Hugging Face's logo

ta4tsering
/
Lhasa_Kanjur_TrOCR_model

Text Generation
Transformers
Safetensors
trocr
Inference Endpoints
Model card Files Community
ta4tsering commited on
Commit
d32b580
·
verified ·
1 Parent(s): 32812be

Upload folder using huggingface_hub

Browse files
Files changed (17) hide show
  1. .ipynb_checkpoints/config-checkpoint.json +178 -0
  2. .ipynb_checkpoints/generation_config-checkpoint.json +12 -0
  3. .ipynb_checkpoints/preprocessor_config-checkpoint.json +36 -0
  4. .ipynb_checkpoints/trainer_state-checkpoint.json +3489 -0
  5. config.json +195 -0
  6. generation_config.json +12 -0
  7. merges.txt +0 -0
  8. model.safetensors +3 -0
  9. optimizer.pt +3 -0
  10. preprocessor_config.json +37 -0
  11. rng_state.pth +3 -0
  12. scheduler.pt +3 -0
  13. special_tokens_map.json +51 -0
  14. tokenizer_config.json +59 -0
  15. trainer_state.json +3489 -0
  16. training_args.bin +3 -0
  17. vocab.json +0 -0
.ipynb_checkpoints/config-checkpoint.json ADDED
@@ -0,0 +1,178 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionEncoderDecoderModel"
4
+ ],
5
+ "decoder": {
6
+ "_name_or_path": "sangjeedondrub/tibetan-roberta-base",
7
+ "add_cross_attention": true,
8
+ "architectures": [
9
+ "RobertaForMaskedLM"
10
+ ],
11
+ "attention_probs_dropout_prob": 0.1,
12
+ "bad_words_ids": null,
13
+ "begin_suppress_tokens": null,
14
+ "bos_token_id": 0,
15
+ "chunk_size_feed_forward": 0,
16
+ "classifier_dropout": null,
17
+ "cross_attention_hidden_size": null,
18
+ "decoder_start_token_id": null,
19
+ "diversity_penalty": 0.0,
20
+ "do_sample": false,
21
+ "early_stopping": false,
22
+ "encoder_no_repeat_ngram_size": 0,
23
+ "eos_token_id": 2,
24
+ "exponential_decay_length_penalty": null,
25
+ "finetuning_task": null,
26
+ "forced_bos_token_id": null,
27
+ "forced_eos_token_id": null,
28
+ "hidden_act": "gelu",
29
+ "hidden_dropout_prob": 0.1,
30
+ "hidden_size": 768,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1"
34
+ },
35
+ "initializer_range": 0.02,
36
+ "intermediate_size": 3072,
37
+ "is_decoder": true,
38
+ "is_encoder_decoder": false,
39
+ "label2id": {
40
+ "LABEL_0": 0,
41
+ "LABEL_1": 1
42
+ },
43
+ "layer_norm_eps": 1e-12,
44
+ "length_penalty": 1.0,
45
+ "max_length": 20,
46
+ "max_position_embeddings": 514,
47
+ "min_length": 0,
48
+ "model_type": "roberta",
49
+ "no_repeat_ngram_size": 0,
50
+ "num_attention_heads": 12,
51
+ "num_beam_groups": 1,
52
+ "num_beams": 1,
53
+ "num_hidden_layers": 6,
54
+ "num_return_sequences": 1,
55
+ "output_attentions": false,
56
+ "output_hidden_states": false,
57
+ "output_scores": false,
58
+ "pad_token_id": 1,
59
+ "position_embedding_type": "absolute",
60
+ "prefix": null,
61
+ "problem_type": null,
62
+ "pruned_heads": {},
63
+ "remove_invalid_values": false,
64
+ "repetition_penalty": 1.0,
65
+ "return_dict": true,
66
+ "return_dict_in_generate": false,
67
+ "sep_token_id": null,
68
+ "suppress_tokens": null,
69
+ "task_specific_params": null,
70
+ "temperature": 1.0,
71
+ "tf_legacy_loss": false,
72
+ "tie_encoder_decoder": false,
73
+ "tie_word_embeddings": true,
74
+ "tokenizer_class": null,
75
+ "top_k": 50,
76
+ "top_p": 1.0,
77
+ "torch_dtype": "float32",
78
+ "torchscript": false,
79
+ "type_vocab_size": 1,
80
+ "typical_p": 1.0,
81
+ "use_bfloat16": false,
82
+ "use_cache": true,
83
+ "vocab_size": 52000
84
+ },
85
+ "decoder_start_token_id": 0,
86
+ "early_stopping": true,
87
+ "encoder": {
88
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
89
+ "add_cross_attention": false,
90
+ "architectures": [
91
+ "ViTModel"
92
+ ],
93
+ "attention_probs_dropout_prob": 0.0,
94
+ "bad_words_ids": null,
95
+ "begin_suppress_tokens": null,
96
+ "bos_token_id": null,
97
+ "chunk_size_feed_forward": 0,
98
+ "cross_attention_hidden_size": null,
99
+ "decoder_start_token_id": null,
100
+ "diversity_penalty": 0.0,
101
+ "do_sample": false,
102
+ "early_stopping": false,
103
+ "encoder_no_repeat_ngram_size": 0,
104
+ "encoder_stride": 16,
105
+ "eos_token_id": null,
106
+ "exponential_decay_length_penalty": null,
107
+ "finetuning_task": null,
108
+ "forced_bos_token_id": null,
109
+ "forced_eos_token_id": null,
110
+ "hidden_act": "gelu",
111
+ "hidden_dropout_prob": 0.0,
112
+ "hidden_size": 768,
113
+ "id2label": {
114
+ "0": "LABEL_0",
115
+ "1": "LABEL_1"
116
+ },
117
+ "image_size": 224,
118
+ "initializer_range": 0.02,
119
+ "intermediate_size": 3072,
120
+ "is_decoder": false,
121
+ "is_encoder_decoder": false,
122
+ "label2id": {
123
+ "LABEL_0": 0,
124
+ "LABEL_1": 1
125
+ },
126
+ "layer_norm_eps": 1e-12,
127
+ "length_penalty": 1.0,
128
+ "max_length": 20,
129
+ "min_length": 0,
130
+ "model_type": "vit",
131
+ "no_repeat_ngram_size": 0,
132
+ "num_attention_heads": 12,
133
+ "num_beam_groups": 1,
134
+ "num_beams": 1,
135
+ "num_channels": 3,
136
+ "num_hidden_layers": 12,
137
+ "num_return_sequences": 1,
138
+ "output_attentions": false,
139
+ "output_hidden_states": false,
140
+ "output_scores": false,
141
+ "pad_token_id": null,
142
+ "patch_size": 16,
143
+ "prefix": null,
144
+ "problem_type": null,
145
+ "pruned_heads": {},
146
+ "qkv_bias": true,
147
+ "remove_invalid_values": false,
148
+ "repetition_penalty": 1.0,
149
+ "return_dict": true,
150
+ "return_dict_in_generate": false,
151
+ "sep_token_id": null,
152
+ "suppress_tokens": null,
153
+ "task_specific_params": null,
154
+ "temperature": 1.0,
155
+ "tf_legacy_loss": false,
156
+ "tie_encoder_decoder": false,
157
+ "tie_word_embeddings": true,
158
+ "tokenizer_class": null,
159
+ "top_k": 50,
160
+ "top_p": 1.0,
161
+ "torch_dtype": null,
162
+ "torchscript": false,
163
+ "typical_p": 1.0,
164
+ "use_bfloat16": false
165
+ },
166
+ "eos_token_id": 2,
167
+ "is_encoder_decoder": true,
168
+ "length_penalty": 2.0,
169
+ "max_length": 170,
170
+ "model_type": "vision-encoder-decoder",
171
+ "no_repeat_ngram_size": 3,
172
+ "num_beams": 4,
173
+ "pad_token_id": 1,
174
+ "tie_word_embeddings": false,
175
+ "torch_dtype": "float32",
176
+ "transformers_version": "4.40.2",
177
+ "vocab_size": 52000
178
+ }
.ipynb_checkpoints/generation_config-checkpoint.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 0,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "length_penalty": 2.0,
7
+ "max_length": 170,
8
+ "no_repeat_ngram_size": 3,
9
+ "num_beams": 4,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.40.2"
12
+ }
.ipynb_checkpoints/preprocessor_config-checkpoint.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_processor_keys": [
3
+ "images",
4
+ "do_resize",
5
+ "size",
6
+ "resample",
7
+ "do_rescale",
8
+ "rescale_factor",
9
+ "do_normalize",
10
+ "image_mean",
11
+ "image_std",
12
+ "return_tensors",
13
+ "data_format",
14
+ "input_data_format"
15
+ ],
16
+ "do_normalize": true,
17
+ "do_rescale": true,
18
+ "do_resize": true,
19
+ "image_mean": [
20
+ 0.5,
21
+ 0.5,
22
+ 0.5
23
+ ],
24
+ "image_processor_type": "ViTImageProcessor",
25
+ "image_std": [
26
+ 0.5,
27
+ 0.5,
28
+ 0.5
29
+ ],
30
+ "resample": 2,
31
+ "rescale_factor": 0.00392156862745098,
32
+ "size": {
33
+ "height": 224,
34
+ "width": 224
35
+ }
36
+ }
.ipynb_checkpoints/trainer_state-checkpoint.json ADDED
@@ -0,0 +1,3489 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.8866971373586723,
5
+ "eval_steps": 4000,
6
+ "global_step": 48000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0060139523694972335,
13
+ "grad_norm": 2.5106329917907715,
14
+ "learning_rate": 4.99007697859033e-05,
15
+ "loss": 3.082,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.012027904738994467,
20
+ "grad_norm": 2.184039831161499,
21
+ "learning_rate": 4.9800537246411676e-05,
22
+ "loss": 1.8328,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.0180418571084917,
27
+ "grad_norm": 2.2055342197418213,
28
+ "learning_rate": 4.970030470692006e-05,
29
+ "loss": 1.6836,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.024055809477988934,
34
+ "grad_norm": 2.0695838928222656,
35
+ "learning_rate": 4.960007216742843e-05,
36
+ "loss": 1.6098,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.03006976184748617,
41
+ "grad_norm": 1.9105968475341797,
42
+ "learning_rate": 4.949983962793682e-05,
43
+ "loss": 1.516,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 0.0360837142169834,
48
+ "grad_norm": 1.882120132446289,
49
+ "learning_rate": 4.9399607088445195e-05,
50
+ "loss": 1.4876,
51
+ "step": 600
52
+ },
53
+ {
54
+ "epoch": 0.042097666586480634,
55
+ "grad_norm": 1.937444806098938,
56
+ "learning_rate": 4.9299374548953576e-05,
57
+ "loss": 1.4493,
58
+ "step": 700
59
+ },
60
+ {
61
+ "epoch": 0.04811161895597787,
62
+ "grad_norm": 2.1085917949676514,
63
+ "learning_rate": 4.919914200946195e-05,
64
+ "loss": 1.4128,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 0.0541255713254751,
69
+ "grad_norm": 1.9272162914276123,
70
+ "learning_rate": 4.909890946997033e-05,
71
+ "loss": 1.4105,
72
+ "step": 900
73
+ },
74
+ {
75
+ "epoch": 0.06013952369497234,
76
+ "grad_norm": 1.818624496459961,
77
+ "learning_rate": 4.899867693047871e-05,
78
+ "loss": 1.3912,
79
+ "step": 1000
80
+ },
81
+ {
82
+ "epoch": 0.06615347606446957,
83
+ "grad_norm": 1.644021987915039,
84
+ "learning_rate": 4.8898444390987094e-05,
85
+ "loss": 1.3397,
86
+ "step": 1100
87
+ },
88
+ {
89
+ "epoch": 0.0721674284339668,
90
+ "grad_norm": 1.7671422958374023,
91
+ "learning_rate": 4.8798211851495476e-05,
92
+ "loss": 1.3873,
93
+ "step": 1200
94
+ },
95
+ {
96
+ "epoch": 0.07818138080346404,
97
+ "grad_norm": 2.032087564468384,
98
+ "learning_rate": 4.869797931200385e-05,
99
+ "loss": 1.3112,
100
+ "step": 1300
101
+ },
102
+ {
103
+ "epoch": 0.08419533317296127,
104
+ "grad_norm": 1.7175960540771484,
105
+ "learning_rate": 4.859774677251223e-05,
106
+ "loss": 1.3089,
107
+ "step": 1400
108
+ },
109
+ {
110
+ "epoch": 0.09020928554245851,
111
+ "grad_norm": 1.7282887697219849,
112
+ "learning_rate": 4.8497514233020606e-05,
113
+ "loss": 1.2983,
114
+ "step": 1500
115
+ },
116
+ {
117
+ "epoch": 0.09622323791195574,
118
+ "grad_norm": 1.836075782775879,
119
+ "learning_rate": 4.839728169352899e-05,
120
+ "loss": 1.3092,
121
+ "step": 1600
122
+ },
123
+ {
124
+ "epoch": 0.10223719028145298,
125
+ "grad_norm": 1.931219220161438,
126
+ "learning_rate": 4.829704915403737e-05,
127
+ "loss": 1.2528,
128
+ "step": 1700
129
+ },
130
+ {
131
+ "epoch": 0.1082511426509502,
132
+ "grad_norm": 1.9112814664840698,
133
+ "learning_rate": 4.819681661454575e-05,
134
+ "loss": 1.25,
135
+ "step": 1800
136
+ },
137
+ {
138
+ "epoch": 0.11426509502044743,
139
+ "grad_norm": 2.1694140434265137,
140
+ "learning_rate": 4.809658407505413e-05,
141
+ "loss": 1.253,
142
+ "step": 1900
143
+ },
144
+ {
145
+ "epoch": 0.12027904738994467,
146
+ "grad_norm": 1.9220771789550781,
147
+ "learning_rate": 4.7996351535562506e-05,
148
+ "loss": 1.2278,
149
+ "step": 2000
150
+ },
151
+ {
152
+ "epoch": 0.12629299975944192,
153
+ "grad_norm": 1.7120180130004883,
154
+ "learning_rate": 4.789611899607089e-05,
155
+ "loss": 1.2483,
156
+ "step": 2100
157
+ },
158
+ {
159
+ "epoch": 0.13230695212893914,
160
+ "grad_norm": 1.917104959487915,
161
+ "learning_rate": 4.779588645657926e-05,
162
+ "loss": 1.2384,
163
+ "step": 2200
164
+ },
165
+ {
166
+ "epoch": 0.13832090449843637,
167
+ "grad_norm": 2.0323758125305176,
168
+ "learning_rate": 4.7695653917087644e-05,
169
+ "loss": 1.2292,
170
+ "step": 2300
171
+ },
172
+ {
173
+ "epoch": 0.1443348568679336,
174
+ "grad_norm": 1.779888391494751,
175
+ "learning_rate": 4.7595421377596025e-05,
176
+ "loss": 1.2258,
177
+ "step": 2400
178
+ },
179
+ {
180
+ "epoch": 0.15034880923743085,
181
+ "grad_norm": 1.7667876482009888,
182
+ "learning_rate": 4.7495188838104406e-05,
183
+ "loss": 1.1875,
184
+ "step": 2500
185
+ },
186
+ {
187
+ "epoch": 0.15636276160692808,
188
+ "grad_norm": 1.8692760467529297,
189
+ "learning_rate": 4.739495629861279e-05,
190
+ "loss": 1.2082,
191
+ "step": 2600
192
+ },
193
+ {
194
+ "epoch": 0.1623767139764253,
195
+ "grad_norm": 1.8578811883926392,
196
+ "learning_rate": 4.729472375912116e-05,
197
+ "loss": 1.1939,
198
+ "step": 2700
199
+ },
200
+ {
201
+ "epoch": 0.16839066634592254,
202
+ "grad_norm": 1.8867884874343872,
203
+ "learning_rate": 4.7194491219629543e-05,
204
+ "loss": 1.2108,
205
+ "step": 2800
206
+ },
207
+ {
208
+ "epoch": 0.17440461871541976,
209
+ "grad_norm": 1.72930109500885,
210
+ "learning_rate": 4.709425868013792e-05,
211
+ "loss": 1.1943,
212
+ "step": 2900
213
+ },
214
+ {
215
+ "epoch": 0.18041857108491702,
216
+ "grad_norm": 1.7612700462341309,
217
+ "learning_rate": 4.69940261406463e-05,
218
+ "loss": 1.1818,
219
+ "step": 3000
220
+ },
221
+ {
222
+ "epoch": 0.18643252345441425,
223
+ "grad_norm": 1.6063175201416016,
224
+ "learning_rate": 4.689379360115468e-05,
225
+ "loss": 1.1926,
226
+ "step": 3100
227
+ },
228
+ {
229
+ "epoch": 0.19244647582391147,
230
+ "grad_norm": 1.7698125839233398,
231
+ "learning_rate": 4.679356106166306e-05,
232
+ "loss": 1.2029,
233
+ "step": 3200
234
+ },
235
+ {
236
+ "epoch": 0.1984604281934087,
237
+ "grad_norm": 1.6680766344070435,
238
+ "learning_rate": 4.6693328522171443e-05,
239
+ "loss": 1.185,
240
+ "step": 3300
241
+ },
242
+ {
243
+ "epoch": 0.20447438056290596,
244
+ "grad_norm": 1.7791478633880615,
245
+ "learning_rate": 4.659309598267982e-05,
246
+ "loss": 1.1663,
247
+ "step": 3400
248
+ },
249
+ {
250
+ "epoch": 0.21048833293240318,
251
+ "grad_norm": 1.6491518020629883,
252
+ "learning_rate": 4.64928634431882e-05,
253
+ "loss": 1.1492,
254
+ "step": 3500
255
+ },
256
+ {
257
+ "epoch": 0.2165022853019004,
258
+ "grad_norm": 1.6312377452850342,
259
+ "learning_rate": 4.6392630903696574e-05,
260
+ "loss": 1.1696,
261
+ "step": 3600
262
+ },
263
+ {
264
+ "epoch": 0.22251623767139764,
265
+ "grad_norm": 1.7078893184661865,
266
+ "learning_rate": 4.6292398364204955e-05,
267
+ "loss": 1.139,
268
+ "step": 3700
269
+ },
270
+ {
271
+ "epoch": 0.22853019004089486,
272
+ "grad_norm": 2.0885114669799805,
273
+ "learning_rate": 4.619216582471334e-05,
274
+ "loss": 1.1399,
275
+ "step": 3800
276
+ },
277
+ {
278
+ "epoch": 0.23454414241039212,
279
+ "grad_norm": 1.8433787822723389,
280
+ "learning_rate": 4.609193328522172e-05,
281
+ "loss": 1.1376,
282
+ "step": 3900
283
+ },
284
+ {
285
+ "epoch": 0.24055809477988935,
286
+ "grad_norm": 1.527872085571289,
287
+ "learning_rate": 4.59917007457301e-05,
288
+ "loss": 1.1094,
289
+ "step": 4000
290
+ },
291
+ {
292
+ "epoch": 0.24055809477988935,
293
+ "eval_cer": 0.7748897683360889,
294
+ "eval_loss": 1.0861139297485352,
295
+ "eval_runtime": 5003.1551,
296
+ "eval_samples_per_second": 3.326,
297
+ "eval_steps_per_second": 0.416,
298
+ "step": 4000
299
+ },
300
+ {
301
+ "epoch": 0.24657204714938658,
302
+ "grad_norm": 1.4162925481796265,
303
+ "learning_rate": 4.5891468206238474e-05,
304
+ "loss": 1.1353,
305
+ "step": 4100
306
+ },
307
+ {
308
+ "epoch": 0.25258599951888383,
309
+ "grad_norm": 1.5863301753997803,
310
+ "learning_rate": 4.5791235666746855e-05,
311
+ "loss": 1.1257,
312
+ "step": 4200
313
+ },
314
+ {
315
+ "epoch": 0.25859995188838103,
316
+ "grad_norm": 1.6928209066390991,
317
+ "learning_rate": 4.569100312725523e-05,
318
+ "loss": 1.1086,
319
+ "step": 4300
320
+ },
321
+ {
322
+ "epoch": 0.2646139042578783,
323
+ "grad_norm": 1.2880722284317017,
324
+ "learning_rate": 4.559077058776362e-05,
325
+ "loss": 1.119,
326
+ "step": 4400
327
+ },
328
+ {
329
+ "epoch": 0.2706278566273755,
330
+ "grad_norm": 1.6384778022766113,
331
+ "learning_rate": 4.549053804827199e-05,
332
+ "loss": 1.1474,
333
+ "step": 4500
334
+ },
335
+ {
336
+ "epoch": 0.27664180899687274,
337
+ "grad_norm": 1.7176462411880493,
338
+ "learning_rate": 4.5390305508780374e-05,
339
+ "loss": 1.0925,
340
+ "step": 4600
341
+ },
342
+ {
343
+ "epoch": 0.28265576136637,
344
+ "grad_norm": 1.6025973558425903,
345
+ "learning_rate": 4.529007296928875e-05,
346
+ "loss": 1.112,
347
+ "step": 4700
348
+ },
349
+ {
350
+ "epoch": 0.2886697137358672,
351
+ "grad_norm": 1.7415896654129028,
352
+ "learning_rate": 4.518984042979713e-05,
353
+ "loss": 1.1218,
354
+ "step": 4800
355
+ },
356
+ {
357
+ "epoch": 0.29468366610536445,
358
+ "grad_norm": 1.310402750968933,
359
+ "learning_rate": 4.508960789030551e-05,
360
+ "loss": 1.0879,
361
+ "step": 4900
362
+ },
363
+ {
364
+ "epoch": 0.3006976184748617,
365
+ "grad_norm": 1.7458600997924805,
366
+ "learning_rate": 4.498937535081389e-05,
367
+ "loss": 1.1088,
368
+ "step": 5000
369
+ },
370
+ {
371
+ "epoch": 0.3067115708443589,
372
+ "grad_norm": 1.8199197053909302,
373
+ "learning_rate": 4.4889142811322274e-05,
374
+ "loss": 1.0987,
375
+ "step": 5100
376
+ },
377
+ {
378
+ "epoch": 0.31272552321385616,
379
+ "grad_norm": 1.5931302309036255,
380
+ "learning_rate": 4.478891027183065e-05,
381
+ "loss": 1.0802,
382
+ "step": 5200
383
+ },
384
+ {
385
+ "epoch": 0.31873947558335336,
386
+ "grad_norm": 1.5596023797988892,
387
+ "learning_rate": 4.468867773233903e-05,
388
+ "loss": 1.1009,
389
+ "step": 5300
390
+ },
391
+ {
392
+ "epoch": 0.3247534279528506,
393
+ "grad_norm": 1.7147523164749146,
394
+ "learning_rate": 4.4588445192847404e-05,
395
+ "loss": 1.1144,
396
+ "step": 5400
397
+ },
398
+ {
399
+ "epoch": 0.33076738032234787,
400
+ "grad_norm": 1.7172225713729858,
401
+ "learning_rate": 4.4488212653355786e-05,
402
+ "loss": 1.0928,
403
+ "step": 5500
404
+ },
405
+ {
406
+ "epoch": 0.33678133269184507,
407
+ "grad_norm": 1.5675067901611328,
408
+ "learning_rate": 4.438798011386417e-05,
409
+ "loss": 1.0421,
410
+ "step": 5600
411
+ },
412
+ {
413
+ "epoch": 0.3427952850613423,
414
+ "grad_norm": 1.755364179611206,
415
+ "learning_rate": 4.428774757437255e-05,
416
+ "loss": 1.0897,
417
+ "step": 5700
418
+ },
419
+ {
420
+ "epoch": 0.3488092374308395,
421
+ "grad_norm": 1.5678260326385498,
422
+ "learning_rate": 4.418751503488093e-05,
423
+ "loss": 1.0703,
424
+ "step": 5800
425
+ },
426
+ {
427
+ "epoch": 0.3548231898003368,
428
+ "grad_norm": 1.8641386032104492,
429
+ "learning_rate": 4.4087282495389304e-05,
430
+ "loss": 1.07,
431
+ "step": 5900
432
+ },
433
+ {
434
+ "epoch": 0.36083714216983404,
435
+ "grad_norm": 1.571837306022644,
436
+ "learning_rate": 4.3987049955897686e-05,
437
+ "loss": 1.0493,
438
+ "step": 6000
439
+ },
440
+ {
441
+ "epoch": 0.36685109453933124,
442
+ "grad_norm": 1.5046846866607666,
443
+ "learning_rate": 4.388681741640606e-05,
444
+ "loss": 1.0935,
445
+ "step": 6100
446
+ },
447
+ {
448
+ "epoch": 0.3728650469088285,
449
+ "grad_norm": 1.88533353805542,
450
+ "learning_rate": 4.378658487691444e-05,
451
+ "loss": 1.0801,
452
+ "step": 6200
453
+ },
454
+ {
455
+ "epoch": 0.3788789992783257,
456
+ "grad_norm": 1.5626581907272339,
457
+ "learning_rate": 4.368635233742282e-05,
458
+ "loss": 1.0442,
459
+ "step": 6300
460
+ },
461
+ {
462
+ "epoch": 0.38489295164782295,
463
+ "grad_norm": 1.4634822607040405,
464
+ "learning_rate": 4.3586119797931204e-05,
465
+ "loss": 1.0552,
466
+ "step": 6400
467
+ },
468
+ {
469
+ "epoch": 0.3909069040173202,
470
+ "grad_norm": 1.8044239282608032,
471
+ "learning_rate": 4.3485887258439586e-05,
472
+ "loss": 1.035,
473
+ "step": 6500
474
+ },
475
+ {
476
+ "epoch": 0.3969208563868174,
477
+ "grad_norm": 1.6868014335632324,
478
+ "learning_rate": 4.338565471894796e-05,
479
+ "loss": 1.0497,
480
+ "step": 6600
481
+ },
482
+ {
483
+ "epoch": 0.40293480875631466,
484
+ "grad_norm": 1.6949145793914795,
485
+ "learning_rate": 4.328542217945634e-05,
486
+ "loss": 1.0369,
487
+ "step": 6700
488
+ },
489
+ {
490
+ "epoch": 0.4089487611258119,
491
+ "grad_norm": 1.7868014574050903,
492
+ "learning_rate": 4.3186191965359634e-05,
493
+ "loss": 1.0385,
494
+ "step": 6800
495
+ },
496
+ {
497
+ "epoch": 0.4149627134953091,
498
+ "grad_norm": 1.550017237663269,
499
+ "learning_rate": 4.3085959425868015e-05,
500
+ "loss": 1.0326,
501
+ "step": 6900
502
+ },
503
+ {
504
+ "epoch": 0.42097666586480637,
505
+ "grad_norm": 1.431362509727478,
506
+ "learning_rate": 4.298572688637639e-05,
507
+ "loss": 1.0154,
508
+ "step": 7000
509
+ },
510
+ {
511
+ "epoch": 0.42699061823430356,
512
+ "grad_norm": 1.946357011795044,
513
+ "learning_rate": 4.288549434688478e-05,
514
+ "loss": 1.0532,
515
+ "step": 7100
516
+ },
517
+ {
518
+ "epoch": 0.4330045706038008,
519
+ "grad_norm": 1.6147353649139404,
520
+ "learning_rate": 4.278526180739316e-05,
521
+ "loss": 1.0633,
522
+ "step": 7200
523
+ },
524
+ {
525
+ "epoch": 0.4390185229732981,
526
+ "grad_norm": 1.655043363571167,
527
+ "learning_rate": 4.2685029267901534e-05,
528
+ "loss": 1.0558,
529
+ "step": 7300
530
+ },
531
+ {
532
+ "epoch": 0.4450324753427953,
533
+ "grad_norm": 1.6090620756149292,
534
+ "learning_rate": 4.2584796728409915e-05,
535
+ "loss": 1.0289,
536
+ "step": 7400
537
+ },
538
+ {
539
+ "epoch": 0.45104642771229253,
540
+ "grad_norm": 1.8263063430786133,
541
+ "learning_rate": 4.248456418891829e-05,
542
+ "loss": 1.0275,
543
+ "step": 7500
544
+ },
545
+ {
546
+ "epoch": 0.45706038008178973,
547
+ "grad_norm": 1.6156238317489624,
548
+ "learning_rate": 4.238433164942667e-05,
549
+ "loss": 0.9976,
550
+ "step": 7600
551
+ },
552
+ {
553
+ "epoch": 0.463074332451287,
554
+ "grad_norm": 1.4604992866516113,
555
+ "learning_rate": 4.228409910993505e-05,
556
+ "loss": 1.0607,
557
+ "step": 7700
558
+ },
559
+ {
560
+ "epoch": 0.46908828482078424,
561
+ "grad_norm": 1.5745257139205933,
562
+ "learning_rate": 4.2183866570443434e-05,
563
+ "loss": 1.0297,
564
+ "step": 7800
565
+ },
566
+ {
567
+ "epoch": 0.47510223719028144,
568
+ "grad_norm": 1.852102518081665,
569
+ "learning_rate": 4.2083634030951815e-05,
570
+ "loss": 1.0173,
571
+ "step": 7900
572
+ },
573
+ {
574
+ "epoch": 0.4811161895597787,
575
+ "grad_norm": 1.6735659837722778,
576
+ "learning_rate": 4.198340149146019e-05,
577
+ "loss": 1.0145,
578
+ "step": 8000
579
+ },
580
+ {
581
+ "epoch": 0.4811161895597787,
582
+ "eval_cer": 0.7706982683703557,
583
+ "eval_loss": 0.9832878708839417,
584
+ "eval_runtime": 5013.5334,
585
+ "eval_samples_per_second": 3.319,
586
+ "eval_steps_per_second": 0.415,
587
+ "step": 8000
588
+ },
589
+ {
590
+ "epoch": 0.4871301419292759,
591
+ "grad_norm": 1.7805702686309814,
592
+ "learning_rate": 4.188316895196857e-05,
593
+ "loss": 1.0279,
594
+ "step": 8100
595
+ },
596
+ {
597
+ "epoch": 0.49314409429877315,
598
+ "grad_norm": 1.4825282096862793,
599
+ "learning_rate": 4.1782936412476945e-05,
600
+ "loss": 1.0124,
601
+ "step": 8200
602
+ },
603
+ {
604
+ "epoch": 0.4991580466682704,
605
+ "grad_norm": 1.5925073623657227,
606
+ "learning_rate": 4.168270387298533e-05,
607
+ "loss": 1.0026,
608
+ "step": 8300
609
+ },
610
+ {
611
+ "epoch": 0.5051719990377677,
612
+ "grad_norm": 1.5965441465377808,
613
+ "learning_rate": 4.158247133349371e-05,
614
+ "loss": 1.0238,
615
+ "step": 8400
616
+ },
617
+ {
618
+ "epoch": 0.5111859514072649,
619
+ "grad_norm": 1.5980632305145264,
620
+ "learning_rate": 4.148223879400209e-05,
621
+ "loss": 0.9909,
622
+ "step": 8500
623
+ },
624
+ {
625
+ "epoch": 0.5171999037767621,
626
+ "grad_norm": 1.7301024198532104,
627
+ "learning_rate": 4.1382006254510464e-05,
628
+ "loss": 1.0172,
629
+ "step": 8600
630
+ },
631
+ {
632
+ "epoch": 0.5232138561462594,
633
+ "grad_norm": 1.6884981393814087,
634
+ "learning_rate": 4.1281773715018845e-05,
635
+ "loss": 1.0326,
636
+ "step": 8700
637
+ },
638
+ {
639
+ "epoch": 0.5292278085157566,
640
+ "grad_norm": 1.6702126264572144,
641
+ "learning_rate": 4.118154117552723e-05,
642
+ "loss": 1.016,
643
+ "step": 8800
644
+ },
645
+ {
646
+ "epoch": 0.5352417608852538,
647
+ "grad_norm": 1.5463213920593262,
648
+ "learning_rate": 4.108231096143052e-05,
649
+ "loss": 1.0282,
650
+ "step": 8900
651
+ },
652
+ {
653
+ "epoch": 0.541255713254751,
654
+ "grad_norm": 1.9977515935897827,
655
+ "learning_rate": 4.09820784219389e-05,
656
+ "loss": 1.0027,
657
+ "step": 9000
658
+ },
659
+ {
660
+ "epoch": 0.5472696656242483,
661
+ "grad_norm": 1.4702314138412476,
662
+ "learning_rate": 4.088184588244728e-05,
663
+ "loss": 1.0138,
664
+ "step": 9100
665
+ },
666
+ {
667
+ "epoch": 0.5532836179937455,
668
+ "grad_norm": 1.5981062650680542,
669
+ "learning_rate": 4.078161334295566e-05,
670
+ "loss": 1.0333,
671
+ "step": 9200
672
+ },
673
+ {
674
+ "epoch": 0.5592975703632427,
675
+ "grad_norm": 1.4159762859344482,
676
+ "learning_rate": 4.068138080346404e-05,
677
+ "loss": 0.9956,
678
+ "step": 9300
679
+ },
680
+ {
681
+ "epoch": 0.56531152273274,
682
+ "grad_norm": 1.4920343160629272,
683
+ "learning_rate": 4.058114826397242e-05,
684
+ "loss": 0.9978,
685
+ "step": 9400
686
+ },
687
+ {
688
+ "epoch": 0.5713254751022372,
689
+ "grad_norm": 1.7335036993026733,
690
+ "learning_rate": 4.048091572448079e-05,
691
+ "loss": 1.0337,
692
+ "step": 9500
693
+ },
694
+ {
695
+ "epoch": 0.5773394274717344,
696
+ "grad_norm": 1.4422550201416016,
