Hugging Face's logo

kiritan
/
iteboshi

Automatic Speech Recognition
Transformers
Safetensors
whisper
Generated from Trainer
Model card Files
xet
 Community
kiritan commited on
Commit
fc8c3cf
·
verified ·
1 Parent(s): fcab880

训练结束，上传最终模型

Browse files
Files changed (5) hide show
  1. README.md +72 -0
  2. all_results.json +8 -0
  3. generation_config.json +167 -0
  4. train_results.json +8 -0
  5. trainer_state.json +2932 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - wer
7
+ model-index:
8
+ - name: iteboshi
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # iteboshi
16
+
17
+ This model was trained from scratch on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 1.1515
20
+ - Wer: inf
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 2e-05
40
+ - train_batch_size: 12
41
+ - eval_batch_size: 12
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 4
44
+ - total_train_batch_size: 48
45
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 500
48
+ - training_steps: 10000
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-------:|:-----:|:---------------:|:---:|
55
+ | 1.834 | 1.6507 | 1000 | 1.9117 | inf |
56
+ | 0.9955 | 3.3006 | 2000 | 1.2766 | inf |
57
+ | 0.7584 | 4.9513 | 3000 | 1.1081 | inf |
58
+ | 0.5473 | 6.6012 | 4000 | 1.0569 | inf |
59
+ | 0.4191 | 8.2510 | 5000 | 1.0568 | inf |
60
+ | 0.3167 | 9.9017 | 6000 | 1.0609 | inf |
61
+ | 0.2322 | 11.5516 | 7000 | 1.0933 | inf |
62
+ | 0.1913 | 13.2015 | 8000 | 1.1227 | inf |
63
+ | 0.1424 | 14.8522 | 9000 | 1.1418 | inf |
64
+ | 0.1615 | 16.5021 | 10000 | 1.1515 | inf |
65
+
66
+
67
+ ### Framework versions
68
+
69
+ - Transformers 4.48.3
70
+ - Pytorch 2.6.0+cu124
71
+ - Datasets 3.5.0
72
+ - Tokenizers 0.21.1
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 16.50206440957886,
3
+ "total_flos": 1.949150849531904e+19,
4
+ "train_loss": 0.8644906110763549,
5
+ "train_runtime": 22776.5193,
6
+ "train_samples_per_second": 21.074,
7
+ "train_steps_per_second": 0.439
8
+ }
generation_config.json ADDED
@@ -0,0 +1,167 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 3,
5
+ 1
6
+ ],
7
+ [
8
+ 4,
9
+ 2
10
+ ],
11
+ [
12
+ 4,
13
+ 3
14
+ ],
15
+ [
16
+ 4,
17
+ 7
18
+ ],
19
+ [
20
+ 5,
21
+ 1
22
+ ],
23
+ [
24
+ 5,
25
+ 2
26
+ ],
27
+ [
28
+ 5,
29
+ 4
30
+ ],
31
+ [
32
+ 5,
33
+ 6
34
+ ]
35
+ ],
36
+ "begin_suppress_tokens": [
37
+ 220,
38
+ 50257
39
+ ],
40
+ "bos_token_id": 50257,
41
+ "decoder_start_token_id": 50258,
42
+ "eos_token_id": 50257,
43
+ "forced_decoder_ids": [
44
+ [
45
+ 1,
46
+ null
47
+ ],
48
+ [
49
+ 2,
50
+ 50359
51
+ ]
52
+ ],
53
+ "is_multilingual": true,
54
+ "lang_to_id": {
55
+ "<|af|>": 50327,
56
+ "<|am|>": 50334,
57
+ "<|ar|>": 50272,
58
+ "<|as|>": 50350,
59
+ "<|az|>": 50304,
60
+ "<|ba|>": 50355,
61
+ "<|be|>": 50330,
62
+ "<|bg|>": 50292,
63
+ "<|bn|>": 50302,
64
+ "<|bo|>": 50347,
65
+ "<|br|>": 50309,
66
+ "<|bs|>": 50315,
67
+ "<|ca|>": 50270,
68
+ "<|cs|>": 50283,
69
+ "<|cy|>": 50297,
70
+ "<|da|>": 50285,
71
+ "<|de|>": 50261,
72
+ "<|el|>": 50281,
73
+ "<|en|>": 50259,
74
+ "<|es|>": 50262,
75
+ "<|et|>": 50307,
76
+ "<|eu|>": 50310,
77
+ "<|fa|>": 50300,
78
+ "<|fi|>": 50277,
79
+ "<|fo|>": 50338,
80
+ "<|fr|>": 50265,
81
+ "<|gl|>": 50319,
82
+ "<|gu|>": 50333,
83
+ "<|haw|>": 50352,
84
+ "<|ha|>": 50354,
85
+ "<|he|>": 50279,
86
+ "<|hi|>": 50276,
87
+ "<|hr|>": 50291,
88
+ "<|ht|>": 50339,
89
+ "<|hu|>": 50286,
90
+ "<|hy|>": 50312,
91
+ "<|id|>": 50275,
92
+ "<|is|>": 50311,
93
+ "<|it|>": 50274,
94
+ "<|ja|>": 50266,
95
+ "<|jw|>": 50356,
96
+ "<|ka|>": 50329,
97
+ "<|kk|>": 50316,
98
+ "<|km|>": 50323,
99
+ "<|kn|>": 50306,
100
+ "<|ko|>": 50264,
101
+ "<|la|>": 50294,
102
+ "<|lb|>": 50345,
103
+ "<|ln|>": 50353,
104
+ "<|lo|>": 50336,
105
+ "<|lt|>": 50293,
106
+ "<|lv|>": 50301,
107
+ "<|mg|>": 50349,
108
+ "<|mi|>": 50295,
109
+ "<|mk|>": 50308,
110
+ "<|ml|>": 50296,
111
+ "<|mn|>": 50314,
112
+ "<|mr|>": 50320,
113
+ "<|ms|>": 50282,
114
+ "<|mt|>": 50343,
115
+ "<|my|>": 50346,
116
+ "<|ne|>": 50313,
117
+ "<|nl|>": 50271,
118
+ "<|nn|>": 50342,
119
+ "<|no|>": 50288,
120
+ "<|oc|>": 50328,
121
+ "<|pa|>": 50321,
122
+ "<|pl|>": 50269,
123
+ "<|ps|>": 50340,
124
+ "<|pt|>": 50267,
125
+ "<|ro|>": 50284,
126
+ "<|ru|>": 50263,
127
+ "<|sa|>": 50344,
128
+ "<|sd|>": 50332,
129
+ "<|si|>": 50322,
130
+ "<|sk|>": 50298,
131
+ "<|sl|>": 50305,
132
+ "<|sn|>": 50324,
133
+ "<|so|>": 50326,
134
+ "<|sq|>": 50317,
135
+ "<|sr|>": 50303,
136
+ "<|su|>": 50357,
137
+ "<|sv|>": 50273,
138
+ "<|sw|>": 50318,
139
+ "<|ta|>": 50287,
140
+ "<|te|>": 50299,
141
+ "<|tg|>": 50331,
142
+ "<|th|>": 50289,
143
+ "<|tk|>": 50341,
144
+ "<|tl|>": 50348,
145
+ "<|tr|>": 50268,
146
+ "<|tt|>": 50351,
147
+ "<|uk|>": 50280,
148
+ "<|ur|>": 50290,
149
+ "<|uz|>": 50337,
150
+ "<|vi|>": 50278,
151
+ "<|yi|>": 50335,
152
+ "<|yo|>": 50325,
153
+ "<|zh|>": 50260
154
+ },
155
+ "max_initial_timestamp_index": 50,
156
+ "max_length": 448,
157
+ "no_timestamps_token_id": 50363,
158
+ "pad_token_id": 50257,
159
