": 445,
450
+ "딧": 446,
451
+ "딨": 447,
452
+ "딩": 448,
453
+ "딪": 449,
454
+ "따": 450,
455
+ "딱": 451,
456
+ "딴": 452,
457
+ "딸": 453,
458
+ "땀": 454,
459
+ "땁": 455,
460
+ "땃": 456,
461
+ "땄": 457,
462
+ "땅": 458,
463
+ "땋": 459,
464
+ "때": 460,
465
+ "땍": 461,
466
+ "땐": 462,
467
+ "땔": 463,
468
+ "땜": 464,
469
+ "땟": 465,
470
+ "땠": 466,
471
+ "땡": 467,
472
+ "떄": 468,
473
+ "떠": 469,
474
+ "떡": 470,
475
+ "떤": 471,
476
+ "떨": 472,
477
+ "떰": 473,
478
+ "떱": 474,
479
+ "떳": 475,
480
+ "떴": 476,
481
+ "떻": 477,
482
+ "떼": 478,
483
+ "떽": 479,
484
+ "뗀": 480,
485
+ "뗄": 481,
486
+ "뗌": 482,
487
+ "뗍": 483,
488
+ "뗐": 484,
489
+ "뗴": 485,
490
+ "또": 486,
491
+ "똑": 487,
492
+ "똘": 488,
493
+ "똣": 489,
494
+ "똥": 490,
495
+ "뚜": 491,
496
+ "뚝": 492,
497
+ "뚤": 493,
498
+ "뚫": 494,
499
+ "뚱": 495,
500
+ "뛌": 496,
501
+ "뛰": 497,
502
+ "뛴": 498,
503
+ "뛸": 499,
504
+ "뜄": 500,
505
+ "뜨": 501,
506
+ "뜩": 502,
507
+ "뜬": 503,
508
+ "뜯": 504,
509
+ "뜰": 505,
510
+ "뜸": 506,
511
+ "뜹": 507,
512
+ "뜻": 508,
513
+ "띄": 509,
514
+ "띈": 510,
515
+ "띌": 511,
516
+ "띔": 512,
517
+ "띕": 513,
518
+ "띠": 514,
519
+ "띡": 515,
520
+ "띨": 516,
521
+ "띵": 517,
522
+ "라": 518,
523
+ "락": 519,
524
+ "란": 520,
525
+ "랄": 521,
526
+ "람": 522,
527
+ "랍": 523,
528
+ "랏": 524,
529
+ "랐": 525,
530
+ "랑": 526,
531
+ "랖": 527,
532
+ "랗": 528,
533
+ "래": 529,
534
+ "랙": 530,
535
+ "랜": 531,
536
+ "랠": 532,
537
+ "램": 533,
538
+ "랩": 534,
539
+ "랫": 535,
540
+ "랬": 536,
541
+ "랭": 537,
542
+ "랴": 538,
543
+ "략": 539,
544
+ "량": 540,
545
+ "러": 541,
546
+ "럭": 542,
547
+ "런": 543,
548
+ "럴": 544,
549
+ "럼": 545,
550
+ "럽": 546,
551
+ "럿": 547,
552
+ "렀": 548,
553
+ "렁": 549,
554
+ "렇": 550,
555
+ "레": 551,
556
+ "렉": 552,
557
+ "렌": 553,
558
+ "렐": 554,
559
+ "렘": 555,
560
+ "렙": 556,
561
+ "렛": 557,
562
+ "렜": 558,
563
+ "려": 559,
564
+ "력": 560,
565
+ "련": 561,
566
+ "렬": 562,
567
+ "렴": 563,
568
+ "렵": 564,
569
+ "렷": 565,
570
+ "렸": 566,
571
+ "령": 567,
572
+ "렿": 568,
573
+ "례": 569,
574
+ "로": 570,
575
+ "록": 571,
576
+ "론": 572,
577
+ "롤": 573,
578
+ "롬": 574,
579
+ "롭": 575,
580
+ "롯": 576,
581
+ "롱": 577,
582
+ "롷": 578,
583
+ "롸": 579,
584
+ "뢰": 580,
585
+ "료": 581,
586
+ "룡": 582,
587
+ "루": 583,
588
+ "룩": 584,
589
+ "룬": 585,
590
+ "룰": 586,
591
+ "룸": 587,
592
+ "룹": 588,
593
+ "룻": 589,
594
+ "룽": 590,
595
+ "뤄": 591,
596
+ "뤘": 592,
597
+ "류": 593,
598
+ "륙": 594,
599
+ "륜": 595,
600
+ "률": 596,
601
+ "륨": 597,
602
+ "륭": 598,
603
+ "르": 599,
604
+ "륵": 600,
605
+ "른": 601,
606
+ "를": 602,
607
+ "름": 603,
608
+ "릅": 604,
609
+ "릇": 605,
610
+ "릉": 606,
611
+ "릎": 607,
612
+ "리": 608,
613
+ "릭": 609,
614
+ "린": 610,
615
+ "릴": 611,
616
+ "림": 612,
617
+ "립": 613,
618
+ "릿": 614,
619
+ "맀": 615,
620
+ "링": 616,
621
+ "마": 617,
622
+ "막": 618,