697
+ "learning_rate": 4.0380683184989175e-05,
698
+ "loss": 1.0185,
699
+ "step": 9600
700
+ },
701
+ {
702
+ "epoch": 0.5833533798412317,
703
+ "grad_norm": 1.776282787322998,
704
+ "learning_rate": 4.0280450645497556e-05,
705
+ "loss": 0.9941,
706
+ "step": 9700
707
+ },
708
+ {
709
+ "epoch": 0.5893673322107289,
710
+ "grad_norm": 1.806261420249939,
711
+ "learning_rate": 4.018021810600594e-05,
712
+ "loss": 0.9772,
713
+ "step": 9800
714
+ },
715
+ {
716
+ "epoch": 0.5953812845802261,
717
+ "grad_norm": 1.6673755645751953,
718
+ "learning_rate": 4.007998556651432e-05,
719
+ "loss": 0.9904,
720
+ "step": 9900
721
+ },
722
+ {
723
+ "epoch": 0.6013952369497234,
724
+ "grad_norm": 1.5652821063995361,
725
+ "learning_rate": 3.997975302702269e-05,
726
+ "loss": 1.0008,
727
+ "step": 10000
728
+ },
729
+ {
730
+ "epoch": 0.6074091893192206,
731
+ "grad_norm": 1.568590760231018,
732
+ "learning_rate": 3.9879520487531075e-05,
733
+ "loss": 1.0402,
734
+ "step": 10100
735
+ },
736
+ {
737
+ "epoch": 0.6134231416887178,
738
+ "grad_norm": 1.505454659461975,
739
+ "learning_rate": 3.977928794803945e-05,
740
+ "loss": 1.0009,
741
+ "step": 10200
742
+ },
743
+ {
744
+ "epoch": 0.619437094058215,
745
+ "grad_norm": 1.4691787958145142,
746
+ "learning_rate": 3.967905540854783e-05,
747
+ "loss": 1.0115,
748
+ "step": 10300
749
+ },
750
+ {
751
+ "epoch": 0.6254510464277123,
752
+ "grad_norm": 1.5172970294952393,
753
+ "learning_rate": 3.957882286905621e-05,
754
+ "loss": 1.0209,
755
+ "step": 10400
756
+ },
757
+ {
758
+ "epoch": 0.6314649987972095,
759
+ "grad_norm": 1.8390883207321167,
760
+ "learning_rate": 3.947859032956459e-05,
761
+ "loss": 1.0101,
762
+ "step": 10500
763
+ },
764
+ {
765
+ "epoch": 0.6374789511667067,
766
+ "grad_norm": 1.5624459981918335,
767
+ "learning_rate": 3.9378357790072975e-05,
768
+ "loss": 1.0085,
769
+ "step": 10600
770
+ },
771
+ {
772
+ "epoch": 0.643492903536204,
773
+ "grad_norm": 1.4316383600234985,
774
+ "learning_rate": 3.927812525058135e-05,
775
+ "loss": 1.0025,
776
+ "step": 10700
777
+ },
778
+ {
779
+ "epoch": 0.6495068559057012,
780
+ "grad_norm": 1.5541032552719116,
781
+ "learning_rate": 3.917789271108973e-05,
782
+ "loss": 0.9938,
783
+ "step": 10800
784
+ },
785
+ {
786
+ "epoch": 0.6555208082751984,
787
+ "grad_norm": 1.7416138648986816,
788
+ "learning_rate": 3.907866249699302e-05,
789
+ "loss": 0.9923,
790
+ "step": 10900
791
+ },
792
+ {
793
+ "epoch": 0.6615347606446957,
794
+ "grad_norm": 1.7461607456207275,
795
+ "learning_rate": 3.8978429957501404e-05,
796
+ "loss": 0.9983,
797
+ "step": 11000
798
+ },
799
+ {
800
+ "epoch": 0.6675487130141929,
801
+ "grad_norm": 1.5705294609069824,
802
+ "learning_rate": 3.887819741800978e-05,
803
+ "loss": 0.9809,
804
+ "step": 11100
805
+ },
806
+ {
807
+ "epoch": 0.6735626653836901,
808
+ "grad_norm": 1.802960753440857,
809
+ "learning_rate": 3.877796487851817e-05,
810
+ "loss": 0.9675,
811
+ "step": 11200
812
+ },
813
+ {
814
+ "epoch": 0.6795766177531873,
815
+ "grad_norm": 1.70058274269104,
816
+ "learning_rate": 3.867773233902655e-05,
817
+ "loss": 0.966,
818
+ "step": 11300
819
+ },
820
+ {
821
+ "epoch": 0.6855905701226847,
822
+ "grad_norm": 1.5391656160354614,
823
+ "learning_rate": 3.857749979953492e-05,
824
+ "loss": 0.9412,
825
+ "step": 11400
826
+ },
827
+ {
828
+ "epoch": 0.6916045224921819,
829
+ "grad_norm": 1.8226732015609741,
830
+ "learning_rate": 3.8477267260043304e-05,
831
+ "loss": 0.96,
832
+ "step": 11500
833
+ },
834
+ {
835
+ "epoch": 0.697618474861679,
836
+ "grad_norm": 1.3753610849380493,
837
+ "learning_rate": 3.837703472055168e-05,
838
+ "loss": 0.9665,
839
+ "step": 11600
840
+ },
841
+ {
842
+ "epoch": 0.7036324272311764,
843
+ "grad_norm": 1.6595444679260254,
844
+ "learning_rate": 3.827680218106006e-05,
845
+ "loss": 0.9851,
846
+ "step": 11700
847
+ },
848
+ {
849
+ "epoch": 0.7096463796006736,
850
+ "grad_norm": 1.6920074224472046,
851
+ "learning_rate": 3.817656964156844e-05,
852
+ "loss": 0.983,
853
+ "step": 11800
854
+ },
855
+ {
856
+ "epoch": 0.7156603319701708,
857
+ "grad_norm": 1.4369592666625977,
858
+ "learning_rate": 3.807633710207682e-05,
859
+ "loss": 0.9418,
860
+ "step": 11900
861
+ },
862
+ {
863
+ "epoch": 0.7216742843396681,
864
+ "grad_norm": 1.7257956266403198,
865
+ "learning_rate": 3.7976104562585204e-05,
866
+ "loss": 0.9877,
867
+ "step": 12000
868
+ },
869
+ {
870
+ "epoch": 0.7216742843396681,
871
+ "eval_cer": 0.7783561907148705,
872
+ "eval_loss": 0.9317989349365234,
873
+ "eval_runtime": 5220.1413,
874
+ "eval_samples_per_second": 3.188,
875
+ "eval_steps_per_second": 0.398,
876
+ "step": 12000
877
+ },
878
+ {
879
+ "epoch": 0.7276882367091653,
880
+ "grad_norm": 1.6520166397094727,
881
+ "learning_rate": 3.787587202309358e-05,
882
+ "loss": 0.9415,
883
+ "step": 12100
884
+ },
885
+ {
886
+ "epoch": 0.7337021890786625,
887
+ "grad_norm": 1.7703155279159546,
888
+ "learning_rate": 3.777563948360196e-05,
889
+ "loss": 0.9653,
890
+ "step": 12200
891
+ },
892
+ {
893
+ "epoch": 0.7397161414481598,
894
+ "grad_norm": 1.4374829530715942,
895
+ "learning_rate": 3.7675406944110334e-05,
896
+ "loss": 0.9541,
897
+ "step": 12300
898
+ },
899
+ {
900
+ "epoch": 0.745730093817657,
901
+ "grad_norm": 1.790893793106079,
902
+ "learning_rate": 3.7575174404618716e-05,
903
+ "loss": 0.9951,
904
+ "step": 12400
905
+ },
906
+ {
907
+ "epoch": 0.7517440461871542,
908
+ "grad_norm": 1.6874678134918213,
909
+ "learning_rate": 3.74749418651271e-05,
910
+ "loss": 0.9704,
911
+ "step": 12500
912
+ },
913
+ {
914
+ "epoch": 0.7577579985566514,
915
+ "grad_norm": 1.7682390213012695,
916
+ "learning_rate": 3.737470932563548e-05,
917
+ "loss": 0.9461,
918
+ "step": 12600
919
+ },
920
+ {
921
+ "epoch": 0.7637719509261487,
922
+ "grad_norm": 1.6911418437957764,
923
+ "learning_rate": 3.727447678614386e-05,
924
+ "loss": 0.9952,
925
+ "step": 12700
926
+ },
927
+ {
928
+ "epoch": 0.7697859032956459,
929
+ "grad_norm": 1.727547526359558,
930
+ "learning_rate": 3.7174244246652234e-05,
931
+ "loss": 0.9442,
932
+ "step": 12800
933
+ },
934
+ {
935
+ "epoch": 0.7757998556651431,
936
+ "grad_norm": 1.4808145761489868,
937
+ "learning_rate": 3.7074011707160616e-05,
938
+ "loss": 0.9742,
939
+ "step": 12900
940
+ },
941
+ {
942
+ "epoch": 0.7818138080346404,
943
+ "grad_norm": 1.876441240310669,
944
+ "learning_rate": 3.697377916766899e-05,
945
+ "loss": 0.9766,
946
+ "step": 13000
947
+ },
948
+ {
949
+ "epoch": 0.7878277604041376,
950
+ "grad_norm": 1.4900315999984741,
951
+ "learning_rate": 3.687354662817737e-05,
952
+ "loss": 0.9612,
953
+ "step": 13100
954
+ },
955
+ {
956
+ "epoch": 0.7938417127736348,
957
+ "grad_norm": 1.4132503271102905,
958
+ "learning_rate": 3.677331408868575e-05,
959
+ "loss": 0.9651,
960
+ "step": 13200
961
+ },
962
+ {
963
+ "epoch": 0.7998556651431321,
964
+ "grad_norm": 1.6306148767471313,
965
+ "learning_rate": 3.6673081549194134e-05,
966
+ "loss": 0.9368,
967
+ "step": 13300
968
+ },
969
+ {
970
+ "epoch": 0.8058696175126293,
971
+ "grad_norm": 1.7364792823791504,
972
+ "learning_rate": 3.657284900970251e-05,
973
+ "loss": 0.9427,
974
+ "step": 13400
975
+ },
976
+ {
977
+ "epoch": 0.8118835698821265,
978
+ "grad_norm": 1.6632161140441895,
979
+ "learning_rate": 3.647261647021089e-05,
980
+ "loss": 0.9505,
981
+ "step": 13500
982
+ },
983
+ {
984
+ "epoch": 0.8178975222516238,
985
+ "grad_norm": 1.5362128019332886,
986
+ "learning_rate": 3.637238393071927e-05,
987
+ "loss": 0.9692,
988
+ "step": 13600
989
+ },
990
+ {
991
+ "epoch": 0.823911474621121,
992
+ "grad_norm": 1.3290611505508423,
993
+ "learning_rate": 3.6272151391227646e-05,
994
+ "loss": 0.9585,
995
+ "step": 13700
996
+ },
997
+ {
998
+ "epoch": 0.8299254269906182,
999
+ "grad_norm": 1.6647266149520874,
1000
+ "learning_rate": 3.617191885173603e-05,
1001
+ "loss": 0.9725,
1002
+ "step": 13800
1003
+ },
1004
+ {
1005
+ "epoch": 0.8359393793601154,
1006
+ "grad_norm": 1.736165165901184,
1007
+ "learning_rate": 3.607168631224441e-05,
1008
+ "loss": 0.9611,
1009
+ "step": 13900
1010
+ },
1011
+ {
1012
+ "epoch": 0.8419533317296127,
1013
+ "grad_norm": 1.6238832473754883,
1014
+ "learning_rate": 3.597145377275279e-05,
1015
+ "loss": 0.9698,
1016
+ "step": 14000
1017
+ },
1018
+ {
1019
+ "epoch": 0.8479672840991099,
1020
+ "grad_norm": 1.741194486618042,
1021
+ "learning_rate": 3.5871221233261165e-05,
1022
+ "loss": 0.9338,
1023
+ "step": 14100
1024
+ },
1025
+ {
1026
+ "epoch": 0.8539812364686071,
1027
+ "grad_norm": 1.7224496603012085,
1028
+ "learning_rate": 3.5770988693769546e-05,
1029
+ "loss": 0.9715,
1030
+ "step": 14200
1031
+ },
1032
+ {
1033
+ "epoch": 0.8599951888381044,
1034
+ "grad_norm": 1.872253179550171,
1035
+ "learning_rate": 3.5671758479672845e-05,
1036
+ "loss": 0.9656,
1037
+ "step": 14300
1038
+ },
1039
+ {
1040
+ "epoch": 0.8660091412076016,
1041
+ "grad_norm": 1.5153071880340576,
1042
+ "learning_rate": 3.557152594018122e-05,
1043
+ "loss": 0.984,
1044
+ "step": 14400
1045
+ },
1046
+ {
1047
+ "epoch": 0.8720230935770988,
1048
+ "grad_norm": 1.667662262916565,
1049
+ "learning_rate": 3.54712934006896e-05,
1050
+ "loss": 0.9386,
1051
+ "step": 14500
1052
+ },
1053
+ {
1054
+ "epoch": 0.8780370459465962,
1055
+ "grad_norm": 1.7471551895141602,
1056
+ "learning_rate": 3.537106086119798e-05,
1057
+ "loss": 0.9486,
1058
+ "step": 14600
1059
+ },
1060
+ {
1061
+ "epoch": 0.8840509983160934,
1062
+ "grad_norm": 1.5477312803268433,
1063
+ "learning_rate": 3.5270828321706364e-05,
1064
+ "loss": 0.9451,
1065
+ "step": 14700
1066
+ },
1067
+ {
1068
+ "epoch": 0.8900649506855906,
1069
+ "grad_norm": 1.5079952478408813,
1070
+ "learning_rate": 3.517059578221474e-05,
1071
+ "loss": 0.9381,
1072
+ "step": 14800
1073
+ },
1074
+ {
1075
+ "epoch": 0.8960789030550877,
1076
+ "grad_norm": 1.5696821212768555,
1077
+ "learning_rate": 3.507036324272312e-05,
1078
+ "loss": 0.9418,
1079
+ "step": 14900
1080
+ },
1081
+ {
1082
+ "epoch": 0.9020928554245851,
1083
+ "grad_norm": 1.552612066268921,
1084
+ "learning_rate": 3.4970130703231494e-05,
1085
+ "loss": 0.9596,
1086
+ "step": 15000
1087
+ },
1088
+ {
1089
+ "epoch": 0.9081068077940823,
1090
+ "grad_norm": 1.7802802324295044,
1091
+ "learning_rate": 3.4869898163739876e-05,
1092
+ "loss": 0.9298,
1093
+ "step": 15100
1094
+ },
1095
+ {
1096
+ "epoch": 0.9141207601635795,
1097
+ "grad_norm": 1.5589861869812012,
1098
+ "learning_rate": 3.476966562424826e-05,
1099
+ "loss": 0.9621,
1100
+ "step": 15200
1101
+ },
1102
+ {
1103
+ "epoch": 0.9201347125330768,
1104
+ "grad_norm": 1.3050284385681152,
1105
+ "learning_rate": 3.466943308475664e-05,
1106
+ "loss": 0.9315,
1107
+ "step": 15300
1108
+ },
1109
+ {
1110
+ "epoch": 0.926148664902574,
1111
+ "grad_norm": 1.5091936588287354,
1112
+ "learning_rate": 3.456920054526502e-05,
1113
+ "loss": 0.9107,
1114
+ "step": 15400
1115
+ },
1116
+ {
1117
+ "epoch": 0.9321626172720712,
1118
+ "grad_norm": 1.5565592050552368,
1119
+ "learning_rate": 3.4468968005773394e-05,
1120
+ "loss": 0.9552,
1121
+ "step": 15500
1122
+ },
1123
+ {
1124
+ "epoch": 0.9381765696415685,
1125
+ "grad_norm": 1.7788596153259277,
1126
+ "learning_rate": 3.4368735466281775e-05,
1127
+ "loss": 0.9459,
1128
+ "step": 15600
1129
+ },
1130
+ {
1131
+ "epoch": 0.9441905220110657,
1132
+ "grad_norm": 1.8757565021514893,
1133
+ "learning_rate": 3.426850292679015e-05,
1134
+ "loss": 0.9254,
1135
+ "step": 15700
1136
+ },
1137
+ {
1138
+ "epoch": 0.9502044743805629,
1139
+ "grad_norm": 1.6978222131729126,
1140
+ "learning_rate": 3.416827038729853e-05,
1141
+ "loss": 0.9083,
1142
+ "step": 15800
1143
+ },
1144
+ {
1145
+ "epoch": 0.9562184267500602,
1146
+ "grad_norm": 1.6449016332626343,
1147
+ "learning_rate": 3.406803784780692e-05,
1148
+ "loss": 0.9767,
1149
+ "step": 15900
1150
+ },
1151
+ {
1152
+ "epoch": 0.9622323791195574,
1153
+ "grad_norm": 1.8677603006362915,
1154
+ "learning_rate": 3.3967805308315294e-05,
1155
+ "loss": 0.9469,
1156
+ "step": 16000
1157
+ },
1158
+ {
1159
+ "epoch": 0.9622323791195574,
1160
+ "eval_cer": 0.7716098871709525,
1161
+ "eval_loss": 0.8905403017997742,
1162
+ "eval_runtime": 5089.7823,
1163
+ "eval_samples_per_second": 3.269,
1164
+ "eval_steps_per_second": 0.409,
1165
+ "step": 16000
1166
+ },
1167
+ {
1168
+ "epoch": 0.9682463314890546,
1169
+ "grad_norm": 1.6178405284881592,
1170
+ "learning_rate": 3.3867572768823675e-05,
1171
+ "loss": 0.9441,
1172
+ "step": 16100
1173
+ },
1174
+ {
1175
+ "epoch": 0.9742602838585518,
1176
+ "grad_norm": 1.7407509088516235,
1177
+ "learning_rate": 3.376734022933205e-05,
1178
+ "loss": 0.9264,
1179
+ "step": 16200
1180
+ },
1181
+ {
1182
+ "epoch": 0.9802742362280491,
1183
+ "grad_norm": 1.5831618309020996,
1184
+ "learning_rate": 3.366710768984043e-05,
1185
+ "loss": 0.9583,
1186
+ "step": 16300
1187
+ },
1188
+ {
1189
+ "epoch": 0.9862881885975463,
1190
+ "grad_norm": 1.6853969097137451,
1191
+ "learning_rate": 3.3566875150348806e-05,
1192
+ "loss": 0.9226,
1193
+ "step": 16400
1194
+ },
1195
+ {
1196
+ "epoch": 0.9923021409670435,
1197
+ "grad_norm": 1.8264875411987305,
1198
+ "learning_rate": 3.3466642610857194e-05,
1199
+ "loss": 0.9492,
1200
+ "step": 16500
1201
+ },
1202
+ {
1203
+ "epoch": 0.9983160933365408,
1204
+ "grad_norm": 1.5579068660736084,
1205
+ "learning_rate": 3.336641007136557e-05,
1206
+ "loss": 0.8872,
1207
+ "step": 16600
1208
+ },
1209
+ {
1210
+ "epoch": 1.004330045706038,
1211
+ "grad_norm": 1.7848278284072876,
1212
+ "learning_rate": 3.326617753187395e-05,
1213
+ "loss": 0.8717,
1214
+ "step": 16700
1215
+ },
1216
+ {
1217
+ "epoch": 1.0103439980755353,
1218
+ "grad_norm": 1.5830621719360352,
1219
+ "learning_rate": 3.316694731777725e-05,
1220
+ "loss": 0.9032,
1221
+ "step": 16800
1222
+ },
1223
+ {
1224
+ "epoch": 1.0163579504450324,
1225
+ "grad_norm": 1.546217441558838,
1226
+ "learning_rate": 3.3066714778285623e-05,
1227
+ "loss": 0.9418,
1228
+ "step": 16900
1229
+ },
1230
+ {
1231
+ "epoch": 1.0223719028145297,
1232
+ "grad_norm": 1.4024384021759033,
1233
+ "learning_rate": 3.2966482238794005e-05,
1234
+ "loss": 0.9075,
1235
+ "step": 17000
1236
+ },
1237
+ {
1238
+ "epoch": 1.028385855184027,
1239
+ "grad_norm": 1.6523361206054688,
1240
+ "learning_rate": 3.286624969930238e-05,
1241
+ "loss": 0.9088,
1242
+ "step": 17100
1243
+ },
1244
+ {
1245
+ "epoch": 1.0343998075535241,
1246
+ "grad_norm": 1.4941192865371704,
1247
+ "learning_rate": 3.276601715981076e-05,
1248
+ "loss": 0.8703,
1249
+ "step": 17200
1250
+ },
1251
+ {
1252
+ "epoch": 1.0404137599230214,
1253
+ "grad_norm": 1.6586402654647827,
1254
+ "learning_rate": 3.266578462031914e-05,
1255
+ "loss": 0.9171,
1256
+ "step": 17300
1257
+ },
1258
+ {
1259
+ "epoch": 1.0464277122925187,
1260
+ "grad_norm": 1.5614475011825562,
1261
+ "learning_rate": 3.2565552080827523e-05,
1262
+ "loss": 0.8819,
1263
+ "step": 17400
1264
+ },
1265
+ {
1266
+ "epoch": 1.0524416646620158,
1267
+ "grad_norm": 1.6588680744171143,
1268
+ "learning_rate": 3.2465319541335905e-05,
1269
+ "loss": 0.9165,
1270
+ "step": 17500
1271
+ },
1272
+ {
1273
+ "epoch": 1.0584556170315131,
1274
+ "grad_norm": 1.4571171998977661,
1275
+ "learning_rate": 3.236508700184428e-05,
1276
+ "loss": 0.9056,
1277
+ "step": 17600
1278
+ },
1279
+ {
1280
+ "epoch": 1.0644695694010102,
1281
+ "grad_norm": 1.7484580278396606,
1282
+ "learning_rate": 3.226485446235266e-05,
1283
+ "loss": 0.9072,
1284
+ "step": 17700
1285
+ },
1286
+ {
1287
+ "epoch": 1.0704835217705075,
1288
+ "grad_norm": 1.388741374015808,
1289
+ "learning_rate": 3.2164621922861035e-05,
1290
+ "loss": 0.8645,
1291
+ "step": 17800
1292
+ },
1293
+ {
1294
+ "epoch": 1.0764974741400049,
1295
+ "grad_norm": 1.5871518850326538,
1296
+ "learning_rate": 3.206438938336942e-05,
1297
+ "loss": 0.908,
1298
+ "step": 17900
1299
+ },
1300
+ {
1301
+ "epoch": 1.0825114265095022,
1302
+ "grad_norm": 1.4603219032287598,
1303
+ "learning_rate": 3.19641568438778e-05,
1304
+ "loss": 0.8714,
1305
+ "step": 18000
1306
+ },
1307
+ {
1308
+ "epoch": 1.0885253788789993,
1309
+ "grad_norm": 1.443608283996582,
1310
+ "learning_rate": 3.186392430438618e-05,
1311
+ "loss": 0.8879,
1312
+ "step": 18100
1313
+ },
1314
+ {
1315
+ "epoch": 1.0945393312484966,
1316
+ "grad_norm": 1.4648326635360718,
1317
+ "learning_rate": 3.1763691764894554e-05,
1318
+ "loss": 0.923,
1319
+ "step": 18200
1320
+ },
1321
+ {
1322
+ "epoch": 1.1005532836179936,
1323
+ "grad_norm": 1.9082708358764648,
1324
+ "learning_rate": 3.1663459225402935e-05,
1325
+ "loss": 0.8898,
1326
+ "step": 18300
1327
+ },
1328
+ {
1329
+ "epoch": 1.106567235987491,
1330
+ "grad_norm": 1.740161418914795,
1331
+ "learning_rate": 3.1563226685911317e-05,
1332
+ "loss": 0.8913,
1333
+ "step": 18400
1334
+ },
1335
+ {
1336
+ "epoch": 1.1125811883569883,
1337
+ "grad_norm": 1.4581352472305298,
1338
+ "learning_rate": 3.146299414641969e-05,
1339
+ "loss": 0.9202,
1340
+ "step": 18500
1341
+ },
1342
+ {
1343
+ "epoch": 1.1185951407264854,
1344
+ "grad_norm": 1.5199153423309326,
1345
+ "learning_rate": 3.136276160692808e-05,
1346
+ "loss": 0.9032,
1347
+ "step": 18600
1348
+ },
1349
+ {
1350
+ "epoch": 1.1246090930959827,
1351
+ "grad_norm": 1.4630061388015747,
1352
+ "learning_rate": 3.1262529067436454e-05,
1353
+ "loss": 0.8771,
1354
+ "step": 18700
1355
+ },
1356
+ {
1357
+ "epoch": 1.13062304546548,
1358
+ "grad_norm": 1.7790659666061401,
1359
+ "learning_rate": 3.1162296527944835e-05,
1360
+ "loss": 0.8673,
1361
+ "step": 18800
1362
+ },
1363
+ {
1364
+ "epoch": 1.136636997834977,
1365
+ "grad_norm": 1.610372543334961,
1366
+ "learning_rate": 3.106206398845321e-05,
1367
+ "loss": 0.9019,
1368
+ "step": 18900
1369
+ },
1370
+ {
1371
+ "epoch": 1.1426509502044744,
1372
+ "grad_norm": 1.847385048866272,
1373
+ "learning_rate": 3.096183144896159e-05,
1374
+ "loss": 0.8907,
1375
+ "step": 19000
1376
+ },
1377
+ {
1378
+ "epoch": 1.1486649025739717,
1379
+ "grad_norm": 1.664432168006897,
1380
+ "learning_rate": 3.086159890946997e-05,
1381
+ "loss": 0.8745,
1382
+ "step": 19100
1383
+ },
1384
+ {
1385
+ "epoch": 1.1546788549434688,
1386
+ "grad_norm": 1.5932984352111816,
1387
+ "learning_rate": 3.0761366369978354e-05,
1388
+ "loss": 0.8697,
1389
+ "step": 19200
1390
+ },
1391
+ {
1392
+ "epoch": 1.160692807312966,
1393
+ "grad_norm": 1.4511469602584839,
1394
+ "learning_rate": 3.0662136155881646e-05,
1395
+ "loss": 0.8748,
1396
+ "step": 19300
1397
+ },
1398
+ {
1399
+ "epoch": 1.1667067596824634,
1400
+ "grad_norm": 1.9143450260162354,
1401
+ "learning_rate": 3.056190361639003e-05,
1402
+ "loss": 0.9045,
1403
+ "step": 19400
1404
+ },
1405
+ {
1406
+ "epoch": 1.1727207120519605,
1407
+ "grad_norm": 1.3927329778671265,
1408
+ "learning_rate": 3.046167107689841e-05,
1409
+ "loss": 0.8927,
1410
+ "step": 19500
1411
+ },
1412
+ {
1413
+ "epoch": 1.1787346644214578,
1414
+ "grad_norm": 1.8178434371948242,
1415
+ "learning_rate": 3.0361438537406783e-05,
1416
+ "loss": 0.9258,
1417
+ "step": 19600
1418
+ },
1419
+ {
1420
+ "epoch": 1.1847486167909551,
1421
+ "grad_norm": 1.6006417274475098,
1422
+ "learning_rate": 3.0261205997915165e-05,
1423
+ "loss": 0.886,
1424
+ "step": 19700
1425
+ },
1426
+ {
1427
+ "epoch": 1.1907625691604522,
1428
+ "grad_norm": 1.6698856353759766,
1429
+ "learning_rate": 3.0160973458423543e-05,
1430
+ "loss": 0.9188,
1431
+ "step": 19800
1432
+ },
1433
+ {
1434
+ "epoch": 1.1967765215299495,
1435
+ "grad_norm": 1.5293818712234497,
1436
+ "learning_rate": 3.0060740918931924e-05,
1437
+ "loss": 0.896,
1438
+ "step": 19900
1439
+ },
1440
+ {
1441
+ "epoch": 1.2027904738994466,
1442
+ "grad_norm": 1.6966098546981812,
1443
+ "learning_rate": 2.9960508379440305e-05,
1444
+ "loss": 0.8677,
1445
+ "step": 20000
1446
+ },
1447
+ {
1448
+ "epoch": 1.2027904738994466,
1449
+ "eval_cer": 0.7689350483770306,
1450
+ "eval_loss": 0.8620118498802185,
1451
+ "eval_runtime": 5134.4813,
1452
+ "eval_samples_per_second": 3.241,
1453
+ "eval_steps_per_second": 0.405,
1454
+ "step": 20000
1455
+ },
1456
+ {
1457
+ "epoch": 1.208804426268944,
1458
+ "grad_norm": 1.6066211462020874,
1459
+ "learning_rate": 2.9860275839948683e-05,
1460
+ "loss": 0.8648,
1461
+ "step": 20100
1462
+ },
1463
+ {
1464
+ "epoch": 1.2148183786384412,
1465
+ "grad_norm": 1.4755396842956543,
1466
+ "learning_rate": 2.9760043300457065e-05,
1467
+ "loss": 0.8683,
1468
+ "step": 20200
1469
+ },
1470
+ {
1471
+ "epoch": 1.2208323310079385,
1472
+ "grad_norm": 1.3547738790512085,
1473
+ "learning_rate": 2.965981076096544e-05,
1474
+ "loss": 0.8786,
1475
+ "step": 20300
1476
+ },
1477
+ {
1478
+ "epoch": 1.2268462833774356,
1479
+ "grad_norm": 1.6254231929779053,
1480
+ "learning_rate": 2.955957822147382e-05,
1481
+ "loss": 0.8758,
1482
+ "step": 20400
1483
+ },
1484
+ {
1485
+ "epoch": 1.232860235746933,
1486
+ "grad_norm": 1.5725833177566528,
1487
+ "learning_rate": 2.94593456819822e-05,
1488
+ "loss": 0.8831,
1489
+ "step": 20500
1490
+ },
1491
+ {
1492
+ "epoch": 1.23887418811643,
1493
+ "grad_norm": 1.6321443319320679,
1494
+ "learning_rate": 2.935911314249058e-05,
1495
+ "loss": 0.868,
1496
+ "step": 20600
1497
+ },
1498
+ {
1499
+ "epoch": 1.2448881404859273,
1500
+ "grad_norm": 1.544110894203186,
1501
+ "learning_rate": 2.925888060299896e-05,
1502
+ "loss": 0.9033,
1503
+ "step": 20700
1504
+ },
1505
+ {
1506
+ "epoch": 1.2509020928554246,
1507
+ "grad_norm": 1.5690948963165283,
1508
+ "learning_rate": 2.915864806350734e-05,
1509
+ "loss": 0.8736,
1510
+ "step": 20800
1511
+ },
1512
+ {
1513
+ "epoch": 1.256916045224922,
1514
+ "grad_norm": 1.3950625658035278,
1515
+ "learning_rate": 2.905841552401572e-05,
1516
+ "loss": 0.9012,
1517
+ "step": 20900
1518
+ },
1519
+ {
1520
+ "epoch": 1.262929997594419,
1521
+ "grad_norm": 1.4699276685714722,
1522
+ "learning_rate": 2.8958182984524095e-05,
1523
+ "loss": 0.8545,
1524
+ "step": 21000
1525
+ },
1526
+ {
1527
+ "epoch": 1.2689439499639164,
1528
+ "grad_norm": 1.607750654220581,
1529
+ "learning_rate": 2.8857950445032476e-05,
1530
+ "loss": 0.9101,
1531
+ "step": 21100
1532
+ },
1533
+ {
1534
+ "epoch": 1.2749579023334134,
1535
+ "grad_norm": 2.1910347938537598,
1536
+ "learning_rate": 2.8757717905540854e-05,
1537
+ "loss": 0.8803,
1538
+ "step": 21200
1539
+ },
1540
+ {
1541
+ "epoch": 1.2809718547029108,
1542
+ "grad_norm": 1.604390025138855,
1543
+ "learning_rate": 2.8657485366049236e-05,
1544
+ "loss": 0.8751,
1545
+ "step": 21300
1546
+ },
1547
+ {
1548
+ "epoch": 1.286985807072408,
1549
+ "grad_norm": 1.5971423387527466,
1550
+ "learning_rate": 2.8557252826557617e-05,
1551
+ "loss": 0.8543,
1552
+ "step": 21400
1553
+ },
1554
+ {
1555
+ "epoch": 1.2929997594419051,
1556
+ "grad_norm": 1.6325972080230713,
1557
+ "learning_rate": 2.8457020287065995e-05,
1558
+ "loss": 0.8612,
1559
+ "step": 21500
1560
+ },
1561
+ {
1562
+ "epoch": 1.2990137118114025,
1563
+ "grad_norm": 1.7952935695648193,
1564
+ "learning_rate": 2.8356787747574376e-05,
1565
+ "loss": 0.8399,
1566
+ "step": 21600
1567
+ },
1568
+ {
1569
+ "epoch": 1.3050276641808995,
1570
+ "grad_norm": 1.683236002922058,
1571
+ "learning_rate": 2.825655520808275e-05,
1572
+ "loss": 0.8585,
1573
+ "step": 21700
1574
+ },
1575
+ {
1576
+ "epoch": 1.3110416165503969,
1577
+ "grad_norm": 1.630666971206665,
1578
+ "learning_rate": 2.8156322668591136e-05,
1579
+ "loss": 0.8891,
1580
+ "step": 21800
1581
+ },
1582
+ {
1583
+ "epoch": 1.3170555689198942,
1584
+ "grad_norm": 1.7404346466064453,
1585
+ "learning_rate": 2.805609012909951e-05,
1586
+ "loss": 0.8514,
1587
+ "step": 21900
1588
+ },
1589
+ {
1590
+ "epoch": 1.3230695212893915,
1591
+ "grad_norm": 1.4314298629760742,
1592
+ "learning_rate": 2.795585758960789e-05,
1593
+ "loss": 0.8561,
1594
+ "step": 22000
1595
+ },
1596
+ {
1597
+ "epoch": 1.3290834736588886,
1598
+ "grad_norm": 1.7691779136657715,
1599
+ "learning_rate": 2.785562505011627e-05,
1600
+ "loss": 0.8474,
1601
+ "step": 22100
1602
+ },
1603
+ {
1604
+ "epoch": 1.3350974260283859,
1605
+ "grad_norm": 1.7155267000198364,
1606
+ "learning_rate": 2.775539251062465e-05,
1607
+ "loss": 0.8289,
1608
+ "step": 22200
1609
+ },
1610
+ {
1611
+ "epoch": 1.341111378397883,
1612
+ "grad_norm": 1.7087023258209229,
1613
+ "learning_rate": 2.7655159971133032e-05,
1614
+ "loss": 0.8544,
1615
+ "step": 22300
1616
+ },
1617
+ {
1618
+ "epoch": 1.3471253307673803,
1619
+ "grad_norm": 1.617749571800232,
1620
+ "learning_rate": 2.755492743164141e-05,
1621
+ "loss": 0.8781,
1622
+ "step": 22400
1623
+ },
1624
+ {
1625
+ "epoch": 1.3531392831368776,
1626
+ "grad_norm": 1.6493247747421265,
1627
+ "learning_rate": 2.745469489214979e-05,
1628
+ "loss": 0.8392,
1629
+ "step": 22500
1630
+ },
1631
+ {
1632
+ "epoch": 1.359153235506375,
1633
+ "grad_norm": 1.809634804725647,
1634
+ "learning_rate": 2.7354462352658166e-05,
1635
+ "loss": 0.8721,
1636
+ "step": 22600
1637
+ },
1638
+ {