+ "prev_sot_token_id": 50361,
160
+ "return_timestamps": false,
161
+ "suppress_tokens": [],
162
+ "task_to_id": {
163
+ "transcribe": 50359,
164
+ "translate": 50358
165
+ },
166
+ "transformers_version": "4.48.3"
167
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 16.50206440957886,
3
+ "total_flos": 1.949150849531904e+19,
4
+ "train_loss": 0.8644906110763549,
5
+ "train_runtime": 22776.5193,
6
+ "train_samples_per_second": 21.074,
7
+ "train_steps_per_second": 0.439
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,2932 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": Infinity,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 16.50206440957886,
5
+ "eval_steps": 1000,
6
+ "global_step": 10000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04128819157720892,
13
+ "grad_norm": 48.47208786010742,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 17.3809,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.08257638315441784,
20
+ "grad_norm": 52.7760009765625,
21
+ "learning_rate": 1.9600000000000003e-06,
22
+ "loss": 15.9672,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.12386457473162675,
27
+ "grad_norm": 64.02676391601562,
28
+ "learning_rate": 2.96e-06,
29
+ "loss": 12.6576,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.16515276630883569,
34
+ "grad_norm": 26.37363052368164,
35
+ "learning_rate": 3.96e-06,
36
+ "loss": 8.293,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.20644095788604458,
41
+ "grad_norm": 12.368265151977539,
42
+ "learning_rate": 4.960000000000001e-06,
43
+ "loss": 6.2484,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.2477291494632535,
48
+ "grad_norm": 9.140127182006836,
49
+ "learning_rate": 5.9600000000000005e-06,
50
+ "loss": 5.0845,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.28901734104046245,
55
+ "grad_norm": 8.285837173461914,
56
+ "learning_rate": 6.96e-06,
57
+ "loss": 4.5217,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.33030553261767137,
62
+ "grad_norm": 8.110857009887695,
63
+ "learning_rate": 7.960000000000002e-06,
64
+ "loss": 4.2152,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.37159372419488024,
69
+ "grad_norm": 8.587676048278809,
70
+ "learning_rate": 8.96e-06,
71
+ "loss": 4.095,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.41288191577208916,
76
+ "grad_norm": 8.696348190307617,
77
+ "learning_rate": 9.960000000000001e-06,
78
+ "loss": 3.9265,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.4541701073492981,
83
+ "grad_norm": 8.708649635314941,
84
+ "learning_rate": 1.0960000000000002e-05,
85
+ "loss": 3.7976,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.495458298926507,
90
+ "grad_norm": 8.679261207580566,
91
+ "learning_rate": 1.196e-05,
92
+ "loss": 3.7627,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.5367464905037159,
97
+ "grad_norm": 9.209630012512207,
98
+ "learning_rate": 1.2960000000000001e-05,
99
+ "loss": 3.6967,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.5780346820809249,
104
+ "grad_norm": 8.386543273925781,
105
+ "learning_rate": 1.396e-05,
106
+ "loss": 3.6216,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.6193228736581338,
111
+ "grad_norm": 8.898401260375977,
112
+ "learning_rate": 1.496e-05,
113
+ "loss": 3.5456,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.6606110652353427,
118
+ "grad_norm": 7.772883415222168,
119
+ "learning_rate": 1.5960000000000003e-05,
120
+ "loss": 3.4954,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.7018992568125516,
125
+ "grad_norm": 8.508569717407227,
126
+ "learning_rate": 1.696e-05,
127
+ "loss": 3.4144,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.7431874483897605,
132
+ "grad_norm": 7.907943248748779,
133
+ "learning_rate": 1.796e-05,
134
+ "loss": 3.3481,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.7844756399669695,
139
+ "grad_norm": 8.553926467895508,
140
+ "learning_rate": 1.896e-05,
141
+ "loss": 3.3021,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.8257638315441783,
146
+ "grad_norm": 9.033528327941895,
147
+ "learning_rate": 1.9960000000000002e-05,
148
+ "loss": 3.2553,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.8670520231213873,
153
+ "grad_norm": 8.92698860168457,
154
+ "learning_rate": 1.9949473684210527e-05,
155
+ "loss": 3.1524,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.9083402146985962,
160
+ "grad_norm": 9.119793891906738,
161
+ "learning_rate": 1.989684210526316e-05,
162
+ "loss": 3.034,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.9496284062758051,
167
+ "grad_norm": 10.161808967590332,
168
+ "learning_rate": 1.984421052631579e-05,
169
+ "loss": 2.9658,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.990916597853014,
174
+ "grad_norm": 9.213955879211426,
175
+ "learning_rate": 1.9791578947368423e-05,
176
+ "loss": 2.8887,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 1.0313790255986788,
181
+ "grad_norm": 8.941823959350586,
182
+ "learning_rate": 1.9738947368421057e-05,
183