623
+ "만": 619,
624
+ "많": 620,
625
+ "맏": 621,
626
+ "말": 622,
627
+ "맑": 623,
628
+ "맘": 624,
629
+ "맙": 625,
630
+ "맛": 626,
631
+ "맜": 627,
632
+ "망": 628,
633
+ "맞": 629,
634
+ "맡": 630,
635
+ "맣": 631,
636
+ "매": 632,
637
+ "맥": 633,
638
+ "맨": 634,
639
+ "맴": 635,
640
+ "맵": 636,
641
+ "맷": 637,
642
+ "맸": 638,
643
+ "맹": 639,
644
+ "맺": 640,
645
+ "맻": 641,
646
+ "머": 642,
647
+ "먹": 643,
648
+ "먼": 644,
649
+ "멀": 645,
650
+ "멈": 646,
651
+ "멉": 647,
652
+ "멋": 648,
653
+ "멍": 649,
654
+ "멓": 650,
655
+ "메": 651,
656
+ "멕": 652,
657
+ "멘": 653,
658
+ "멜": 654,
659
+ "멤": 655,
660
+ "멧": 656,
661
+ "멨": 657,
662
+ "멩": 658,
663
+ "며": 659,
664
+ "멱": 660,
665
+ "면": 661,
666
+ "멸": 662,
667
+ "몄": 663,
668
+ "명": 664,
669
+ "몇": 665,
670
+ "모": 666,
671
+ "목": 667,
672
+ "몫": 668,
673
+ "몬": 669,
674
+ "몰": 670,
675
+ "몸": 671,
676
+ "몹": 672,
677
+ "못": 673,
678
+ "몽": 674,
679
+ "묘": 675,
680
+ "무": 676,
681
+ "묵": 677,
682
+ "묶": 678,
683
+ "문": 679,
684
+ "묻": 680,
685
+ "물": 681,
686
+ "묽": 682,
687
+ "뭅": 683,
688
+ "뭇": 684,
689
+ "뭉": 685,
690
+ "뭏": 686,
691
+ "뭐": 687,
692
+ "뭔": 688,
693
+ "뭘": 689,
694
+ "뭡": 690,
695
+ "��": 691,
696
+ "뮤": 692,
697
+ "뮨": 693,
698
+ "므": 694,
699
+ "믄": 695,
700
+ "믈": 696,
701
+ "미": 697,
702
+ "믹": 698,
703
+ "민": 699,
704
+ "믿": 700,
705
+ "밀": 701,
706
+ "밉": 702,
707
+ "밋": 703,
708
+ "밌": 704,
709
+ "밍": 705,
710
+ "및": 706,
711
+ "밑": 707,
712
+ "바": 708,
713
+ "박": 709,
714
+ "밖": 710,
715
+ "반": 711,
716
+ "받": 712,
717
+ "발": 713,
718
+ "밝": 714,
719
+ "밟": 715,
720
+ "밤": 716,
721
+ "밥": 717,
722
+ "밧": 718,
723
+ "방": 719,
724
+ "밭": 720,
725
+ "배": 721,
726
+ "백": 722,
727
+ "밴": 723,
728
+ "밸": 724,
729
+ "뱀": 725,
730
+ "뱁": 726,
731
+ "뱃": 727,
732
+ "뱄": 728,
733
+ "뱅": 729,
734
+ "뱉": 730,
735
+ "버": 731,
736
+ "벅": 732,
737
+ "번": 733,
738
+ "벋": 734,
739
+ "벌": 735,
740
+ "범": 736,
741
+ "법": 737,
742
+ "벗": 738,
743
+ "벙": 739,
744
+ "벚": 740,
745
+ "베": 741,
746
+ "벡": 742,
747
+ "벤": 743,
748
+ "벨": 744,
749
+ "벰": 745,
750
+ "벳": 746,
751
+ "벼": 747,
752
+ "벽": 748,
753
+ "변": 749,
754
+ "별": 750,
755
+ "볌": 751,
756
+ "볍": 752,
757
+ "볐": 753,
758
+ "병": 754,
759
+ "볕": 755,
760
+ "보": 756,
761
+ "복": 757,
762
+ "볶": 758,
763
+ "본": 759,
764
+ "볼": 760,
765
+ "봄": 761,
766
+ "봅": 762,
767
+ "봇": 763,
768
+ "봉": 764,
769
+ "봐": 765,
770
+ "봤": 766,
771
+ "봬": 767,
772
+ "뵀": 768,
773
+ "뵈": 769,
774
+ "뵌": 770,
775
+ "뵐": 771,
776
+ "뵙": 772,
777
+ "부": 773,
778
+ "북": 774,
779
+ "분": 775,
780
+ "붇": 776,
781
+ "불": 777,
782
+ "붉": 778,
783
+ "붐": 779,
784
+ "붑": 780,
785
+ "붓": 781,
786
+ "붕": 782,
787
+ "붙": 783,
788
+ "뷔": 784,
789
+ "뷰": 785,
790
+ "뷸": 786,
791
+ "브": 787,
792
+ "븐": 788,
793
+ "블": 789,
794
+ "비": 790,
795
+ "빅": 791,
796