1639
+ "epoch": 1.365167187875872,
1640
+ "grad_norm": 1.3698049783706665,
1641
+ "learning_rate": 2.7254229813166547e-05,
1642
+ "loss": 0.8533,
1643
+ "step": 22700
1644
+ },
1645
+ {
1646
+ "epoch": 1.3711811402453693,
1647
+ "grad_norm": 1.7568131685256958,
1648
+ "learning_rate": 2.7153997273674925e-05,
1649
+ "loss": 0.886,
1650
+ "step": 22800
1651
+ },
1652
+ {
1653
+ "epoch": 1.3771950926148664,
1654
+ "grad_norm": 1.867412805557251,
1655
+ "learning_rate": 2.7053764734183307e-05,
1656
+ "loss": 0.8637,
1657
+ "step": 22900
1658
+ },
1659
+ {
1660
+ "epoch": 1.3832090449843637,
1661
+ "grad_norm": 2.0730977058410645,
1662
+ "learning_rate": 2.6953532194691688e-05,
1663
+ "loss": 0.8626,
1664
+ "step": 23000
1665
+ },
1666
+ {
1667
+ "epoch": 1.389222997353861,
1668
+ "grad_norm": 1.8011558055877686,
1669
+ "learning_rate": 2.6853299655200066e-05,
1670
+ "loss": 0.8784,
1671
+ "step": 23100
1672
+ },
1673
+ {
1674
+ "epoch": 1.3952369497233583,
1675
+ "grad_norm": 1.6936458349227905,
1676
+ "learning_rate": 2.6753067115708447e-05,
1677
+ "loss": 0.8634,
1678
+ "step": 23200
1679
+ },
1680
+ {
1681
+ "epoch": 1.4012509020928554,
1682
+ "grad_norm": 1.7492289543151855,
1683
+ "learning_rate": 2.665383690161174e-05,
1684
+ "loss": 0.8678,
1685
+ "step": 23300
1686
+ },
1687
+ {
1688
+ "epoch": 1.4072648544623527,
1689
+ "grad_norm": 1.8972880840301514,
1690
+ "learning_rate": 2.655360436212012e-05,
1691
+ "loss": 0.8939,
1692
+ "step": 23400
1693
+ },
1694
+ {
1695
+ "epoch": 1.4132788068318498,
1696
+ "grad_norm": 1.6961406469345093,
1697
+ "learning_rate": 2.64533718226285e-05,
1698
+ "loss": 0.8727,
1699
+ "step": 23500
1700
+ },
1701
+ {
1702
+ "epoch": 1.4192927592013471,
1703
+ "grad_norm": 1.583854079246521,
1704
+ "learning_rate": 2.635313928313688e-05,
1705
+ "loss": 0.8332,
1706
+ "step": 23600
1707
+ },
1708
+ {
1709
+ "epoch": 1.4253067115708444,
1710
+ "grad_norm": 1.6541253328323364,
1711
+ "learning_rate": 2.6252906743645255e-05,
1712
+ "loss": 0.8798,
1713
+ "step": 23700
1714
+ },
1715
+ {
1716
+ "epoch": 1.4313206639403415,
1717
+ "grad_norm": 1.7607979774475098,
1718
+ "learning_rate": 2.6152674204153636e-05,
1719
+ "loss": 0.8472,
1720
+ "step": 23800
1721
+ },
1722
+ {
1723
+ "epoch": 1.4373346163098388,
1724
+ "grad_norm": 1.5591400861740112,
1725
+ "learning_rate": 2.605244166466202e-05,
1726
+ "loss": 0.8355,
1727
+ "step": 23900
1728
+ },
1729
+ {
1730
+ "epoch": 1.4433485686793361,
1731
+ "grad_norm": 1.4700669050216675,
1732
+ "learning_rate": 2.5952209125170395e-05,
1733
+ "loss": 0.8521,
1734
+ "step": 24000
1735
+ },
1736
+ {
1737
+ "epoch": 1.4433485686793361,
1738
+ "eval_cer": 0.7597912847133936,
1739
+ "eval_loss": 0.8321590423583984,
1740
+ "eval_runtime": 5084.7673,
1741
+ "eval_samples_per_second": 3.273,
1742
+ "eval_steps_per_second": 0.409,
1743
+ "step": 24000
1744
+ },
1745
+ {
1746
+ "epoch": 1.4493625210488332,
1747
+ "grad_norm": 1.4155552387237549,
1748
+ "learning_rate": 2.5851976585678777e-05,
1749
+ "loss": 0.8797,
1750
+ "step": 24100
1751
+ },
1752
+ {
1753
+ "epoch": 1.4553764734183305,
1754
+ "grad_norm": 1.4764596223831177,
1755
+ "learning_rate": 2.5751744046187155e-05,
1756
+ "loss": 0.8657,
1757
+ "step": 24200
1758
+ },
1759
+ {
1760
+ "epoch": 1.4613904257878279,
1761
+ "grad_norm": 1.463333010673523,
1762
+ "learning_rate": 2.5651511506695536e-05,
1763
+ "loss": 0.8746,
1764
+ "step": 24300
1765
+ },
1766
+ {
1767
+ "epoch": 1.467404378157325,
1768
+ "grad_norm": 1.5392202138900757,
1769
+ "learning_rate": 2.555127896720391e-05,
1770
+ "loss": 0.8512,
1771
+ "step": 24400
1772
+ },
1773
+ {
1774
+ "epoch": 1.4734183305268223,
1775
+ "grad_norm": 1.8480241298675537,
1776
+ "learning_rate": 2.5451046427712295e-05,
1777
+ "loss": 0.8562,
1778
+ "step": 24500
1779
+ },
1780
+ {
1781
+ "epoch": 1.4794322828963193,
1782
+ "grad_norm": 1.7533873319625854,
1783
+ "learning_rate": 2.5350813888220677e-05,
1784
+ "loss": 0.8534,
1785
+ "step": 24600
1786
+ },
1787
+ {
1788
+ "epoch": 1.4854462352658167,
1789
+ "grad_norm": 1.6647679805755615,
1790
+ "learning_rate": 2.525058134872905e-05,
1791
+ "loss": 0.8335,
1792
+ "step": 24700
1793
+ },
1794
+ {
1795
+ "epoch": 1.491460187635314,
1796
+ "grad_norm": 1.8899763822555542,
1797
+ "learning_rate": 2.5150348809237433e-05,
1798
+ "loss": 0.8486,
1799
+ "step": 24800
1800
+ },
1801
+ {
1802
+ "epoch": 1.4974741400048113,
1803
+ "grad_norm": 1.7569955587387085,
1804
+ "learning_rate": 2.505011626974581e-05,
1805
+ "loss": 0.8415,
1806
+ "step": 24900
1807
+ },
1808
+ {
1809
+ "epoch": 1.5034880923743084,
1810
+ "grad_norm": 1.5141854286193848,
1811
+ "learning_rate": 2.4949883730254192e-05,
1812
+ "loss": 0.8107,
1813
+ "step": 25000
1814
+ },
1815
+ {
1816
+ "epoch": 1.5095020447438057,
1817
+ "grad_norm": 1.8239057064056396,
1818
+ "learning_rate": 2.484965119076257e-05,
1819
+ "loss": 0.8841,
1820
+ "step": 25100
1821
+ },
1822
+ {
1823
+ "epoch": 1.5155159971133028,
1824
+ "grad_norm": 1.433118224143982,
1825
+ "learning_rate": 2.474941865127095e-05,
1826
+ "loss": 0.8408,
1827
+ "step": 25200
1828
+ },
1829
+ {
1830
+ "epoch": 1.5215299494828,
1831
+ "grad_norm": 1.6874032020568848,
1832
+ "learning_rate": 2.4650188437174247e-05,
1833
+ "loss": 0.8387,
1834
+ "step": 25300
1835
+ },
1836
+ {
1837
+ "epoch": 1.5275439018522974,
1838
+ "grad_norm": 1.4172905683517456,
1839
+ "learning_rate": 2.4549955897682625e-05,
1840
+ "loss": 0.8357,
1841
+ "step": 25400
1842
+ },
1843
+ {
1844
+ "epoch": 1.5335578542217947,
1845
+ "grad_norm": 1.4848599433898926,
1846
+ "learning_rate": 2.4449723358191003e-05,
1847
+ "loss": 0.8451,
1848
+ "step": 25500
1849
+ },
1850
+ {
1851
+ "epoch": 1.5395718065912918,
1852
+ "grad_norm": 1.4670342206954956,
1853
+ "learning_rate": 2.4349490818699384e-05,
1854
+ "loss": 0.8505,
1855
+ "step": 25600
1856
+ },
1857
+ {
1858
+ "epoch": 1.5455857589607889,
1859
+ "grad_norm": 1.5914552211761475,
1860
+ "learning_rate": 2.4249258279207762e-05,
1861
+ "loss": 0.8635,
1862
+ "step": 25700
1863
+ },
1864
+ {
1865
+ "epoch": 1.5515997113302862,
1866
+ "grad_norm": 1.5550841093063354,
1867
+ "learning_rate": 2.414902573971614e-05,
1868
+ "loss": 0.8763,
1869
+ "step": 25800
1870
+ },
1871
+ {
1872
+ "epoch": 1.5576136636997835,
1873
+ "grad_norm": 1.5907316207885742,
1874
+ "learning_rate": 2.4048793200224525e-05,
1875
+ "loss": 0.8664,
1876
+ "step": 25900
1877
+ },
1878
+ {
1879
+ "epoch": 1.5636276160692808,
1880
+ "grad_norm": 1.4494388103485107,
1881
+ "learning_rate": 2.3948560660732903e-05,
1882
+ "loss": 0.819,
1883
+ "step": 26000
1884
+ },
1885
+ {
1886
+ "epoch": 1.5696415684387781,
1887
+ "grad_norm": 1.599004864692688,
1888
+ "learning_rate": 2.384832812124128e-05,
1889
+ "loss": 0.8363,
1890
+ "step": 26100
1891
+ },
1892
+ {
1893
+ "epoch": 1.5756555208082752,
1894
+ "grad_norm": 1.887817621231079,
1895
+ "learning_rate": 2.3748095581749662e-05,
1896
+ "loss": 0.8845,
1897
+ "step": 26200
1898
+ },
1899
+ {
1900
+ "epoch": 1.5816694731777723,
1901
+ "grad_norm": 1.3124005794525146,
1902
+ "learning_rate": 2.364786304225804e-05,
1903
+ "loss": 0.858,
1904
+ "step": 26300
1905
+ },
1906
+ {
1907
+ "epoch": 1.5876834255472696,
1908
+ "grad_norm": 1.6560554504394531,
1909
+ "learning_rate": 2.3547630502766418e-05,
1910
+ "loss": 0.8553,
1911
+ "step": 26400
1912
+ },
1913
+ {
1914
+ "epoch": 1.593697377916767,
1915
+ "grad_norm": 1.5678675174713135,
1916
+ "learning_rate": 2.34473979632748e-05,
1917
+ "loss": 0.8488,
1918
+ "step": 26500
1919
+ },
1920
+ {
1921
+ "epoch": 1.5997113302862642,
1922
+ "grad_norm": 1.4168376922607422,
1923
+ "learning_rate": 2.334716542378318e-05,
1924
+ "loss": 0.8263,
1925
+ "step": 26600
1926
+ },
1927
+ {
1928
+ "epoch": 1.6057252826557615,
1929
+ "grad_norm": 1.6189205646514893,
1930
+ "learning_rate": 2.324693288429156e-05,
1931
+ "loss": 0.8489,
1932
+ "step": 26700
1933
+ },
1934
+ {
1935
+ "epoch": 1.6117392350252586,
1936
+ "grad_norm": 1.7498302459716797,
1937
+ "learning_rate": 2.3146700344799936e-05,
1938
+ "loss": 0.8069,
1939
+ "step": 26800
1940
+ },
1941
+ {
1942
+ "epoch": 1.6177531873947557,
1943
+ "grad_norm": 1.5609160661697388,
1944
+ "learning_rate": 2.3046467805308318e-05,
1945
+ "loss": 0.8337,
1946
+ "step": 26900
1947
+ },
1948
+ {
1949
+ "epoch": 1.623767139764253,
1950
+ "grad_norm": 1.7673338651657104,
1951
+ "learning_rate": 2.2946235265816696e-05,
1952
+ "loss": 0.8264,
1953
+ "step": 27000
1954
+ },
1955
+ {
1956
+ "epoch": 1.6297810921337503,
1957
+ "grad_norm": 1.593299150466919,
1958
+ "learning_rate": 2.2846002726325074e-05,
1959
+ "loss": 0.828,
1960
+ "step": 27100
1961
+ },
1962
+ {
1963
+ "epoch": 1.6357950445032476,
1964
+ "grad_norm": 1.5802645683288574,
1965
+ "learning_rate": 2.274677251222837e-05,
1966
+ "loss": 0.8428,
1967
+ "step": 27200
1968
+ },
1969
+ {
1970
+ "epoch": 1.6418089968727447,
1971
+ "grad_norm": 1.7584878206253052,
1972
+ "learning_rate": 2.264653997273675e-05,
1973
+ "loss": 0.8378,
1974
+ "step": 27300
1975
+ },
1976
+ {
1977
+ "epoch": 1.647822949242242,
1978
+ "grad_norm": 1.5360692739486694,
1979
+ "learning_rate": 2.254630743324513e-05,
1980
+ "loss": 0.8288,
1981
+ "step": 27400
1982
+ },
1983
+ {
1984
+ "epoch": 1.6538369016117391,
1985
+ "grad_norm": 1.5635976791381836,
1986
+ "learning_rate": 2.244607489375351e-05,
1987
+ "loss": 0.8208,
1988
+ "step": 27500
1989
+ },
1990
+ {
1991
+ "epoch": 1.6598508539812364,
1992
+ "grad_norm": 1.778735876083374,
1993
+ "learning_rate": 2.2345842354261888e-05,
1994
+ "loss": 0.847,
1995
+ "step": 27600
1996
+ },
1997
+ {
1998
+ "epoch": 1.6658648063507338,
1999
+ "grad_norm": 1.5961335897445679,
2000
+ "learning_rate": 2.224560981477027e-05,
2001
+ "loss": 0.8059,
2002
+ "step": 27700
2003
+ },
2004
+ {
2005
+ "epoch": 1.671878758720231,
2006
+ "grad_norm": 1.396517038345337,
2007
+ "learning_rate": 2.2145377275278647e-05,
2008
+ "loss": 0.8406,
2009
+ "step": 27800
2010
+ },
2011
+ {
2012
+ "epoch": 1.6778927110897282,
2013
+ "grad_norm": 1.554319977760315,
2014
+ "learning_rate": 2.2045144735787025e-05,
2015
+ "loss": 0.836,
2016
+ "step": 27900
2017
+ },
2018
+ {
2019
+ "epoch": 1.6839066634592252,
2020
+ "grad_norm": 1.5663318634033203,
2021
+ "learning_rate": 2.1944912196295406e-05,
2022
+ "loss": 0.8386,
2023
+ "step": 28000
2024
+ },
2025
+ {
2026
+ "epoch": 1.6839066634592252,
2027
+ "eval_cer": 0.7604478286805494,
2028
+ "eval_loss": 0.8103429079055786,
2029
+ "eval_runtime": 5152.8312,
2030
+ "eval_samples_per_second": 3.229,
2031
+ "eval_steps_per_second": 0.404,
2032
+ "step": 28000
2033
+ },
2034
+ {
2035
+ "epoch": 1.6899206158287225,
2036
+ "grad_norm": 1.7991820573806763,
2037
+ "learning_rate": 2.1844679656803784e-05,
2038
+ "loss": 0.8428,
2039
+ "step": 28100
2040
+ },
2041
+ {
2042
+ "epoch": 1.6959345681982199,
2043
+ "grad_norm": 1.6566849946975708,
2044
+ "learning_rate": 2.1744447117312162e-05,
2045
+ "loss": 0.8331,
2046
+ "step": 28200
2047
+ },
2048
+ {
2049
+ "epoch": 1.7019485205677172,
2050
+ "grad_norm": 1.650564432144165,
2051
+ "learning_rate": 2.1644214577820547e-05,
2052
+ "loss": 0.8379,
2053
+ "step": 28300
2054
+ },
2055
+ {
2056
+ "epoch": 1.7079624729372145,
2057
+ "grad_norm": 1.6355592012405396,
2058
+ "learning_rate": 2.1543982038328925e-05,
2059
+ "loss": 0.8576,
2060
+ "step": 28400
2061
+ },
2062
+ {
2063
+ "epoch": 1.7139764253067116,
2064
+ "grad_norm": 1.7112095355987549,
2065
+ "learning_rate": 2.1443749498837303e-05,
2066
+ "loss": 0.8173,
2067
+ "step": 28500
2068
+ },
2069
+ {
2070
+ "epoch": 1.7199903776762087,
2071
+ "grad_norm": 1.7781462669372559,
2072
+ "learning_rate": 2.1343516959345684e-05,
2073
+ "loss": 0.8292,
2074
+ "step": 28600
2075
+ },
2076
+ {
2077
+ "epoch": 1.726004330045706,
2078
+ "grad_norm": 1.708770513534546,
2079
+ "learning_rate": 2.1243284419854062e-05,
2080
+ "loss": 0.8312,
2081
+ "step": 28700
2082
+ },
2083
+ {
2084
+ "epoch": 1.7320182824152033,
2085
+ "grad_norm": 1.869710087776184,
2086
+ "learning_rate": 2.114305188036244e-05,
2087
+ "loss": 0.8529,
2088
+ "step": 28800
2089
+ },
2090
+ {
2091
+ "epoch": 1.7380322347847006,
2092
+ "grad_norm": 1.4506940841674805,
2093
+ "learning_rate": 2.104281934087082e-05,
2094
+ "loss": 0.8454,
2095
+ "step": 28900
2096
+ },
2097
+ {
2098
+ "epoch": 1.744046187154198,
2099
+ "grad_norm": 1.5264636278152466,
2100
+ "learning_rate": 2.0942586801379203e-05,
2101
+ "loss": 0.8281,
2102
+ "step": 29000
2103
+ },
2104
+ {
2105
+ "epoch": 1.750060139523695,
2106
+ "grad_norm": 1.9614264965057373,
2107
+ "learning_rate": 2.084235426188758e-05,
2108
+ "loss": 0.8328,
2109
+ "step": 29100
2110
+ },
2111
+ {
2112
+ "epoch": 1.756074091893192,
2113
+ "grad_norm": 1.4223591089248657,
2114
+ "learning_rate": 2.074212172239596e-05,
2115
+ "loss": 0.8142,
2116
+ "step": 29200
2117
+ },
2118
+ {
2119
+ "epoch": 1.7620880442626894,
2120
+ "grad_norm": 1.6863304376602173,
2121
+ "learning_rate": 2.064188918290434e-05,
2122
+ "loss": 0.8306,
2123
+ "step": 29300
2124
+ },
2125
+ {
2126
+ "epoch": 1.7681019966321867,
2127
+ "grad_norm": 1.5096672773361206,
2128
+ "learning_rate": 2.0541656643412718e-05,
2129
+ "loss": 0.7964,
2130
+ "step": 29400
2131
+ },
2132
+ {
2133
+ "epoch": 1.774115949001684,
2134
+ "grad_norm": 1.9181997776031494,
2135
+ "learning_rate": 2.0441424103921096e-05,
2136
+ "loss": 0.8221,
2137
+ "step": 29500
2138
+ },
2139
+ {
2140
+ "epoch": 1.780129901371181,
2141
+ "grad_norm": 1.9824730157852173,
2142
+ "learning_rate": 2.0341191564429478e-05,
2143
+ "loss": 0.8379,
2144
+ "step": 29600
2145
+ },
2146
+ {
2147
+ "epoch": 1.7861438537406784,
2148
+ "grad_norm": 1.4306327104568481,
2149
+ "learning_rate": 2.0240959024937855e-05,
2150
+ "loss": 0.805,
2151
+ "step": 29700
2152
+ },
2153
+ {
2154
+ "epoch": 1.7921578061101755,
2155
+ "grad_norm": 1.6249910593032837,
2156
+ "learning_rate": 2.0140726485446237e-05,
2157
+ "loss": 0.8145,
2158
+ "step": 29800
2159
+ },
2160
+ {
2161
+ "epoch": 1.7981717584796728,
2162
+ "grad_norm": 1.551161766052246,
2163
+ "learning_rate": 2.0040493945954618e-05,
2164
+ "loss": 0.8436,
2165
+ "step": 29900
2166
+ },
2167
+ {
2168
+ "epoch": 1.8041857108491701,
2169
+ "grad_norm": 1.5218690633773804,
2170
+ "learning_rate": 1.9940261406462996e-05,
2171
+ "loss": 0.8181,
2172
+ "step": 30000
2173
+ },
2174
+ {
2175
+ "epoch": 1.8101996632186674,
2176
+ "grad_norm": 1.5932899713516235,
2177
+ "learning_rate": 1.9840028866971374e-05,
2178
+ "loss": 0.8478,
2179
+ "step": 30100
2180
+ },
2181
+ {
2182
+ "epoch": 1.8162136155881645,
2183
+ "grad_norm": 1.4991642236709595,
2184
+ "learning_rate": 1.9739796327479752e-05,
2185
+ "loss": 0.8254,
2186
+ "step": 30200
2187
+ },
2188
+ {
2189
+ "epoch": 1.8222275679576616,
2190
+ "grad_norm": 1.647438883781433,
2191
+ "learning_rate": 1.9639563787988133e-05,
2192
+ "loss": 0.8342,
2193
+ "step": 30300
2194
+ },
2195
+ {
2196
+ "epoch": 1.828241520327159,
2197
+ "grad_norm": 1.6653351783752441,
2198
+ "learning_rate": 1.953933124849651e-05,
2199
+ "loss": 0.8249,
2200
+ "step": 30400
2201
+ },
2202
+ {
2203
+ "epoch": 1.8342554726966562,
2204
+ "grad_norm": 1.6969921588897705,
2205
+ "learning_rate": 1.943909870900489e-05,
2206
+ "loss": 0.8544,
2207
+ "step": 30500
2208
+ },
2209
+ {
2210
+ "epoch": 1.8402694250661535,
2211
+ "grad_norm": 1.7201420068740845,
2212
+ "learning_rate": 1.9338866169513274e-05,
2213
+ "loss": 0.8323,
2214
+ "step": 30600
2215
+ },
2216
+ {
2217
+ "epoch": 1.8462833774356509,
2218
+ "grad_norm": 1.7479013204574585,
2219
+ "learning_rate": 1.9238633630021652e-05,
2220
+ "loss": 0.8307,
2221
+ "step": 30700
2222
+ },
2223
+ {
2224
+ "epoch": 1.852297329805148,
2225
+ "grad_norm": 1.6966118812561035,
2226
+ "learning_rate": 1.913840109053003e-05,
2227
+ "loss": 0.8291,
2228
+ "step": 30800
2229
+ },
2230
+ {
2231
+ "epoch": 1.858311282174645,
2232
+ "grad_norm": 1.6381275653839111,
2233
+ "learning_rate": 1.903816855103841e-05,
2234
+ "loss": 0.8121,
2235
+ "step": 30900
2236
+ },
2237
+ {
2238
+ "epoch": 1.8643252345441423,
2239
+ "grad_norm": 1.4601831436157227,
2240
+ "learning_rate": 1.893793601154679e-05,
2241
+ "loss": 0.8174,
2242
+ "step": 31000
2243
+ },
2244
+ {
2245
+ "epoch": 1.8703391869136397,
2246
+ "grad_norm": 1.8310879468917847,
2247
+ "learning_rate": 1.8837703472055167e-05,
2248
+ "loss": 0.8023,
2249
+ "step": 31100
2250
+ },
2251
+ {
2252
+ "epoch": 1.876353139283137,
2253
+ "grad_norm": 1.4689913988113403,
2254
+ "learning_rate": 1.8738473257958463e-05,
2255
+ "loss": 0.8035,
2256
+ "step": 31200
2257
+ },
2258
+ {
2259
+ "epoch": 1.8823670916526343,
2260
+ "grad_norm": 1.4330099821090698,
2261
+ "learning_rate": 1.8638240718466844e-05,
2262
+ "loss": 0.8185,
2263
+ "step": 31300
2264
+ },
2265
+ {
2266
+ "epoch": 1.8883810440221314,
2267
+ "grad_norm": 1.8419737815856934,
2268
+ "learning_rate": 1.8538008178975225e-05,
2269
+ "loss": 0.8144,
2270
+ "step": 31400
2271
+ },
2272
+ {
2273
+ "epoch": 1.8943949963916284,
2274
+ "grad_norm": 1.2664531469345093,
2275
+ "learning_rate": 1.8437775639483603e-05,
2276
+ "loss": 0.816,
2277
+ "step": 31500
2278
+ },
2279
+ {
2280
+ "epoch": 1.9004089487611258,
2281
+ "grad_norm": 1.6704432964324951,
2282
+ "learning_rate": 1.833754309999198e-05,
2283
+ "loss": 0.817,
2284
+ "step": 31600
2285
+ },
2286
+ {
2287
+ "epoch": 1.906422901130623,
2288
+ "grad_norm": 1.7487777471542358,
2289
+ "learning_rate": 1.8237310560500363e-05,
2290
+ "loss": 0.8041,
2291
+ "step": 31700
2292
+ },
2293
+ {
2294
+ "epoch": 1.9124368535001204,
2295
+ "grad_norm": 1.2405571937561035,
2296
+ "learning_rate": 1.813707802100874e-05,
2297
+ "loss": 0.825,
2298
+ "step": 31800
2299
+ },
2300
+ {
2301
+ "epoch": 1.9184508058696175,
2302
+ "grad_norm": 1.543731689453125,
2303
+ "learning_rate": 1.803684548151712e-05,
2304
+ "loss": 0.8147,
2305
+ "step": 31900
2306
+ },
2307
+ {
2308
+ "epoch": 1.9244647582391148,
2309
+ "grad_norm": 1.5452948808670044,
2310
+ "learning_rate": 1.79366129420255e-05,
2311
+ "loss": 0.7973,
2312
+ "step": 32000
2313
+ },
2314
+ {
2315
+ "epoch": 1.9244647582391148,
2316
+ "eval_cer": 0.7489049827996835,
2317
+ "eval_loss": 0.7830217480659485,
2318
+ "eval_runtime": 5092.4712,
2319
+ "eval_samples_per_second": 3.268,
2320
+ "eval_steps_per_second": 0.408,
2321
+ "step": 32000
2322
+ },
2323
+ {
2324
+ "epoch": 1.9304787106086119,
2325
+ "grad_norm": 1.640507459640503,
2326
+ "learning_rate": 1.7836380402533878e-05,
2327
+ "loss": 0.831,
2328
+ "step": 32100
2329
+ },
2330
+ {
2331
+ "epoch": 1.9364926629781092,
2332
+ "grad_norm": 1.919505000114441,
2333
+ "learning_rate": 1.773614786304226e-05,
2334
+ "loss": 0.8216,
2335
+ "step": 32200
2336
+ },
2337
+ {
2338
+ "epoch": 1.9425066153476065,
2339
+ "grad_norm": 1.5534350872039795,
2340
+ "learning_rate": 1.763591532355064e-05,
2341
+ "loss": 0.8157,
2342
+ "step": 32300
2343
+ },
2344
+ {
2345
+ "epoch": 1.9485205677171038,
2346
+ "grad_norm": 1.5943905115127563,
2347
+ "learning_rate": 1.753568278405902e-05,
2348
+ "loss": 0.8026,
2349
+ "step": 32400
2350
+ },
2351
+ {
2352
+ "epoch": 1.954534520086601,
2353
+ "grad_norm": 1.788720965385437,
2354
+ "learning_rate": 1.7435450244567397e-05,
2355
+ "loss": 0.7798,
2356
+ "step": 32500
2357
+ },
2358
+ {
2359
+ "epoch": 1.9605484724560982,
2360
+ "grad_norm": 1.707412600517273,
2361
+ "learning_rate": 1.7335217705075778e-05,
2362
+ "loss": 0.8125,
2363
+ "step": 32600
2364
+ },
2365
+ {
2366
+ "epoch": 1.9665624248255953,
2367
+ "grad_norm": 1.6318702697753906,
2368
+ "learning_rate": 1.7234985165584156e-05,
2369
+ "loss": 0.8184,
2370
+ "step": 32700
2371
+ },
2372
+ {
2373
+ "epoch": 1.9725763771950926,
2374
+ "grad_norm": 1.6191486120224,
2375
+ "learning_rate": 1.7134752626092534e-05,
2376
+ "loss": 0.8056,
2377
+ "step": 32800
2378
+ },
2379
+ {
2380
+ "epoch": 1.97859032956459,
2381
+ "grad_norm": 1.6070526838302612,
2382
+ "learning_rate": 1.7034520086600915e-05,
2383
+ "loss": 0.8134,
2384
+ "step": 32900
2385
+ },
2386
+ {
2387
+ "epoch": 1.9846042819340872,
2388
+ "grad_norm": 1.3369784355163574,
2389
+ "learning_rate": 1.6934287547109297e-05,
2390
+ "loss": 0.8086,
2391
+ "step": 33000
2392
+ },
2393
+ {
2394
+ "epoch": 1.9906182343035843,
2395
+ "grad_norm": 1.7080943584442139,
2396
+ "learning_rate": 1.6834055007617674e-05,
2397
+ "loss": 0.8304,
2398
+ "step": 33100
2399
+ },
2400
+ {
2401
+ "epoch": 1.9966321866730814,
2402
+ "grad_norm": 1.3051142692565918,
2403
+ "learning_rate": 1.673482479352097e-05,
2404
+ "loss": 0.778,
2405
+ "step": 33200
2406
+ },
2407
+ {
2408
+ "epoch": 2.0026461390425787,
2409
+ "grad_norm": 1.9086428880691528,
2410
+ "learning_rate": 1.6634592254029348e-05,
2411
+ "loss": 0.7972,
2412
+ "step": 33300
2413
+ },
2414
+ {
2415
+ "epoch": 2.008660091412076,
2416
+ "grad_norm": 1.6656805276870728,
2417
+ "learning_rate": 1.653435971453773e-05,
2418
+ "loss": 0.7701,
2419
+ "step": 33400
2420
+ },
2421
+ {
2422
+ "epoch": 2.0146740437815733,
2423
+ "grad_norm": 1.5769175291061401,
2424
+ "learning_rate": 1.6434127175046107e-05,
2425
+ "loss": 0.7602,
2426
+ "step": 33500
2427
+ },
2428
+ {
2429
+ "epoch": 2.0206879961510706,
2430
+ "grad_norm": 1.6363704204559326,
2431
+ "learning_rate": 1.6333894635554485e-05,
2432
+ "loss": 0.7826,
2433
+ "step": 33600
2434
+ },
2435
+ {
2436
+ "epoch": 2.0267019485205675,
2437
+ "grad_norm": 1.784424066543579,
2438
+ "learning_rate": 1.6233662096062867e-05,
2439
+ "loss": 0.7805,
2440
+ "step": 33700
2441
+ },
2442
+ {
2443
+ "epoch": 2.032715900890065,
2444
+ "grad_norm": 1.5795265436172485,
2445
+ "learning_rate": 1.6133429556571248e-05,
2446
+ "loss": 0.7289,
2447
+ "step": 33800
2448
+ },
2449
+ {
2450
+ "epoch": 2.038729853259562,
2451
+ "grad_norm": 1.382318377494812,
2452
+ "learning_rate": 1.6033197017079626e-05,
2453
+ "loss": 0.7869,
2454
+ "step": 33900
2455
+ },
2456
+ {
2457
+ "epoch": 2.0447438056290594,
2458
+ "grad_norm": 2.0357506275177,
2459
+ "learning_rate": 1.5932964477588004e-05,
2460
+ "loss": 0.7477,
2461
+ "step": 34000
2462
+ },
2463
+ {
2464
+ "epoch": 2.0507577579985568,
2465
+ "grad_norm": 1.6945205926895142,
2466
+ "learning_rate": 1.5832731938096385e-05,
2467
+ "loss": 0.7582,
2468
+ "step": 34100
2469
+ },
2470
+ {
2471
+ "epoch": 2.056771710368054,
2472
+ "grad_norm": 1.4343385696411133,
2473
+ "learning_rate": 1.5732499398604763e-05,
2474
+ "loss": 0.7536,
2475
+ "step": 34200
2476
+ },
2477
+ {
2478
+ "epoch": 2.062785662737551,
2479
+ "grad_norm": 1.7783600091934204,
2480
+ "learning_rate": 1.563226685911314e-05,
2481
+ "loss": 0.8331,
2482
+ "step": 34300
2483
+ },
2484
+ {
2485
+ "epoch": 2.0687996151070482,
2486
+ "grad_norm": 1.8440674543380737,
2487
+ "learning_rate": 1.5532034319621522e-05,
2488
+ "loss": 0.7352,
2489
+ "step": 34400
2490
+ },
2491
+ {
2492
+ "epoch": 2.0748135674765456,
2493
+ "grad_norm": 1.8098615407943726,
2494
+ "learning_rate": 1.54318017801299e-05,
2495
+ "loss": 0.7951,
2496
+ "step": 34500
2497
+ },
2498
+ {
2499
+ "epoch": 2.080827519846043,
2500
+ "grad_norm": 1.521584391593933,
2501
+ "learning_rate": 1.5331569240638282e-05,
2502
+ "loss": 0.7684,
2503
+ "step": 34600
2504
+ },
2505
+ {
2506
+ "epoch": 2.08684147221554,
2507
+ "grad_norm": 1.738142490386963,
2508
+ "learning_rate": 1.5231336701146661e-05,
2509
+ "loss": 0.7589,
2510
+ "step": 34700
2511
+ },
2512
+ {
2513
+ "epoch": 2.0928554245850375,
2514
+ "grad_norm": 1.5031851530075073,
2515
+ "learning_rate": 1.5131104161655041e-05,
2516
+ "loss": 0.7508,
2517
+ "step": 34800
2518
+ },
2519
+ {
2520
+ "epoch": 2.0988693769545343,
2521
+ "grad_norm": 1.8703136444091797,
2522
+ "learning_rate": 1.503087162216342e-05,
2523
+ "loss": 0.803,
2524
+ "step": 34900
2525
+ },
2526
+ {
2527
+ "epoch": 2.1048833293240317,
2528
+ "grad_norm": 1.6643913984298706,
2529
+ "learning_rate": 1.4930639082671799e-05,
2530
+ "loss": 0.7689,
2531
+ "step": 35000
2532
+ },
2533
+ {
2534
+ "epoch": 2.110897281693529,
2535
+ "grad_norm": 1.4073503017425537,
2536
+ "learning_rate": 1.4830406543180178e-05,
2537
+ "loss": 0.7575,
2538
+ "step": 35100
2539
+ },
2540
+ {
2541
+ "epoch": 2.1169112340630263,
2542
+ "grad_norm": 1.6007989645004272,
2543
+ "learning_rate": 1.4731176329083474e-05,
2544
+ "loss": 0.7317,
2545
+ "step": 35200
2546
+ },
2547
+ {
2548
+ "epoch": 2.1229251864325236,
2549
+ "grad_norm": 1.7965283393859863,
2550
+ "learning_rate": 1.4630943789591854e-05,
2551
+ "loss": 0.7142,
2552
+ "step": 35300
2553
+ },
2554
+ {
2555
+ "epoch": 2.1289391388020205,
2556
+ "grad_norm": 1.3655446767807007,
2557
+ "learning_rate": 1.4530711250100235e-05,
2558
+ "loss": 0.7613,
2559
+ "step": 35400
2560
+ },
2561
+ {
2562
+ "epoch": 2.1349530911715178,
2563
+ "grad_norm": 1.779159426689148,
2564
+ "learning_rate": 1.4430478710608613e-05,
2565
+ "loss": 0.7827,
2566
+ "step": 35500
2567
+ },