+ "loss": 2.719,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 1.0726672171758878,
188
+ "grad_norm": 8.994600296020508,
189
+ "learning_rate": 1.9686315789473687e-05,
190
+ "loss": 2.6422,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 1.1139554087530965,
195
+ "grad_norm": 10.292838096618652,
196
+ "learning_rate": 1.9633684210526316e-05,
197
+ "loss": 2.5674,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 1.1552436003303055,
202
+ "grad_norm": 10.515816688537598,
203
+ "learning_rate": 1.958105263157895e-05,
204
+ "loss": 2.5031,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 1.1965317919075145,
209
+ "grad_norm": 9.79793643951416,
210
+ "learning_rate": 1.952842105263158e-05,
211
+ "loss": 2.3914,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 1.2378199834847234,
216
+ "grad_norm": 9.563103675842285,
217
+ "learning_rate": 1.9475789473684213e-05,
218
+ "loss": 2.3986,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 1.2791081750619324,
223
+ "grad_norm": 9.516156196594238,
224
+ "learning_rate": 1.9423157894736843e-05,
225
+ "loss": 2.2832,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 1.3203963666391412,
230
+ "grad_norm": 9.650139808654785,
231
+ "learning_rate": 1.9370526315789476e-05,
232
+ "loss": 2.252,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 1.3616845582163501,
237
+ "grad_norm": 9.165299415588379,
238
+ "learning_rate": 1.9317894736842106e-05,
239
+ "loss": 2.1517,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 1.402972749793559,
244
+ "grad_norm": 9.827438354492188,
245
+ "learning_rate": 1.926526315789474e-05,
246
+ "loss": 2.1643,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 1.4442609413707679,
251
+ "grad_norm": 9.822501182556152,
252
+ "learning_rate": 1.921263157894737e-05,
253
+ "loss": 2.0519,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 1.4855491329479769,
258
+ "grad_norm": 10.055537223815918,
259
+ "learning_rate": 1.916e-05,
260
+ "loss": 2.0872,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 1.5268373245251858,
265
+ "grad_norm": 9.642505645751953,
266
+ "learning_rate": 1.9107368421052633e-05,
267
+ "loss": 1.999,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 1.5681255161023948,
272
+ "grad_norm": 9.428027153015137,
273
+ "learning_rate": 1.9054736842105263e-05,
274
+ "loss": 1.954,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 1.6094137076796038,
279
+ "grad_norm": 8.909976959228516,
280
+ "learning_rate": 1.9002105263157896e-05,
281
+ "loss": 1.9201,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 1.6507018992568125,
286
+ "grad_norm": 9.313642501831055,
287
+ "learning_rate": 1.894947368421053e-05,
288
+ "loss": 1.834,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 1.6507018992568125,
293
+ "eval_loss": 1.9117295742034912,
294
+ "eval_runtime": 784.2462,
295
+ "eval_samples_per_second": 13.492,
296
+ "eval_steps_per_second": 1.125,
297
+ "eval_wer": Infinity,
298
+ "step": 1000
299
+ },
300
+ {
301
+ "epoch": 1.6919900908340215,
302
+ "grad_norm": 9.633246421813965,
303
+ "learning_rate": 1.889684210526316e-05,
304
+ "loss": 1.8354,
305
+ "step": 1025
306
+ },
307
+ {
308
+ "epoch": 1.7332782824112303,
309
+ "grad_norm": 9.896079063415527,
310
+ "learning_rate": 1.8844210526315793e-05,
311
+ "loss": 1.7921,
312
+ "step": 1050
313
+ },
314
+ {
315
+ "epoch": 1.7745664739884393,
316
+ "grad_norm": 9.53538703918457,
317
+ "learning_rate": 1.8791578947368423e-05,
318
+ "loss": 1.7655,
319
+ "step": 1075
320
+ },
321
+ {
322
+ "epoch": 1.8158546655656482,
323
+ "grad_norm": 9.284989356994629,
324
+ "learning_rate": 1.8738947368421056e-05,
325
+ "loss": 1.7743,
326
+ "step": 1100
327
+ },
328
+ {
329
+ "epoch": 1.8571428571428572,
330
+ "grad_norm": 10.101766586303711,
331
+ "learning_rate": 1.8686315789473686e-05,
332
+ "loss": 1.7346,
333
+ "step": 1125
334
+ },
335
+ {
336
+ "epoch": 1.8984310487200662,
337
+ "grad_norm": 8.92200756072998,
338
+ "learning_rate": 1.8633684210526316e-05,
339
+ "loss": 1.6885,
340
+ "step": 1150
341
+ },
342
+ {
343
+ "epoch": 1.939719240297275,
344
+ "grad_norm": 9.907315254211426,
345
+ "learning_rate": 1.858105263157895e-05,
346
+ "loss": 1.6912,
347
+ "step": 1175
348
+ },
349
+ {
350
+ "epoch": 1.981007431874484,
351
+ "grad_norm": 9.401750564575195,
352
+ "learning_rate": 1.852842105263158e-05,
353
+ "loss": 1.6279,
354
+ "step": 1200
355
+ },
356
+ {
357
+ "epoch": 2.0214698596201486,
358
+ "grad_norm": 10.461494445800781,
359
+ "learning_rate": 1.8475789473684212e-05,
360
+ "loss": 1.5268,
361
+ "step": 1225
362
+ },
363
+ {
364
+ "epoch": 2.0627580511973576,
365
+ "grad_norm": 8.997374534606934,
366
+ "learning_rate": 1.8423157894736842e-05,
367
+ "loss": 1.4142,
368
+ "step": 1250
369
+ },
370
+ {
371
+ "epoch": 2.1040462427745665,
372
+ "grad_norm": 9.4984712600708,
373
+ "learning_rate": 1.8370526315789476e-05,
374
+ "loss": 1.4769,
375
+ "step": 1275
376
+ },
377
+ {
378
+ "epoch": 2.1453344343517755,
379
+ "grad_norm": 9.436307907104492,
380