+ "빈": 792,
797
+ "빌": 793,
798
+ "빔": 794,
799
+ "빕": 795,
800
+ "빗": 796,
801
+ "빙": 797,
802
+ "빚": 798,
803
+ "빛": 799,
804
+ "빠": 800,
805
+ "빡": 801,
806
+ "빤": 802,
807
+ "빨": 803,
808
+ "빳": 804,
809
+ "빴": 805,
810
+ "빵": 806,
811
+ "빻": 807,
812
+ "빼": 808,
813
+ "빽": 809,
814
+ "뺀": 810,
815
+ "뺄": 811,
816
+ "뺌": 812,
817
+ "뺍": 813,
818
+ "뺏": 814,
819
+ "뺐": 815,
820
+ "뺑": 816,
821
+ "뺘": 817,
822
+ "뺨": 818,
823
+ "뺼": 819,
824
+ "뻐": 820,
825
+ "뻑": 821,
826
+ "뻔": 822,
827
+ "뻗": 823,
828
+ "뻘": 824,
829
+ "뻣": 825,
830
+ "뻤": 826,
831
+ "뻥": 827,
832
+ "뻬": 828,
833
+ "뼀": 829,
834
+ "뼈": 830,
835
+ "뼘": 831,
836
+ "뼛": 832,
837
+ "뽀": 833,
838
+ "뽁": 834,
839
+ "뽂": 835,
840
+ "뽄": 836,
841
+ "뽈": 837,
842
+ "뽐": 838,
843
+ "뽑": 839,
844
+ "뽕": 840,
845
+ "뽜": 841,
846
+ "뽝": 842,
847
+ "뾰": 843,
848
+ "뿅": 844,
849
+ "뿌": 845,
850
+ "뿍": 846,
851
+ "뿐": 847,
852
+ "뿔": 848,
853
+ "뿜": 849,
854
+ "뿝": 850,
855
+ "뿟": 851,
856
+ "뿡": 852,
857
+ "뿨": 853,
858
+ "쁘": 854,
859
+ "쁜": 855,
860
+ "쁠": 856,
861
+ "쁨": 857,
862
+ "쁩": 858,
863
+ "삐": 859,
864
+ "삑": 860,
865
+ "삔": 861,
866
+ "삘": 862,
867
+ "삥": 863,
868
+ "사": 864,
869
+ "삭": 865,
870
+ "산": 866,
871
+ "살": 867,
872
+ "삶": 868,
873
+ "삼": 869,
874
+ "삽": 870,
875
+ "삿": 871,
876
+ "샀": 872,
877
+ "상": 873,
878
+ "새": 874,
879
+ "색": 875,
880
+ "샌": 876,
881
+ "샐": 877,
882
+ "샘": 878,
883
+ "샙": 879,
884
+ "샜": 880,
885
+ "생": 881,
886
+ "샤": 882,
887
+ "샥": 883,
888
+ "샴": 884,
889
+ "샵": 885,
890
+ "샷": 886,
891
+ "샹": 887,
892
+ "섀": 888,
893
+ "서": 889,
894
+ "석": 890,
895
+ "섞": 891,
896
+ "선": 892,
897
+ "설": 893,
898
+ "섬": 894,
899
+ "섭": 895,
900
+ "섯": 896,
901
+ "섰": 897,
902
+ "성": 898,
903
+ "세": 899,
904
+ "섹": 900,
905
+ "센": 901,
906
+ "셀": 902,
907
+ "셈": 903,
908
+ "셉": 904,
909
+ "셋": 905,
910
+ "셔": 906,
911
+ "션": 907,
912
+ "셜": 908,
913
+ "셧": 909,
914
+ "셨": 910,
915
+ "셰": 911,
916
+ "솁": 912,
917
+ "소": 913,
918
+ "속": 914,
919
+ "손": 915,
920
+ "솔": 916,
921
+ "솜": 917,
922
+ "솝": 918,
923
+ "솟": 919,
924
+ "송": 920,
925
+ "솥": 921,
926
+ "솨": 922,
927
+ "쇄": 923,
928
+ "쇠": 924,
929
+ "쇼": 925,
930
+ "쇽": 926,
931
+ "숄": 927,
932
+ "숍": 928,
933
+ "숏": 929,
934
+ "숑": 930,
935
+ "수": 931,
936
+ "숙": 932,
937
+ "순": 933,
938
+ "숟": 934,
939
+ "술": 935,
940
+ "숨": 936,
941
+ "숩": 937,
942
+ "숫": 938,
943
+ "숭": 939,
944
+ "숯": 940,
945
+ "숱": 941,
946
+ "숲": 942,
947
+ "숴": 943,
948
+ "쉈": 944,
949
+ "쉐": 945,
950
+ "쉘": 946,
951
+ "쉣": 947,
952
+ "쉬": 948,
953
+ "쉰": 949,
954
+ "쉴": 950,
955
+ "쉼": 951,
956
+ "쉽": 952,
957
+ "슈": 953,
958
+ "슉": 954,
959
+ "슐": 955,
960
+ "슘": 956,
961
+ "슛": 957,
962
+ "슝": 958,
963
+ "스": 959,
964
+ "슥": 960,
965
+ "슨": 961,
966
+ "슬": 962,
967
+ "슴": 963,
968
+ "습": 964,
969
+ "슷": 965,