2568
+ {
2569
+ "epoch": 2.140967043541015,
2570
+ "grad_norm": 1.9253307580947876,
2571
+ "learning_rate": 1.4330246171116993e-05,
2572
+ "loss": 0.7899,
2573
+ "step": 35600
2574
+ },
2575
+ {
2576
+ "epoch": 2.1469809959105124,
2577
+ "grad_norm": 1.3449054956436157,
2578
+ "learning_rate": 1.4230013631625372e-05,
2579
+ "loss": 0.7494,
2580
+ "step": 35700
2581
+ },
2582
+ {
2583
+ "epoch": 2.1529949482800097,
2584
+ "grad_norm": 1.9044090509414673,
2585
+ "learning_rate": 1.412978109213375e-05,
2586
+ "loss": 0.7714,
2587
+ "step": 35800
2588
+ },
2589
+ {
2590
+ "epoch": 2.159008900649507,
2591
+ "grad_norm": 1.674017071723938,
2592
+ "learning_rate": 1.402954855264213e-05,
2593
+ "loss": 0.7478,
2594
+ "step": 35900
2595
+ },
2596
+ {
2597
+ "epoch": 2.1650228530190043,
2598
+ "grad_norm": 1.8109982013702393,
2599
+ "learning_rate": 1.392931601315051e-05,
2600
+ "loss": 0.7541,
2601
+ "step": 36000
2602
+ },
2603
+ {
2604
+ "epoch": 2.1650228530190043,
2605
+ "eval_cer": 0.7396310458108724,
2606
+ "eval_loss": 0.7524659633636475,
2607
+ "eval_runtime": 5154.2928,
2608
+ "eval_samples_per_second": 3.228,
2609
+ "eval_steps_per_second": 0.404,
2610
+ "step": 36000
2611
+ },
2612
+ {
2613
+ "epoch": 2.171036805388501,
2614
+ "grad_norm": 1.316389560699463,
2615
+ "learning_rate": 1.3829083473658887e-05,
2616
+ "loss": 0.7647,
2617
+ "step": 36100
2618
+ },
2619
+ {
2620
+ "epoch": 2.1770507577579985,
2621
+ "grad_norm": 1.7067075967788696,
2622
+ "learning_rate": 1.372885093416727e-05,
2623
+ "loss": 0.7635,
2624
+ "step": 36200
2625
+ },
2626
+ {
2627
+ "epoch": 2.183064710127496,
2628
+ "grad_norm": 1.8793973922729492,
2629
+ "learning_rate": 1.3628618394675648e-05,
2630
+ "loss": 0.7313,
2631
+ "step": 36300
2632
+ },
2633
+ {
2634
+ "epoch": 2.189078662496993,
2635
+ "grad_norm": 1.769338607788086,
2636
+ "learning_rate": 1.3528385855184028e-05,
2637
+ "loss": 0.7573,
2638
+ "step": 36400
2639
+ },
2640
+ {
2641
+ "epoch": 2.1950926148664904,
2642
+ "grad_norm": 1.6032990217208862,
2643
+ "learning_rate": 1.3428153315692408e-05,
2644
+ "loss": 0.76,
2645
+ "step": 36500
2646
+ },
2647
+ {
2648
+ "epoch": 2.2011065672359873,
2649
+ "grad_norm": 1.5864907503128052,
2650
+ "learning_rate": 1.3327920776200786e-05,
2651
+ "loss": 0.7587,
2652
+ "step": 36600
2653
+ },
2654
+ {
2655
+ "epoch": 2.2071205196054846,
2656
+ "grad_norm": 1.2785674333572388,
2657
+ "learning_rate": 1.3227688236709165e-05,
2658
+ "loss": 0.7403,
2659
+ "step": 36700
2660
+ },
2661
+ {
2662
+ "epoch": 2.213134471974982,
2663
+ "grad_norm": 1.4437572956085205,
2664
+ "learning_rate": 1.3127455697217545e-05,
2665
+ "loss": 0.7358,
2666
+ "step": 36800
2667
+ },
2668
+ {
2669
+ "epoch": 2.2191484243444792,
2670
+ "grad_norm": 1.8562610149383545,
2671
+ "learning_rate": 1.3027223157725923e-05,
2672
+ "loss": 0.7429,
2673
+ "step": 36900
2674
+ },
2675
+ {
2676
+ "epoch": 2.2251623767139765,
2677
+ "grad_norm": 1.6878858804702759,
2678
+ "learning_rate": 1.2926990618234306e-05,
2679
+ "loss": 0.7526,
2680
+ "step": 37000
2681
+ },
2682
+ {
2683
+ "epoch": 2.231176329083474,
2684
+ "grad_norm": 1.9118757247924805,
2685
+ "learning_rate": 1.2826758078742684e-05,
2686
+ "loss": 0.7513,
2687
+ "step": 37100
2688
+ },
2689
+ {
2690
+ "epoch": 2.2371902814529707,
2691
+ "grad_norm": 1.3607146739959717,
2692
+ "learning_rate": 1.272752786464598e-05,
2693
+ "loss": 0.747,
2694
+ "step": 37200
2695
+ },
2696
+ {
2697
+ "epoch": 2.243204233822468,
2698
+ "grad_norm": 1.8414541482925415,
2699
+ "learning_rate": 1.2627295325154359e-05,
2700
+ "loss": 0.757,
2701
+ "step": 37300
2702
+ },
2703
+ {
2704
+ "epoch": 2.2492181861919653,
2705
+ "grad_norm": 1.5014030933380127,
2706
+ "learning_rate": 1.2527062785662739e-05,
2707
+ "loss": 0.7704,
2708
+ "step": 37400
2709
+ },
2710
+ {
2711
+ "epoch": 2.2552321385614627,
2712
+ "grad_norm": 1.850203514099121,
2713
+ "learning_rate": 1.2426830246171117e-05,
2714
+ "loss": 0.7532,
2715
+ "step": 37500
2716
+ },
2717
+ {
2718
+ "epoch": 2.26124609093096,
2719
+ "grad_norm": 1.9308381080627441,
2720
+ "learning_rate": 1.2326597706679498e-05,
2721
+ "loss": 0.7658,
2722
+ "step": 37600
2723
+ },
2724
+ {
2725
+ "epoch": 2.267260043300457,
2726
+ "grad_norm": 1.8409243822097778,
2727
+ "learning_rate": 1.2226365167187876e-05,
2728
+ "loss": 0.7307,
2729
+ "step": 37700
2730
+ },
2731
+ {
2732
+ "epoch": 2.273273995669954,
2733
+ "grad_norm": 1.7760223150253296,
2734
+ "learning_rate": 1.2126132627696256e-05,
2735
+ "loss": 0.7548,
2736
+ "step": 37800
2737
+ },
2738
+ {
2739
+ "epoch": 2.2792879480394515,
2740
+ "grad_norm": 1.4862762689590454,
2741
+ "learning_rate": 1.2025900088204635e-05,
2742
+ "loss": 0.7196,
2743
+ "step": 37900
2744
+ },
2745
+ {
2746
+ "epoch": 2.2853019004089488,
2747
+ "grad_norm": 1.6604909896850586,
2748
+ "learning_rate": 1.1925667548713015e-05,
2749
+ "loss": 0.7306,
2750
+ "step": 38000
2751
+ },
2752
+ {
2753
+ "epoch": 2.291315852778446,
2754
+ "grad_norm": 1.6279034614562988,
2755
+ "learning_rate": 1.1825435009221395e-05,
2756
+ "loss": 0.7683,
2757
+ "step": 38100
2758
+ },
2759
+ {
2760
+ "epoch": 2.2973298051479434,
2761
+ "grad_norm": 1.5816621780395508,
2762
+ "learning_rate": 1.1725202469729774e-05,
2763
+ "loss": 0.723,
2764
+ "step": 38200
2765
+ },
2766
+ {
2767
+ "epoch": 2.3033437575174407,
2768
+ "grad_norm": 1.7849699258804321,
2769
+ "learning_rate": 1.1624969930238152e-05,
2770
+ "loss": 0.7831,
2771
+ "step": 38300
2772
+ },
2773
+ {
2774
+ "epoch": 2.3093577098869376,
2775
+ "grad_norm": 1.851671814918518,
2776
+ "learning_rate": 1.1524737390746534e-05,
2777
+ "loss": 0.7162,
2778
+ "step": 38400
2779
+ },
2780
+ {
2781
+ "epoch": 2.315371662256435,
2782
+ "grad_norm": 1.5026549100875854,
2783
+ "learning_rate": 1.1424504851254912e-05,
2784
+ "loss": 0.7803,
2785
+ "step": 38500
2786
+ },
2787
+ {
2788
+ "epoch": 2.321385614625932,
2789
+ "grad_norm": 1.6620761156082153,
2790
+ "learning_rate": 1.1324272311763291e-05,
2791
+ "loss": 0.733,
2792
+ "step": 38600
2793
+ },
2794
+ {
2795
+ "epoch": 2.3273995669954295,
2796
+ "grad_norm": 1.2507511377334595,
2797
+ "learning_rate": 1.1224039772271671e-05,
2798
+ "loss": 0.733,
2799
+ "step": 38700
2800
+ },
2801
+ {
2802
+ "epoch": 2.333413519364927,
2803
+ "grad_norm": 1.938541293144226,
2804
+ "learning_rate": 1.112380723278005e-05,
2805
+ "loss": 0.7499,
2806
+ "step": 38800
2807
+ },
2808
+ {
2809
+ "epoch": 2.3394274717344237,
2810
+ "grad_norm": 1.796823263168335,
2811
+ "learning_rate": 1.102357469328843e-05,
2812
+ "loss": 0.7399,
2813
+ "step": 38900
2814
+ },
2815
+ {
2816
+ "epoch": 2.345441424103921,
2817
+ "grad_norm": 1.834004521369934,
2818
+ "learning_rate": 1.0923342153796808e-05,
2819
+ "loss": 0.7322,
2820
+ "step": 39000
2821
+ },
2822
+ {
2823
+ "epoch": 2.3514553764734183,
2824
+ "grad_norm": 1.7822822332382202,
2825
+ "learning_rate": 1.0823109614305188e-05,
2826
+ "loss": 0.7315,
2827
+ "step": 39100
2828
+ },
2829
+ {
2830
+ "epoch": 2.3574693288429156,
2831
+ "grad_norm": 1.7883449792861938,
2832
+ "learning_rate": 1.0722877074813569e-05,
2833
+ "loss": 0.7205,
2834
+ "step": 39200
2835
+ },
2836
+ {
2837
+ "epoch": 2.363483281212413,
2838
+ "grad_norm": 1.966545581817627,
2839
+ "learning_rate": 1.0623646860716865e-05,
2840
+ "loss": 0.768,
2841
+ "step": 39300
2842
+ },
2843
+ {
2844
+ "epoch": 2.3694972335819102,
2845
+ "grad_norm": 1.722288966178894,
2846
+ "learning_rate": 1.0523414321225243e-05,
2847
+ "loss": 0.7554,
2848
+ "step": 39400
2849
+ },
2850
+ {
2851
+ "epoch": 2.375511185951407,
2852
+ "grad_norm": 1.7346769571304321,
2853
+ "learning_rate": 1.0423181781733622e-05,
2854
+ "loss": 0.7158,
2855
+ "step": 39500
2856
+ },
2857
+ {
2858
+ "epoch": 2.3815251383209044,
2859
+ "grad_norm": 1.603703498840332,
2860
+ "learning_rate": 1.0322949242242002e-05,
2861
+ "loss": 0.7467,
2862
+ "step": 39600
2863
+ },
2864
+ {
2865
+ "epoch": 2.3875390906904017,
2866
+ "grad_norm": 1.8487290143966675,
2867
+ "learning_rate": 1.0222716702750382e-05,
2868
+ "loss": 0.7492,
2869
+ "step": 39700
2870
+ },
2871
+ {
2872
+ "epoch": 2.393553043059899,
2873
+ "grad_norm": 1.7814853191375732,
2874
+ "learning_rate": 1.0122484163258761e-05,
2875
+ "loss": 0.7153,
2876
+ "step": 39800
2877
+ },
2878
+ {
2879
+ "epoch": 2.3995669954293963,
2880
+ "grad_norm": 1.745309591293335,
2881
+ "learning_rate": 1.002225162376714e-05,
2882
+ "loss": 0.7417,
2883
+ "step": 39900
2884
+ },
2885
+ {
2886
+ "epoch": 2.405580947798893,
2887
+ "grad_norm": 1.8568013906478882,
2888
+ "learning_rate": 9.92201908427552e-06,
2889
+ "loss": 0.7176,
2890
+ "step": 40000
2891
+ },
2892
+ {
2893
+ "epoch": 2.405580947798893,
2894
+ "eval_cer": 0.7346033545787555,
2895
+ "eval_loss": 0.7291049957275391,
2896
+ "eval_runtime": 5180.4578,
2897
+ "eval_samples_per_second": 3.212,
2898
+ "eval_steps_per_second": 0.402,
2899
+ "step": 40000
2900
+ },
2901
+ {
2902
+ "epoch": 2.4115949001683905,
2903
+ "grad_norm": 2.15785813331604,
2904
+ "learning_rate": 9.8217865447839e-06,
2905
+ "loss": 0.7234,
2906
+ "step": 40100
2907
+ },
2908
+ {
2909
+ "epoch": 2.417608852537888,
2910
+ "grad_norm": 1.7856571674346924,
2911
+ "learning_rate": 9.721554005292278e-06,
2912
+ "loss": 0.7176,
2913
+ "step": 40200
2914
+ },
2915
+ {
2916
+ "epoch": 2.423622804907385,
2917
+ "grad_norm": 1.682504415512085,
2918
+ "learning_rate": 9.621321465800658e-06,
2919
+ "loss": 0.7231,
2920
+ "step": 40300
2921
+ },
2922
+ {
2923
+ "epoch": 2.4296367572768824,
2924
+ "grad_norm": 1.3437491655349731,
2925
+ "learning_rate": 9.521088926309037e-06,
2926
+ "loss": 0.7232,
2927
+ "step": 40400
2928
+ },
2929
+ {
2930
+ "epoch": 2.4356507096463798,
2931
+ "grad_norm": 1.6993297338485718,
2932
+ "learning_rate": 9.420856386817417e-06,
2933
+ "loss": 0.7539,
2934
+ "step": 40500
2935
+ },
2936
+ {
2937
+ "epoch": 2.441664662015877,
2938
+ "grad_norm": 1.942014455795288,
2939
+ "learning_rate": 9.320623847325797e-06,
2940
+ "loss": 0.7519,
2941
+ "step": 40600
2942
+ },
2943
+ {
2944
+ "epoch": 2.447678614385374,
2945
+ "grad_norm": 1.7039834260940552,
2946
+ "learning_rate": 9.220391307834175e-06,
2947
+ "loss": 0.7362,
2948
+ "step": 40700
2949
+ },
2950
+ {
2951
+ "epoch": 2.4536925667548712,
2952
+ "grad_norm": 1.7124940156936646,
2953
+ "learning_rate": 9.120158768342556e-06,
2954
+ "loss": 0.7457,
2955
+ "step": 40800
2956
+ },
2957
+ {
2958
+ "epoch": 2.4597065191243686,
2959
+ "grad_norm": 1.910249948501587,
2960
+ "learning_rate": 9.019926228850934e-06,
2961
+ "loss": 0.7236,
2962
+ "step": 40900
2963
+ },
2964
+ {
2965
+ "epoch": 2.465720471493866,
2966
+ "grad_norm": 1.5514588356018066,
2967
+ "learning_rate": 8.919693689359314e-06,
2968
+ "loss": 0.721,
2969
+ "step": 41000
2970
+ },
2971
+ {
2972
+ "epoch": 2.471734423863363,
2973
+ "grad_norm": 1.744598150253296,
2974
+ "learning_rate": 8.819461149867693e-06,
2975
+ "loss": 0.7008,
2976
+ "step": 41100
2977
+ },
2978
+ {
2979
+ "epoch": 2.47774837623286,
2980
+ "grad_norm": 1.7827790975570679,
2981
+ "learning_rate": 8.719228610376073e-06,
2982
+ "loss": 0.7196,
2983
+ "step": 41200
2984
+ },
2985
+ {
2986
+ "epoch": 2.4837623286023573,
2987
+ "grad_norm": 1.7352166175842285,
2988
+ "learning_rate": 8.619998396279369e-06,
2989
+ "loss": 0.7322,
2990
+ "step": 41300
2991
+ },
2992
+ {
2993
+ "epoch": 2.4897762809718547,
2994
+ "grad_norm": 1.7378534078598022,
2995
+ "learning_rate": 8.519765856787748e-06,
2996
+ "loss": 0.7434,
2997
+ "step": 41400
2998
+ },
2999
+ {
3000
+ "epoch": 2.495790233341352,
3001
+ "grad_norm": 1.712098240852356,
3002
+ "learning_rate": 8.419533317296128e-06,
3003
+ "loss": 0.6939,
3004
+ "step": 41500
3005
+ },
3006
+ {
3007
+ "epoch": 2.5018041857108493,
3008
+ "grad_norm": 1.7368323802947998,
3009
+ "learning_rate": 8.319300777804506e-06,
3010
+ "loss": 0.726,
3011
+ "step": 41600
3012
+ },
3013
+ {
3014
+ "epoch": 2.5078181380803466,
3015
+ "grad_norm": 1.7819844484329224,
3016
+ "learning_rate": 8.219068238312887e-06,
3017
+ "loss": 0.7271,
3018
+ "step": 41700
3019
+ },
3020
+ {
3021
+ "epoch": 2.513832090449844,
3022
+ "grad_norm": 1.892824411392212,
3023
+ "learning_rate": 8.118835698821265e-06,
3024
+ "loss": 0.7146,
3025
+ "step": 41800
3026
+ },
3027
+ {
3028
+ "epoch": 2.5198460428193408,
3029
+ "grad_norm": 1.8355358839035034,
3030
+ "learning_rate": 8.018603159329645e-06,
3031
+ "loss": 0.7276,
3032
+ "step": 41900
3033
+ },
3034
+ {
3035
+ "epoch": 2.525859995188838,
3036
+ "grad_norm": 1.428154468536377,
3037
+ "learning_rate": 7.918370619838024e-06,
3038
+ "loss": 0.726,
3039
+ "step": 42000
3040
+ },
3041
+ {
3042
+ "epoch": 2.5318739475583354,
3043
+ "grad_norm": 1.8524224758148193,
3044
+ "learning_rate": 7.818138080346404e-06,
3045
+ "loss": 0.7178,
3046
+ "step": 42100
3047
+ },
3048
+ {
3049
+ "epoch": 2.5378878999278327,
3050
+ "grad_norm": 2.0265612602233887,
3051
+ "learning_rate": 7.717905540854784e-06,
3052
+ "loss": 0.714,
3053
+ "step": 42200
3054
+ },
3055
+ {
3056
+ "epoch": 2.5439018522973296,
3057
+ "grad_norm": 1.721408486366272,
3058
+ "learning_rate": 7.6176730013631625e-06,
3059
+ "loss": 0.7319,
3060
+ "step": 42300
3061
+ },
3062
+ {
3063
+ "epoch": 2.549915804666827,
3064
+ "grad_norm": 1.725827932357788,
3065
+ "learning_rate": 7.517440461871543e-06,
3066
+ "loss": 0.7663,
3067
+ "step": 42400
3068
+ },
3069
+ {
3070
+ "epoch": 2.555929757036324,
3071
+ "grad_norm": 1.6882712841033936,
3072
+ "learning_rate": 7.417207922379922e-06,
3073
+ "loss": 0.7191,
3074
+ "step": 42500
3075
+ },
3076
+ {
3077
+ "epoch": 2.5619437094058215,
3078
+ "grad_norm": 1.743870496749878,
3079
+ "learning_rate": 7.3169753828883015e-06,
3080
+ "loss": 0.7372,
3081
+ "step": 42600
3082
+ },
3083
+ {
3084
+ "epoch": 2.567957661775319,
3085
+ "grad_norm": 1.8450312614440918,
3086
+ "learning_rate": 7.21674284339668e-06,
3087
+ "loss": 0.6971,
3088
+ "step": 42700
3089
+ },
3090
+ {
3091
+ "epoch": 2.573971614144816,
3092
+ "grad_norm": 1.6685088872909546,
3093
+ "learning_rate": 7.116510303905061e-06,
3094
+ "loss": 0.7331,
3095
+ "step": 42800
3096
+ },
3097
+ {
3098
+ "epoch": 2.5799855665143134,
3099
+ "grad_norm": 1.406771183013916,
3100
+ "learning_rate": 7.01627776441344e-06,
3101
+ "loss": 0.7292,
3102
+ "step": 42900
3103
+ },
3104
+ {
3105
+ "epoch": 2.5859995188838103,
3106
+ "grad_norm": 1.750435471534729,
3107
+ "learning_rate": 6.916045224921818e-06,
3108
+ "loss": 0.7298,
3109
+ "step": 43000
3110
+ },
3111
+ {
3112
+ "epoch": 2.5920134712533076,
3113
+ "grad_norm": 1.4334131479263306,
3114
+ "learning_rate": 6.815812685430198e-06,
3115
+ "loss": 0.7332,
3116
+ "step": 43100
3117
+ },
3118
+ {
3119
+ "epoch": 2.598027423622805,
3120
+ "grad_norm": 1.4097541570663452,
3121
+ "learning_rate": 6.7155801459385786e-06,
3122
+ "loss": 0.7183,
3123
+ "step": 43200
3124
+ },
3125
+ {
3126
+ "epoch": 2.6040413759923022,
3127
+ "grad_norm": 1.5977458953857422,
3128
+ "learning_rate": 6.616349931841874e-06,
3129
+ "loss": 0.7227,
3130
+ "step": 43300
3131
+ },
3132
+ {
3133
+ "epoch": 2.610055328361799,
3134
+ "grad_norm": 1.766050100326538,
3135
+ "learning_rate": 6.516117392350253e-06,
3136
+ "loss": 0.7275,
3137
+ "step": 43400
3138
+ },
3139
+ {
3140
+ "epoch": 2.6160692807312964,
3141
+ "grad_norm": 1.6626147031784058,
3142
+ "learning_rate": 6.415884852858632e-06,
3143
+ "loss": 0.7101,
3144
+ "step": 43500
3145
+ },
3146
+ {
3147
+ "epoch": 2.6220832331007937,
3148
+ "grad_norm": 2.104124069213867,
3149
+ "learning_rate": 6.315652313367011e-06,
3150
+ "loss": 0.7267,
3151
+ "step": 43600
3152
+ },
3153
+ {
3154
+ "epoch": 2.628097185470291,
3155
+ "grad_norm": 1.66257905960083,
3156
+ "learning_rate": 6.215419773875391e-06,
3157
+ "loss": 0.6873,
3158
+ "step": 43700
3159
+ },
3160
+ {
3161
+ "epoch": 2.6341111378397883,
3162
+ "grad_norm": 2.1178150177001953,
3163
+ "learning_rate": 6.115187234383771e-06,
3164
+ "loss": 0.7533,
3165
+ "step": 43800
3166
+ },
3167
+ {
3168
+ "epoch": 2.6401250902092857,
3169
+ "grad_norm": 1.5612884759902954,
3170
+ "learning_rate": 6.0149546948921495e-06,
3171
+ "loss": 0.6981,
3172
+ "step": 43900
3173
+ },
3174
+ {
3175
+ "epoch": 2.646139042578783,
3176
+ "grad_norm": 1.9699441194534302,
3177
+ "learning_rate": 5.914722155400529e-06,
3178
+ "loss": 0.7477,
3179
+ "step": 44000
3180
+ },
3181
+ {
3182
+ "epoch": 2.646139042578783,
3183
+ "eval_cer": 0.7222449991747579,
3184
+ "eval_loss": 0.7039721608161926,
3185
+ "eval_runtime": 5169.4179,
3186
+ "eval_samples_per_second": 3.219,
3187
+ "eval_steps_per_second": 0.402,
3188
+ "step": 44000
3189
+ },
3190
+ {
3191
+ "epoch": 2.6521529949482803,
3192
+ "grad_norm": 1.6843528747558594,
3193
+ "learning_rate": 5.814489615908909e-06,
3194
+ "loss": 0.7235,
3195
+ "step": 44100
3196
+ },
3197
+ {
3198
+ "epoch": 2.658166947317777,
3199
+ "grad_norm": 1.4490200281143188,
3200
+ "learning_rate": 5.7142570764172885e-06,
3201
+ "loss": 0.7022,
3202
+ "step": 44200
3203
+ },
3204
+ {
3205
+ "epoch": 2.6641808996872745,
3206
+ "grad_norm": 1.6219418048858643,
3207
+ "learning_rate": 5.614024536925667e-06,
3208
+ "loss": 0.7215,
3209
+ "step": 44300
3210
+ },
3211
+ {
3212
+ "epoch": 2.6701948520567718,
3213
+ "grad_norm": 1.7598456144332886,
3214
+ "learning_rate": 5.513791997434047e-06,
3215
+ "loss": 0.6884,
3216
+ "step": 44400
3217
+ },
3218
+ {
3219
+ "epoch": 2.676208804426269,
3220
+ "grad_norm": 1.4096354246139526,
3221
+ "learning_rate": 5.413559457942427e-06,
3222
+ "loss": 0.7142,
3223
+ "step": 44500
3224
+ },
3225
+ {
3226
+ "epoch": 2.682222756795766,
3227
+ "grad_norm": 1.890046238899231,
3228
+ "learning_rate": 5.313326918450806e-06,
3229
+ "loss": 0.7346,
3230
+ "step": 44600
3231
+ },
3232
+ {
3233
+ "epoch": 2.6882367091652632,
3234
+ "grad_norm": 1.9413307905197144,
3235
+ "learning_rate": 5.213094378959186e-06,
3236
+ "loss": 0.688,
3237
+ "step": 44700
3238
+ },
3239
+ {
3240
+ "epoch": 2.6942506615347606,
3241
+ "grad_norm": 1.8647221326828003,
3242
+ "learning_rate": 5.112861839467565e-06,
3243
+ "loss": 0.732,
3244
+ "step": 44800
3245
+ },
3246
+ {
3247
+ "epoch": 2.700264613904258,
3248
+ "grad_norm": 1.6978052854537964,
3249
+ "learning_rate": 5.012629299975944e-06,
3250
+ "loss": 0.7075,
3251
+ "step": 44900
3252
+ },
3253
+ {
3254
+ "epoch": 2.706278566273755,
3255
+ "grad_norm": 1.3552030324935913,
3256
+ "learning_rate": 4.912396760484324e-06,
3257
+ "loss": 0.7074,
3258
+ "step": 45000
3259
+ },
3260
+ {
3261
+ "epoch": 2.7122925186432525,
3262
+ "grad_norm": 1.762635350227356,
3263
+ "learning_rate": 4.812164220992704e-06,
3264
+ "loss": 0.7219,
3265
+ "step": 45100
3266
+ },
3267
+ {
3268
+ "epoch": 2.71830647101275,
3269
+ "grad_norm": 1.7568144798278809,
3270
+ "learning_rate": 4.7119316815010824e-06,
3271
+ "loss": 0.6925,
3272
+ "step": 45200
3273
+ },
3274
+ {
3275
+ "epoch": 2.7243204233822467,
3276
+ "grad_norm": 1.5333248376846313,
3277
+ "learning_rate": 4.612701467404378e-06,
3278
+ "loss": 0.7133,
3279
+ "step": 45300
3280
+ },
3281
+ {
3282
+ "epoch": 2.730334375751744,
3283
+ "grad_norm": 1.8916515111923218,
3284
+ "learning_rate": 4.512468927912758e-06,
3285
+ "loss": 0.717,
3286
+ "step": 45400
3287
+ },
3288
+ {
3289
+ "epoch": 2.7363483281212413,
3290
+ "grad_norm": 1.8751533031463623,
3291
+ "learning_rate": 4.412236388421137e-06,
3292
+ "loss": 0.7306,
3293
+ "step": 45500
3294
+ },
3295
+ {
3296
+ "epoch": 2.7423622804907386,
3297
+ "grad_norm": 1.8414188623428345,
3298
+ "learning_rate": 4.312003848929517e-06,
3299
+ "loss": 0.7241,
3300
+ "step": 45600
3301
+ },
3302
+ {
3303
+ "epoch": 2.748376232860236,
3304
+ "grad_norm": 1.4498494863510132,
3305
+ "learning_rate": 4.211771309437896e-06,
3306
+ "loss": 0.7358,
3307
+ "step": 45700
3308
+ },
3309
+ {
3310
+ "epoch": 2.754390185229733,
3311
+ "grad_norm": 1.519124984741211,
3312
+ "learning_rate": 4.111538769946275e-06,
3313
+ "loss": 0.7059,
3314
+ "step": 45800
3315
+ },
3316
+ {
3317
+ "epoch": 2.76040413759923,
3318
+ "grad_norm": 1.9624022245407104,
3319
+ "learning_rate": 4.011306230454655e-06,
3320
+ "loss": 0.7185,
3321
+ "step": 45900
3322
+ },
3323
+ {
3324
+ "epoch": 2.7664180899687274,
3325
+ "grad_norm": 1.7385069131851196,
3326
+ "learning_rate": 3.911073690963035e-06,
3327
+ "loss": 0.7065,
3328
+ "step": 46000
3329
+ },
3330
+ {
3331
+ "epoch": 2.7724320423382247,
3332
+ "grad_norm": 1.5801081657409668,
3333
+ "learning_rate": 3.8108411514714135e-06,
3334
+ "loss": 0.7242,
3335
+ "step": 46100
3336
+ },
3337
+ {
3338
+ "epoch": 2.778445994707722,
3339
+ "grad_norm": 2.89750337600708,
3340
+ "learning_rate": 3.710608611979793e-06,
3341
+ "loss": 0.7023,
3342
+ "step": 46200
3343
+ },
3344
+ {
3345
+ "epoch": 2.7844599470772193,
3346
+ "grad_norm": 1.7838506698608398,
3347
+ "learning_rate": 3.6103760724881724e-06,
3348
+ "loss": 0.709,
3349
+ "step": 46300
3350
+ },
3351
+ {
3352
+ "epoch": 2.7904738994467166,
3353
+ "grad_norm": 1.7187174558639526,
3354
+ "learning_rate": 3.510143532996552e-06,
3355
+ "loss": 0.7292,
3356
+ "step": 46400
3357
+ },
3358
+ {
3359
+ "epoch": 2.7964878518162135,
3360
+ "grad_norm": 1.692230463027954,
3361
+ "learning_rate": 3.4099109935049313e-06,
3362
+ "loss": 0.6929,
3363
+ "step": 46500
3364
+ },
3365
+ {
3366
+ "epoch": 2.802501804185711,
3367
+ "grad_norm": 1.8588602542877197,
3368
+ "learning_rate": 3.309678454013311e-06,
3369
+ "loss": 0.7182,
3370
+ "step": 46600
3371
+ },
3372
+ {
3373
+ "epoch": 2.808515756555208,
3374
+ "grad_norm": 1.7374218702316284,
3375
+ "learning_rate": 3.20944591452169e-06,
3376
+ "loss": 0.6878,
3377
+ "step": 46700
3378
+ },
3379
+ {
3380
+ "epoch": 2.8145297089247054,
3381
+ "grad_norm": 1.6396222114562988,
3382
+ "learning_rate": 3.10921337503007e-06,
3383
+ "loss": 0.7035,
3384
+ "step": 46800
3385
+ },
3386
+ {
3387
+ "epoch": 2.8205436612942023,
3388
+ "grad_norm": 1.4402636289596558,
3389
+ "learning_rate": 3.0089808355384495e-06,
3390
+ "loss": 0.7131,
3391
+ "step": 46900
3392
+ },
3393
+ {
3394
+ "epoch": 2.8265576136636996,
3395
+ "grad_norm": 2.006782293319702,
3396
+ "learning_rate": 2.9087482960468287e-06,
3397
+ "loss": 0.7222,
3398
+ "step": 47000
3399
+ },
3400
+ {
3401
+ "epoch": 2.832571566033197,
3402
+ "grad_norm": 1.543579339981079,
3403
+ "learning_rate": 2.8085157565552084e-06,
3404
+ "loss": 0.683,
3405
+ "step": 47100
3406
+ },
3407
+ {
3408
+ "epoch": 2.8385855184026942,
3409
+ "grad_norm": 1.7819281816482544,
3410
+ "learning_rate": 2.7082832170635876e-06,
3411
+ "loss": 0.7095,
3412
+ "step": 47200
3413
+ },
3414
+ {
3415
+ "epoch": 2.8445994707721916,
3416
+ "grad_norm": 1.6801820993423462,
3417
+ "learning_rate": 2.608050677571967e-06,
3418
+ "loss": 0.7021,
3419
+ "step": 47300
3420
+ },
3421
+ {
3422
+ "epoch": 2.850613423141689,
3423
+ "grad_norm": 1.8617347478866577,
3424
+ "learning_rate": 2.508820463475263e-06,
3425
+ "loss": 0.7094,
3426
+ "step": 47400
3427
+ },
3428
+ {
3429
+ "epoch": 2.856627375511186,
3430
+ "grad_norm": 1.8291690349578857,
3431
+ "learning_rate": 2.408587923983642e-06,
3432
+ "loss": 0.7004,
3433
+ "step": 47500
3434
+ },
3435
+ {
3436
+ "epoch": 2.862641327880683,
3437
+ "grad_norm": 1.8136513233184814,
3438
+ "learning_rate": 2.3083553844920213e-06,
3439
+ "loss": 0.6936,
3440
+ "step": 47600
3441
+ },
3442
+ {
3443
+ "epoch": 2.8686552802501804,
3444
+ "grad_norm": 2.1329762935638428,
3445
+ "learning_rate": 2.208122845000401e-06,
3446
+ "loss": 0.6904,
3447
+ "step": 47700
3448
+ },
3449
+ {
3450
+ "epoch": 2.8746692326196777,
3451
+ "grad_norm": 1.9445267915725708,
3452
+ "learning_rate": 2.1078903055087806e-06,
3453
+ "loss": 0.6832,
3454
+ "step": 47800
3455
+ },
3456
+ {
3457
+ "epoch": 2.880683184989175,
3458
+ "grad_norm": 1.6533479690551758,
3459
+ "learning_rate": 2.0076577660171602e-06,
3460
+ "loss": 0.6754,
3461
+ "step": 47900
3462
+ },
3463
+ {
3464
+ "epoch": 2.8866971373586723,
3465
+ "grad_norm": 2.0631463527679443,
3466
+ "learning_rate": 1.9074252265255394e-06,
3467
+ "loss": 0.6882,
3468
+ "step": 48000
3469
+ },
3470
+ {
3471
+ "epoch": 2.8866971373586723,
3472
+ "eval_cer": 0.7184241484900502,
3473
+ "eval_loss": 0.6888573169708252,
3474
+ "eval_runtime": 5167.0943,
3475
+ "eval_samples_per_second": 3.22,
3476
+ "eval_steps_per_second": 0.403,
3477
+ "step": 48000
3478
+ }
3479
+ ],
3480
+ "logging_steps": 100,
3481
+ "max_steps": 49884,
3482
+ "num_input_tokens_seen": 0,
3483
+ "num_train_epochs": 3,
3484
+ "save_steps": 4000,
3485
+ "total_flos": 4.985333650921882e+19,
3486
+ "train_batch_size": 8,
3487
+ "trial_name": null,
3488
+ "trial_params": null
3489
+ }
config.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/TrOCR-Pipeline/TrOCR-Tibetan/checkpoint-48000",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "TrOCRForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 0,
10
+ "d_model": 1024,