+ "learning_rate": 1.831789473684211e-05,
381
+ "loss": 1.4284,
382
+ "step": 1300
383
+ },
384
+ {
385
+ "epoch": 2.1866226259289845,
386
+ "grad_norm": 9.324549674987793,
387
+ "learning_rate": 1.826526315789474e-05,
388
+ "loss": 1.4117,
389
+ "step": 1325
390
+ },
391
+ {
392
+ "epoch": 2.227910817506193,
393
+ "grad_norm": 9.608540534973145,
394
+ "learning_rate": 1.821263157894737e-05,
395
+ "loss": 1.4093,
396
+ "step": 1350
397
+ },
398
+ {
399
+ "epoch": 2.269199009083402,
400
+ "grad_norm": 8.661190032958984,
401
+ "learning_rate": 1.8160000000000002e-05,
402
+ "loss": 1.3663,
403
+ "step": 1375
404
+ },
405
+ {
406
+ "epoch": 2.310487200660611,
407
+ "grad_norm": 9.19803524017334,
408
+ "learning_rate": 1.8107368421052632e-05,
409
+ "loss": 1.3647,
410
+ "step": 1400
411
+ },
412
+ {
413
+ "epoch": 2.35177539223782,
414
+ "grad_norm": 9.229018211364746,
415
+ "learning_rate": 1.8054736842105266e-05,
416
+ "loss": 1.3282,
417
+ "step": 1425
418
+ },
419
+ {
420
+ "epoch": 2.393063583815029,
421
+ "grad_norm": 9.667059898376465,
422
+ "learning_rate": 1.8002105263157896e-05,
423
+ "loss": 1.3098,
424
+ "step": 1450
425
+ },
426
+ {
427
+ "epoch": 2.434351775392238,
428
+ "grad_norm": 8.727728843688965,
429
+ "learning_rate": 1.794947368421053e-05,
430
+ "loss": 1.3246,
431
+ "step": 1475
432
+ },
433
+ {
434
+ "epoch": 2.475639966969447,
435
+ "grad_norm": 9.386205673217773,
436
+ "learning_rate": 1.789684210526316e-05,
437
+ "loss": 1.2703,
438
+ "step": 1500
439
+ },
440
+ {
441
+ "epoch": 2.516928158546656,
442
+ "grad_norm": 9.435502052307129,
443
+ "learning_rate": 1.7844210526315792e-05,
444
+ "loss": 1.3239,
445
+ "step": 1525
446
+ },
447
+ {
448
+ "epoch": 2.558216350123865,
449
+ "grad_norm": 8.652278900146484,
450
+ "learning_rate": 1.7791578947368422e-05,
451
+ "loss": 1.2833,
452
+ "step": 1550
453
+ },
454
+ {
455
+ "epoch": 2.5995045417010734,
456
+ "grad_norm": 8.806912422180176,
457
+ "learning_rate": 1.7738947368421052e-05,
458
+ "loss": 1.2917,
459
+ "step": 1575
460
+ },
461
+ {
462
+ "epoch": 2.6407927332782823,
463
+ "grad_norm": 9.386215209960938,
464
+ "learning_rate": 1.7686315789473685e-05,
465
+ "loss": 1.2775,
466
+ "step": 1600
467
+ },
468
+ {
469
+ "epoch": 2.6820809248554913,
470
+ "grad_norm": 8.931626319885254,
471
+ "learning_rate": 1.7633684210526315e-05,
472
+ "loss": 1.2501,
473
+ "step": 1625
474
+ },
475
+ {
476
+ "epoch": 2.7233691164327003,
477
+ "grad_norm": 8.861407279968262,
478
+ "learning_rate": 1.758105263157895e-05,
479
+ "loss": 1.2031,
480
+ "step": 1650
481
+ },
482
+ {
483
+ "epoch": 2.7646573080099093,
484
+ "grad_norm": 8.485048294067383,
485
+ "learning_rate": 1.7528421052631582e-05,
486
+ "loss": 1.2482,
487
+ "step": 1675
488
+ },
489
+ {
490
+ "epoch": 2.805945499587118,
491
+ "grad_norm": 9.203420639038086,
492
+ "learning_rate": 1.7475789473684212e-05,
493
+ "loss": 1.2105,
494
+ "step": 1700
495
+ },
496
+ {
497
+ "epoch": 2.847233691164327,
498
+ "grad_norm": 9.27082633972168,
499
+ "learning_rate": 1.7423157894736845e-05,
500
+ "loss": 1.2441,
501
+ "step": 1725
502
+ },
503
+ {
504
+ "epoch": 2.8885218827415358,
505
+ "grad_norm": 9.1829833984375,
506
+ "learning_rate": 1.7370526315789475e-05,
507
+ "loss": 1.2006,
508
+ "step": 1750
509
+ },
510
+ {
511
+ "epoch": 2.9298100743187447,
512
+ "grad_norm": 8.873353958129883,
513
+ "learning_rate": 1.731789473684211e-05,
514
+ "loss": 1.167,
515
+ "step": 1775
516
+ },
517
+ {
518
+ "epoch": 2.9710982658959537,
519
+ "grad_norm": 8.414481163024902,
520
+ "learning_rate": 1.726526315789474e-05,
521
+ "loss": 1.162,
522
+ "step": 1800
523
+ },
524
+ {
525
+ "epoch": 3.0115606936416186,
526
+ "grad_norm": 8.181883811950684,
527
+ "learning_rate": 1.721263157894737e-05,
528
+ "loss": 1.1539,
529
+ "step": 1825
530
+ },
531
+ {
532
+ "epoch": 3.0528488852188276,
533
+ "grad_norm": 8.511463165283203,
534
+ "learning_rate": 1.7160000000000002e-05,
535
+ "loss": 1.0424,
536
+ "step": 1850
537
+ },
538
+ {
539
+ "epoch": 3.094137076796036,
540
+ "grad_norm": 8.532715797424316,
541
+ "learning_rate": 1.710736842105263e-05,
542
+ "loss": 1.0293,
543
+ "step": 1875
544
+ },
545
+ {
546
+ "epoch": 3.135425268373245,
547
+ "grad_norm": 7.531360149383545,
548
+ "learning_rate": 1.7054736842105265e-05,
549
+ "loss": 0.997,
550
+ "step": 1900
551
+ },
552
+ {
553
+ "epoch": 3.176713459950454,
554
+ "grad_norm": 8.686123847961426,
555
+ "learning_rate": 1.7002105263157895e-05,
556
+ "loss": 1.0264,
557
+ "step": 1925
558
+ },
559
+ {
560
+ "epoch": 3.218001651527663,
561
+ "grad_norm": 8.670825958251953,
562
+ "learning_rate": 1.6949473684210528e-05,
563
+ "loss": 1.0235,
564
+ "step": 1950
565
+ },
566
+ {
567
+ "epoch": 3.259289843104872,
568
+ "grad_norm": 9.212960243225098,
569
+ "learning_rate": 1.689684210526316e-05,
570
+ "loss": 1.0109,
571
+ "step": 1975
572
+ },
573
+ {
574
+ "epoch": 3.300578034682081,
575
+ "grad_norm": 7.840266704559326,
576
+ "learning_rate": 1.684421052631579e-05,
577
+ "loss": 0.9955,
578