970
+ "승": 966,
971
+ "시": 967,
972
+ "식": 968,
973
+ "신": 969,
974
+ "싣": 970,
975
+ "실": 971,
976
+ "싫": 972,
977
+ "심": 973,
978
+ "십": 974,
979
+ "싯": 975,
980
+ "싰": 976,
981
+ "싱": 977,
982
+ "싶": 978,
983
+ "싸": 979,
984
+ "싹": 980,
985
+ "싼": 981,
986
+ "쌀": 982,
987
+ "쌈": 983,
988
+ "쌉": 984,
989
+ "쌌": 985,
990
+ "쌍": 986,
991
+ "쌓": 987,
992
+ "쌔": 988,
993
+ "쌜": 989,
994
+ "쌤": 990,
995
+ "쌩": 991,
996
+ "쌰": 992,
997
+ "써": 993,
998
+ "썩": 994,
999
+ "썪": 995,
1000
+ "썬": 996,
1001
+ "썰": 997,
1002
+ "썸": 998,
1003
+ "썹": 999,
1004
+ "썻": 1000,
1005
+ "썼": 1001,
1006
+ "썽": 1002,
1007
+ "쎄": 1003,
1008
+ "쎅": 1004,
1009
+ "쎈": 1005,
1010
+ "쎕": 1006,
1011
+ "쎼": 1007,
1012
+ "쏘": 1008,
1013
+ "쏙": 1009,
1014
+ "쏜": 1010,
1015
+ "쏟": 1011,
1016
+ "쏠": 1012,
1017
+ "쏩": 1013,
1018
+ "쏭": 1014,
1019
+ "쏴": 1015,
1020
+ "쏵": 1016,
1021
+ "쐈": 1017,
1022
+ "쐬": 1018,
1023
+ "쐴": 1019,
1024
+ "쑈": 1020,
1025
+ "쑉": 1021,
1026
+ "쑐": 1022,
1027
+ "쑤": 1023,
1028
+ "쑥": 1024,
1029
+ "쓰": 1025,
1030
+ "쓱": 1026,
1031
+ "쓴": 1027,
1032
+ "쓸": 1028,
1033
+ "쓽": 1029,
1034
+ "씀": 1030,
1035
+ "씁": 1031,
1036
+ "씌": 1032,
1037
+ "씐": 1033,
1038
+ "씨": 1034,
1039
+ "씩": 1035,
1040
+ "씬": 1036,
1041
+ "씰": 1037,
1042
+ "씸": 1038,
1043
+ "씹": 1039,
1044
+ "씻": 1040,
1045
+ "씼": 1041,
1046
+ "씽": 1042,
1047
+ "아": 1043,
1048
+ "악": 1044,
1049
+ "안": 1045,
1050
+ "앉": 1046,
1051
+ "않": 1047,
1052
+ "알": 1048,
1053
+ "앎": 1049,
1054
+ "앓": 1050,
1055
+ "암": 1051,
1056
+ "압": 1052,
1057
+ "앗": 1053,
1058
+ "았": 1054,
1059
+ "앙": 1055,
1060
+ "앞": 1056,
1061
+ "애": 1057,
1062
+ "액": 1058,
1063
+ "앤": 1059,
1064
+ "앨": 1060,
1065
+ "앰": 1061,
1066
+ "앱": 1062,
1067
+ "앴": 1063,
1068
+ "앵": 1064,
1069
+ "야": 1065,
1070
+ "약": 1066,
1071
+ "얀": 1067,
1072
+ "얄": 1068,
1073
+ "얇": 1069,
1074
+ "얌": 1070,
1075
+ "얍": 1071,
1076
+ "얏": 1072,
1077
+ "양": 1073,
1078
+ "얕": 1074,
1079
+ "얗": 1075,
1080
+ "얘": 1076,
1081
+ "얜": 1077,
1082
+ "얬": 1078,
1083
+ "얳": 1079,
1084
+ "어": 1080,
1085
+ "억": 1081,
1086
+ "언": 1082,
1087
+ "얹": 1083,
1088
+ "얻": 1084,
1089
+ "얼": 1085,
1090
+ "엄": 1086,
1091
+ "업": 1087,
1092
+ "없": 1088,
1093
+ "엇": 1089,
1094
+ "었": 1090,
1095
+ "엉": 1091,
1096
+ "엊": 1092,
1097
+ "엌": 1093,
1098
+ "엍": 1094,
1099
+ "엎": 1095,
1100
+ "에": 1096,
1101
+ "엑": 1097,
1102
+ "엔": 1098,
1103
+ "엘": 1099,
1104
+ "엠": 1100,
1105
+ "엣": 1101,
1106
+ "엥": 1102,
1107
+ "여": 1103,
1108
+ "역": 1104,
1109
+ "엮": 1105,
1110
+ "연": 1106,
1111
+ "열": 1107,
1112
+ "염": 1108,
1113
+ "엽": 1109,
1114
+ "엿": 1110,
1115
+ "였": 1111,
1116
+ "영": 1112,
1117
+ "옅": 1113,
1118
+ "옆": 1114,
1119
+ "옇": 1115,
1120
+ "예": 1116,
1121
+ "옐": 1117,
1122
+ "옘": 1118,
1123
+ "옙": 1119,
1124
+ "옛": 1120,
1125
+ "옜": 1121,
1126
+ "오": 1122,
1127