11
+ "decoder": {
12
+ "_name_or_path": "sangjeedondrub/tibetan-roberta-base",
13
+ "add_cross_attention": true,
14
+ "architectures": [
15
+ "RobertaForMaskedLM"
16
+ ],
17
+ "attention_probs_dropout_prob": 0.1,
18
+ "bad_words_ids": null,
19
+ "begin_suppress_tokens": null,
20
+ "bos_token_id": 0,
21
+ "chunk_size_feed_forward": 0,
22
+ "classifier_dropout": null,
23
+ "cross_attention_hidden_size": null,
24
+ "decoder_start_token_id": null,
25
+ "diversity_penalty": 0.0,
26
+ "do_sample": false,
27
+ "early_stopping": false,
28
+ "encoder_no_repeat_ngram_size": 0,
29
+ "eos_token_id": 2,
30
+ "exponential_decay_length_penalty": null,
31
+ "finetuning_task": null,
32
+ "forced_bos_token_id": null,
33
+ "forced_eos_token_id": null,
34
+ "hidden_act": "gelu",
35
+ "hidden_dropout_prob": 0.1,
36
+ "hidden_size": 768,
37
+ "id2label": {
38
+ "0": "LABEL_0",
39
+ "1": "LABEL_1"
40
+ },
41
+ "initializer_range": 0.02,
42
+ "intermediate_size": 3072,
43
+ "is_decoder": true,
44
+ "is_encoder_decoder": false,
45
+ "label2id": {
46
+ "LABEL_0": 0,
47
+ "LABEL_1": 1
48
+ },
49
+ "layer_norm_eps": 1e-12,
50
+ "length_penalty": 1.0,
51
+ "max_length": 20,
52
+ "max_position_embeddings": 514,
53
+ "min_length": 0,
54
+ "model_type": "roberta",
55
+ "no_repeat_ngram_size": 0,
56
+ "num_attention_heads": 12,
57
+ "num_beam_groups": 1,
58
+ "num_beams": 1,
59
+ "num_hidden_layers": 6,
60
+ "num_return_sequences": 1,
61
+ "output_attentions": false,
62
+ "output_hidden_states": false,
63
+ "output_scores": false,
64
+ "pad_token_id": 1,
65
+ "position_embedding_type": "absolute",
66
+ "prefix": null,
67
+ "problem_type": null,
68
+ "pruned_heads": {},
69
+ "remove_invalid_values": false,
70
+ "repetition_penalty": 1.0,
71
+ "return_dict": true,
72
+ "return_dict_in_generate": false,
73
+ "sep_token_id": null,
74
+ "suppress_tokens": null,
75
+ "task_specific_params": null,
76
+ "temperature": 1.0,
77
+ "tf_legacy_loss": false,
78
+ "tie_encoder_decoder": false,
79
+ "tie_word_embeddings": true,
80
+ "tokenizer_class": null,
81
+ "top_k": 50,
82
+ "top_p": 1.0,
83
+ "torch_dtype": "float32",
84
+ "torchscript": false,
85
+ "type_vocab_size": 1,
86
+ "typical_p": 1.0,
87
+ "use_bfloat16": false,
88
+ "use_cache": true,
89
+ "vocab_size": 52000
90
+ },
91
+ "decoder_attention_heads": 16,
92
+ "decoder_ffn_dim": 4096,
93
+ "decoder_layerdrop": 0.0,
94
+ "decoder_layers": 12,
95
+ "decoder_start_token_id": 0,
96
+ "dropout": 0.1,
97
+ "early_stopping": true,
98
+ "encoder": {
99
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
100
+ "add_cross_attention": false,
101
+ "architectures": [
102
+ "ViTModel"
103
+ ],
104
+ "attention_probs_dropout_prob": 0.0,
105
+ "bad_words_ids": null,
106
+ "begin_suppress_tokens": null,
107
+ "bos_token_id": null,
108
+ "chunk_size_feed_forward": 0,
109
+ "cross_attention_hidden_size": null,
110
+ "decoder_start_token_id": null,
111
+ "diversity_penalty": 0.0,
112
+ "do_sample": false,
113
+ "early_stopping": false,
114
+ "encoder_no_repeat_ngram_size": 0,
115
+ "encoder_stride": 16,
116
+ "eos_token_id": null,
117
+ "exponential_decay_length_penalty": null,
118
+ "finetuning_task": null,
119
+ "forced_bos_token_id": null,
120
+ "forced_eos_token_id": null,
121
+ "hidden_act": "gelu",
122
+ "hidden_dropout_prob": 0.0,
123
+ "hidden_size": 768,
124
+ "id2label": {
125
+ "0": "LABEL_0",
126
+ "1": "LABEL_1"
127
+ },
128
+ "image_size": 224,
129
+ "initializer_range": 0.02,
130
+ "intermediate_size": 3072,
131
+ "is_decoder": false,
132
+ "is_encoder_decoder": false,
133
+ "label2id": {
134
+ "LABEL_0": 0,
135
+ "LABEL_1": 1
136
+ },
137
+ "layer_norm_eps": 1e-12,
138
+ "length_penalty": 1.0,
139
+ "max_length": 20,
140
+ "min_length": 0,
141
+ "model_type": "vit",
142
+ "no_repeat_ngram_size": 0,
143
+ "num_attention_heads": 12,
144
+ "num_beam_groups": 1,
145
+ "num_beams": 1,
146
+ "num_channels": 3,
147
+ "num_hidden_layers": 12,
148
+ "num_return_sequences": 1,
149
+ "output_attentions": false,
150
+ "output_hidden_states": false,
151
+ "output_scores": false,
152
+ "pad_token_id": null,
153
+ "patch_size": 16,
154
+ "prefix": null,
155
+ "problem_type": null,
156
+ "pruned_heads": {},
157
+ "qkv_bias": true,
158
+ "remove_invalid_values": false,
159
+ "repetition_penalty": 1.0,
160
+ "return_dict": true,
161
+ "return_dict_in_generate": false,
162
+ "sep_token_id": null,
163
+ "suppress_tokens": null,
164
+ "task_specific_params": null,
165
+ "temperature": 1.0,
166
+ "tf_legacy_loss": false,
167
+ "tie_encoder_decoder": false,
168
+ "tie_word_embeddings": true,
169
+ "tokenizer_class": null,
170
+ "top_k": 50,
171
+ "top_p": 1.0,
172
+ "torch_dtype": null,
173
+ "torchscript": false,
174
+ "typical_p": 1.0,
175
+ "use_bfloat16": false
176
+ },
177
+ "eos_token_id": 2,
178
+ "init_std": 0.02,
179
+ "is_decoder": true,
180
+ "layernorm_embedding": true,
181
+ "length_penalty": 2.0,
182
+ "max_length": 170,
183
+ "max_position_embeddings": 512,
184
+ "model_type": "trocr",
185
+ "no_repeat_ngram_size": 3,
186
+ "num_beams": 4,
187
+ "pad_token_id": 1,
188
+ "scale_embedding": false,
189
+ "tie_word_embeddings": false,
190
+ "torch_dtype": "float32",
191
+ "transformers_version": "4.40.2",
192
+ "use_cache": true,
193
+ "use_learned_position_embeddings": true,
194
+ "vocab_size": 52000
195
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 0,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "length_penalty": 2.0,
7
+ "max_length": 170,
8
+ "no_repeat_ngram_size": 3,
9
+ "num_beams": 4,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.40.2"
12
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8188a46fe6f44af68210cc7789c0d05ef493fc20ce5a6a9295ff34a0c2838621
3
+ size 1234375264
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2832b751f0c3c6270b73fa1d496cb1e4cac9be3b1848bd4c313451db7a3b44d
3
+ size 1468196502
preprocessor_config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_processor_keys": [
3
+ "images",
4
+ "do_resize",
5
+ "size",
6
+ "resample",
7
+ "do_rescale",
8
+ "rescale_factor",
9
+ "do_normalize",
10
+ "image_mean",
11
+ "image_std",
12
+ "return_tensors",
13
+ "data_format",
14
+ "input_data_format"
15
+ ],
16
+ "do_normalize": true,
17
+ "do_rescale": true,
18
+ "do_resize": true,
19
+ "image_mean": [
20
+ 0.5,
21
+ 0.5,
22
+ 0.5
23
+ ],
24
+ "image_processor_type": "ViTImageProcessor",
25
+ "image_std": [
26
+ 0.5,
27
+ 0.5,
28
+ 0.5
29
+ ],
30
+ "processor_class": "TrOCRProcessor",
31
+ "resample": 2,
32
+ "rescale_factor": 0.00392156862745098,
33
+ "size": {
34
+ "height": 224,
35
+ "width": 224
36
+ }
37
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cab7114f3491328cd60a47d2a29e8e579f8ae0566bebac3227375eb41b2c170
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:043339386d36c3dbb813856dd2cebb20bd2dd305e1854d283c373f2b5df89425
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "4": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "max_len": 512,
52
+ "model_max_length": 512,
53
+ "pad_token": "<pad>",
54
+ "processor_class": "TrOCRProcessor",
55
+ "sep_token": "</s>",
56
+ "tokenizer_class": "RobertaTokenizer",
57
+ "trim_offsets": true,
58
+ "unk_token": "<unk>"
59
+ }
trainer_state.json ADDED
@@ -0,0 +1,3489 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.8866971373586723,
5
+ "eval_steps": 4000,
6
+ "global_step": 48000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0060139523694972335,
13
+ "grad_norm": 2.5106329917907715,
14
+ "learning_rate": 4.99007697859033e-05,
15
+ "loss": 3.082,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.012027904738994467,
20
+ "grad_norm": 2.184039831161499,
21
+ "learning_rate": 4.9800537246411676e-05,
22
+ "loss": 1.8328,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.0180418571084917,
27
+ "grad_norm": 2.2055342197418213,
28
+ "learning_rate": 4.970030470692006e-05,
29
+ "loss": 1.6836,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.024055809477988934,
34
+ "grad_norm": 2.0695838928222656,
35
+ "learning_rate": 4.960007216742843e-05,
36
+ "loss": 1.6098,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.03006976184748617,
41
+ "grad_norm": 1.9105968475341797,
42
+ "learning_rate": 4.949983962793682e-05,
43
+ "loss": 1.516,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 0.0360837142169834,
48
+ "grad_norm": 1.882120132446289,
49
+ "learning_rate": 4.9399607088445195e-05,
50
+ "loss": 1.4876,
51
+ "step": 600
52
+ },
53
+ {
54
+ "epoch": 0.042097666586480634,
55
+ "grad_norm": 1.937444806098938,
56
+ "learning_rate": 4.9299374548953576e-05,
57
+ "loss": 1.4493,
58
+ "step": 700
59
+ },
60
+ {
61
+ "epoch": 0.04811161895597787,
62
+ "grad_norm": 2.1085917949676514,
63
+ "learning_rate": 4.919914200946195e-05,
64
+ "loss": 1.4128,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 0.0541255713254751,
69
+ "grad_norm": 1.9272162914276123,
70
+ "learning_rate": 4.909890946997033e-05,
71
+ "loss": 1.4105,
72
+ "step": 900
73
+ },
74
+ {
75
+ "epoch": 0.06013952369497234,
76
+ "grad_norm": 1.818624496459961,
77
+ "learning_rate": 4.899867693047871e-05,
78
+ "loss": 1.3912,
79
+ "step": 1000
80
+ },
81
+ {
82
+ "epoch": 0.06615347606446957,
83
+ "grad_norm": 1.644021987915039,
84
+ "learning_rate": 4.8898444390987094e-05,
85
+ "loss": 1.3397,
86
+ "step": 1100
87
+ },
88
+ {
89
+ "epoch": 0.0721674284339668,
90
+ "grad_norm": 1.7671422958374023,
91
+ "learning_rate": 4.8798211851495476e-05,
92
+ "loss": 1.3873,
93
+ "step": 1200
94
+ },
95
+ {
96
+ "epoch": 0.07818138080346404,
97
+ "grad_norm": 2.032087564468384,
98
+ "learning_rate": 4.869797931200385e-05,
99
+ "loss": 1.3112,
100
+ "step": 1300
101
+ },
102
+ {
103
+ "epoch": 0.08419533317296127,
104
+ "grad_norm": 1.7175960540771484,
105
+ "learning_rate": 4.859774677251223e-05,
106
+ "loss": 1.3089,
107
+ "step": 1400
108
+ },
109
+ {
110
+ "epoch": 0.09020928554245851,
111
+ "grad_norm": 1.7282887697219849,
112
+ "learning_rate": 4.8497514233020606e-05,
113
+ "loss": 1.2983,
114
+ "step": 1500
115
+ },
116
+ {
117
+ "epoch": 0.09622323791195574,
118
+ "grad_norm": 1.836075782775879,
119
+ "learning_rate": 4.839728169352899e-05,
120
+ "loss": 1.3092,
121
+ "step": 1600
122
+ },
123
+ {
124
+ "epoch": 0.10223719028145298,
125
+ "grad_norm": 1.931219220161438,
126
+ "learning_rate": 4.829704915403737e-05,
127
+ "loss": 1.2528,
128
+ "step": 1700
129
+ },
130
+ {
131
+ "epoch": 0.1082511426509502,
132
+ "grad_norm": 1.9112814664840698,
133
+ "learning_rate": 4.819681661454575e-05,
134
+ "loss": 1.25,
135
+ "step": 1800
136
+ },
137
+ {
138
+ "epoch": 0.11426509502044743,
139
+ "grad_norm": 2.1694140434265137,
140
+ "learning_rate": 4.809658407505413e-05,
141
+ "loss": 1.253,
142
+ "step": 1900
143
+ },
144
+ {
145
+ "epoch": 0.12027904738994467,
146
+ "grad_norm": 1.9220771789550781,
147
+ "learning_rate": 4.7996351535562506e-05,
148
+ "loss": 1.2278,
149
+ "step": 2000
150
+ },
151
+ {
152
+ "epoch": 0.12629299975944192,
153
+ "grad_norm": 1.7120180130004883,
154
+ "learning_rate": 4.789611899607089e-05,
155
+ "loss": 1.2483,
156
+ "step": 2100
157
+ },
158
+ {
159
+ "epoch": 0.13230695212893914,
160
+ "grad_norm": 1.917104959487915,
161
+ "learning_rate": 4.779588645657926e-05,
162
+ "loss": 1.2384,
163
+ "step": 2200
164
+ },
165
+ {
166
+ "epoch": 0.13832090449843637,
167
+ "grad_norm": 2.0323758125305176,
168
+ "learning_rate": 4.7695653917087644e-05,
169
+ "loss": 1.2292,
170
+ "step": 2300
171
+ },
172
+ {
173
+ "epoch": 0.1443348568679336,
174
+ "grad_norm": 1.779888391494751,
175
+ "learning_rate": 4.7595421377596025e-05,
176
+ "loss": 1.2258,
177
+ "step": 2400
178
+ },
179
+ {
180
+ "epoch": 0.15034880923743085,
181
+ "grad_norm": 1.7667876482009888,
182
+ "learning_rate": 4.7495188838104406e-05,
183
+ "loss": 1.1875,
184
+ "step": 2500
185
+ },
186
+ {
187
+ "epoch": 0.15636276160692808,
188
+ "grad_norm": 1.8692760467529297,
189
+ "learning_rate": 4.739495629861279e-05,
190
+ "loss": 1.2082,
191
+ "step": 2600
192
+ },
193
+ {
194
+ "epoch": 0.1623767139764253,
195
+ "grad_norm": 1.8578811883926392,
196
+ "learning_rate": 4.729472375912116e-05,
197
+ "loss": 1.1939,
198
+ "step": 2700
199
+ },
200
+ {
201
+ "epoch": 0.16839066634592254,
202
+ "grad_norm": 1.8867884874343872,
203
+ "learning_rate": 4.7194491219629543e-05,
204
+ "loss": 1.2108,
205
+ "step": 2800
206
+ },
207
+ {
208
+ "epoch": 0.17440461871541976,
209
+ "grad_norm": 1.72930109500885,
210
+ "learning_rate": 4.709425868013792e-05,
211
+ "loss": 1.1943,
212
+ "step": 2900
213
+ },
214
+ {
215
+ "epoch": 0.18041857108491702,
216
+ "grad_norm": 1.7612700462341309,
217
+ "learning_rate": 4.69940261406463e-05,
218
+ "loss": 1.1818,
219
+ "step": 3000
220
+ },
221
+ {
222
+ "epoch": 0.18643252345441425,
223
+ "grad_norm": 1.6063175201416016,
224
+ "learning_rate": 4.689379360115468e-05,
225
+ "loss": 1.1926,
226
+ "step": 3100
227
+ },
228
+ {
229
+ "epoch": 0.19244647582391147,
230
+ "grad_norm": 1.7698125839233398,
231
+ "learning_rate": 4.679356106166306e-05,
232
+ "loss": 1.2029,
233
+ "step": 3200
234
+ },
235
+ {
236
+ "epoch": 0.1984604281934087,
237
+ "grad_norm": 1.6680766344070435,
238
+ "learning_rate": 4.6693328522171443e-05,
239
+ "loss": 1.185,
240
+ "step": 3300
241
+ },
242
+ {
243
+ "epoch": 0.20447438056290596,
244
+ "grad_norm": 1.7791478633880615,
245
+ "learning_rate": 4.659309598267982e-05,
246
+ "loss": 1.1663,
247
+ "step": 3400
248
+ },
249
+ {
250
+ "epoch": 0.21048833293240318,
251
+ "grad_norm": 1.6491518020629883,
252
+ "learning_rate": 4.64928634431882e-05,
253
+ "loss": 1.1492,
254
+ "step": 3500
255
+ },
256
+ {
257
+ "epoch": 0.2165022853019004,
258
+ "grad_norm": 1.6312377452850342,
259
+ "learning_rate": 4.6392630903696574e-05,
260
+ "loss": 1.1696,
261
+ "step": 3600
262
+ },
263
+ {
264
+ "epoch": 0.22251623767139764,
265
+ "grad_norm": 1.7078893184661865,
266
+ "learning_rate": 4.6292398364204955e-05,
267
+ "loss": 1.139,
268
+ "step": 3700
269
+ },
270
+ {
271
+ "epoch": 0.22853019004089486,
272
+ "grad_norm": 2.0885114669799805,
273
+ "learning_rate": 4.619216582471334e-05,
274
+ "loss": 1.1399,
275
+ "step": 3800
276
+ },
277
+ {
278
+ "epoch": 0.23454414241039212,
279
+ "grad_norm": 1.8433787822723389,
280
+ "learning_rate": 4.609193328522172e-05,
281
+ "loss": 1.1376,
282
+ "step": 3900
283
+ },
284
+ {
285
+ "epoch": 0.24055809477988935,
286
+ "grad_norm": 1.527872085571289,
287
+ "learning_rate": 4.59917007457301e-05,
288
+ "loss": 1.1094,
289
+ "step": 4000
290
+ },
291
+ {
292
+ "epoch": 0.24055809477988935,
293
+ "eval_cer": 0.7748897683360889,
294
+ "eval_loss": 1.0861139297485352,
295
+ "eval_runtime": 5003.1551,
296
+ "eval_samples_per_second": 3.326,
297
+ "eval_steps_per_second": 0.416,
298
+ "step": 4000
299
+ },
300
+ {
301
+ "epoch": 0.24657204714938658,
302
+ "grad_norm": 1.4162925481796265,
303
+ "learning_rate": 4.5891468206238474e-05,
304
+ "loss": 1.1353,
305
+ "step": 4100
306
+ },
307
+ {
308
+ "epoch": 0.25258599951888383,
309
+ "grad_norm": 1.5863301753997803,
310
+ "learning_rate": 4.5791235666746855e-05,
311
+ "loss": 1.1257,
312
+ "step": 4200
313
+ },
314
+ {
315
+ "epoch": 0.25859995188838103,
316
+ "grad_norm": 1.6928209066390991,
317
+ "learning_rate": 4.569100312725523e-05,
318
+ "loss": 1.1086,
319
+ "step": 4300
320
+ },
321
+ {
322
+ "epoch": 0.2646139042578783,
323
+ "grad_norm": 1.2880722284317017,
324
+ "learning_rate": 4.559077058776362e-05,
325
+ "loss": 1.119,
326
+ "step": 4400
327
+ },
328
+ {
329
+ "epoch": 0.2706278566273755,
330
+ "grad_norm": 1.6384778022766113,
331
+ "learning_rate": 4.549053804827199e-05,
332
+ "loss": 1.1474,
333
+ "step": 4500
334
+ },
335
+ {
336
+ "epoch": 0.27664180899687274,
337
+ "grad_norm": 1.7176462411880493,
338
+ "learning_rate": 4.5390305508780374e-05,
339
+ "loss": 1.0925,
340
+ "step": 4600
341
+ },
342
+ {
343
+ "epoch": 0.28265576136637,
344
+ "grad_norm": 1.6025973558425903,
345
+ "learning_rate": 4.529007296928875e-05,
346
+ "loss": 1.112,
347
+ "step": 4700
348
+ },
349
+ {
350
+ "epoch": 0.2886697137358672,
351
+ "grad_norm": 1.7415896654129028,
352
+ "learning_rate": 4.518984042979713e-05,
353
+ "loss": 1.1218,
354
+ "step": 4800
355
+ },
356
+ {
357
+ "epoch": 0.29468366610536445,
358
+ "grad_norm": 1.310402750968933,
359
+ "learning_rate": 4.508960789030551e-05,
360
+ "loss": 1.0879,
361
+ "step": 4900
362
+ },
363
+ {
364
+ "epoch": 0.3006976184748617,
365
+ "grad_norm": 1.7458600997924805,
366
+ "learning_rate": 4.498937535081389e-05,
367
+ "loss": 1.1088,
368
+ "step": 5000
369
+ },
370
+ {
371
+ "epoch": 0.3067115708443589,
372
+ "grad_norm": 1.8199197053909302,
373
+ "learning_rate": 4.4889142811322274e-05,
374
+ "loss": 1.0987,
375
+ "step": 5100
376
+ },
377
+ {
378
+ "epoch": 0.31272552321385616,
379
+ "grad_norm": 1.5931302309036255,
380
+ "learning_rate": 4.478891027183065e-05,
381
+ "loss": 1.0802,
382
+ "step": 5200
383
+ },
384
+ {
385
+ "epoch": 0.31873947558335336,
386
+ "grad_norm": 1.5596023797988892,
387
+ "learning_rate": 4.468867773233903e-05,
388
+ "loss": 1.1009,
389
+ "step": 5300
390
+ },
391
+ {
392
+ "epoch": 0.3247534279528506,
393
+ "grad_norm": 1.7147523164749146,
394
+ "learning_rate": 4.4588445192847404e-05,
395
+ "loss": 1.1144,
396
+ "step": 5400
397
+ },
398
+ {
399
+ "epoch": 0.33076738032234787,
400
+ "grad_norm": 1.7172225713729858,
401
+ "learning_rate": 4.4488212653355786e-05,
402
+ "loss": 1.0928,
403
+ "step": 5500
404
+ },
405
+ {
406
+ "epoch": 0.33678133269184507,
407
+ "grad_norm": 1.5675067901611328,
408
+ "learning_rate": 4.438798011386417e-05,
409
+ "loss": 1.0421,
410
+ "step": 5600
411
+ },
412
+ {
413
+ "epoch": 0.3427952850613423,
414
+ "grad_norm": 1.755364179611206,
415
+ "learning_rate": 4.428774757437255e-05,
416
+ "loss": 1.0897,
417
+ "step": 5700
418
+ },
419
+ {
420
+ "epoch": 0.3488092374308395,
421
+ "grad_norm": 1.5678260326385498,
422
+ "learning_rate": 4.418751503488093e-05,
423
+ "loss": 1.0703,
424
+ "step": 5800
425
+ },
426
+ {
427
+ "epoch": 0.3548231898003368,
428
+ "grad_norm": 1.8641386032104492,
429
+ "learning_rate": 4.4087282495389304e-05,
430
+ "loss": 1.07,
431
+ "step": 5900
432
+ },
433
+ {
434
+ "epoch": 0.36083714216983404,
435
+ "grad_norm": 1.571837306022644,
436
+ "learning_rate": 4.3987049955897686e-05,
437
+ "loss": 1.0493,
438
+ "step": 6000
439
+ },
440
+ {
441
+ "epoch": 0.36685109453933124,
442
+ "grad_norm": 1.5046846866607666,
443
+ "learning_rate": 4.388681741640606e-05,
444
+ "loss": 1.0935,
445
+ "step": 6100
446
+ },
447
+ {
448
+ "epoch": 0.3728650469088285,
449
+ "grad_norm": 1.88533353805542,
450
+ "learning_rate": 4.378658487691444e-05,
451
+ "loss": 1.0801,
452
+ "step": 6200
453
+ },
454
+ {
455
+ "epoch": 0.3788789992783257,
456
+ "grad_norm": 1.5626581907272339,
457
+ "learning_rate": 4.368635233742282e-05,
458
+ "loss": 1.0442,
459
+ "step": 6300
460
+ },
461
+ {
462
+ "epoch": 0.38489295164782295,
463
+ "grad_norm": 1.4634822607040405,
464
+ "learning_rate": 4.3586119797931204e-05,
465
+ "loss": 1.0552,
466
+ "step": 6400
467
+ },
468
+ {
469
+ "epoch": 0.3909069040173202,
470
+ "grad_norm": 1.8044239282608032,
471
+ "learning_rate": 4.3485887258439586e-05,
472
+ "loss": 1.035,
473
+ "step": 6500
474
+ },
475
+ {
476
+ "epoch": 0.3969208563868174,
477
+ "grad_norm": 1.6868014335632324,
478
+ "learning_rate": 4.338565471894796e-05,
479
+ "loss": 1.0497,
480
+ "step": 6600
481
+ },
482
+ {
483
+ "epoch": 0.40293480875631466,
484
+ "grad_norm": 1.6949145793914795,
485
+ "learning_rate": 4.328542217945634e-05,
486
+ "loss": 1.0369,
487
+ "step": 6700
488
+ },
489
+ {
490
+ "epoch": 0.4089487611258119,
491
+ "grad_norm": 1.7868014574050903,
492
+ "learning_rate": 4.3186191965359634e-05,
493
+ "loss": 1.0385,
494
+ "step": 6800
495
+ },
496
+ {
497
+ "epoch": 0.4149627134953091,
498
+ "grad_norm": 1.550017237663269,
499
+ "learning_rate": 4.3085959425868015e-05,
500
+ "loss": 1.0326,
501
+ "step": 6900
502
+ },
503
+ {
504
+ "epoch": 0.42097666586480637,
505
+ "grad_norm": 1.431362509727478,
506
+ "learning_rate": 4.298572688637639e-05,
507
+ "loss": 1.0154,
508
+ "step": 7000
509
+ },
510
+ {
511
+ "epoch": 0.42699061823430356,
512
+ "grad_norm": 1.946357011795044,
513
+ "learning_rate": 4.288549434688478e-05,
514
+ "loss": 1.0532,
515
+ "step": 7100
516
+ },
517
+ {
518
+ "epoch": 0.4330045706038008,
519
+ "grad_norm": 1.6147353649139404,
520
+ "learning_rate": 4.278526180739316e-05,
521
+ "loss": 1.0633,
522
+ "step": 7200
523
+ },
524
+ {
525
+ "epoch": 0.4390185229732981,
526
+ "grad_norm": 1.655043363571167,
527
+ "learning_rate": 4.2685029267901534e-05,
528
+ "loss": 1.0558,
529
+ "step": 7300
530
+ },
531
+ {
532
+ "epoch": 0.4450324753427953,
533
+ "grad_norm": 1.6090620756149292,
534
+ "learning_rate": 4.2584796728409915e-05,
535
+ "loss": 1.0289,
536
+ "step": 7400
537
+ },
538
+ {
539
+ "epoch": 0.45104642771229253,
540
+ "grad_norm": 1.8263063430786133,
541
+ "learning_rate": 4.248456418891829e-05,
542
+ "loss": 1.0275,
543
+ "step": 7500
544
+ },
545
+ {
546
+ "epoch": 0.45706038008178973,
547
+ "grad_norm": 1.6156238317489624,
548
+ "learning_rate": 4.238433164942667e-05,
549
+ "loss": 0.9976,
550
+ "step": 7600
551
+ },
552
+ {
553
+ "epoch": 0.463074332451287,
554
+ "grad_norm": 1.4604992866516113,
555
+ "learning_rate": 4.228409910993505e-05,
556
+ "loss": 1.0607,
557
+ "step": 7700
558
+ },
559
+ {
560
+ "epoch": 0.46908828482078424,
561
+ "grad_norm": 1.5745257139205933,
562
+ "learning_rate": 4.2183866570443434e-05,
563
+ "loss": 1.0297,
564
+ "step": 7800
565
+ },
566
+ {
567
+ "epoch": 0.47510223719028144,
568
+ "grad_norm": 1.852102518081665,
569
+ "learning_rate": 4.2083634030951815e-05,
570
+ "loss": 1.0173,
571
+ "step": 7900
572
+ },
573
+ {
574
+ "epoch": 0.4811161895597787,
575
+ "grad_norm": 1.6735659837722778,
576
+ "learning_rate": 4.198340149146019e-05,
577
+ "loss": 1.0145,
578
+ "step": 8000
579
+ },
580
+ {
581
+ "epoch": 0.4811161895597787,
582
+ "eval_cer": 0.7706982683703557,
583
+ "eval_loss": 0.9832878708839417,
584
+ "eval_runtime": 5013.5334,
585
+ "eval_samples_per_second": 3.319,
586
+ "eval_steps_per_second": 0.415,
587
+ "step": 8000
588
+ },
589
+ {
590
+ "epoch": 0.4871301419292759,
591
+ "grad_norm": 1.7805702686309814,
592
+ "learning_rate": 4.188316895196857e-05,
593
+ "loss": 1.0279,
594
+ "step": 8100
595
+ },
596
+ {
597
+ "epoch": 0.49314409429877315,
598
+ "grad_norm": 1.4825282096862793,
599
+ "learning_rate": 4.1782936412476945e-05,
600
+ "loss": 1.0124,
601
+ "step": 8200
602
+ },
603
+ {
604
+ "epoch": 0.4991580466682704,
605
+ "grad_norm": 1.5925073623657227,
606
+ "learning_rate": 4.168270387298533e-05,
607
+ "loss": 1.0026,
608
+ "step": 8300
609
+ },
610
+ {
611
+ "epoch": 0.5051719990377677,
612
+ "grad_norm": 1.5965441465377808,
613
+ "learning_rate": 4.158247133349371e-05,
614
+ "loss": 1.0238,
615
+ "step": 8400
616
+ },
617
+ {
618
+ "epoch": 0.5111859514072649,
619
+ "grad_norm": 1.5980632305145264,
620
+ "learning_rate": 4.148223879400209e-05,
621
+ "loss": 0.9909,
622
+ "step": 8500
623
+ },
624
+ {
625
+ "epoch": 0.5171999037767621,
626
+ "grad_norm": 1.7301024198532104,
627
+ "learning_rate": 4.1382006254510464e-05,
628
+ "loss": 1.0172,
629
+ "step": 8600
630
+ },
631
+ {
632
+ "epoch": 0.5232138561462594,
633
+ "grad_norm": 1.6884981393814087,
634
+ "learning_rate": 4.1281773715018845e-05,
635
+ "loss": 1.0326,
636
+ "step": 8700
637
+ },
638
+ {
639
+ "epoch": 0.5292278085157566,
640
+ "grad_norm": 1.6702126264572144,
641
+ "learning_rate": 4.118154117552723e-05,
642
+ "loss": 1.016,
643
+ "step": 8800
644
+ },
645
+ {
646
+ "epoch": 0.5352417608852538,
647
+ "grad_norm": 1.5463213920593262,
648
+ "learning_rate": 4.108231096143052e-05,
649
+ "loss": 1.0282,
650
+ "step": 8900
651
+ },
652
+ {
653
+ "epoch": 0.541255713254751,
654
+ "grad_norm": 1.9977515935897827,
655
+ "learning_rate": 4.09820784219389e-05,
656
+ "loss": 1.0027,
657
+ "step": 9000
658
+ },
659
+ {
660
+ "epoch": 0.5472696656242483,
661
+ "grad_norm": 1.4702314138412476,
662
+ "learning_rate": 4.088184588244728e-05,
663
+ "loss": 1.0138,
664
+ "step": 9100
665
+ },
666
+ {
667
+ "epoch": 0.5532836179937455,
668
+ "grad_norm": 1.5981062650680542,
669
+ "learning_rate": 4.078161334295566e-05,
670
+ "loss": 1.0333,
671
+ "step": 9200
672
+ },
673
+ {
674
+ "epoch": 0.5592975703632427,
675
+ "grad_norm": 1.4159762859344482,
676
+ "learning_rate": 4.068138080346404e-05,
677
+ "loss": 0.9956,
678
+ "step": 9300
679
+ },
680
+ {
681
+ "epoch": 0.56531152273274,
682
+ "grad_norm": 1.4920343160629272,
683
+ "learning_rate": 4.058114826397242e-05,
684
+ "loss": 0.9978,
685
+ "step": 9400
686
+ },
687
+ {
688
+ "epoch": 0.5713254751022372,
689
+ "grad_norm": 1.7335036993026733,
690
+ "learning_rate": 4.048091572448079e-05,
691
+ "loss": 1.0337,
692
+ "step": 9500
693
+ },
694
+ {
695