+ "step": 2000
579
+ },
580
+ {
581
+ "epoch": 3.300578034682081,
582
+ "eval_loss": 1.2765947580337524,
583
+ "eval_runtime": 710.8083,
584
+ "eval_samples_per_second": 14.886,
585
+ "eval_steps_per_second": 1.241,
586
+ "eval_wer": Infinity,
587
+ "step": 2000
588
+ },
589
+ {
590
+ "epoch": 3.34186622625929,
591
+ "grad_norm": 7.762492656707764,
592
+ "learning_rate": 1.6791578947368425e-05,
593
+ "loss": 0.9837,
594
+ "step": 2025
595
+ },
596
+ {
597
+ "epoch": 3.383154417836499,
598
+ "grad_norm": 8.323661804199219,
599
+ "learning_rate": 1.6738947368421055e-05,
600
+ "loss": 1.0037,
601
+ "step": 2050
602
+ },
603
+ {
604
+ "epoch": 3.424442609413708,
605
+ "grad_norm": 9.538601875305176,
606
+ "learning_rate": 1.6686315789473685e-05,
607
+ "loss": 1.0539,
608
+ "step": 2075
609
+ },
610
+ {
611
+ "epoch": 3.4657308009909165,
612
+ "grad_norm": 7.49852180480957,
613
+ "learning_rate": 1.6633684210526318e-05,
614
+ "loss": 0.9676,
615
+ "step": 2100
616
+ },
617
+ {
618
+ "epoch": 3.5070189925681254,
619
+ "grad_norm": 8.388888359069824,
620
+ "learning_rate": 1.6581052631578948e-05,
621
+ "loss": 1.0012,
622
+ "step": 2125
623
+ },
624
+ {
625
+ "epoch": 3.5483071841453344,
626
+ "grad_norm": 8.11563491821289,
627
+ "learning_rate": 1.652842105263158e-05,
628
+ "loss": 1.0151,
629
+ "step": 2150
630
+ },
631
+ {
632
+ "epoch": 3.5895953757225434,
633
+ "grad_norm": 8.094681739807129,
634
+ "learning_rate": 1.647578947368421e-05,
635
+ "loss": 0.9746,
636
+ "step": 2175
637
+ },
638
+ {
639
+ "epoch": 3.6308835672997524,
640
+ "grad_norm": 7.915094375610352,
641
+ "learning_rate": 1.6423157894736845e-05,
642
+ "loss": 1.0166,
643
+ "step": 2200
644
+ },
645
+ {
646
+ "epoch": 3.6721717588769613,
647
+ "grad_norm": 7.892699718475342,
648
+ "learning_rate": 1.6370526315789475e-05,
649
+ "loss": 0.9443,
650
+ "step": 2225
651
+ },
652
+ {
653
+ "epoch": 3.71345995045417,
654
+ "grad_norm": 8.693673133850098,
655
+ "learning_rate": 1.6317894736842105e-05,
656
+ "loss": 0.9299,
657
+ "step": 2250
658
+ },
659
+ {
660
+ "epoch": 3.754748142031379,
661
+ "grad_norm": 7.691691875457764,
662
+ "learning_rate": 1.6265263157894738e-05,
663
+ "loss": 0.9333,
664
+ "step": 2275
665
+ },
666
+ {
667
+ "epoch": 3.796036333608588,
668
+ "grad_norm": 8.604223251342773,
669
+ "learning_rate": 1.6212631578947368e-05,
670
+ "loss": 0.9782,
671
+ "step": 2300
672
+ },
673
+ {
674
+ "epoch": 3.837324525185797,
675
+ "grad_norm": 8.439273834228516,
676
+ "learning_rate": 1.616e-05,
677
+ "loss": 0.9229,
678
+ "step": 2325
679
+ },
680
+ {
681
+ "epoch": 3.878612716763006,
682
+ "grad_norm": 8.69174575805664,
683
+ "learning_rate": 1.6107368421052634e-05,
684
+ "loss": 0.9848,
685
+ "step": 2350
686
+ },
687
+ {
688
+ "epoch": 3.9199009083402148,
689
+ "grad_norm": 8.314153671264648,
690
+ "learning_rate": 1.6054736842105264e-05,
691
+ "loss": 0.9285,
692
+ "step": 2375
693
+ },
694
+ {
695
+ "epoch": 3.9611890999174237,
696
+ "grad_norm": 8.562870979309082,
697
+ "learning_rate": 1.6002105263157898e-05,
698
+ "loss": 0.9432,
699
+ "step": 2400
700
+ },
701
+ {
702
+ "epoch": 4.001651527663088,
703
+ "grad_norm": 7.700297832489014,
704
+ "learning_rate": 1.5949473684210528e-05,
705
+ "loss": 0.9091,
706
+ "step": 2425
707
+ },
708
+ {
709
+ "epoch": 4.042939719240297,
710
+ "grad_norm": 8.673540115356445,
711
+ "learning_rate": 1.589684210526316e-05,
712
+ "loss": 0.8088,
713
+ "step": 2450
714
+ },
715
+ {
716
+ "epoch": 4.084227910817506,
717
+ "grad_norm": 7.37542724609375,
718
+ "learning_rate": 1.584421052631579e-05,
719
+ "loss": 0.7986,
720
+ "step": 2475
721
+ },
722
+ {
723
+ "epoch": 4.125516102394715,
724
+ "grad_norm": 6.779449462890625,
725
+ "learning_rate": 1.579157894736842e-05,
726
+ "loss": 0.8114,
727
+ "step": 2500
728
+ },
729
+ {
730
+ "epoch": 4.166804293971924,
731
+ "grad_norm": 7.313191890716553,
732
+ "learning_rate": 1.5738947368421054e-05,
733
+ "loss": 0.8089,
734
+ "step": 2525
735
+ },
736
+ {
737
+ "epoch": 4.208092485549133,
738
+ "grad_norm": 6.947272777557373,
739
+ "learning_rate": 1.5686315789473684e-05,
740
+ "loss": 0.778,
741
+ "step": 2550
742
+ },
743
+ {
744
+ "epoch": 4.249380677126342,
745
+ "grad_norm": 8.322178840637207,
746
+ "learning_rate": 1.5633684210526318e-05,
747
+ "loss": 0.7897,
748
+ "step": 2575
749
+ },
750
+ {
751
+ "epoch": 4.290668868703551,
752
+ "grad_norm": 8.319843292236328,
753
+ "learning_rate": 1.5581052631578947e-05,
754
+ "loss": 0.8583,
755
+ "step": 2600
756
+ },
757
+ {
758
+ "epoch": 4.33195706028076,
759
+ "grad_norm": 7.782474994659424,
760
+ "learning_rate": 1.552842105263158e-05,
761
+ "loss": 0.8077,
762
+ "step": 2625
763
+ },
764
+ {
765
+ "epoch": 4.373245251857969,
766
+ "grad_norm": 8.119930267333984,
767
+ "learning_rate": 1.547578947368421e-05,
768
+ "loss": 0.807,
769
+ "step": 2650
770
+ },
771
+ {
772
+ "epoch": 4.414533443435178,
773
+ "grad_norm": 10.038588523864746,
774
+ "learning_rate": 1.5423157894736844e-05,
775
+ "loss": 0.8248,
776