+ "옥": 1123,
1128
+ "온": 1124,
1129
+ "올": 1125,
1130
+ "옮": 1126,
1131
+ "옳": 1127,
1132
+ "옴": 1128,
1133
+ "옵": 1129,
1134
+ "옷": 1130,
1135
+ "옹": 1131,
1136
+ "옽": 1132,
1137
+ "와": 1133,
1138
+ "왁": 1134,
1139
+ "완": 1135,
1140
+ "왈": 1136,
1141
+ "왓": 1137,
1142
+ "왔": 1138,
1143
+ "왕": 1139,
1144
+ "왜": 1140,
1145
+ "왠": 1141,
1146
+ "왯": 1142,
1147
+ "외": 1143,
1148
+ "왼": 1144,
1149
+ "요": 1145,
1150
+ "욕": 1146,
1151
+ "욘": 1147,
1152
+ "욜": 1148,
1153
+ "욤": 1149,
1154
+ "용": 1150,
1155
+ "우": 1151,
1156
+ "욱": 1152,
1157
+ "운": 1153,
1158
+ "울": 1154,
1159
+ "움": 1155,
1160
+ "웁": 1156,
1161
+ "웃": 1157,
1162
+ "웅": 1158,
1163
+ "워": 1159,
1164
+ "웍": 1160,
1165
+ "원": 1161,
1166
+ "월": 1162,
1167
+ "웠": 1163,
1168
+ "웡": 1164,
1169
+ "웨": 1165,
1170
+ "웬": 1166,
1171
+ "웰": 1167,
1172
+ "웸": 1168,
1173
+ "웹": 1169,
1174
+ "웻": 1170,
1175
+ "위": 1171,
1176
+ "윅": 1172,
1177
+ "윈": 1173,
1178
+ "윌": 1174,
1179
+ "윕": 1175,
1180
+ "윗": 1176,
1181
+ "윙": 1177,
1182
+ "유": 1178,
1183
+ "육": 1179,
1184
+ "윤": 1180,
1185
+ "율": 1181,
1186
+ "융": 1182,
1187
+ "윷": 1183,
1188
+ "으": 1184,
1189
+ "윽": 1185,
1190
+ "은": 1186,
1191
+ "을": 1187,
1192
+ "읊": 1188,
1193
+ "음": 1189,
1194
+ "읍": 1190,
1195
+ "읎": 1191,
1196
+ "읏": 1192,
1197
+ "응": 1193,
1198
+ "읓": 1194,
1199
+ "의": 1195,
1200
+ "이": 1196,
1201
+ "익": 1197,
1202
+ "인": 1198,
1203
+ "일": 1199,
1204
+ "읽": 1200,
1205
+ "잃": 1201,
1206
+ "임": 1202,
1207
+ "입": 1203,
1208
+ "잇": 1204,
1209
+ "있": 1205,
1210
+ "잉": 1206,
1211
+ "잊": 1207,
1212
+ "잌": 1208,
1213
+ "잎": 1209,
1214
+ "자": 1210,
1215
+ "작": 1211,
1216
+ "잔": 1212,
1217
+ "잖": 1213,
1218
+ "잘": 1214,
1219
+ "잠": 1215,
1220
+ "잡": 1216,
1221
+ "잣": 1217,
1222
+ "잤": 1218,
1223
+ "장": 1219,
1224
+ "잦": 1220,
1225
+ "재": 1221,
1226
+ "잭": 1222,
1227
+ "잰": 1223,
1228
+ "잴": 1224,
1229
+ "잼": 1225,
1230
+ "잽": 1226,
1231
+ "잿": 1227,
1232
+ "쟀": 1228,
1233
+ "쟁": 1229,
1234
+ "쟈": 1230,
1235
+ "쟤": 1231,
1236
+ "쟨": 1232,
1237
+ "저": 1233,
1238
+ "적": 1234,
1239
+ "전": 1235,
1240
+ "절": 1236,
1241
+ "젊": 1237,
1242
+ "점": 1238,
1243
+ "접": 1239,
1244
+ "젓": 1240,
1245
+ "정": 1241,
1246
+ "젖": 1242,
1247
+ "제": 1243,
1248
+ "젝": 1244,
1249
+ "젠": 1245,
1250
+ "젤": 1246,
1251
+ "젭": 1247,
1252
+ "젯": 1248,
1253
+ "져": 1249,
1254
+ "젼": 1250,
1255
+ "졌": 1251,
1256
+ "조": 1252,
1257
+ "족": 1253,
1258
+ "존": 1254,
1259
+ "졸": 1255,
1260
+ "좀": 1256,
1261
+ "좁": 1257,
1262
+ "좃": 1258,
1263
+ "종": 1259,
1264
+ "좆": 1260,
1265
+ "좋": 1261,
1266
+ "좌": 1262,
1267
+ "좔": 1263,
1268
+ "죄": 1264,
1269
+ "죈": 1265,
1270
+ "죙": 1266,
1271
+ "죠": 1267,
1272
+ "죵": 1268,
1273
+ "주": 1269,
1274
+ "죽": 1270,
1275
+ "준": 1271,
1276
+ "줄": 1272,
1277
+ "줌": 1273,
1278
+ "줍": 1274,
1279
+ "줏": 1275,
1280
+ "중": 1276,
1281
+ "줘": 1277,