+ "epoch": 0.5773394274717344,
696
+ "grad_norm": 1.4422550201416016,
697
+ "learning_rate": 4.0380683184989175e-05,
698
+ "loss": 1.0185,
699
+ "step": 9600
700
+ },
701
+ {
702
+ "epoch": 0.5833533798412317,
703
+ "grad_norm": 1.776282787322998,
704
+ "learning_rate": 4.0280450645497556e-05,
705
+ "loss": 0.9941,
706
+ "step": 9700
707
+ },
708
+ {
709
+ "epoch": 0.5893673322107289,
710
+ "grad_norm": 1.806261420249939,
711
+ "learning_rate": 4.018021810600594e-05,
712
+ "loss": 0.9772,
713
+ "step": 9800
714
+ },
715
+ {
716
+ "epoch": 0.5953812845802261,
717
+ "grad_norm": 1.6673755645751953,
718
+ "learning_rate": 4.007998556651432e-05,
719
+ "loss": 0.9904,
720
+ "step": 9900
721
+ },
722
+ {
723
+ "epoch": 0.6013952369497234,
724
+ "grad_norm": 1.5652821063995361,
725
+ "learning_rate": 3.997975302702269e-05,
726
+ "loss": 1.0008,
727
+ "step": 10000
728
+ },
729
+ {
730
+ "epoch": 0.6074091893192206,
731
+ "grad_norm": 1.568590760231018,
732
+ "learning_rate": 3.9879520487531075e-05,
733
+ "loss": 1.0402,
734
+ "step": 10100
735
+ },
736
+ {
737
+ "epoch": 0.6134231416887178,
738
+ "grad_norm": 1.505454659461975,
739
+ "learning_rate": 3.977928794803945e-05,
740
+ "loss": 1.0009,
741
+ "step": 10200
742
+ },
743
+ {
744
+ "epoch": 0.619437094058215,
745
+ "grad_norm": 1.4691787958145142,
746
+ "learning_rate": 3.967905540854783e-05,
747
+ "loss": 1.0115,
748
+ "step": 10300
749
+ },
750
+ {
751
+ "epoch": 0.6254510464277123,
752
+ "grad_norm": 1.5172970294952393,
753
+ "learning_rate": 3.957882286905621e-05,
754
+ "loss": 1.0209,
755
+ "step": 10400
756
+ },
757
+ {
758
+ "epoch": 0.6314649987972095,
759
+ "grad_norm": 1.8390883207321167,
760
+ "learning_rate": 3.947859032956459e-05,
761
+ "loss": 1.0101,
762
+ "step": 10500
763
+ },
764
+ {
765
+ "epoch": 0.6374789511667067,
766
+ "grad_norm": 1.5624459981918335,
767
+ "learning_rate": 3.9378357790072975e-05,
768
+ "loss": 1.0085,
769
+ "step": 10600
770
+ },
771
+ {
772
+ "epoch": 0.643492903536204,
773
+ "grad_norm": 1.4316383600234985,
774
+ "learning_rate": 3.927812525058135e-05,
775
+ "loss": 1.0025,
776
+ "step": 10700
777
+ },
778
+ {
779
+ "epoch": 0.6495068559057012,
780
+ "grad_norm": 1.5541032552719116,
781
+ "learning_rate": 3.917789271108973e-05,
782
+ "loss": 0.9938,
783
+ "step": 10800
784
+ },
785
+ {
786
+ "epoch": 0.6555208082751984,
787
+ "grad_norm": 1.7416138648986816,
788
+ "learning_rate": 3.907866249699302e-05,
789
+ "loss": 0.9923,
790
+ "step": 10900
791
+ },
792
+ {
793
+ "epoch": 0.6615347606446957,
794
+ "grad_norm": 1.7461607456207275,
795
+ "learning_rate": 3.8978429957501404e-05,
796
+ "loss": 0.9983,
797
+ "step": 11000
798
+ },
799
+ {
800
+ "epoch": 0.6675487130141929,
801
+ "grad_norm": 1.5705294609069824,
802
+ "learning_rate": 3.887819741800978e-05,
803
+ "loss": 0.9809,
804
+ "step": 11100
805
+ },
806
+ {
807
+ "epoch": 0.6735626653836901,
808
+ "grad_norm": 1.802960753440857,
809
+ "learning_rate": 3.877796487851817e-05,
810
+ "loss": 0.9675,
811
+ "step": 11200
812
+ },
813
+ {
814
+ "epoch": 0.6795766177531873,
815
+ "grad_norm": 1.70058274269104,
816
+ "learning_rate": 3.867773233902655e-05,
817
+ "loss": 0.966,
818
+ "step": 11300
819
+ },
820
+ {
821
+ "epoch": 0.6855905701226847,
822
+ "grad_norm": 1.5391656160354614,
823
+ "learning_rate": 3.857749979953492e-05,
824
+ "loss": 0.9412,
825
+ "step": 11400
826
+ },
827
+ {
828
+ "epoch": 0.6916045224921819,
829
+ "grad_norm": 1.8226732015609741,
830
+ "learning_rate": 3.8477267260043304e-05,
831
+ "loss": 0.96,
832
+ "step": 11500
833
+ },
834
+ {
835
+ "epoch": 0.697618474861679,
836
+ "grad_norm": 1.3753610849380493,
837
+ "learning_rate": 3.837703472055168e-05,
838
+ "loss": 0.9665,
839
+ "step": 11600
840
+ },
841
+ {
842
+ "epoch": 0.7036324272311764,
843
+ "grad_norm": 1.6595444679260254,
844
+ "learning_rate": 3.827680218106006e-05,
845
+ "loss": 0.9851,
846
+ "step": 11700
847
+ },
848
+ {
849
+ "epoch": 0.7096463796006736,
850
+ "grad_norm": 1.6920074224472046,
851
+ "learning_rate": 3.817656964156844e-05,
852
+ "loss": 0.983,
853
+ "step": 11800
854
+ },
855
+ {
856
+ "epoch": 0.7156603319701708,
857
+ "grad_norm": 1.4369592666625977,
858
+ "learning_rate": 3.807633710207682e-05,
859
+ "loss": 0.9418,
860
+ "step": 11900
861
+ },
862
+ {
863
+ "epoch": 0.7216742843396681,
864
+ "grad_norm": 1.7257956266403198,
865
+ "learning_rate": 3.7976104562585204e-05,
866
+ "loss": 0.9877,
867
+ "step": 12000
868
+ },
869
+ {
870
+ "epoch": 0.7216742843396681,
871
+ "eval_cer": 0.7783561907148705,
872
+ "eval_loss": 0.9317989349365234,
873
+ "eval_runtime": 5220.1413,
874
+ "eval_samples_per_second": 3.188,
875
+ "eval_steps_per_second": 0.398,
876
+ "step": 12000
877
+ },
878
+ {
879
+ "epoch": 0.7276882367091653,
880
+ "grad_norm": 1.6520166397094727,
881
+ "learning_rate": 3.787587202309358e-05,
882
+ "loss": 0.9415,
883
+ "step": 12100
884
+ },
885
+ {
886
+ "epoch": 0.7337021890786625,
887
+ "grad_norm": 1.7703155279159546,
888
+ "learning_rate": 3.777563948360196e-05,
889
+ "loss": 0.9653,
890
+ "step": 12200
891
+ },
892
+ {
893
+ "epoch": 0.7397161414481598,
894
+ "grad_norm": 1.4374829530715942,
895
+ "learning_rate": 3.7675406944110334e-05,
896
+ "loss": 0.9541,
897
+ "step": 12300
898
+ },
899
+ {
900
+ "epoch": 0.745730093817657,
901
+ "grad_norm": 1.790893793106079,
902
+ "learning_rate": 3.7575174404618716e-05,
903
+ "loss": 0.9951,
904
+ "step": 12400
905
+ },
906
+ {
907
+ "epoch": 0.7517440461871542,
908
+ "grad_norm": 1.6874678134918213,
909
+ "learning_rate": 3.74749418651271e-05,
910
+ "loss": 0.9704,
911
+ "step": 12500
912
+ },
913
+ {
914
+ "epoch": 0.7577579985566514,
915
+ "grad_norm": 1.7682390213012695,
916
+ "learning_rate": 3.737470932563548e-05,
917
+ "loss": 0.9461,
918
+ "step": 12600
919
+ },
920
+ {
921
+ "epoch": 0.7637719509261487,
922
+ "grad_norm": 1.6911418437957764,
923
+ "learning_rate": 3.727447678614386e-05,
924
+ "loss": 0.9952,
925
+ "step": 12700
926
+ },
927
+ {
928
+ "epoch": 0.7697859032956459,
929
+ "grad_norm": 1.727547526359558,
930
+ "learning_rate": 3.7174244246652234e-05,
931
+ "loss": 0.9442,
932
+ "step": 12800
933
+ },
934
+ {
935
+ "epoch": 0.7757998556651431,
936
+ "grad_norm": 1.4808145761489868,
937
+ "learning_rate": 3.7074011707160616e-05,
938
+ "loss": 0.9742,
939
+ "step": 12900
940
+ },
941
+ {
942
+ "epoch": 0.7818138080346404,
943
+ "grad_norm": 1.876441240310669,
944
+ "learning_rate": 3.697377916766899e-05,
945
+ "loss": 0.9766,
946
+ "step": 13000
947
+ },
948
+ {
949
+ "epoch": 0.7878277604041376,
950
+ "grad_norm": 1.4900315999984741,
951
+ "learning_rate": 3.687354662817737e-05,
952
+ "loss": 0.9612,
953
+ "step": 13100
954
+ },
955
+ {
956
+ "epoch": 0.7938417127736348,
957
+ "grad_norm": 1.4132503271102905,
958
+ "learning_rate": 3.677331408868575e-05,
959
+ "loss": 0.9651,
960
+ "step": 13200
961
+ },
962
+ {
963
+ "epoch": 0.7998556651431321,
964
+ "grad_norm": 1.6306148767471313,
965
+ "learning_rate": 3.6673081549194134e-05,
966
+ "loss": 0.9368,
967
+ "step": 13300
968
+ },
969
+ {
970
+ "epoch": 0.8058696175126293,
971
+ "grad_norm": 1.7364792823791504,
972
+ "learning_rate": 3.657284900970251e-05,
973
+ "loss": 0.9427,
974
+ "step": 13400
975
+ },
976
+ {
977
+ "epoch": 0.8118835698821265,
978
+ "grad_norm": 1.6632161140441895,
979
+ "learning_rate": 3.647261647021089e-05,
980
+ "loss": 0.9505,
981
+ "step": 13500
982
+ },
983
+ {
984
+ "epoch": 0.8178975222516238,
985
+ "grad_norm": 1.5362128019332886,
986
+ "learning_rate": 3.637238393071927e-05,
987
+ "loss": 0.9692,
988
+ "step": 13600
989
+ },
990
+ {
991
+ "epoch": 0.823911474621121,
992
+ "grad_norm": 1.3290611505508423,
993
+ "learning_rate": 3.6272151391227646e-05,
994
+ "loss": 0.9585,
995
+ "step": 13700
996
+ },
997
+ {
998
+ "epoch": 0.8299254269906182,
999
+ "grad_norm": 1.6647266149520874,
1000
+ "learning_rate": 3.617191885173603e-05,
1001
+ "loss": 0.9725,
1002
+ "step": 13800
1003
+ },
1004
+ {
1005
+ "epoch": 0.8359393793601154,
1006
+ "grad_norm": 1.736165165901184,
1007
+ "learning_rate": 3.607168631224441e-05,
1008
+ "loss": 0.9611,
1009
+ "step": 13900
1010
+ },
1011
+ {
1012
+ "epoch": 0.8419533317296127,
1013
+ "grad_norm": 1.6238832473754883,
1014
+ "learning_rate": 3.597145377275279e-05,
1015
+ "loss": 0.9698,
1016
+ "step": 14000
1017
+ },
1018
+ {
1019
+ "epoch": 0.8479672840991099,
1020
+ "grad_norm": 1.741194486618042,
1021
+ "learning_rate": 3.5871221233261165e-05,
1022
+ "loss": 0.9338,
1023
+ "step": 14100
1024
+ },
1025
+ {
1026
+ "epoch": 0.8539812364686071,
1027
+ "grad_norm": 1.7224496603012085,
1028
+ "learning_rate": 3.5770988693769546e-05,
1029
+ "loss": 0.9715,
1030
+ "step": 14200
1031
+ },
1032
+ {
1033
+ "epoch": 0.8599951888381044,
1034
+ "grad_norm": 1.872253179550171,
1035
+ "learning_rate": 3.5671758479672845e-05,
1036
+ "loss": 0.9656,
1037
+ "step": 14300
1038
+ },
1039
+ {
1040
+ "epoch": 0.8660091412076016,
1041
+ "grad_norm": 1.5153071880340576,
1042
+ "learning_rate": 3.557152594018122e-05,
1043
+ "loss": 0.984,
1044
+ "step": 14400
1045
+ },
1046
+ {
1047
+ "epoch": 0.8720230935770988,
1048
+ "grad_norm": 1.667662262916565,
1049
+ "learning_rate": 3.54712934006896e-05,
1050
+ "loss": 0.9386,
1051
+ "step": 14500
1052
+ },
1053
+ {
1054
+ "epoch": 0.8780370459465962,
1055
+ "grad_norm": 1.7471551895141602,
1056
+ "learning_rate": 3.537106086119798e-05,
1057
+ "loss": 0.9486,
1058
+ "step": 14600
1059
+ },
1060
+ {
1061
+ "epoch": 0.8840509983160934,
1062
+ "grad_norm": 1.5477312803268433,
1063
+ "learning_rate": 3.5270828321706364e-05,
1064
+ "loss": 0.9451,
1065
+ "step": 14700
1066
+ },
1067
+ {
1068
+ "epoch": 0.8900649506855906,
1069
+ "grad_norm": 1.5079952478408813,
1070
+ "learning_rate": 3.517059578221474e-05,
1071
+ "loss": 0.9381,
1072
+ "step": 14800
1073
+ },
1074
+ {
1075
+ "epoch": 0.8960789030550877,
1076
+ "grad_norm": 1.5696821212768555,
1077
+ "learning_rate": 3.507036324272312e-05,
1078
+ "loss": 0.9418,
1079
+ "step": 14900
1080
+ },
1081
+ {
1082
+ "epoch": 0.9020928554245851,
1083
+ "grad_norm": 1.552612066268921,
1084
+ "learning_rate": 3.4970130703231494e-05,
1085
+ "loss": 0.9596,
1086
+ "step": 15000
1087
+ },
1088
+ {
1089
+ "epoch": 0.9081068077940823,
1090
+ "grad_norm": 1.7802802324295044,
1091
+ "learning_rate": 3.4869898163739876e-05,
1092
+ "loss": 0.9298,
1093
+ "step": 15100
1094
+ },
1095
+ {
1096
+ "epoch": 0.9141207601635795,
1097
+ "grad_norm": 1.5589861869812012,
1098
+ "learning_rate": 3.476966562424826e-05,
1099
+ "loss": 0.9621,
1100
+ "step": 15200
1101
+ },
1102
+ {
1103
+ "epoch": 0.9201347125330768,
1104
+ "grad_norm": 1.3050284385681152,
1105
+ "learning_rate": 3.466943308475664e-05,
1106
+ "loss": 0.9315,
1107
+ "step": 15300
1108
+ },
1109
+ {
1110
+ "epoch": 0.926148664902574,
1111
+ "grad_norm": 1.5091936588287354,
1112
+ "learning_rate": 3.456920054526502e-05,
1113
+ "loss": 0.9107,
1114
+ "step": 15400
1115
+ },
1116
+ {
1117
+ "epoch": 0.9321626172720712,
1118
+ "grad_norm": 1.5565592050552368,
1119
+ "learning_rate": 3.4468968005773394e-05,
1120
+ "loss": 0.9552,
1121
+ "step": 15500
1122
+ },
1123
+ {
1124
+ "epoch": 0.9381765696415685,
1125
+ "grad_norm": 1.7788596153259277,
1126
+ "learning_rate": 3.4368735466281775e-05,
1127
+ "loss": 0.9459,
1128
+ "step": 15600
1129
+ },
1130
+ {
1131
+ "epoch": 0.9441905220110657,
1132
+ "grad_norm": 1.8757565021514893,
1133
+ "learning_rate": 3.426850292679015e-05,
1134
+ "loss": 0.9254,
1135
+ "step": 15700
1136
+ },
1137
+ {
1138
+ "epoch": 0.9502044743805629,
1139
+ "grad_norm": 1.6978222131729126,
1140
+ "learning_rate": 3.416827038729853e-05,
1141
+ "loss": 0.9083,
1142
+ "step": 15800
1143
+ },
1144
+ {
1145
+ "epoch": 0.9562184267500602,
1146
+ "grad_norm": 1.6449016332626343,
1147
+ "learning_rate": 3.406803784780692e-05,
1148
+ "loss": 0.9767,
1149
+ "step": 15900
1150
+ },
1151
+ {
1152
+ "epoch": 0.9622323791195574,
1153
+ "grad_norm": 1.8677603006362915,
1154
+ "learning_rate": 3.3967805308315294e-05,
1155
+ "loss": 0.9469,
1156
+ "step": 16000
1157
+ },
1158
+ {
1159
+ "epoch": 0.9622323791195574,
1160
+ "eval_cer": 0.7716098871709525,
1161
+ "eval_loss": 0.8905403017997742,
1162
+ "eval_runtime": 5089.7823,
1163
+ "eval_samples_per_second": 3.269,
1164
+ "eval_steps_per_second": 0.409,
1165
+ "step": 16000
1166
+ },
1167
+ {
1168
+ "epoch": 0.9682463314890546,
1169
+ "grad_norm": 1.6178405284881592,
1170
+ "learning_rate": 3.3867572768823675e-05,
1171
+ "loss": 0.9441,
1172
+ "step": 16100
1173
+ },
1174
+ {
1175
+ "epoch": 0.9742602838585518,
1176
+ "grad_norm": 1.7407509088516235,
1177
+ "learning_rate": 3.376734022933205e-05,
1178
+ "loss": 0.9264,
1179
+ "step": 16200
1180
+ },
1181
+ {
1182
+ "epoch": 0.9802742362280491,
1183
+ "grad_norm": 1.5831618309020996,
1184
+ "learning_rate": 3.366710768984043e-05,
1185
+ "loss": 0.9583,
1186
+ "step": 16300
1187
+ },
1188
+ {
1189
+ "epoch": 0.9862881885975463,
1190
+ "grad_norm": 1.6853969097137451,
1191
+ "learning_rate": 3.3566875150348806e-05,
1192
+ "loss": 0.9226,
1193
+ "step": 16400
1194
+ },
1195
+ {
1196
+ "epoch": 0.9923021409670435,
1197
+ "grad_norm": 1.8264875411987305,
1198
+ "learning_rate": 3.3466642610857194e-05,
1199
+ "loss": 0.9492,
1200
+ "step": 16500
1201
+ },
1202
+ {
1203
+ "epoch": 0.9983160933365408,
1204
+ "grad_norm": 1.5579068660736084,
1205
+ "learning_rate": 3.336641007136557e-05,
1206
+ "loss": 0.8872,
1207
+ "step": 16600
1208
+ },
1209
+ {
1210
+ "epoch": 1.004330045706038,
1211
+ "grad_norm": 1.7848278284072876,
1212
+ "learning_rate": 3.326617753187395e-05,
1213
+ "loss": 0.8717,
1214
+ "step": 16700
1215
+ },
1216
+ {
1217
+ "epoch": 1.0103439980755353,
1218
+ "grad_norm": 1.5830621719360352,
1219
+ "learning_rate": 3.316694731777725e-05,
1220
+ "loss": 0.9032,
1221
+ "step": 16800
1222
+ },
1223
+ {
1224
+ "epoch": 1.0163579504450324,
1225
+ "grad_norm": 1.546217441558838,
1226
+ "learning_rate": 3.3066714778285623e-05,
1227
+ "loss": 0.9418,
1228
+ "step": 16900
1229
+ },
1230
+ {
1231
+ "epoch": 1.0223719028145297,
1232
+ "grad_norm": 1.4024384021759033,
1233
+ "learning_rate": 3.2966482238794005e-05,
1234
+ "loss": 0.9075,
1235
+ "step": 17000
1236
+ },
1237
+ {
1238
+ "epoch": 1.028385855184027,
1239
+ "grad_norm": 1.6523361206054688,
1240
+ "learning_rate": 3.286624969930238e-05,
1241
+ "loss": 0.9088,
1242
+ "step": 17100
1243
+ },
1244
+ {
1245
+ "epoch": 1.0343998075535241,
1246
+ "grad_norm": 1.4941192865371704,
1247
+ "learning_rate": 3.276601715981076e-05,
1248
+ "loss": 0.8703,
1249
+ "step": 17200
1250
+ },
1251
+ {
1252
+ "epoch": 1.0404137599230214,
1253
+ "grad_norm": 1.6586402654647827,
1254
+ "learning_rate": 3.266578462031914e-05,
1255
+ "loss": 0.9171,
1256
+ "step": 17300
1257
+ },
1258
+ {
1259
+ "epoch": 1.0464277122925187,
1260
+ "grad_norm": 1.5614475011825562,
1261
+ "learning_rate": 3.2565552080827523e-05,
1262
+ "loss": 0.8819,
1263
+ "step": 17400
1264
+ },
1265
+ {
1266
+ "epoch": 1.0524416646620158,
1267
+ "grad_norm": 1.6588680744171143,
1268
+ "learning_rate": 3.2465319541335905e-05,
1269
+ "loss": 0.9165,
1270
+ "step": 17500
1271
+ },
1272
+ {
1273
+ "epoch": 1.0584556170315131,
1274
+ "grad_norm": 1.4571171998977661,
1275
+ "learning_rate": 3.236508700184428e-05,
1276
+ "loss": 0.9056,
1277
+ "step": 17600
1278
+ },
1279
+ {
1280
+ "epoch": 1.0644695694010102,
1281
+ "grad_norm": 1.7484580278396606,
1282
+ "learning_rate": 3.226485446235266e-05,
1283
+ "loss": 0.9072,
1284
+ "step": 17700
1285
+ },
1286
+ {
1287
+ "epoch": 1.0704835217705075,
1288
+ "grad_norm": 1.388741374015808,
1289
+ "learning_rate": 3.2164621922861035e-05,
1290
+ "loss": 0.8645,
1291
+ "step": 17800
1292
+ },
1293
+ {
1294
+ "epoch": 1.0764974741400049,
1295
+ "grad_norm": 1.5871518850326538,
1296
+ "learning_rate": 3.206438938336942e-05,
1297
+ "loss": 0.908,
1298
+ "step": 17900
1299
+ },
1300
+ {
1301
+ "epoch": 1.0825114265095022,
1302
+ "grad_norm": 1.4603219032287598,
1303
+ "learning_rate": 3.19641568438778e-05,
1304
+ "loss": 0.8714,
1305
+ "step": 18000
1306
+ },
1307
+ {
1308
+ "epoch": 1.0885253788789993,
1309
+ "grad_norm": 1.443608283996582,
1310
+ "learning_rate": 3.186392430438618e-05,
1311
+ "loss": 0.8879,
1312
+ "step": 18100
1313
+ },
1314
+ {
1315
+ "epoch": 1.0945393312484966,
1316
+ "grad_norm": 1.4648326635360718,
1317
+ "learning_rate": 3.1763691764894554e-05,
1318
+ "loss": 0.923,
1319
+ "step": 18200
1320
+ },
1321
+ {
1322
+ "epoch": 1.1005532836179936,
1323
+ "grad_norm": 1.9082708358764648,
1324
+ "learning_rate": 3.1663459225402935e-05,
1325
+ "loss": 0.8898,
1326
+ "step": 18300
1327
+ },
1328
+ {
1329
+ "epoch": 1.106567235987491,
1330
+ "grad_norm": 1.740161418914795,
1331
+ "learning_rate": 3.1563226685911317e-05,
1332
+ "loss": 0.8913,
1333
+ "step": 18400
1334
+ },
1335
+ {
1336
+ "epoch": 1.1125811883569883,
1337
+ "grad_norm": 1.4581352472305298,
1338
+ "learning_rate": 3.146299414641969e-05,
1339
+ "loss": 0.9202,
1340
+ "step": 18500
1341
+ },
1342
+ {
1343
+ "epoch": 1.1185951407264854,
1344
+ "grad_norm": 1.5199153423309326,
1345
+ "learning_rate": 3.136276160692808e-05,
1346
+ "loss": 0.9032,
1347
+ "step": 18600
1348
+ },
1349
+ {
1350
+ "epoch": 1.1246090930959827,
1351
+ "grad_norm": 1.4630061388015747,
1352
+ "learning_rate": 3.1262529067436454e-05,
1353
+ "loss": 0.8771,
1354
+ "step": 18700
1355
+ },
1356
+ {
1357
+ "epoch": 1.13062304546548,
1358
+ "grad_norm": 1.7790659666061401,
1359
+ "learning_rate": 3.1162296527944835e-05,
1360
+ "loss": 0.8673,
1361
+ "step": 18800
1362
+ },
1363
+ {
1364
+ "epoch": 1.136636997834977,
1365
+ "grad_norm": 1.610372543334961,
1366
+ "learning_rate": 3.106206398845321e-05,
1367
+ "loss": 0.9019,
1368
+ "step": 18900
1369
+ },
1370
+ {
1371
+ "epoch": 1.1426509502044744,
1372
+ "grad_norm": 1.847385048866272,
1373
+ "learning_rate": 3.096183144896159e-05,
1374
+ "loss": 0.8907,
1375
+ "step": 19000
1376
+ },
1377
+ {
1378
+ "epoch": 1.1486649025739717,
1379
+ "grad_norm": 1.664432168006897,
1380
+ "learning_rate": 3.086159890946997e-05,
1381
+ "loss": 0.8745,
1382
+ "step": 19100
1383
+ },
1384
+ {
1385
+ "epoch": 1.1546788549434688,
1386
+ "grad_norm": 1.5932984352111816,
1387
+ "learning_rate": 3.0761366369978354e-05,
1388
+ "loss": 0.8697,
1389
+ "step": 19200
1390
+ },
1391
+ {
1392
+ "epoch": 1.160692807312966,
1393
+ "grad_norm": 1.4511469602584839,
1394
+ "learning_rate": 3.0662136155881646e-05,
1395
+ "loss": 0.8748,
1396
+ "step": 19300
1397
+ },
1398
+ {
1399
+ "epoch": 1.1667067596824634,
1400
+ "grad_norm": 1.9143450260162354,
1401
+ "learning_rate": 3.056190361639003e-05,
1402
+ "loss": 0.9045,
1403
+ "step": 19400
1404
+ },
1405
+ {
1406
+ "epoch": 1.1727207120519605,
1407
+ "grad_norm": 1.3927329778671265,
1408
+ "learning_rate": 3.046167107689841e-05,
1409
+ "loss": 0.8927,
1410
+ "step": 19500
1411
+ },
1412
+ {
1413
+ "epoch": 1.1787346644214578,
1414
+ "grad_norm": 1.8178434371948242,
1415
+ "learning_rate": 3.0361438537406783e-05,
1416
+ "loss": 0.9258,
1417
+ "step": 19600
1418
+ },
1419
+ {
1420
+ "epoch": 1.1847486167909551,
1421
+ "grad_norm": 1.6006417274475098,
1422
+ "learning_rate": 3.0261205997915165e-05,
1423
+ "loss": 0.886,
1424
+ "step": 19700
1425
+ },
1426
+ {
1427
+ "epoch": 1.1907625691604522,
1428
+ "grad_norm": 1.6698856353759766,
1429
+ "learning_rate": 3.0160973458423543e-05,
1430
+ "loss": 0.9188,
1431
+ "step": 19800
1432
+ },
1433
+ {
1434
+ "epoch": 1.1967765215299495,
1435
+ "grad_norm": 1.5293818712234497,
1436
+ "learning_rate": 3.0060740918931924e-05,
1437
+ "loss": 0.896,
1438
+ "step": 19900
1439
+ },
1440
+ {
1441
+ "epoch": 1.2027904738994466,
1442
+ "grad_norm": 1.6966098546981812,
1443
+ "learning_rate": 2.9960508379440305e-05,
1444
+ "loss": 0.8677,
1445
+ "step": 20000
1446
+ },
1447
+ {
1448
+ "epoch": 1.2027904738994466,
1449
+ "eval_cer": 0.7689350483770306,
1450
+ "eval_loss": 0.8620118498802185,
1451
+ "eval_runtime": 5134.4813,
1452
+ "eval_samples_per_second": 3.241,
1453
+ "eval_steps_per_second": 0.405,
1454
+ "step": 20000
1455
+ },
1456
+ {
1457
+ "epoch": 1.208804426268944,
1458
+ "grad_norm": 1.6066211462020874,
1459
+ "learning_rate": 2.9860275839948683e-05,
1460
+ "loss": 0.8648,
1461
+ "step": 20100
1462
+ },
1463
+ {
1464
+ "epoch": 1.2148183786384412,
1465
+ "grad_norm": 1.4755396842956543,
1466
+ "learning_rate": 2.9760043300457065e-05,
1467
+ "loss": 0.8683,
1468
+ "step": 20200
1469
+ },
1470
+ {
1471
+ "epoch": 1.2208323310079385,
1472
+ "grad_norm": 1.3547738790512085,
1473
+ "learning_rate": 2.965981076096544e-05,
1474
+ "loss": 0.8786,
1475
+ "step": 20300
1476
+ },
1477
+ {
1478
+ "epoch": 1.2268462833774356,
1479
+ "grad_norm": 1.6254231929779053,
1480
+ "learning_rate": 2.955957822147382e-05,
1481
+ "loss": 0.8758,
1482
+ "step": 20400
1483
+ },
1484
+ {
1485
+ "epoch": 1.232860235746933,
1486
+ "grad_norm": 1.5725833177566528,
1487
+ "learning_rate": 2.94593456819822e-05,
1488
+ "loss": 0.8831,
1489
+ "step": 20500
1490
+ },
1491
+ {
1492
+ "epoch": 1.23887418811643,
1493
+ "grad_norm": 1.6321443319320679,
1494
+ "learning_rate": 2.935911314249058e-05,
1495
+ "loss": 0.868,
1496
+ "step": 20600
1497
+ },
1498
+ {
1499
+ "epoch": 1.2448881404859273,
1500
+ "grad_norm": 1.544110894203186,
1501
+ "learning_rate": 2.925888060299896e-05,
1502
+ "loss": 0.9033,
1503
+ "step": 20700
1504
+ },
1505
+ {
1506
+ "epoch": 1.2509020928554246,
1507
+ "grad_norm": 1.5690948963165283,
1508
+ "learning_rate": 2.915864806350734e-05,
1509
+ "loss": 0.8736,
1510
+ "step": 20800
1511
+ },
1512
+ {
1513
+ "epoch": 1.256916045224922,
1514
+ "grad_norm": 1.3950625658035278,
1515
+ "learning_rate": 2.905841552401572e-05,
1516
+ "loss": 0.9012,
1517
+ "step": 20900
1518
+ },
1519
+ {
1520
+ "epoch": 1.262929997594419,
1521
+ "grad_norm": 1.4699276685714722,
1522
+ "learning_rate": 2.8958182984524095e-05,
1523
+ "loss": 0.8545,
1524
+ "step": 21000
1525
+ },
1526
+ {
1527
+ "epoch": 1.2689439499639164,
1528
+ "grad_norm": 1.607750654220581,
1529
+ "learning_rate": 2.8857950445032476e-05,
1530
+ "loss": 0.9101,
1531
+ "step": 21100
1532
+ },
1533
+ {
1534
+ "epoch": 1.2749579023334134,
1535
+ "grad_norm": 2.1910347938537598,
1536
+ "learning_rate": 2.8757717905540854e-05,
1537
+ "loss": 0.8803,
1538
+ "step": 21200
1539
+ },
1540
+ {
1541
+ "epoch": 1.2809718547029108,
1542
+ "grad_norm": 1.604390025138855,
1543
+ "learning_rate": 2.8657485366049236e-05,
1544
+ "loss": 0.8751,
1545
+ "step": 21300
1546
+ },
1547
+ {
1548
+ "epoch": 1.286985807072408,
1549
+ "grad_norm": 1.5971423387527466,
1550
+ "learning_rate": 2.8557252826557617e-05,
1551
+ "loss": 0.8543,
1552
+ "step": 21400
1553
+ },
1554
+ {
1555
+ "epoch": 1.2929997594419051,
1556
+ "grad_norm": 1.6325972080230713,
1557
+ "learning_rate": 2.8457020287065995e-05,
1558
+ "loss": 0.8612,
1559
+ "step": 21500
1560
+ },
1561
+ {
1562
+ "epoch": 1.2990137118114025,
1563
+ "grad_norm": 1.7952935695648193,
1564
+ "learning_rate": 2.8356787747574376e-05,
1565
+ "loss": 0.8399,
1566
+ "step": 21600
1567
+ },
1568
+ {
1569
+ "epoch": 1.3050276641808995,
1570
+ "grad_norm": 1.683236002922058,
1571
+ "learning_rate": 2.825655520808275e-05,
1572
+ "loss": 0.8585,
1573
+ "step": 21700
1574
+ },
1575
+ {
1576
+ "epoch": 1.3110416165503969,
1577
+ "grad_norm": 1.630666971206665,
1578
+ "learning_rate": 2.8156322668591136e-05,
1579
+ "loss": 0.8891,
1580
+ "step": 21800
1581
+ },
1582
+ {
1583
+ "epoch": 1.3170555689198942,
1584
+ "grad_norm": 1.7404346466064453,
1585
+ "learning_rate": 2.805609012909951e-05,
1586
+ "loss": 0.8514,
1587
+ "step": 21900
1588
+ },
1589
+ {
1590
+ "epoch": 1.3230695212893915,
1591
+ "grad_norm": 1.4314298629760742,
1592
+ "learning_rate": 2.795585758960789e-05,
1593
+ "loss": 0.8561,
1594
+ "step": 22000
1595
+ },
1596
+ {
1597
+ "epoch": 1.3290834736588886,
1598
+ "grad_norm": 1.7691779136657715,
1599
+ "learning_rate": 2.785562505011627e-05,
1600
+ "loss": 0.8474,
1601
+ "step": 22100
1602
+ },
1603
+ {
1604
+ "epoch": 1.3350974260283859,
1605
+ "grad_norm": 1.7155267000198364,
1606
+ "learning_rate": 2.775539251062465e-05,
1607
+ "loss": 0.8289,
1608
+ "step": 22200
1609
+ },
1610
+ {
1611
+ "epoch": 1.341111378397883,
1612
+ "grad_norm": 1.7087023258209229,
1613
+ "learning_rate": 2.7655159971133032e-05,
1614
+ "loss": 0.8544,
1615
+ "step": 22300
1616
+ },
1617
+ {
1618
+ "epoch": 1.3471253307673803,
1619
+ "grad_norm": 1.617749571800232,
1620
+ "learning_rate": 2.755492743164141e-05,
1621
+ "loss": 0.8781,
1622
+ "step": 22400
1623
+ },
1624
+ {
1625
+ "epoch": 1.3531392831368776,
1626
+ "grad_norm": 1.6493247747421265,
1627
+ "learning_rate": 2.745469489214979e-05,
1628
+ "loss": 0.8392,
1629
+ "step": 22500
1630
+ },
1631
+ {
1632
+ "epoch": 1.359153235506375,
1633
+ "grad_norm": 1.809634804725647,
1634
+ "learning_rate": 2.7354462352658166e-05,
1635
+ "loss": 0.8721,