+ "step": 2675
777
+ },
778
+ {
779
+ "epoch": 4.455821635012386,
780
+ "grad_norm": 7.1363115310668945,
781
+ "learning_rate": 1.5370526315789477e-05,
782
+ "loss": 0.8092,
783
+ "step": 2700
784
+ },
785
+ {
786
+ "epoch": 4.497109826589595,
787
+ "grad_norm": 8.094152450561523,
788
+ "learning_rate": 1.5317894736842107e-05,
789
+ "loss": 0.8386,
790
+ "step": 2725
791
+ },
792
+ {
793
+ "epoch": 4.538398018166804,
794
+ "grad_norm": 8.467181205749512,
795
+ "learning_rate": 1.5265263157894737e-05,
796
+ "loss": 0.802,
797
+ "step": 2750
798
+ },
799
+ {
800
+ "epoch": 4.579686209744013,
801
+ "grad_norm": 9.26181697845459,
802
+ "learning_rate": 1.521263157894737e-05,
803
+ "loss": 0.8057,
804
+ "step": 2775
805
+ },
806
+ {
807
+ "epoch": 4.620974401321222,
808
+ "grad_norm": 8.754571914672852,
809
+ "learning_rate": 1.516e-05,
810
+ "loss": 0.7905,
811
+ "step": 2800
812
+ },
813
+ {
814
+ "epoch": 4.662262592898431,
815
+ "grad_norm": 7.060452461242676,
816
+ "learning_rate": 1.5107368421052632e-05,
817
+ "loss": 0.7644,
818
+ "step": 2825
819
+ },
820
+ {
821
+ "epoch": 4.70355078447564,
822
+ "grad_norm": 8.613765716552734,
823
+ "learning_rate": 1.5054736842105264e-05,
824
+ "loss": 0.7676,
825
+ "step": 2850
826
+ },
827
+ {
828
+ "epoch": 4.744838976052849,
829
+ "grad_norm": 7.17141056060791,
830
+ "learning_rate": 1.5002105263157895e-05,
831
+ "loss": 0.7648,
832
+ "step": 2875
833
+ },
834
+ {
835
+ "epoch": 4.786127167630058,
836
+ "grad_norm": 8.803303718566895,
837
+ "learning_rate": 1.4949473684210527e-05,
838
+ "loss": 0.8048,
839
+ "step": 2900
840
+ },
841
+ {
842
+ "epoch": 4.827415359207267,
843
+ "grad_norm": 7.4490203857421875,
844
+ "learning_rate": 1.4896842105263159e-05,
845
+ "loss": 0.78,
846
+ "step": 2925
847
+ },
848
+ {
849
+ "epoch": 4.868703550784476,
850
+ "grad_norm": 8.119257926940918,
851
+ "learning_rate": 1.484421052631579e-05,
852
+ "loss": 0.7647,
853
+ "step": 2950
854
+ },
855
+ {
856
+ "epoch": 4.909991742361685,
857
+ "grad_norm": 7.520335674285889,
858
+ "learning_rate": 1.4791578947368422e-05,
859
+ "loss": 0.772,
860
+ "step": 2975
861
+ },
862
+ {
863
+ "epoch": 4.951279933938894,
864
+ "grad_norm": 7.8852219581604,
865
+ "learning_rate": 1.4738947368421055e-05,
866
+ "loss": 0.7584,
867
+ "step": 3000
868
+ },
869
+ {
870
+ "epoch": 4.951279933938894,
871
+ "eval_loss": 1.1080505847930908,
872
+ "eval_runtime": 699.2603,
873
+ "eval_samples_per_second": 15.132,
874
+ "eval_steps_per_second": 1.261,
875
+ "eval_wer": Infinity,
876
+ "step": 3000
877
+ },
878
+ {
879
+ "epoch": 4.992568125516103,
880
+ "grad_norm": 8.253033638000488,
881
+ "learning_rate": 1.4686315789473687e-05,
882
+ "loss": 0.7497,
883
+ "step": 3025
884
+ },
885
+ {
886
+ "epoch": 5.033030553261767,
887
+ "grad_norm": 7.215710163116455,
888
+ "learning_rate": 1.4633684210526317e-05,
889
+ "loss": 0.6874,
890
+ "step": 3050
891
+ },
892
+ {
893
+ "epoch": 5.074318744838976,
894
+ "grad_norm": 7.491812229156494,
895
+ "learning_rate": 1.4581052631578949e-05,
896
+ "loss": 0.6607,
897
+ "step": 3075
898
+ },
899
+ {
900
+ "epoch": 5.115606936416185,
901
+ "grad_norm": 7.644138813018799,
902
+ "learning_rate": 1.452842105263158e-05,
903
+ "loss": 0.6834,
904
+ "step": 3100
905
+ },
906
+ {
907
+ "epoch": 5.156895127993394,
908
+ "grad_norm": 7.1401686668396,
909
+ "learning_rate": 1.4475789473684212e-05,
910
+ "loss": 0.6667,
911
+ "step": 3125
912
+ },
913
+ {
914
+ "epoch": 5.198183319570603,
915
+ "grad_norm": 7.159341335296631,
916
+ "learning_rate": 1.4423157894736843e-05,
917
+ "loss": 0.6706,
918
+ "step": 3150
919
+ },
920
+ {
921
+ "epoch": 5.239471511147812,
922
+ "grad_norm": 6.769131660461426,
923
+ "learning_rate": 1.4370526315789475e-05,
924
+ "loss": 0.6508,
925
+ "step": 3175
926
+ },
927
+ {
928
+ "epoch": 5.280759702725021,
929
+ "grad_norm": 7.549663543701172,
930
+ "learning_rate": 1.4317894736842107e-05,
931
+ "loss": 0.6674,
932
+ "step": 3200
933
+ },
934
+ {
935
+ "epoch": 5.32204789430223,
936
+ "grad_norm": 7.3631720542907715,
937
+ "learning_rate": 1.4265263157894738e-05,
938
+ "loss": 0.6401,
939
+ "step": 3225
940
+ },
941
+ {
942
+ "epoch": 5.363336085879438,
943
+ "grad_norm": 7.147627353668213,
944
+ "learning_rate": 1.4212631578947368e-05,
945
+ "loss": 0.6805,
946
+ "step": 3250
947
+ },
948
+ {
949
+ "epoch": 5.404624277456647,
950
+ "grad_norm": 7.807270050048828,
951
+ "learning_rate": 1.416e-05,
952
+ "loss": 0.6333,
953
+ "step": 3275
954
+ },
955
+ {
956
+ "epoch": 5.445912469033856,
957
+ "grad_norm": 7.820711612701416,
958
+ "learning_rate": 1.4107368421052632e-05,
959
+ "loss": 0.7001,
960
+ "step": 3300
961
+ },
962
+ {
963
+ "epoch": 5.487200660611065,
964
+ "grad_norm": 8.526030540466309,
965
+ "learning_rate": 1.4054736842105263e-05,
966
+ "loss": 0.6728,
967
+ "step": 3325
968
+ },
969
+ {
970
+ "epoch": 5.528488852188274,
971
+ "grad_norm": 6.857508659362793,
972
+ "learning_rate": 1.4002105263157897e-05,
973
+ "loss": 0.6425,
974
+ "step": 3350