1282
+ "줬": 1278,
1283
+ "쥐": 1279,
1284
+ "쥑": 1280,
1285
+ "쥬": 1281,
1286
+ "즈": 1282,
1287
+ "즉": 1283,
1288
+ "즌": 1284,
1289
+ "즐": 1285,
1290
+ "즘": 1286,
1291
+ "즙": 1287,
1292
+ "증": 1288,
1293
+ "지": 1289,
1294
+ "직": 1290,
1295
+ "진": 1291,
1296
+ "질": 1292,
1297
+ "짐": 1293,
1298
+ "집": 1294,
1299
+ "짓": 1295,
1300
+ "징": 1296,
1301
+ "짖": 1297,
1302
+ "짙": 1298,
1303
+ "짚": 1299,
1304
+ "짜": 1300,
1305
+ "짝": 1301,
1306
+ "짠": 1302,
1307
+ "짤": 1303,
1308
+ "짧": 1304,
1309
+ "짬": 1305,
1310
+ "짭": 1306,
1311
+ "짰": 1307,
1312
+ "짱": 1308,
1313
+ "째": 1309,
1314
+ "짼": 1310,
1315
+ "쨋": 1311,
1316
+ "쨌": 1312,
1317
+ "쨍": 1313,
1318
+ "쨔": 1314,
1319
+ "쨰": 1315,
1320
+ "쩌": 1316,
1321
+ "쩍": 1317,
1322
+ "쩐": 1318,
1323
+ "쩔": 1319,
1324
+ "쩜": 1320,
1325
+ "쩝": 1321,
1326
+ "쩠": 1322,
1327
+ "쩡": 1323,
1328
+ "쩨": 1324,
1329
+ "쩬": 1325,
1330
+ "쪄": 1326,
1331
+ "쪘": 1327,
1332
+ "쪠": 1328,
1333
+ "쪼": 1329,
1334
+ "쪽": 1330,
1335
+ "쫀": 1331,
1336
+ "쫄": 1332,
1337
+ "쫌": 1333,
1338
+ "쫍": 1334,
1339
+ "쫑": 1335,
1340
+ "쫒": 1336,
1341
+ "쫓": 1337,
1342
+ "쫘": 1338,
1343
+ "쫙": 1339,
1344
+ "쬐": 1340,
1345
+ "쭈": 1341,
1346
+ "쭉": 1342,
1347
+ "쭐": 1343,
1348
+ "쭘": 1344,
1349
+ "쭝": 1345,
1350
+ "쭤": 1346,
1351
+ "쮸": 1347,
1352
+ "쯔": 1348,
1353
+ "쯤": 1349,
1354
+ "쯥": 1350,
1355
+ "찌": 1351,
1356
+ "찍": 1352,
1357
+ "찐": 1353,
1358
+ "찔": 1354,
1359
+ "찜": 1355,
1360
+ "찝": 1356,
1361
+ "찡": 1357,
1362
+ "찢": 1358,
1363
+ "차": 1359,
1364
+ "착": 1360,
1365
+ "찬": 1361,
1366
+ "찮": 1362,
1367
+ "찰": 1363,
1368
+ "참": 1364,
1369
+ "찹": 1365,
1370
+ "찼": 1366,
1371
+ "창": 1367,
1372
+ "찾": 1368,
1373
+ "채": 1369,
1374
+ "책": 1370,
1375
+ "챌": 1371,
1376
+ "챔": 1372,
1377
+ "챕": 1373,
1378
+ "챗": 1374,
1379
+ "챘": 1375,
1380
+ "챙": 1376,
1381
+ "처": 1377,
1382
+ "척": 1378,
1383
+ "천": 1379,
1384
+ "철": 1380,
1385
+ "첨": 1381,
1386
+ "첩": 1382,
1387
+ "첫": 1383,
1388
+ "청": 1384,
1389
+ "체": 1385,
1390
+ "첵": 1386,
1391
+ "첸": 1387,
1392
+ "첼": 1388,
1393
+ "쳇": 1389,
1394
+ "쳐": 1390,
1395
+ "쳤": 1391,
1396
+ "초": 1392,
1397
+ "촉": 1393,
1398
+ "촌": 1394,
1399
+ "촐": 1395,
1400
+ "촛": 1396,
1401
+ "총": 1397,
1402
+ "촤": 1398,
1403
+ "촥": 1399,
1404
+ "촬": 1400,
1405
+ "최": 1401,
1406
+ "쵸": 1402,
1407
+ "추": 1403,
1408
+ "축": 1404,
1409
+ "춘": 1405,
1410
+ "출": 1406,
1411
+ "춤": 1407,
1412
+ "춥": 1408,
1413
+ "춧": 1409,
1414
+ "충": 1410,
1415
+ "춰": 1411,
1416
+ "췄": 1412,
1417
+ "췌": 1413,
1418
+ "취": 1414,
1419
+ "츄": 1415,
1420
+ "츠": 1416,
1421
+ "측": 1417,
1422
+ "층": 1418,
1423
+ "치": 1419,
1424
+ "칙": 1420,
1425
+ "친": 1421,
1426
+ "칠": 1422,
1427
+ "침": 1423,
1428
+ "칩": 1424,
1429
+ "칫": 1425,
1430
+ "칬": 1426,
1431
+ "칭": 1427,
1432
+ "카": 1428,
1433
+ "칵": 1429,
1434
+ "칸": 1430,
1435
+ "칼": 1431,