1636
+ "step": 22600
1637
+ },
1638
+ {
1639
+ "epoch": 1.365167187875872,
1640
+ "grad_norm": 1.3698049783706665,
1641
+ "learning_rate": 2.7254229813166547e-05,
1642
+ "loss": 0.8533,
1643
+ "step": 22700
1644
+ },
1645
+ {
1646
+ "epoch": 1.3711811402453693,
1647
+ "grad_norm": 1.7568131685256958,
1648
+ "learning_rate": 2.7153997273674925e-05,
1649
+ "loss": 0.886,
1650
+ "step": 22800
1651
+ },
1652
+ {
1653
+ "epoch": 1.3771950926148664,
1654
+ "grad_norm": 1.867412805557251,
1655
+ "learning_rate": 2.7053764734183307e-05,
1656
+ "loss": 0.8637,
1657
+ "step": 22900
1658
+ },
1659
+ {
1660
+ "epoch": 1.3832090449843637,
1661
+ "grad_norm": 2.0730977058410645,
1662
+ "learning_rate": 2.6953532194691688e-05,
1663
+ "loss": 0.8626,
1664
+ "step": 23000
1665
+ },
1666
+ {
1667
+ "epoch": 1.389222997353861,
1668
+ "grad_norm": 1.8011558055877686,
1669
+ "learning_rate": 2.6853299655200066e-05,
1670
+ "loss": 0.8784,
1671
+ "step": 23100
1672
+ },
1673
+ {
1674
+ "epoch": 1.3952369497233583,
1675
+ "grad_norm": 1.6936458349227905,
1676
+ "learning_rate": 2.6753067115708447e-05,
1677
+ "loss": 0.8634,
1678
+ "step": 23200
1679
+ },
1680
+ {
1681
+ "epoch": 1.4012509020928554,
1682
+ "grad_norm": 1.7492289543151855,
1683
+ "learning_rate": 2.665383690161174e-05,
1684
+ "loss": 0.8678,
1685
+ "step": 23300
1686
+ },
1687
+ {
1688
+ "epoch": 1.4072648544623527,
1689
+ "grad_norm": 1.8972880840301514,
1690
+ "learning_rate": 2.655360436212012e-05,
1691
+ "loss": 0.8939,
1692
+ "step": 23400
1693
+ },
1694
+ {
1695
+ "epoch": 1.4132788068318498,
1696
+ "grad_norm": 1.6961406469345093,
1697
+ "learning_rate": 2.64533718226285e-05,
1698
+ "loss": 0.8727,
1699
+ "step": 23500
1700
+ },
1701
+ {
1702
+ "epoch": 1.4192927592013471,
1703
+ "grad_norm": 1.583854079246521,
1704
+ "learning_rate": 2.635313928313688e-05,
1705
+ "loss": 0.8332,
1706
+ "step": 23600
1707
+ },
1708
+ {
1709
+ "epoch": 1.4253067115708444,
1710
+ "grad_norm": 1.6541253328323364,
1711
+ "learning_rate": 2.6252906743645255e-05,
1712
+ "loss": 0.8798,
1713
+ "step": 23700
1714
+ },
1715
+ {
1716
+ "epoch": 1.4313206639403415,
1717
+ "grad_norm": 1.7607979774475098,
1718
+ "learning_rate": 2.6152674204153636e-05,
1719
+ "loss": 0.8472,
1720
+ "step": 23800
1721
+ },
1722
+ {
1723
+ "epoch": 1.4373346163098388,
1724
+ "grad_norm": 1.5591400861740112,
1725
+ "learning_rate": 2.605244166466202e-05,
1726
+ "loss": 0.8355,
1727
+ "step": 23900
1728
+ },
1729
+ {
1730
+ "epoch": 1.4433485686793361,
1731
+ "grad_norm": 1.4700669050216675,
1732
+ "learning_rate": 2.5952209125170395e-05,
1733
+ "loss": 0.8521,
1734
+ "step": 24000
1735
+ },
1736
+ {
1737
+ "epoch": 1.4433485686793361,
1738
+ "eval_cer": 0.7597912847133936,
1739
+ "eval_loss": 0.8321590423583984,
1740
+ "eval_runtime": 5084.7673,
1741
+ "eval_samples_per_second": 3.273,
1742
+ "eval_steps_per_second": 0.409,
1743
+ "step": 24000
1744
+ },
1745
+ {
1746
+ "epoch": 1.4493625210488332,
1747
+ "grad_norm": 1.4155552387237549,
1748
+ "learning_rate": 2.5851976585678777e-05,
1749
+ "loss": 0.8797,
1750
+ "step": 24100
1751
+ },
1752
+ {
1753
+ "epoch": 1.4553764734183305,
1754
+ "grad_norm": 1.4764596223831177,
1755
+ "learning_rate": 2.5751744046187155e-05,
1756
+ "loss": 0.8657,
1757
+ "step": 24200
1758
+ },
1759
+ {
1760
+ "epoch": 1.4613904257878279,
1761
+ "grad_norm": 1.463333010673523,
1762
+ "learning_rate": 2.5651511506695536e-05,
1763
+ "loss": 0.8746,
1764
+ "step": 24300
1765
+ },
1766
+ {
1767
+ "epoch": 1.467404378157325,
1768
+ "grad_norm": 1.5392202138900757,
1769
+ "learning_rate": 2.555127896720391e-05,
1770
+ "loss": 0.8512,
1771
+ "step": 24400
1772
+ },
1773
+ {
1774
+ "epoch": 1.4734183305268223,
1775
+ "grad_norm": 1.8480241298675537,
1776
+ "learning_rate": 2.5451046427712295e-05,
1777
+ "loss": 0.8562,
1778
+ "step": 24500
1779
+ },
1780
+ {
1781
+ "epoch": 1.4794322828963193,
1782
+ "grad_norm": 1.7533873319625854,
1783
+ "learning_rate": 2.5350813888220677e-05,
1784
+ "loss": 0.8534,
1785
+ "step": 24600
1786
+ },
1787
+ {
1788
+ "epoch": 1.4854462352658167,
1789
+ "grad_norm": 1.6647679805755615,
1790
+ "learning_rate": 2.525058134872905e-05,
1791
+ "loss": 0.8335,
1792
+ "step": 24700
1793
+ },
1794
+ {
1795
+ "epoch": 1.491460187635314,
1796
+ "grad_norm": 1.8899763822555542,
1797
+ "learning_rate": 2.5150348809237433e-05,
1798
+ "loss": 0.8486,
1799
+ "step": 24800
1800
+ },
1801
+ {
1802
+ "epoch": 1.4974741400048113,
1803
+ "grad_norm": 1.7569955587387085,
1804
+ "learning_rate": 2.505011626974581e-05,
1805
+ "loss": 0.8415,
1806
+ "step": 24900
1807
+ },
1808
+ {
1809
+ "epoch": 1.5034880923743084,
1810
+ "grad_norm": 1.5141854286193848,
1811
+ "learning_rate": 2.4949883730254192e-05,
1812
+ "loss": 0.8107,
1813
+ "step": 25000
1814
+ },
1815
+ {
1816
+ "epoch": 1.5095020447438057,
1817
+ "grad_norm": 1.8239057064056396,
1818
+ "learning_rate": 2.484965119076257e-05,
1819
+ "loss": 0.8841,
1820
+ "step": 25100
1821
+ },
1822
+ {
1823
+ "epoch": 1.5155159971133028,
1824
+ "grad_norm": 1.433118224143982,
1825
+ "learning_rate": 2.474941865127095e-05,
1826
+ "loss": 0.8408,
1827
+ "step": 25200
1828
+ },
1829
+ {
1830
+ "epoch": 1.5215299494828,
1831
+ "grad_norm": 1.6874032020568848,
1832
+ "learning_rate": 2.4650188437174247e-05,
1833
+ "loss": 0.8387,
1834
+ "step": 25300
1835
+ },
1836
+ {
1837
+ "epoch": 1.5275439018522974,
1838
+ "grad_norm": 1.4172905683517456,
1839
+ "learning_rate": 2.4549955897682625e-05,
1840
+ "loss": 0.8357,
1841
+ "step": 25400
1842
+ },
1843
+ {
1844
+ "epoch": 1.5335578542217947,
1845
+ "grad_norm": 1.4848599433898926,
1846
+ "learning_rate": 2.4449723358191003e-05,
1847
+ "loss": 0.8451,
1848
+ "step": 25500
1849
+ },
1850
+ {
1851
+ "epoch": 1.5395718065912918,
1852
+ "grad_norm": 1.4670342206954956,
1853
+ "learning_rate": 2.4349490818699384e-05,
1854
+ "loss": 0.8505,
1855
+ "step": 25600
1856
+ },
1857
+ {
1858
+ "epoch": 1.5455857589607889,
1859
+ "grad_norm": 1.5914552211761475,
1860
+ "learning_rate": 2.4249258279207762e-05,
1861
+ "loss": 0.8635,
1862
+ "step": 25700
1863
+ },
1864
+ {
1865
+ "epoch": 1.5515997113302862,
1866
+ "grad_norm": 1.5550841093063354,
1867
+ "learning_rate": 2.414902573971614e-05,
1868
+ "loss": 0.8763,
1869
+ "step": 25800
1870
+ },
1871
+ {
1872
+ "epoch": 1.5576136636997835,
1873
+ "grad_norm": 1.5907316207885742,
1874
+ "learning_rate": 2.4048793200224525e-05,
1875
+ "loss": 0.8664,
1876
+ "step": 25900
1877
+ },
1878
+ {
1879
+ "epoch": 1.5636276160692808,
1880
+ "grad_norm": 1.4494388103485107,
1881
+ "learning_rate": 2.3948560660732903e-05,
1882
+ "loss": 0.819,
1883
+ "step": 26000
1884
+ },
1885
+ {
1886
+ "epoch": 1.5696415684387781,
1887
+ "grad_norm": 1.599004864692688,
1888
+ "learning_rate": 2.384832812124128e-05,
1889
+ "loss": 0.8363,
1890
+ "step": 26100
1891
+ },
1892
+ {
1893
+ "epoch": 1.5756555208082752,
1894
+ "grad_norm": 1.887817621231079,
1895
+ "learning_rate": 2.3748095581749662e-05,
1896
+ "loss": 0.8845,
1897
+ "step": 26200
1898
+ },
1899
+ {
1900
+ "epoch": 1.5816694731777723,
1901
+ "grad_norm": 1.3124005794525146,
1902
+ "learning_rate": 2.364786304225804e-05,
1903
+ "loss": 0.858,
1904
+ "step": 26300
1905
+ },
1906
+ {
1907
+ "epoch": 1.5876834255472696,
1908
+ "grad_norm": 1.6560554504394531,
1909
+ "learning_rate": 2.3547630502766418e-05,
1910
+ "loss": 0.8553,
1911
+ "step": 26400
1912
+ },
1913
+ {
1914
+ "epoch": 1.593697377916767,
1915
+ "grad_norm": 1.5678675174713135,
1916
+ "learning_rate": 2.34473979632748e-05,
1917
+ "loss": 0.8488,
1918
+ "step": 26500
1919
+ },
1920
+ {
1921
+ "epoch": 1.5997113302862642,
1922
+ "grad_norm": 1.4168376922607422,
1923
+ "learning_rate": 2.334716542378318e-05,
1924
+ "loss": 0.8263,
1925
+ "step": 26600
1926
+ },
1927
+ {
1928
+ "epoch": 1.6057252826557615,
1929
+ "grad_norm": 1.6189205646514893,
1930
+ "learning_rate": 2.324693288429156e-05,
1931
+ "loss": 0.8489,
1932
+ "step": 26700
1933
+ },
1934
+ {
1935
+ "epoch": 1.6117392350252586,
1936
+ "grad_norm": 1.7498302459716797,
1937
+ "learning_rate": 2.3146700344799936e-05,
1938
+ "loss": 0.8069,
1939
+ "step": 26800
1940
+ },
1941
+ {
1942
+ "epoch": 1.6177531873947557,
1943
+ "grad_norm": 1.5609160661697388,
1944
+ "learning_rate": 2.3046467805308318e-05,
1945
+ "loss": 0.8337,
1946
+ "step": 26900
1947
+ },
1948
+ {
1949
+ "epoch": 1.623767139764253,
1950
+ "grad_norm": 1.7673338651657104,
1951
+ "learning_rate": 2.2946235265816696e-05,
1952
+ "loss": 0.8264,
1953
+ "step": 27000
1954
+ },
1955
+ {
1956
+ "epoch": 1.6297810921337503,
1957
+ "grad_norm": 1.593299150466919,
1958
+ "learning_rate": 2.2846002726325074e-05,
1959
+ "loss": 0.828,
1960
+ "step": 27100
1961
+ },
1962
+ {
1963
+ "epoch": 1.6357950445032476,
1964
+ "grad_norm": 1.5802645683288574,
1965
+ "learning_rate": 2.274677251222837e-05,
1966
+ "loss": 0.8428,
1967
+ "step": 27200
1968
+ },
1969
+ {
1970
+ "epoch": 1.6418089968727447,
1971
+ "grad_norm": 1.7584878206253052,
1972
+ "learning_rate": 2.264653997273675e-05,
1973
+ "loss": 0.8378,
1974
+ "step": 27300
1975
+ },
1976
+ {
1977
+ "epoch": 1.647822949242242,
1978
+ "grad_norm": 1.5360692739486694,
1979
+ "learning_rate": 2.254630743324513e-05,
1980
+ "loss": 0.8288,
1981
+ "step": 27400
1982
+ },
1983
+ {
1984
+ "epoch": 1.6538369016117391,
1985
+ "grad_norm": 1.5635976791381836,
1986
+ "learning_rate": 2.244607489375351e-05,
1987
+ "loss": 0.8208,
1988
+ "step": 27500
1989
+ },
1990
+ {
1991
+ "epoch": 1.6598508539812364,
1992
+ "grad_norm": 1.778735876083374,
1993
+ "learning_rate": 2.2345842354261888e-05,
1994
+ "loss": 0.847,
1995
+ "step": 27600
1996
+ },
1997
+ {
1998
+ "epoch": 1.6658648063507338,
1999
+ "grad_norm": 1.5961335897445679,
2000
+ "learning_rate": 2.224560981477027e-05,
2001
+ "loss": 0.8059,
2002
+ "step": 27700
2003
+ },
2004
+ {
2005
+ "epoch": 1.671878758720231,
2006
+ "grad_norm": 1.396517038345337,
2007
+ "learning_rate": 2.2145377275278647e-05,
2008
+ "loss": 0.8406,
2009
+ "step": 27800
2010
+ },
2011
+ {
2012
+ "epoch": 1.6778927110897282,
2013
+ "grad_norm": 1.554319977760315,
2014
+ "learning_rate": 2.2045144735787025e-05,
2015
+ "loss": 0.836,
2016
+ "step": 27900
2017
+ },
2018
+ {
2019
+ "epoch": 1.6839066634592252,
2020
+ "grad_norm": 1.5663318634033203,
2021
+ "learning_rate": 2.1944912196295406e-05,
2022
+ "loss": 0.8386,
2023
+ "step": 28000
2024
+ },
2025
+ {
2026
+ "epoch": 1.6839066634592252,
2027
+ "eval_cer": 0.7604478286805494,
2028
+ "eval_loss": 0.8103429079055786,
2029
+ "eval_runtime": 5152.8312,
2030
+ "eval_samples_per_second": 3.229,
2031
+ "eval_steps_per_second": 0.404,
2032
+ "step": 28000
2033
+ },
2034
+ {
2035
+ "epoch": 1.6899206158287225,
2036
+ "grad_norm": 1.7991820573806763,
2037
+ "learning_rate": 2.1844679656803784e-05,
2038
+ "loss": 0.8428,
2039
+ "step": 28100
2040
+ },
2041
+ {
2042
+ "epoch": 1.6959345681982199,
2043
+ "grad_norm": 1.6566849946975708,
2044
+ "learning_rate": 2.1744447117312162e-05,
2045
+ "loss": 0.8331,
2046
+ "step": 28200
2047
+ },
2048
+ {
2049
+ "epoch": 1.7019485205677172,
2050
+ "grad_norm": 1.650564432144165,
2051
+ "learning_rate": 2.1644214577820547e-05,
2052
+ "loss": 0.8379,
2053
+ "step": 28300
2054
+ },
2055
+ {
2056
+ "epoch": 1.7079624729372145,
2057
+ "grad_norm": 1.6355592012405396,
2058
+ "learning_rate": 2.1543982038328925e-05,
2059
+ "loss": 0.8576,
2060
+ "step": 28400
2061
+ },
2062
+ {
2063
+ "epoch": 1.7139764253067116,
2064
+ "grad_norm": 1.7112095355987549,
2065
+ "learning_rate": 2.1443749498837303e-05,
2066
+ "loss": 0.8173,
2067
+ "step": 28500
2068
+ },
2069
+ {
2070
+ "epoch": 1.7199903776762087,
2071
+ "grad_norm": 1.7781462669372559,
2072
+ "learning_rate": 2.1343516959345684e-05,
2073
+ "loss": 0.8292,
2074
+ "step": 28600
2075
+ },
2076
+ {
2077
+ "epoch": 1.726004330045706,
2078
+ "grad_norm": 1.708770513534546,
2079
+ "learning_rate": 2.1243284419854062e-05,
2080
+ "loss": 0.8312,
2081
+ "step": 28700
2082
+ },
2083
+ {
2084
+ "epoch": 1.7320182824152033,
2085
+ "grad_norm": 1.869710087776184,
2086
+ "learning_rate": 2.114305188036244e-05,
2087
+ "loss": 0.8529,
2088
+ "step": 28800
2089
+ },
2090
+ {
2091
+ "epoch": 1.7380322347847006,
2092
+ "grad_norm": 1.4506940841674805,
2093
+ "learning_rate": 2.104281934087082e-05,
2094
+ "loss": 0.8454,
2095
+ "step": 28900
2096
+ },
2097
+ {
2098
+ "epoch": 1.744046187154198,
2099
+ "grad_norm": 1.5264636278152466,
2100
+ "learning_rate": 2.0942586801379203e-05,
2101
+ "loss": 0.8281,
2102
+ "step": 29000
2103
+ },
2104
+ {
2105
+ "epoch": 1.750060139523695,
2106
+ "grad_norm": 1.9614264965057373,
2107
+ "learning_rate": 2.084235426188758e-05,
2108
+ "loss": 0.8328,
2109
+ "step": 29100
2110
+ },
2111
+ {
2112
+ "epoch": 1.756074091893192,
2113
+ "grad_norm": 1.4223591089248657,
2114
+ "learning_rate": 2.074212172239596e-05,
2115
+ "loss": 0.8142,
2116
+ "step": 29200
2117
+ },
2118
+ {
2119
+ "epoch": 1.7620880442626894,
2120
+ "grad_norm": 1.6863304376602173,
2121
+ "learning_rate": 2.064188918290434e-05,
2122
+ "loss": 0.8306,
2123
+ "step": 29300
2124
+ },
2125
+ {
2126
+ "epoch": 1.7681019966321867,
2127
+ "grad_norm": 1.5096672773361206,
2128
+ "learning_rate": 2.0541656643412718e-05,
2129
+ "loss": 0.7964,
2130
+ "step": 29400
2131
+ },
2132
+ {
2133
+ "epoch": 1.774115949001684,
2134
+ "grad_norm": 1.9181997776031494,
2135
+ "learning_rate": 2.0441424103921096e-05,
2136
+ "loss": 0.8221,
2137
+ "step": 29500
2138
+ },
2139
+ {
2140
+ "epoch": 1.780129901371181,
2141
+ "grad_norm": 1.9824730157852173,
2142
+ "learning_rate": 2.0341191564429478e-05,
2143
+ "loss": 0.8379,
2144
+ "step": 29600
2145
+ },
2146
+ {
2147
+ "epoch": 1.7861438537406784,
2148
+ "grad_norm": 1.4306327104568481,
2149
+ "learning_rate": 2.0240959024937855e-05,
2150
+ "loss": 0.805,
2151
+ "step": 29700
2152
+ },
2153
+ {
2154
+ "epoch": 1.7921578061101755,
2155
+ "grad_norm": 1.6249910593032837,
2156
+ "learning_rate": 2.0140726485446237e-05,
2157
+ "loss": 0.8145,
2158
+ "step": 29800
2159
+ },
2160
+ {
2161
+ "epoch": 1.7981717584796728,
2162
+ "grad_norm": 1.551161766052246,
2163
+ "learning_rate": 2.0040493945954618e-05,
2164
+ "loss": 0.8436,
2165
+ "step": 29900
2166
+ },
2167
+ {
2168
+ "epoch": 1.8041857108491701,
2169
+ "grad_norm": 1.5218690633773804,
2170
+ "learning_rate": 1.9940261406462996e-05,
2171
+ "loss": 0.8181,
2172
+ "step": 30000
2173
+ },
2174
+ {
2175
+ "epoch": 1.8101996632186674,
2176
+ "grad_norm": 1.5932899713516235,
2177
+ "learning_rate": 1.9840028866971374e-05,
2178
+ "loss": 0.8478,
2179
+ "step": 30100
2180
+ },
2181
+ {
2182
+ "epoch": 1.8162136155881645,
2183
+ "grad_norm": 1.4991642236709595,
2184
+ "learning_rate": 1.9739796327479752e-05,
2185
+ "loss": 0.8254,
2186
+ "step": 30200
2187
+ },
2188
+ {
2189
+ "epoch": 1.8222275679576616,
2190
+ "grad_norm": 1.647438883781433,
2191
+ "learning_rate": 1.9639563787988133e-05,
2192
+ "loss": 0.8342,
2193
+ "step": 30300
2194
+ },
2195
+ {
2196
+ "epoch": 1.828241520327159,
2197
+ "grad_norm": 1.6653351783752441,
2198
+ "learning_rate": 1.953933124849651e-05,
2199
+ "loss": 0.8249,
2200
+ "step": 30400
2201
+ },
2202
+ {
2203
+ "epoch": 1.8342554726966562,
2204
+ "grad_norm": 1.6969921588897705,
2205
+ "learning_rate": 1.943909870900489e-05,
2206
+ "loss": 0.8544,
2207
+ "step": 30500
2208
+ },
2209
+ {
2210
+ "epoch": 1.8402694250661535,
2211
+ "grad_norm": 1.7201420068740845,
2212
+ "learning_rate": 1.9338866169513274e-05,
2213
+ "loss": 0.8323,
2214
+ "step": 30600
2215
+ },
2216
+ {
2217
+ "epoch": 1.8462833774356509,
2218
+ "grad_norm": 1.7479013204574585,
2219
+ "learning_rate": 1.9238633630021652e-05,
2220
+ "loss": 0.8307,
2221
+ "step": 30700
2222
+ },
2223
+ {
2224
+ "epoch": 1.852297329805148,
2225
+ "grad_norm": 1.6966118812561035,
2226
+ "learning_rate": 1.913840109053003e-05,
2227
+ "loss": 0.8291,
2228
+ "step": 30800
2229
+ },
2230
+ {
2231
+ "epoch": 1.858311282174645,
2232
+ "grad_norm": 1.6381275653839111,
2233
+ "learning_rate": 1.903816855103841e-05,
2234
+ "loss": 0.8121,
2235
+ "step": 30900
2236
+ },
2237
+ {
2238
+ "epoch": 1.8643252345441423,
2239
+ "grad_norm": 1.4601831436157227,
2240
+ "learning_rate": 1.893793601154679e-05,
2241
+ "loss": 0.8174,
2242
+ "step": 31000
2243
+ },
2244
+ {
2245
+ "epoch": 1.8703391869136397,
2246
+ "grad_norm": 1.8310879468917847,
2247
+ "learning_rate": 1.8837703472055167e-05,
2248
+ "loss": 0.8023,
2249
+ "step": 31100
2250
+ },
2251
+ {
2252
+ "epoch": 1.876353139283137,
2253
+ "grad_norm": 1.4689913988113403,
2254
+ "learning_rate": 1.8738473257958463e-05,
2255
+ "loss": 0.8035,
2256
+ "step": 31200
2257
+ },
2258
+ {
2259
+ "epoch": 1.8823670916526343,
2260
+ "grad_norm": 1.4330099821090698,
2261
+ "learning_rate": 1.8638240718466844e-05,
2262
+ "loss": 0.8185,
2263
+ "step": 31300
2264
+ },
2265
+ {
2266
+ "epoch": 1.8883810440221314,
2267
+ "grad_norm": 1.8419737815856934,
2268
+ "learning_rate": 1.8538008178975225e-05,
2269
+ "loss": 0.8144,
2270
+ "step": 31400
2271
+ },
2272
+ {
2273
+ "epoch": 1.8943949963916284,
2274
+ "grad_norm": 1.2664531469345093,
2275
+ "learning_rate": 1.8437775639483603e-05,
2276
+ "loss": 0.816,
2277
+ "step": 31500
2278
+ },
2279
+ {
2280
+ "epoch": 1.9004089487611258,
2281
+ "grad_norm": 1.6704432964324951,
2282
+ "learning_rate": 1.833754309999198e-05,
2283
+ "loss": 0.817,
2284
+ "step": 31600
2285
+ },
2286
+ {
2287
+ "epoch": 1.906422901130623,
2288
+ "grad_norm": 1.7487777471542358,
2289
+ "learning_rate": 1.8237310560500363e-05,
2290
+ "loss": 0.8041,
2291
+ "step": 31700
2292
+ },
2293
+ {
2294
+ "epoch": 1.9124368535001204,
2295
+ "grad_norm": 1.2405571937561035,
2296
+ "learning_rate": 1.813707802100874e-05,
2297
+ "loss": 0.825,
2298
+ "step": 31800
2299
+ },
2300
+ {
2301
+ "epoch": 1.9184508058696175,
2302
+ "grad_norm": 1.543731689453125,
2303
+ "learning_rate": 1.803684548151712e-05,
2304
+ "loss": 0.8147,
2305
+ "step": 31900
2306
+ },
2307
+ {
2308
+ "epoch": 1.9244647582391148,
2309
+ "grad_norm": 1.5452948808670044,
2310
+ "learning_rate": 1.79366129420255e-05,
2311
+ "loss": 0.7973,
2312
+ "step": 32000
2313
+ },
2314
+ {
2315
+ "epoch": 1.9244647582391148,
2316
+ "eval_cer": 0.7489049827996835,
2317
+ "eval_loss": 0.7830217480659485,
2318
+ "eval_runtime": 5092.4712,
2319
+ "eval_samples_per_second": 3.268,
2320
+ "eval_steps_per_second": 0.408,
2321
+ "step": 32000
2322
+ },
2323
+ {
2324
+ "epoch": 1.9304787106086119,
2325
+ "grad_norm": 1.640507459640503,
2326
+ "learning_rate": 1.7836380402533878e-05,
2327
+ "loss": 0.831,
2328
+ "step": 32100
2329
+ },
2330
+ {
2331
+ "epoch": 1.9364926629781092,
2332
+ "grad_norm": 1.919505000114441,
2333
+ "learning_rate": 1.773614786304226e-05,
2334
+ "loss": 0.8216,
2335
+ "step": 32200
2336
+ },
2337
+ {
2338
+ "epoch": 1.9425066153476065,
2339
+ "grad_norm": 1.5534350872039795,
2340
+ "learning_rate": 1.763591532355064e-05,
2341
+ "loss": 0.8157,
2342
+ "step": 32300
2343
+ },
2344
+ {
2345
+ "epoch": 1.9485205677171038,
2346
+ "grad_norm": 1.5943905115127563,
2347
+ "learning_rate": 1.753568278405902e-05,
2348
+ "loss": 0.8026,
2349
+ "step": 32400
2350
+ },
2351
+ {
2352
+ "epoch": 1.954534520086601,
2353
+ "grad_norm": 1.788720965385437,
2354
+ "learning_rate": 1.7435450244567397e-05,
2355
+ "loss": 0.7798,
2356
+ "step": 32500
2357
+ },
2358
+ {
2359
+ "epoch": 1.9605484724560982,
2360
+ "grad_norm": 1.707412600517273,
2361
+ "learning_rate": 1.7335217705075778e-05,
2362
+ "loss": 0.8125,
2363
+ "step": 32600
2364
+ },
2365
+ {
2366
+ "epoch": 1.9665624248255953,
2367
+ "grad_norm": 1.6318702697753906,
2368
+ "learning_rate": 1.7234985165584156e-05,
2369
+ "loss": 0.8184,
2370
+ "step": 32700
2371
+ },
2372
+ {
2373
+ "epoch": 1.9725763771950926,
2374
+ "grad_norm": 1.6191486120224,
2375
+ "learning_rate": 1.7134752626092534e-05,
2376
+ "loss": 0.8056,
2377
+ "step": 32800
2378
+ },
2379
+ {
2380
+ "epoch": 1.97859032956459,
2381
+ "grad_norm": 1.6070526838302612,
2382
+ "learning_rate": 1.7034520086600915e-05,
2383
+ "loss": 0.8134,
2384
+ "step": 32900
2385
+ },
2386
+ {
2387
+ "epoch": 1.9846042819340872,
2388
+ "grad_norm": 1.3369784355163574,
2389
+ "learning_rate": 1.6934287547109297e-05,
2390
+ "loss": 0.8086,
2391
+ "step": 33000
2392
+ },
2393
+ {
2394
+ "epoch": 1.9906182343035843,
2395
+ "grad_norm": 1.7080943584442139,
2396
+ "learning_rate": 1.6834055007617674e-05,
2397
+ "loss": 0.8304,
2398
+ "step": 33100
2399
+ },
2400
+ {
2401
+ "epoch": 1.9966321866730814,
2402
+ "grad_norm": 1.3051142692565918,
2403
+ "learning_rate": 1.673482479352097e-05,
2404
+ "loss": 0.778,
2405
+ "step": 33200
2406
+ },
2407
+ {
2408
+ "epoch": 2.0026461390425787,
2409
+ "grad_norm": 1.9086428880691528,
2410
+ "learning_rate": 1.6634592254029348e-05,
2411
+ "loss": 0.7972,
2412
+ "step": 33300
2413
+ },
2414
+ {
2415
+ "epoch": 2.008660091412076,
2416
+ "grad_norm": 1.6656805276870728,
2417
+ "learning_rate": 1.653435971453773e-05,
2418
+ "loss": 0.7701,
2419
+ "step": 33400
2420
+ },
2421
+ {
2422
+ "epoch": 2.0146740437815733,
2423
+ "grad_norm": 1.5769175291061401,
2424
+ "learning_rate": 1.6434127175046107e-05,
2425
+ "loss": 0.7602,
2426
+ "step": 33500
2427
+ },
2428
+ {
2429
+ "epoch": 2.0206879961510706,
2430
+ "grad_norm": 1.6363704204559326,
2431
+ "learning_rate": 1.6333894635554485e-05,
2432
+ "loss": 0.7826,
2433
+ "step": 33600
2434
+ },
2435
+ {
2436
+ "epoch": 2.0267019485205675,
2437
+ "grad_norm": 1.784424066543579,
2438
+ "learning_rate": 1.6233662096062867e-05,
2439
+ "loss": 0.7805,
2440
+ "step": 33700
2441
+ },
2442
+ {
2443
+ "epoch": 2.032715900890065,
2444
+ "grad_norm": 1.5795265436172485,
2445
+ "learning_rate": 1.6133429556571248e-05,
2446
+ "loss": 0.7289,
2447
+ "step": 33800
2448
+ },
2449
+ {
2450
+ "epoch": 2.038729853259562,
2451
+ "grad_norm": 1.382318377494812,
2452
+ "learning_rate": 1.6033197017079626e-05,
2453
+ "loss": 0.7869,
2454
+ "step": 33900
2455
+ },
2456
+ {
2457
+ "epoch": 2.0447438056290594,
2458
+ "grad_norm": 2.0357506275177,
2459
+ "learning_rate": 1.5932964477588004e-05,
2460
+ "loss": 0.7477,
2461
+ "step": 34000
2462
+ },
2463
+ {
2464
+ "epoch": 2.0507577579985568,
2465
+ "grad_norm": 1.6945205926895142,
2466
+ "learning_rate": 1.5832731938096385e-05,
2467
+ "loss": 0.7582,
2468
+ "step": 34100
2469
+ },
2470
+ {
2471
+ "epoch": 2.056771710368054,
2472
+ "grad_norm": 1.4343385696411133,
2473
+ "learning_rate": 1.5732499398604763e-05,
2474
+ "loss": 0.7536,
2475
+ "step": 34200
2476
+ },
2477
+ {
2478
+ "epoch": 2.062785662737551,
2479
+ "grad_norm": 1.7783600091934204,
2480
+ "learning_rate": 1.563226685911314e-05,
2481
+ "loss": 0.8331,
2482
+ "step": 34300
2483
+ },
2484
+ {
2485
+ "epoch": 2.0687996151070482,
2486
+ "grad_norm": 1.8440674543380737,
2487
+ "learning_rate": 1.5532034319621522e-05,
2488
+ "loss": 0.7352,
2489
+ "step": 34400
2490
+ },
2491
+ {
2492
+ "epoch": 2.0748135674765456,
2493
+ "grad_norm": 1.8098615407943726,
2494
+ "learning_rate": 1.54318017801299e-05,
2495
+ "loss": 0.7951,
2496
+ "step": 34500
2497
+ },
2498
+ {
2499
+ "epoch": 2.080827519846043,
2500
+ "grad_norm": 1.521584391593933,
2501
+ "learning_rate": 1.5331569240638282e-05,
2502
+ "loss": 0.7684,
2503
+ "step": 34600
2504
+ },
2505
+ {
2506
+ "epoch": 2.08684147221554,
2507
+ "grad_norm": 1.738142490386963,
2508
+ "learning_rate": 1.5231336701146661e-05,
2509
+ "loss": 0.7589,
2510
+ "step": 34700
2511
+ },
2512
+ {
2513
+ "epoch": 2.0928554245850375,
2514
+ "grad_norm": 1.5031851530075073,
2515
+ "learning_rate": 1.5131104161655041e-05,
2516
+ "loss": 0.7508,
2517
+ "step": 34800
2518
+ },
2519
+ {
2520
+ "epoch": 2.0988693769545343,
2521
+ "grad_norm": 1.8703136444091797,
2522
+ "learning_rate": 1.503087162216342e-05,
2523
+ "loss": 0.803,
2524
+ "step": 34900
2525
+ },
2526
+ {
2527
+ "epoch": 2.1048833293240317,
2528
+ "grad_norm": 1.6643913984298706,
2529
+ "learning_rate": 1.4930639082671799e-05,
2530
+ "loss": 0.7689,
2531
+ "step": 35000
2532
+ },
2533
+ {
2534
+ "epoch": 2.110897281693529,
2535
+ "grad_norm": 1.4073503017425537,
2536
+ "learning_rate": 1.4830406543180178e-05,
2537
+ "loss": 0.7575,
2538
+ "step": 35100
2539
+ },
2540
+ {
2541
+ "epoch": 2.1169112340630263,
2542
+ "grad_norm": 1.6007989645004272,
2543
+ "learning_rate": 1.4731176329083474e-05,
2544
+ "loss": 0.7317,
2545
+ "step": 35200
2546
+ },
2547
+ {
2548
+ "epoch": 2.1229251864325236,
2549
+ "grad_norm": 1.7965283393859863,
2550
+ "learning_rate": 1.4630943789591854e-05,
2551
+ "loss": 0.7142,
2552
+ "step": 35300
2553
+ },
2554
+ {
2555
+ "epoch": 2.1289391388020205,
2556
+ "grad_norm": 1.3655446767807007,
2557
+ "learning_rate": 1.4530711250100235e-05,
2558
+ "loss": 0.7613,
2559
+ "step": 35400
2560
+ },
2561
+ {
2562
+ "epoch": 2.1349530911715178,
2563
+ "grad_norm": 1.779159426689148,
2564
+ "learning_rate": 1.4430478710608613e-05,
2565