975
+ },
976
+ {
977
+ "epoch": 5.569777043765483,
978
+ "grad_norm": 6.723370552062988,
979
+ "learning_rate": 1.3949473684210528e-05,
980
+ "loss": 0.6616,
981
+ "step": 3375
982
+ },
983
+ {
984
+ "epoch": 5.611065235342692,
985
+ "grad_norm": 7.373733043670654,
986
+ "learning_rate": 1.389684210526316e-05,
987
+ "loss": 0.6609,
988
+ "step": 3400
989
+ },
990
+ {
991
+ "epoch": 5.652353426919901,
992
+ "grad_norm": 7.192294120788574,
993
+ "learning_rate": 1.3844210526315791e-05,
994
+ "loss": 0.6413,
995
+ "step": 3425
996
+ },
997
+ {
998
+ "epoch": 5.69364161849711,
999
+ "grad_norm": 6.876883029937744,
1000
+ "learning_rate": 1.3791578947368423e-05,
1001
+ "loss": 0.648,
1002
+ "step": 3450
1003
+ },
1004
+ {
1005
+ "epoch": 5.734929810074319,
1006
+ "grad_norm": 7.807666301727295,
1007
+ "learning_rate": 1.3738947368421055e-05,
1008
+ "loss": 0.6643,
1009
+ "step": 3475
1010
+ },
1011
+ {
1012
+ "epoch": 5.776218001651528,
1013
+ "grad_norm": 7.850846767425537,
1014
+ "learning_rate": 1.3686315789473685e-05,
1015
+ "loss": 0.6469,
1016
+ "step": 3500
1017
+ },
1018
+ {
1019
+ "epoch": 5.817506193228737,
1020
+ "grad_norm": 7.765756607055664,
1021
+ "learning_rate": 1.3633684210526316e-05,
1022
+ "loss": 0.6642,
1023
+ "step": 3525
1024
+ },
1025
+ {
1026
+ "epoch": 5.858794384805946,
1027
+ "grad_norm": 7.8592915534973145,
1028
+ "learning_rate": 1.3581052631578948e-05,
1029
+ "loss": 0.6531,
1030
+ "step": 3550
1031
+ },
1032
+ {
1033
+ "epoch": 5.900082576383154,
1034
+ "grad_norm": 7.569016456604004,
1035
+ "learning_rate": 1.352842105263158e-05,
1036
+ "loss": 0.646,
1037
+ "step": 3575
1038
+ },
1039
+ {
1040
+ "epoch": 5.941370767960363,
1041
+ "grad_norm": 6.537998199462891,
1042
+ "learning_rate": 1.3475789473684211e-05,
1043
+ "loss": 0.609,
1044
+ "step": 3600
1045
+ },
1046
+ {
1047
+ "epoch": 5.982658959537572,
1048
+ "grad_norm": 7.915237903594971,
1049
+ "learning_rate": 1.3423157894736843e-05,
1050
+ "loss": 0.6304,
1051
+ "step": 3625
1052
+ },
1053
+ {
1054
+ "epoch": 6.023121387283237,
1055
+ "grad_norm": 7.4267683029174805,
1056
+ "learning_rate": 1.3370526315789475e-05,
1057
+ "loss": 0.5556,
1058
+ "step": 3650
1059
+ },
1060
+ {
1061
+ "epoch": 6.064409578860446,
1062
+ "grad_norm": 6.749971389770508,
1063
+ "learning_rate": 1.3317894736842108e-05,
1064
+ "loss": 0.5266,
1065
+ "step": 3675
1066
+ },
1067
+ {
1068
+ "epoch": 6.105697770437655,
1069
+ "grad_norm": 6.487602233886719,
1070
+ "learning_rate": 1.326526315789474e-05,
1071
+ "loss": 0.5717,
1072
+ "step": 3700
1073
+ },
1074
+ {
1075
+ "epoch": 6.146985962014864,
1076
+ "grad_norm": 6.367309093475342,
1077
+ "learning_rate": 1.321263157894737e-05,
1078
+ "loss": 0.5661,
1079
+ "step": 3725
1080
+ },
1081
+ {
1082
+ "epoch": 6.188274153592072,
1083
+ "grad_norm": 6.503593444824219,
1084
+ "learning_rate": 1.3160000000000001e-05,
1085
+ "loss": 0.552,
1086
+ "step": 3750
1087
+ },
1088
+ {
1089
+ "epoch": 6.229562345169281,
1090
+ "grad_norm": 6.83872127532959,
1091
+ "learning_rate": 1.3107368421052633e-05,
1092
+ "loss": 0.57,
1093
+ "step": 3775
1094
+ },
1095
+ {
1096
+ "epoch": 6.27085053674649,
1097
+ "grad_norm": 6.761080741882324,
1098
+ "learning_rate": 1.3054736842105264e-05,
1099
+ "loss": 0.5332,
1100
+ "step": 3800
1101
+ },
1102
+ {
1103
+ "epoch": 6.312138728323699,
1104
+ "grad_norm": 7.023492336273193,
1105
+ "learning_rate": 1.3002105263157896e-05,
1106
+ "loss": 0.5588,
1107
+ "step": 3825
1108
+ },
1109
+ {
1110
+ "epoch": 6.353426919900908,
1111
+ "grad_norm": 7.079690933227539,
1112
+ "learning_rate": 1.2949473684210528e-05,
1113
+ "loss": 0.5471,
1114
+ "step": 3850
1115
+ },
1116
+ {
1117
+ "epoch": 6.394715111478117,
1118
+ "grad_norm": 7.1418633460998535,
1119
+ "learning_rate": 1.289684210526316e-05,
1120
+ "loss": 0.5471,
1121
+ "step": 3875
1122
+ },
1123
+ {
1124
+ "epoch": 6.436003303055326,
1125
+ "grad_norm": 6.428386211395264,
1126
+ "learning_rate": 1.2844210526315791e-05,
1127
+ "loss": 0.5519,
1128
+ "step": 3900
1129
+ },
1130
+ {
1131
+ "epoch": 6.477291494632535,
1132
+ "grad_norm": 6.753655910491943,
1133
+ "learning_rate": 1.279157894736842e-05,
1134
+ "loss": 0.5745,
1135
+ "step": 3925
1136
+ },
1137
+ {
1138
+ "epoch": 6.518579686209744,
1139
+ "grad_norm": 6.977758407592773,
1140
+ "learning_rate": 1.2738947368421052e-05,
1141
+ "loss": 0.5657,
1142
+ "step": 3950
1143
+ },
1144
+ {
1145
+ "epoch": 6.559867877786953,
1146
+ "grad_norm": 6.839511394500732,
1147
+ "learning_rate": 1.2686315789473684e-05,
1148
+ "loss": 0.5432,
1149
+ "step": 3975
1150
+ },
1151
+ {
1152
+ "epoch": 6.601156069364162,
1153
+ "grad_norm": 7.744399547576904,
1154
+ "learning_rate": 1.2633684210526316e-05,
1155
+ "loss": 0.5473,
1156
+ "step": 4000
1157
+ },
1158
+ {
1159
+ "epoch": 6.601156069364162,
1160
+ "eval_loss": 1.056926965713501,
1161
+ "eval_runtime": 696.4132,
1162
+ "eval_samples_per_second": 15.194,
1163
+ "eval_steps_per_second": 1.266,
1164
+ "eval_wer": Infinity,
1165
+ "step": 4000
1166