1436
+ "캅": 1432,
1437
+ "캉": 1433,
1438
+ "캐": 1434,
1439
+ "캔": 1435,
1440
+ "캠": 1436,
1441
+ "캡": 1437,
1442
+ "캣": 1438,
1443
+ "캬": 1439,
1444
+ "커": 1440,
1445
+ "컥": 1441,
1446
+ "컨": 1442,
1447
+ "컬": 1443,
1448
+ "컴": 1444,
1449
+ "컵": 1445,
1450
+ "컷": 1446,
1451
+ "컸": 1447,
1452
+ "컹": 1448,
1453
+ "컽": 1449,
1454
+ "케": 1450,
1455
+ "켄": 1451,
1456
+ "켈": 1452,
1457
+ "켓": 1453,
1458
+ "켜": 1454,
1459
+ "켠": 1455,
1460
+ "켰": 1456,
1461
+ "코": 1457,
1462
+ "콕": 1458,
1463
+ "콘": 1459,
1464
+ "콜": 1460,
1465
+ "콤": 1461,
1466
+ "콧": 1462,
1467
+ "콩": 1463,
1468
+ "콸": 1464,
1469
+ "쾅": 1465,
1470
+ "쾌": 1466,
1471
+ "쿄": 1467,
1472
+ "쿠": 1468,
1473
+ "쿡": 1469,
1474
+ "쿤": 1470,
1475
+ "쿨": 1471,
1476
+ "쿰": 1472,
1477
+ "쿱": 1473,
1478
+ "쿵": 1474,
1479
+ "퀄": 1475,
1480
+ "퀘": 1476,
1481
+ "퀴": 1477,
1482
+ "퀵": 1478,
1483
+ "퀸": 1479,
1484
+ "퀼": 1480,
1485
+ "큐": 1481,
1486
+ "크": 1482,
1487
+ "큰": 1483,
1488
+ "클": 1484,
1489
+ "큼": 1485,
1490
+ "큽": 1486,
1491
+ "킁": 1487,
1492
+ "키": 1488,
1493
+ "킥": 1489,
1494
+ "킨": 1490,
1495
+ "킬": 1491,
1496
+ "킴": 1492,
1497
+ "킵": 1493,
1498
+ "킷": 1494,
1499
+ "킹": 1495,
1500
+ "타": 1496,
1501
+ "탁": 1497,
1502
+ "탄": 1498,
1503
+ "탈": 1499,
1504
+ "탐": 1500,
1505
+ "탑": 1501,
1506
+ "탓": 1502,
1507
+ "탔": 1503,
1508
+ "탕": 1504,
1509
+ "태": 1505,
1510
+ "택": 1506,
1511
+ "탠": 1507,
1512
+ "탬": 1508,
1513
+ "탭": 1509,
1514
+ "탯": 1510,
1515
+ "탱": 1511,
1516
+ "터": 1512,
1517
+ "턱": 1513,
1518
+ "턴": 1514,
1519
+ "털": 1515,
1520
+ "텀": 1516,
1521
+ "텁": 1517,
1522
+ "텃": 1518,
1523
+ "텄": 1519,
1524
+ "텅": 1520,
1525
+ "테": 1521,
1526
+ "텍": 1522,
1527
+ "텐": 1523,
1528
+ "텔": 1524,
1529
+ "템": 1525,
1530
+ "텝": 1526,
1531
+ "텨": 1527,
1532
+ "텼": 1528,
1533
+ "토": 1529,
1534
+ "톡": 1530,
1535
+ "톤": 1531,
1536
+ "톨": 1532,
1537
+ "톰": 1533,
1538
+ "톱": 1534,
1539
+ "통": 1535,
1540
+ "퇴": 1536,
1541
+ "투": 1537,
1542
+ "툭": 1538,
1543
+ "툰": 1539,
1544
+ "툴": 1540,
1545
+ "툼": 1541,
1546
+ "퉁": 1542,
1547
+ "퉤": 1543,
1548
+ "튀": 1544,
1549
+ "튄": 1545,
1550
+ "튈": 1546,
1551
+ "튑": 1547,
1552
+ "튕": 1548,
1553
+ "튜": 1549,
1554
+ "트": 1550,
1555
+ "특": 1551,
1556
+ "튼": 1552,
1557
+ "틀": 1553,
1558
+ "틈": 1554,
1559
+ "틉": 1555,
1560
+ "틋": 1556,
1561
+ "틑": 1557,
1562
+ "티": 1558,
1563
+ "틱": 1559,
1564
+ "틴": 1560,
1565
+ "틸": 1561,
1566
+ "팀": 1562,
1567
+ "팁": 1563,
1568
+ "팅": 1564,
1569
+ "파": 1565,
1570
+ "팍": 1566,
1571
+ "팎": 1567,
1572
+ "판": 1568,
1573
+ "팔": 1569,
1574
+ "팜": 1570,
1575
+ "팝": 1571,
1576
+ "팟": 1572,
1577
+ "팠": 1573,
1578
+ "팡": 1574,
1579
+ "팥": 1575,
1580
+ "패": 1576,
1581
+ "팩": 1577,
1582
+ "팬": 1578,
1583
+ "팸": 1579,
1584
+ "팹": 1580,
1585
+ "팻": 1581,
1586
+ "팽": 1582,
1587
+ "퍼": 1583,
1588
+ "퍽": 1584,
1589
+ "펀": 1585,
1590