+ "loss": 0.7827,
2566
+ "step": 35500
2567
+ },
2568
+ {
2569
+ "epoch": 2.140967043541015,
2570
+ "grad_norm": 1.9253307580947876,
2571
+ "learning_rate": 1.4330246171116993e-05,
2572
+ "loss": 0.7899,
2573
+ "step": 35600
2574
+ },
2575
+ {
2576
+ "epoch": 2.1469809959105124,
2577
+ "grad_norm": 1.3449054956436157,
2578
+ "learning_rate": 1.4230013631625372e-05,
2579
+ "loss": 0.7494,
2580
+ "step": 35700
2581
+ },
2582
+ {
2583
+ "epoch": 2.1529949482800097,
2584
+ "grad_norm": 1.9044090509414673,
2585
+ "learning_rate": 1.412978109213375e-05,
2586
+ "loss": 0.7714,
2587
+ "step": 35800
2588
+ },
2589
+ {
2590
+ "epoch": 2.159008900649507,
2591
+ "grad_norm": 1.674017071723938,
2592
+ "learning_rate": 1.402954855264213e-05,
2593
+ "loss": 0.7478,
2594
+ "step": 35900
2595
+ },
2596
+ {
2597
+ "epoch": 2.1650228530190043,
2598
+ "grad_norm": 1.8109982013702393,
2599
+ "learning_rate": 1.392931601315051e-05,
2600
+ "loss": 0.7541,
2601
+ "step": 36000
2602
+ },
2603
+ {
2604
+ "epoch": 2.1650228530190043,
2605
+ "eval_cer": 0.7396310458108724,
2606
+ "eval_loss": 0.7524659633636475,
2607
+ "eval_runtime": 5154.2928,
2608
+ "eval_samples_per_second": 3.228,
2609
+ "eval_steps_per_second": 0.404,
2610
+ "step": 36000
2611
+ },
2612
+ {
2613
+ "epoch": 2.171036805388501,
2614
+ "grad_norm": 1.316389560699463,
2615
+ "learning_rate": 1.3829083473658887e-05,
2616
+ "loss": 0.7647,
2617
+ "step": 36100
2618
+ },
2619
+ {
2620
+ "epoch": 2.1770507577579985,
2621
+ "grad_norm": 1.7067075967788696,
2622
+ "learning_rate": 1.372885093416727e-05,
2623
+ "loss": 0.7635,
2624
+ "step": 36200
2625
+ },
2626
+ {
2627
+ "epoch": 2.183064710127496,
2628
+ "grad_norm": 1.8793973922729492,
2629
+ "learning_rate": 1.3628618394675648e-05,
2630
+ "loss": 0.7313,
2631
+ "step": 36300
2632
+ },
2633
+ {
2634
+ "epoch": 2.189078662496993,
2635
+ "grad_norm": 1.769338607788086,
2636
+ "learning_rate": 1.3528385855184028e-05,
2637
+ "loss": 0.7573,
2638
+ "step": 36400
2639
+ },
2640
+ {
2641
+ "epoch": 2.1950926148664904,
2642
+ "grad_norm": 1.6032990217208862,
2643
+ "learning_rate": 1.3428153315692408e-05,
2644
+ "loss": 0.76,
2645
+ "step": 36500
2646
+ },
2647
+ {
2648
+ "epoch": 2.2011065672359873,
2649
+ "grad_norm": 1.5864907503128052,
2650
+ "learning_rate": 1.3327920776200786e-05,
2651
+ "loss": 0.7587,
2652
+ "step": 36600
2653
+ },
2654
+ {
2655
+ "epoch": 2.2071205196054846,
2656
+ "grad_norm": 1.2785674333572388,
2657
+ "learning_rate": 1.3227688236709165e-05,
2658
+ "loss": 0.7403,
2659
+ "step": 36700
2660
+ },
2661
+ {
2662
+ "epoch": 2.213134471974982,
2663
+ "grad_norm": 1.4437572956085205,
2664
+ "learning_rate": 1.3127455697217545e-05,
2665
+ "loss": 0.7358,
2666
+ "step": 36800
2667
+ },
2668
+ {
2669
+ "epoch": 2.2191484243444792,
2670
+ "grad_norm": 1.8562610149383545,
2671
+ "learning_rate": 1.3027223157725923e-05,
2672
+ "loss": 0.7429,
2673
+ "step": 36900
2674
+ },
2675
+ {
2676
+ "epoch": 2.2251623767139765,
2677
+ "grad_norm": 1.6878858804702759,
2678
+ "learning_rate": 1.2926990618234306e-05,
2679
+ "loss": 0.7526,
2680
+ "step": 37000
2681
+ },
2682
+ {
2683
+ "epoch": 2.231176329083474,
2684
+ "grad_norm": 1.9118757247924805,
2685
+ "learning_rate": 1.2826758078742684e-05,
2686
+ "loss": 0.7513,
2687
+ "step": 37100
2688
+ },
2689
+ {
2690
+ "epoch": 2.2371902814529707,
2691
+ "grad_norm": 1.3607146739959717,
2692
+ "learning_rate": 1.272752786464598e-05,
2693
+ "loss": 0.747,
2694
+ "step": 37200
2695
+ },
2696
+ {
2697
+ "epoch": 2.243204233822468,
2698
+ "grad_norm": 1.8414541482925415,
2699
+ "learning_rate": 1.2627295325154359e-05,
2700
+ "loss": 0.757,
2701
+ "step": 37300
2702
+ },
2703
+ {
2704
+ "epoch": 2.2492181861919653,
2705
+ "grad_norm": 1.5014030933380127,
2706
+ "learning_rate": 1.2527062785662739e-05,
2707
+ "loss": 0.7704,
2708
+ "step": 37400
2709
+ },
2710
+ {
2711
+ "epoch": 2.2552321385614627,
2712
+ "grad_norm": 1.850203514099121,
2713
+ "learning_rate": 1.2426830246171117e-05,
2714
+ "loss": 0.7532,
2715
+ "step": 37500
2716
+ },
2717
+ {
2718
+ "epoch": 2.26124609093096,
2719
+ "grad_norm": 1.9308381080627441,
2720
+ "learning_rate": 1.2326597706679498e-05,
2721
+ "loss": 0.7658,
2722
+ "step": 37600
2723
+ },
2724
+ {
2725
+ "epoch": 2.267260043300457,
2726
+ "grad_norm": 1.8409243822097778,
2727
+ "learning_rate": 1.2226365167187876e-05,
2728
+ "loss": 0.7307,
2729
+ "step": 37700
2730
+ },
2731
+ {
2732
+ "epoch": 2.273273995669954,
2733
+ "grad_norm": 1.7760223150253296,
2734
+ "learning_rate": 1.2126132627696256e-05,
2735
+ "loss": 0.7548,
2736
+ "step": 37800
2737
+ },
2738
+ {
2739
+ "epoch": 2.2792879480394515,
2740
+ "grad_norm": 1.4862762689590454,
2741
+ "learning_rate": 1.2025900088204635e-05,
2742
+ "loss": 0.7196,
2743
+ "step": 37900
2744
+ },
2745
+ {
2746
+ "epoch": 2.2853019004089488,
2747
+ "grad_norm": 1.6604909896850586,
2748
+ "learning_rate": 1.1925667548713015e-05,
2749
+ "loss": 0.7306,
2750
+ "step": 38000
2751
+ },
2752
+ {
2753
+ "epoch": 2.291315852778446,
2754
+ "grad_norm": 1.6279034614562988,
2755
+ "learning_rate": 1.1825435009221395e-05,
2756
+ "loss": 0.7683,
2757
+ "step": 38100
2758
+ },
2759
+ {
2760
+ "epoch": 2.2973298051479434,
2761
+ "grad_norm": 1.5816621780395508,
2762
+ "learning_rate": 1.1725202469729774e-05,
2763
+ "loss": 0.723,
2764
+ "step": 38200
2765
+ },
2766
+ {
2767
+ "epoch": 2.3033437575174407,
2768
+ "grad_norm": 1.7849699258804321,
2769
+ "learning_rate": 1.1624969930238152e-05,
2770
+ "loss": 0.7831,
2771
+ "step": 38300
2772
+ },
2773
+ {
2774
+ "epoch": 2.3093577098869376,
2775
+ "grad_norm": 1.851671814918518,
2776
+ "learning_rate": 1.1524737390746534e-05,
2777
+ "loss": 0.7162,
2778
+ "step": 38400
2779
+ },
2780
+ {
2781
+ "epoch": 2.315371662256435,
2782
+ "grad_norm": 1.5026549100875854,
2783
+ "learning_rate": 1.1424504851254912e-05,
2784
+ "loss": 0.7803,
2785
+ "step": 38500
2786
+ },
2787
+ {
2788
+ "epoch": 2.321385614625932,
2789
+ "grad_norm": 1.6620761156082153,
2790
+ "learning_rate": 1.1324272311763291e-05,
2791
+ "loss": 0.733,
2792
+ "step": 38600
2793
+ },
2794
+ {
2795
+ "epoch": 2.3273995669954295,
2796
+ "grad_norm": 1.2507511377334595,
2797
+ "learning_rate": 1.1224039772271671e-05,
2798
+ "loss": 0.733,
2799
+ "step": 38700
2800
+ },
2801
+ {
2802
+ "epoch": 2.333413519364927,
2803
+ "grad_norm": 1.938541293144226,
2804
+ "learning_rate": 1.112380723278005e-05,
2805
+ "loss": 0.7499,
2806
+ "step": 38800
2807
+ },
2808
+ {
2809
+ "epoch": 2.3394274717344237,
2810
+ "grad_norm": 1.796823263168335,
2811
+ "learning_rate": 1.102357469328843e-05,
2812
+ "loss": 0.7399,
2813
+ "step": 38900
2814
+ },
2815
+ {
2816
+ "epoch": 2.345441424103921,
2817
+ "grad_norm": 1.834004521369934,
2818
+ "learning_rate": 1.0923342153796808e-05,
2819
+ "loss": 0.7322,
2820
+ "step": 39000
2821
+ },
2822
+ {
2823
+ "epoch": 2.3514553764734183,
2824
+ "grad_norm": 1.7822822332382202,
2825
+ "learning_rate": 1.0823109614305188e-05,
2826
+ "loss": 0.7315,
2827
+ "step": 39100
2828
+ },
2829
+ {
2830
+ "epoch": 2.3574693288429156,
2831
+ "grad_norm": 1.7883449792861938,
2832
+ "learning_rate": 1.0722877074813569e-05,
2833
+ "loss": 0.7205,
2834
+ "step": 39200
2835
+ },
2836
+ {
2837
+ "epoch": 2.363483281212413,
2838
+ "grad_norm": 1.966545581817627,
2839
+ "learning_rate": 1.0623646860716865e-05,
2840
+ "loss": 0.768,
2841
+ "step": 39300
2842
+ },
2843
+ {
2844
+ "epoch": 2.3694972335819102,
2845
+ "grad_norm": 1.722288966178894,
2846
+ "learning_rate": 1.0523414321225243e-05,
2847
+ "loss": 0.7554,
2848
+ "step": 39400
2849
+ },
2850
+ {
2851
+ "epoch": 2.375511185951407,
2852
+ "grad_norm": 1.7346769571304321,
2853
+ "learning_rate": 1.0423181781733622e-05,
2854
+ "loss": 0.7158,
2855
+ "step": 39500
2856
+ },
2857
+ {
2858
+ "epoch": 2.3815251383209044,
2859
+ "grad_norm": 1.603703498840332,
2860
+ "learning_rate": 1.0322949242242002e-05,
2861
+ "loss": 0.7467,
2862
+ "step": 39600
2863
+ },
2864
+ {
2865
+ "epoch": 2.3875390906904017,
2866
+ "grad_norm": 1.8487290143966675,
2867
+ "learning_rate": 1.0222716702750382e-05,
2868
+ "loss": 0.7492,
2869
+ "step": 39700
2870
+ },
2871
+ {
2872
+ "epoch": 2.393553043059899,
2873
+ "grad_norm": 1.7814853191375732,
2874
+ "learning_rate": 1.0122484163258761e-05,
2875
+ "loss": 0.7153,
2876
+ "step": 39800
2877
+ },
2878
+ {
2879
+ "epoch": 2.3995669954293963,
2880
+ "grad_norm": 1.745309591293335,
2881
+ "learning_rate": 1.002225162376714e-05,
2882
+ "loss": 0.7417,
2883
+ "step": 39900
2884
+ },
2885
+ {
2886
+ "epoch": 2.405580947798893,
2887
+ "grad_norm": 1.8568013906478882,
2888
+ "learning_rate": 9.92201908427552e-06,
2889
+ "loss": 0.7176,
2890
+ "step": 40000
2891
+ },
2892
+ {
2893
+ "epoch": 2.405580947798893,
2894
+ "eval_cer": 0.7346033545787555,
2895
+ "eval_loss": 0.7291049957275391,
2896
+ "eval_runtime": 5180.4578,
2897
+ "eval_samples_per_second": 3.212,
2898
+ "eval_steps_per_second": 0.402,
2899
+ "step": 40000
2900
+ },
2901
+ {
2902
+ "epoch": 2.4115949001683905,
2903
+ "grad_norm": 2.15785813331604,
2904
+ "learning_rate": 9.8217865447839e-06,
2905
+ "loss": 0.7234,
2906
+ "step": 40100
2907
+ },
2908
+ {
2909
+ "epoch": 2.417608852537888,
2910
+ "grad_norm": 1.7856571674346924,
2911
+ "learning_rate": 9.721554005292278e-06,
2912
+ "loss": 0.7176,
2913
+ "step": 40200
2914
+ },
2915
+ {
2916
+ "epoch": 2.423622804907385,
2917
+ "grad_norm": 1.682504415512085,
2918
+ "learning_rate": 9.621321465800658e-06,
2919
+ "loss": 0.7231,
2920
+ "step": 40300
2921
+ },
2922
+ {
2923
+ "epoch": 2.4296367572768824,
2924
+ "grad_norm": 1.3437491655349731,
2925
+ "learning_rate": 9.521088926309037e-06,
2926
+ "loss": 0.7232,
2927
+ "step": 40400
2928
+ },
2929
+ {
2930
+ "epoch": 2.4356507096463798,
2931
+ "grad_norm": 1.6993297338485718,
2932
+ "learning_rate": 9.420856386817417e-06,
2933
+ "loss": 0.7539,
2934
+ "step": 40500
2935
+ },
2936
+ {
2937
+ "epoch": 2.441664662015877,
2938
+ "grad_norm": 1.942014455795288,
2939
+ "learning_rate": 9.320623847325797e-06,
2940
+ "loss": 0.7519,
2941
+ "step": 40600
2942
+ },
2943
+ {
2944
+ "epoch": 2.447678614385374,
2945
+ "grad_norm": 1.7039834260940552,
2946
+ "learning_rate": 9.220391307834175e-06,
2947
+ "loss": 0.7362,
2948
+ "step": 40700
2949
+ },
2950
+ {
2951
+ "epoch": 2.4536925667548712,
2952
+ "grad_norm": 1.7124940156936646,
2953
+ "learning_rate": 9.120158768342556e-06,
2954
+ "loss": 0.7457,
2955
+ "step": 40800
2956
+ },
2957
+ {
2958
+ "epoch": 2.4597065191243686,
2959
+ "grad_norm": 1.910249948501587,
2960
+ "learning_rate": 9.019926228850934e-06,
2961
+ "loss": 0.7236,
2962
+ "step": 40900
2963
+ },
2964
+ {
2965
+ "epoch": 2.465720471493866,
2966
+ "grad_norm": 1.5514588356018066,
2967
+ "learning_rate": 8.919693689359314e-06,
2968
+ "loss": 0.721,
2969
+ "step": 41000
2970
+ },
2971
+ {
2972
+ "epoch": 2.471734423863363,
2973
+ "grad_norm": 1.744598150253296,
2974
+ "learning_rate": 8.819461149867693e-06,
2975
+ "loss": 0.7008,
2976
+ "step": 41100
2977
+ },
2978
+ {
2979
+ "epoch": 2.47774837623286,
2980
+ "grad_norm": 1.7827790975570679,
2981
+ "learning_rate": 8.719228610376073e-06,
2982
+ "loss": 0.7196,
2983
+ "step": 41200
2984
+ },
2985
+ {
2986
+ "epoch": 2.4837623286023573,
2987
+ "grad_norm": 1.7352166175842285,
2988
+ "learning_rate": 8.619998396279369e-06,
2989
+ "loss": 0.7322,
2990
+ "step": 41300
2991
+ },
2992
+ {
2993
+ "epoch": 2.4897762809718547,
2994
+ "grad_norm": 1.7378534078598022,
2995
+ "learning_rate": 8.519765856787748e-06,
2996
+ "loss": 0.7434,
2997
+ "step": 41400
2998
+ },
2999
+ {
3000
+ "epoch": 2.495790233341352,
3001
+ "grad_norm": 1.712098240852356,
3002
+ "learning_rate": 8.419533317296128e-06,
3003
+ "loss": 0.6939,
3004
+ "step": 41500
3005
+ },
3006
+ {
3007
+ "epoch": 2.5018041857108493,
3008
+ "grad_norm": 1.7368323802947998,
3009
+ "learning_rate": 8.319300777804506e-06,
3010
+ "loss": 0.726,
3011
+ "step": 41600
3012
+ },
3013
+ {
3014
+ "epoch": 2.5078181380803466,
3015
+ "grad_norm": 1.7819844484329224,
3016
+ "learning_rate": 8.219068238312887e-06,
3017
+ "loss": 0.7271,
3018
+ "step": 41700
3019
+ },
3020
+ {
3021
+ "epoch": 2.513832090449844,
3022
+ "grad_norm": 1.892824411392212,
3023
+ "learning_rate": 8.118835698821265e-06,
3024
+ "loss": 0.7146,
3025
+ "step": 41800
3026
+ },
3027
+ {
3028
+ "epoch": 2.5198460428193408,
3029
+ "grad_norm": 1.8355358839035034,
3030
+ "learning_rate": 8.018603159329645e-06,
3031
+ "loss": 0.7276,
3032
+ "step": 41900
3033
+ },
3034
+ {
3035
+ "epoch": 2.525859995188838,
3036
+ "grad_norm": 1.428154468536377,
3037
+ "learning_rate": 7.918370619838024e-06,
3038
+ "loss": 0.726,
3039
+ "step": 42000
3040
+ },
3041
+ {
3042
+ "epoch": 2.5318739475583354,
3043
+ "grad_norm": 1.8524224758148193,
3044
+ "learning_rate": 7.818138080346404e-06,
3045
+ "loss": 0.7178,
3046
+ "step": 42100
3047
+ },
3048
+ {
3049
+ "epoch": 2.5378878999278327,
3050
+ "grad_norm": 2.0265612602233887,
3051
+ "learning_rate": 7.717905540854784e-06,
3052
+ "loss": 0.714,
3053
+ "step": 42200
3054
+ },
3055
+ {
3056
+ "epoch": 2.5439018522973296,
3057
+ "grad_norm": 1.721408486366272,
3058
+ "learning_rate": 7.6176730013631625e-06,
3059
+ "loss": 0.7319,
3060
+ "step": 42300
3061
+ },
3062
+ {
3063
+ "epoch": 2.549915804666827,
3064
+ "grad_norm": 1.725827932357788,
3065
+ "learning_rate": 7.517440461871543e-06,
3066
+ "loss": 0.7663,
3067
+ "step": 42400
3068
+ },
3069
+ {
3070
+ "epoch": 2.555929757036324,
3071
+ "grad_norm": 1.6882712841033936,
3072
+ "learning_rate": 7.417207922379922e-06,
3073
+ "loss": 0.7191,
3074
+ "step": 42500
3075
+ },
3076
+ {
3077
+ "epoch": 2.5619437094058215,
3078
+ "grad_norm": 1.743870496749878,
3079
+ "learning_rate": 7.3169753828883015e-06,
3080
+ "loss": 0.7372,
3081
+ "step": 42600
3082
+ },
3083
+ {
3084
+ "epoch": 2.567957661775319,
3085
+ "grad_norm": 1.8450312614440918,
3086
+ "learning_rate": 7.21674284339668e-06,
3087
+ "loss": 0.6971,
3088
+ "step": 42700
3089
+ },
3090
+ {
3091
+ "epoch": 2.573971614144816,
3092
+ "grad_norm": 1.6685088872909546,
3093
+ "learning_rate": 7.116510303905061e-06,
3094
+ "loss": 0.7331,
3095
+ "step": 42800
3096
+ },
3097
+ {
3098
+ "epoch": 2.5799855665143134,
3099
+ "grad_norm": 1.406771183013916,
3100
+ "learning_rate": 7.01627776441344e-06,
3101
+ "loss": 0.7292,
3102
+ "step": 42900
3103
+ },
3104
+ {
3105
+ "epoch": 2.5859995188838103,
3106
+ "grad_norm": 1.750435471534729,
3107
+ "learning_rate": 6.916045224921818e-06,
3108
+ "loss": 0.7298,
3109
+ "step": 43000
3110
+ },
3111
+ {
3112
+ "epoch": 2.5920134712533076,
3113
+ "grad_norm": 1.4334131479263306,
3114
+ "learning_rate": 6.815812685430198e-06,
3115
+ "loss": 0.7332,
3116
+ "step": 43100
3117
+ },
3118
+ {
3119
+ "epoch": 2.598027423622805,
3120
+ "grad_norm": 1.4097541570663452,
3121
+ "learning_rate": 6.7155801459385786e-06,
3122
+ "loss": 0.7183,
3123
+ "step": 43200
3124
+ },
3125
+ {
3126
+ "epoch": 2.6040413759923022,
3127
+ "grad_norm": 1.5977458953857422,
3128
+ "learning_rate": 6.616349931841874e-06,
3129
+ "loss": 0.7227,
3130
+ "step": 43300
3131
+ },
3132
+ {
3133
+ "epoch": 2.610055328361799,
3134
+ "grad_norm": 1.766050100326538,
3135
+ "learning_rate": 6.516117392350253e-06,
3136
+ "loss": 0.7275,
3137
+ "step": 43400
3138
+ },
3139
+ {
3140
+ "epoch": 2.6160692807312964,
3141
+ "grad_norm": 1.6626147031784058,
3142
+ "learning_rate": 6.415884852858632e-06,
3143
+ "loss": 0.7101,
3144
+ "step": 43500
3145
+ },
3146
+ {
3147
+ "epoch": 2.6220832331007937,
3148
+ "grad_norm": 2.104124069213867,
3149
+ "learning_rate": 6.315652313367011e-06,
3150
+ "loss": 0.7267,
3151
+ "step": 43600
3152
+ },
3153
+ {
3154
+ "epoch": 2.628097185470291,
3155
+ "grad_norm": 1.66257905960083,
3156
+ "learning_rate": 6.215419773875391e-06,
3157
+ "loss": 0.6873,
3158
+ "step": 43700
3159
+ },
3160
+ {
3161
+ "epoch": 2.6341111378397883,
3162
+ "grad_norm": 2.1178150177001953,
3163
+ "learning_rate": 6.115187234383771e-06,
3164
+ "loss": 0.7533,
3165
+ "step": 43800
3166
+ },
3167
+ {
3168
+ "epoch": 2.6401250902092857,
3169
+ "grad_norm": 1.5612884759902954,
3170
+ "learning_rate": 6.0149546948921495e-06,
3171
+ "loss": 0.6981,
3172
+ "step": 43900
3173
+ },
3174
+ {
3175
+ "epoch": 2.646139042578783,
3176
+ "grad_norm": 1.9699441194534302,
3177
+ "learning_rate": 5.914722155400529e-06,
3178
+ "loss": 0.7477,
3179
+ "step": 44000
3180
+ },
3181
+ {
3182
+ "epoch": 2.646139042578783,
3183
+ "eval_cer": 0.7222449991747579,
3184
+ "eval_loss": 0.7039721608161926,
3185
+ "eval_runtime": 5169.4179,
3186
+ "eval_samples_per_second": 3.219,
3187
+ "eval_steps_per_second": 0.402,
3188
+ "step": 44000
3189
+ },
3190
+ {
3191
+ "epoch": 2.6521529949482803,
3192
+ "grad_norm": 1.6843528747558594,
3193
+ "learning_rate": 5.814489615908909e-06,
3194
+ "loss": 0.7235,
3195
+ "step": 44100
3196
+ },
3197
+ {
3198
+ "epoch": 2.658166947317777,
3199
+ "grad_norm": 1.4490200281143188,
3200
+ "learning_rate": 5.7142570764172885e-06,
3201
+ "loss": 0.7022,
3202
+ "step": 44200
3203
+ },
3204
+ {
3205
+ "epoch": 2.6641808996872745,
3206
+ "grad_norm": 1.6219418048858643,
3207
+ "learning_rate": 5.614024536925667e-06,
3208
+ "loss": 0.7215,
3209
+ "step": 44300
3210
+ },
3211
+ {
3212
+ "epoch": 2.6701948520567718,
3213
+ "grad_norm": 1.7598456144332886,
3214
+ "learning_rate": 5.513791997434047e-06,
3215
+ "loss": 0.6884,
3216
+ "step": 44400
3217
+ },
3218
+ {
3219
+ "epoch": 2.676208804426269,
3220
+ "grad_norm": 1.4096354246139526,
3221
+ "learning_rate": 5.413559457942427e-06,
3222
+ "loss": 0.7142,
3223
+ "step": 44500
3224
+ },
3225
+ {
3226
+ "epoch": 2.682222756795766,
3227
+ "grad_norm": 1.890046238899231,
3228
+ "learning_rate": 5.313326918450806e-06,
3229
+ "loss": 0.7346,
3230
+ "step": 44600
3231
+ },
3232
+ {
3233
+ "epoch": 2.6882367091652632,
3234
+ "grad_norm": 1.9413307905197144,
3235
+ "learning_rate": 5.213094378959186e-06,
3236
+ "loss": 0.688,
3237
+ "step": 44700
3238
+ },
3239
+ {
3240
+ "epoch": 2.6942506615347606,
3241
+ "grad_norm": 1.8647221326828003,
3242
+ "learning_rate": 5.112861839467565e-06,
3243
+ "loss": 0.732,
3244
+ "step": 44800
3245
+ },
3246
+ {
3247
+ "epoch": 2.700264613904258,
3248
+ "grad_norm": 1.6978052854537964,
3249
+ "learning_rate": 5.012629299975944e-06,
3250
+ "loss": 0.7075,
3251
+ "step": 44900
3252
+ },
3253
+ {
3254
+ "epoch": 2.706278566273755,
3255
+ "grad_norm": 1.3552030324935913,
3256
+ "learning_rate": 4.912396760484324e-06,
3257
+ "loss": 0.7074,
3258
+ "step": 45000
3259
+ },
3260
+ {
3261
+ "epoch": 2.7122925186432525,
3262
+ "grad_norm": 1.762635350227356,
3263
+ "learning_rate": 4.812164220992704e-06,
3264
+ "loss": 0.7219,
3265
+ "step": 45100
3266
+ },
3267
+ {
3268
+ "epoch": 2.71830647101275,
3269
+ "grad_norm": 1.7568144798278809,
3270
+ "learning_rate": 4.7119316815010824e-06,
3271
+ "loss": 0.6925,
3272
+ "step": 45200
3273
+ },
3274
+ {
3275
+ "epoch": 2.7243204233822467,
3276
+ "grad_norm": 1.5333248376846313,
3277
+ "learning_rate": 4.612701467404378e-06,
3278
+ "loss": 0.7133,
3279
+ "step": 45300
3280
+ },
3281
+ {
3282
+ "epoch": 2.730334375751744,
3283
+ "grad_norm": 1.8916515111923218,
3284
+ "learning_rate": 4.512468927912758e-06,
3285
+ "loss": 0.717,
3286
+ "step": 45400
3287
+ },
3288
+ {
3289
+ "epoch": 2.7363483281212413,
3290
+ "grad_norm": 1.8751533031463623,
3291
+ "learning_rate": 4.412236388421137e-06,
3292
+ "loss": 0.7306,
3293
+ "step": 45500
3294
+ },
3295
+ {
3296
+ "epoch": 2.7423622804907386,
3297
+ "grad_norm": 1.8414188623428345,
3298
+ "learning_rate": 4.312003848929517e-06,
3299
+ "loss": 0.7241,
3300
+ "step": 45600
3301
+ },
3302
+ {
3303
+ "epoch": 2.748376232860236,
3304
+ "grad_norm": 1.4498494863510132,
3305
+ "learning_rate": 4.211771309437896e-06,
3306
+ "loss": 0.7358,
3307
+ "step": 45700
3308
+ },
3309
+ {
3310
+ "epoch": 2.754390185229733,
3311
+ "grad_norm": 1.519124984741211,
3312
+ "learning_rate": 4.111538769946275e-06,
3313
+ "loss": 0.7059,
3314
+ "step": 45800
3315
+ },
3316
+ {
3317
+ "epoch": 2.76040413759923,
3318
+ "grad_norm": 1.9624022245407104,
3319
+ "learning_rate": 4.011306230454655e-06,
3320
+ "loss": 0.7185,
3321
+ "step": 45900
3322
+ },
3323
+ {
3324
+ "epoch": 2.7664180899687274,
3325
+ "grad_norm": 1.7385069131851196,
3326
+ "learning_rate": 3.911073690963035e-06,
3327
+ "loss": 0.7065,
3328
+ "step": 46000
3329
+ },
3330
+ {
3331
+ "epoch": 2.7724320423382247,
3332
+ "grad_norm": 1.5801081657409668,
3333
+ "learning_rate": 3.8108411514714135e-06,
3334
+ "loss": 0.7242,
3335
+ "step": 46100
3336
+ },
3337
+ {
3338
+ "epoch": 2.778445994707722,
3339
+ "grad_norm": 2.89750337600708,
3340
+ "learning_rate": 3.710608611979793e-06,
3341
+ "loss": 0.7023,
3342
+ "step": 46200
3343
+ },
3344
+ {
3345
+ "epoch": 2.7844599470772193,
3346
+ "grad_norm": 1.7838506698608398,
3347
+ "learning_rate": 3.6103760724881724e-06,
3348
+ "loss": 0.709,
3349
+ "step": 46300
3350
+ },
3351
+ {
3352
+ "epoch": 2.7904738994467166,
3353
+ "grad_norm": 1.7187174558639526,
3354
+ "learning_rate": 3.510143532996552e-06,
3355
+ "loss": 0.7292,
3356
+ "step": 46400
3357
+ },
3358
+ {
3359
+ "epoch": 2.7964878518162135,
3360
+ "grad_norm": 1.692230463027954,
3361
+ "learning_rate": 3.4099109935049313e-06,
3362
+ "loss": 0.6929,
3363
+ "step": 46500
3364
+ },
3365
+ {
3366
+ "epoch": 2.802501804185711,
3367
+ "grad_norm": 1.8588602542877197,
3368
+ "learning_rate": 3.309678454013311e-06,
3369
+ "loss": 0.7182,
3370
+ "step": 46600
3371
+ },
3372
+ {
3373
+ "epoch": 2.808515756555208,
3374
+ "grad_norm": 1.7374218702316284,
3375
+ "learning_rate": 3.20944591452169e-06,
3376
+ "loss": 0.6878,
3377
+ "step": 46700
3378
+ },
3379
+ {
3380
+ "epoch": 2.8145297089247054,
3381
+ "grad_norm": 1.6396222114562988,
3382
+ "learning_rate": 3.10921337503007e-06,
3383
+ "loss": 0.7035,
3384
+ "step": 46800
3385
+ },
3386
+ {
3387
+ "epoch": 2.8205436612942023,
3388
+ "grad_norm": 1.4402636289596558,
3389
+ "learning_rate": 3.0089808355384495e-06,
3390
+ "loss": 0.7131,
3391
+ "step": 46900
3392
+ },
3393
+ {
3394
+ "epoch": 2.8265576136636996,
3395
+ "grad_norm": 2.006782293319702,
3396
+ "learning_rate": 2.9087482960468287e-06,
3397
+ "loss": 0.7222,
3398
+ "step": 47000
3399
+ },
3400
+ {
3401
+ "epoch": 2.832571566033197,
3402
+ "grad_norm": 1.543579339981079,
3403
+ "learning_rate": 2.8085157565552084e-06,
3404
+ "loss": 0.683,
3405
+ "step": 47100
3406
+ },
3407
+ {
3408
+ "epoch": 2.8385855184026942,
3409
+ "grad_norm": 1.7819281816482544,
3410
+ "learning_rate": 2.7082832170635876e-06,
3411
+ "loss": 0.7095,
3412
+ "step": 47200
3413
+ },
3414
+ {
3415
+ "epoch": 2.8445994707721916,
3416
+ "grad_norm": 1.6801820993423462,
3417
+ "learning_rate": 2.608050677571967e-06,
3418
+ "loss": 0.7021,
3419
+ "step": 47300
3420
+ },
3421
+ {
3422
+ "epoch": 2.850613423141689,
3423
+ "grad_norm": 1.8617347478866577,
3424
+ "learning_rate": 2.508820463475263e-06,
3425
+ "loss": 0.7094,
3426
+ "step": 47400
3427
+ },
3428
+ {
3429
+ "epoch": 2.856627375511186,
3430
+ "grad_norm": 1.8291690349578857,
3431
+ "learning_rate": 2.408587923983642e-06,
3432
+ "loss": 0.7004,
3433
+ "step": 47500
3434
+ },
3435
+ {
3436
+ "epoch": 2.862641327880683,
3437
+ "grad_norm": 1.8136513233184814,
3438
+ "learning_rate": 2.3083553844920213e-06,
3439
+ "loss": 0.6936,
3440
+ "step": 47600
3441
+ },
3442
+ {
3443
+ "epoch": 2.8686552802501804,
3444
+ "grad_norm": 2.1329762935638428,
3445
+ "learning_rate": 2.208122845000401e-06,
3446
+ "loss": 0.6904,
3447
+ "step": 47700
3448
+ },
3449
+ {
3450
+ "epoch": 2.8746692326196777,
3451
+ "grad_norm": 1.9445267915725708,
3452
+ "learning_rate": 2.1078903055087806e-06,
3453
+ "loss": 0.6832,
3454
+ "step": 47800
3455
+ },
3456
+ {
3457
+ "epoch": 2.880683184989175,
3458
+ "grad_norm": 1.6533479690551758,
3459
+ "learning_rate": 2.0076577660171602e-06,
3460
+ "loss": 0.6754,
3461
+ "step": 47900
3462
+ },
3463
+ {
3464
+ "epoch": 2.8866971373586723,
3465
+ "grad_norm": 2.0631463527679443,
3466
+ "learning_rate": 1.9074252265255394e-06,
3467
+ "loss": 0.6882,
3468
+ "step": 48000
3469
+ },
3470
+ {
3471
+ "epoch": 2.8866971373586723,
3472
+ "eval_cer": 0.7184241484900502,
3473
+ "eval_loss": 0.6888573169708252,
3474
+ "eval_runtime": 5167.0943,
3475
+ "eval_samples_per_second": 3.22,
3476
+ "eval_steps_per_second": 0.403,
3477
+ "step": 48000
3478
+ }
3479
+ ],
3480
+ "logging_steps": 100,
3481
+ "max_steps": 49884,
3482
+ "num_input_tokens_seen": 0,
3483
+ "num_train_epochs": 3,
3484
+ "save_steps": 4000,
3485
+ "total_flos": 4.985333650921882e+19,
3486
+ "train_batch_size": 8,
3487
+ "trial_name": null,
3488
+ "trial_params": null
3489
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f36f6d0ec03a7732815092becf5540a6fee71c1b1c1f65dfd7432850d7dd639
3
+ size 5112
vocab.json ADDED
The diff for this file is too large to render. See raw diff