+ "펄": 1586,
1591
+ "펌": 1587,
1592
+ "펍": 1588,
1593
+ "펐": 1589,
1594
+ "펑": 1590,
1595
+ "페": 1591,
1596
+ "펙": 1592,
1597
+ "펜": 1593,
1598
+ "펠": 1594,
1599
+ "펩": 1595,
1600
+ "펫": 1596,
1601
+ "펭": 1597,
1602
+ "펴": 1598,
1603
+ "편": 1599,
1604
+ "펼": 1600,
1605
+ "폄": 1601,
1606
+ "폈": 1602,
1607
+ "평": 1603,
1608
+ "폐": 1604,
1609
+ "포": 1605,
1610
+ "폭": 1606,
1611
+ "폰": 1607,
1612
+ "폴": 1608,
1613
+ "폼": 1609,
1614
+ "폽": 1610,
1615
+ "폿": 1611,
1616
+ "퐁": 1612,
1617
+ "표": 1613,
1618
+ "푠": 1614,
1619
+ "푸": 1615,
1620
+ "푹": 1616,
1621
+ "푼": 1617,
1622
+ "풀": 1618,
1623
+ "풂": 1619,
1624
+ "품": 1620,
1625
+ "풉": 1621,
1626
+ "풋": 1622,
1627
+ "풍": 1623,
1628
+ "퓨": 1624,
1629
+ "프": 1625,
1630
+ "픈": 1626,
1631
+ "플": 1627,
1632
+ "픔": 1628,
1633
+ "픕": 1629,
1634
+ "픗": 1630,
1635
+ "피": 1631,
1636
+ "픽": 1632,
1637
+ "핀": 1633,
1638
+ "필": 1634,
1639
+ "핌": 1635,
1640
+ "핍": 1636,
1641
+ "핏": 1637,
1642
+ "핑": 1638,
1643
+ "하": 1639,
1644
+ "학": 1640,
1645
+ "한": 1641,
1646
+ "할": 1642,
1647
+ "핥": 1643,
1648
+ "함": 1644,
1649
+ "합": 1645,
1650
+ "핫": 1646,
1651
+ "핬": 1647,
1652
+ "항": 1648,
1653
+ "해": 1649,
1654
+ "핵": 1650,
1655
+ "핸": 1651,
1656
+ "핼": 1652,
1657
+ "햄": 1653,
1658
+ "햅": 1654,
1659
+ "햇": 1655,
1660
+ "했": 1656,
1661
+ "행": 1657,
1662
+ "햐": 1658,
1663
+ "향": 1659,
1664
+ "허": 1660,
1665
+ "헉": 1661,
1666
+ "헌": 1662,
1667
+ "헐": 1663,
1668
+ "험": 1664,
1669
+ "헙": 1665,
1670
+ "헛": 1666,
1671
+ "헝": 1667,
1672
+ "헤": 1668,
1673
+ "헥": 1669,
1674
+ "헨": 1670,
1675
+ "헬": 1671,
1676
+ "헵": 1672,
1677
+ "헷": 1673,
1678
+ "헸": 1674,
1679
+ "헹": 1675,
1680
+ "혀": 1676,
1681
+ "혁": 1677,
1682
+ "현": 1678,
1683
+ "혈": 1679,
1684
+ "혐": 1680,
1685
+ "협": 1681,
1686
+ "혓": 1682,
1687
+ "혔": 1683,
1688
+ "형": 1684,
1689
+ "혜": 1685,
1690
+ "호": 1686,
1691
+ "혹": 1687,
1692
+ "혼": 1688,
1693
+ "홀": 1689,
1694
+ "홈": 1690,
1695
+ "홉": 1691,
1696
+ "홋": 1692,
1697
+ "홍": 1693,
1698
+ "화": 1694,
1699
+ "확": 1695,
1700
+ "환": 1696,
1701
+ "활": 1697,
1702
+ "황": 1698,
1703
+ "횃": 1699,
1704
+ "회": 1700,
1705
+ "획": 1701,
1706
+ "횝": 1702,
1707
+ "횟": 1703,
1708
+ "횡": 1704,
1709
+ "효": 1705,
1710
+ "후": 1706,
1711
+ "훅": 1707,
1712
+ "훈": 1708,
1713
+ "훌": 1709,
1714
+ "훑": 1710,
1715
+ "훔": 1711,
1716
+ "훤": 1712,
1717
+ "훨": 1713,
1718
+ "훼": 1714,
1719
+ "휑": 1715,
1720
+ "휘": 1716,
1721
+ "휙": 1717,
1722
+ "휜": 1718,
1723
+ "휠": 1719,
1724
+ "휩": 1720,
1725
+ "휴": 1721,
1726
+ "흉": 1722,
1727
+ "흐": 1723,
1728
+ "흑": 1724,
1729
+ "흔": 1725,
1730
+ "흘": 1726,
1731
+ "흙": 1727,
1732
+ "흠": 1728,
1733
+ "흡": 1729,
1734
+ "흥": 1730,
1735
+ "흩": 1731,
1736
+ "희": 1732,
1737
+ "흰": 1733,
1738
+ "히": 1734,
1739
+ "힌": 1735,
1740
+ "힐": 1736,
1741
+ "힘": 1737,
1742
+ "힙": 1738,
1743
+ "힝": 1739
1744
+ }