+ },
1167
+ {
1168
+ "epoch": 6.642444260941371,
1169
+ "grad_norm": 7.011012077331543,
1170
+ "learning_rate": 1.2581052631578949e-05,
1171
+ "loss": 0.5369,
1172
+ "step": 4025
1173
+ },
1174
+ {
1175
+ "epoch": 6.68373245251858,
1176
+ "grad_norm": 8.44078540802002,
1177
+ "learning_rate": 1.252842105263158e-05,
1178
+ "loss": 0.5776,
1179
+ "step": 4050
1180
+ },
1181
+ {
1182
+ "epoch": 6.725020644095789,
1183
+ "grad_norm": 7.306187629699707,
1184
+ "learning_rate": 1.2475789473684212e-05,
1185
+ "loss": 0.5903,
1186
+ "step": 4075
1187
+ },
1188
+ {
1189
+ "epoch": 6.766308835672998,
1190
+ "grad_norm": 6.952062606811523,
1191
+ "learning_rate": 1.2423157894736844e-05,
1192
+ "loss": 0.55,
1193
+ "step": 4100
1194
+ },
1195
+ {
1196
+ "epoch": 6.807597027250207,
1197
+ "grad_norm": 6.906418323516846,
1198
+ "learning_rate": 1.2370526315789476e-05,
1199
+ "loss": 0.528,
1200
+ "step": 4125
1201
+ },
1202
+ {
1203
+ "epoch": 6.848885218827416,
1204
+ "grad_norm": 7.448940277099609,
1205
+ "learning_rate": 1.2317894736842107e-05,
1206
+ "loss": 0.5505,
1207
+ "step": 4150
1208
+ },
1209
+ {
1210
+ "epoch": 6.890173410404624,
1211
+ "grad_norm": 7.480178356170654,
1212
+ "learning_rate": 1.2265263157894737e-05,
1213
+ "loss": 0.5409,
1214
+ "step": 4175
1215
+ },
1216
+ {
1217
+ "epoch": 6.931461601981833,
1218
+ "grad_norm": 7.500922679901123,
1219
+ "learning_rate": 1.2212631578947369e-05,
1220
+ "loss": 0.5136,
1221
+ "step": 4200
1222
+ },
1223
+ {
1224
+ "epoch": 6.972749793559042,
1225
+ "grad_norm": 6.731492042541504,
1226
+ "learning_rate": 1.216e-05,
1227
+ "loss": 0.534,
1228
+ "step": 4225
1229
+ },
1230
+ {
1231
+ "epoch": 7.013212221304707,
1232
+ "grad_norm": 5.806182861328125,
1233
+ "learning_rate": 1.2107368421052632e-05,
1234
+ "loss": 0.4876,
1235
+ "step": 4250
1236
+ },
1237
+ {
1238
+ "epoch": 7.054500412881916,
1239
+ "grad_norm": 6.7710089683532715,
1240
+ "learning_rate": 1.2054736842105264e-05,
1241
+ "loss": 0.4452,
1242
+ "step": 4275
1243
+ },
1244
+ {
1245
+ "epoch": 7.095788604459124,
1246
+ "grad_norm": 7.5482563972473145,
1247
+ "learning_rate": 1.2002105263157895e-05,
1248
+ "loss": 0.4388,
1249
+ "step": 4300
1250
+ },
1251
+ {
1252
+ "epoch": 7.137076796036333,
1253
+ "grad_norm": 5.3507537841796875,
1254
+ "learning_rate": 1.1949473684210527e-05,
1255
+ "loss": 0.4572,
1256
+ "step": 4325
1257
+ },
1258
+ {
1259
+ "epoch": 7.178364987613542,
1260
+ "grad_norm": 6.261397838592529,
1261
+ "learning_rate": 1.189684210526316e-05,
1262
+ "loss": 0.46,
1263
+ "step": 4350
1264
+ },
1265
+ {
1266
+ "epoch": 7.219653179190751,
1267
+ "grad_norm": 7.379156589508057,
1268
+ "learning_rate": 1.1844210526315792e-05,
1269
+ "loss": 0.4595,
1270
+ "step": 4375
1271
+ },
1272
+ {
1273
+ "epoch": 7.26094137076796,
1274
+ "grad_norm": 6.441304683685303,
1275
+ "learning_rate": 1.1791578947368424e-05,
1276
+ "loss": 0.4623,
1277
+ "step": 4400
1278
+ },
1279
+ {
1280
+ "epoch": 7.302229562345169,
1281
+ "grad_norm": 6.100840091705322,
1282
+ "learning_rate": 1.1738947368421054e-05,
1283
+ "loss": 0.4686,
1284
+ "step": 4425
1285
+ },
1286
+ {
1287
+ "epoch": 7.343517753922378,
1288
+ "grad_norm": 6.594038486480713,
1289
+ "learning_rate": 1.1686315789473685e-05,
1290
+ "loss": 0.4638,
1291
+ "step": 4450
1292
+ },
1293
+ {
1294
+ "epoch": 7.384805945499587,
1295
+ "grad_norm": 6.223761558532715,
1296
+ "learning_rate": 1.1633684210526317e-05,
1297
+ "loss": 0.4687,
1298
+ "step": 4475
1299
+ },
1300
+ {
1301
+ "epoch": 7.426094137076796,
1302
+ "grad_norm": 5.7157063484191895,
1303
+ "learning_rate": 1.1581052631578948e-05,
1304
+ "loss": 0.4475,
1305
+ "step": 4500
1306
+ },
1307
+ {
1308
+ "epoch": 7.467382328654005,
1309
+ "grad_norm": 7.293723106384277,
1310
+ "learning_rate": 1.152842105263158e-05,
1311
+ "loss": 0.4725,
1312
+ "step": 4525
1313
+ },
1314
+ {
1315
+ "epoch": 7.508670520231214,
1316
+ "grad_norm": 6.591246604919434,
1317
+ "learning_rate": 1.1475789473684212e-05,
1318
+ "loss": 0.4783,
1319
+ "step": 4550
1320
+ },
1321
+ {
1322
+ "epoch": 7.549958711808423,
1323
+ "grad_norm": 7.094688415527344,
1324
+ "learning_rate": 1.1423157894736843e-05,
1325
+ "loss": 0.4521,
1326
+ "step": 4575
1327
+ },
1328
+ {
1329
+ "epoch": 7.591246903385632,
1330
+ "grad_norm": 7.4666595458984375,
1331
+ "learning_rate": 1.1370526315789473e-05,
1332
+ "loss": 0.4679,
1333
+ "step": 4600
1334
+ },
1335
+ {
1336
+ "epoch": 7.632535094962841,
1337
+ "grad_norm": 6.891570568084717,
1338
+ "learning_rate": 1.1317894736842105e-05,
1339
+ "loss": 0.469,
1340
+ "step": 4625
1341
+ },
1342
+ {
1343
+ "epoch": 7.673823286540049,
1344
+ "grad_norm": 7.736660003662109,
1345
+ "learning_rate": 1.1265263157894737e-05,
1346
+ "loss": 0.4874,
1347
+ "step": 4650
1348
+ },
1349
+ {
1350
+ "epoch": 7.715111478117258,
1351
+ "grad_norm": 7.7448248863220215,
1352
+ "learning_rate": 1.1212631578947368e-05,
1353
+ "loss": 0.4516,
1354
+ "step": 4675
1355
+ },
1356
+ {
1357
+ "epoch": 7.756399669694467,
1358
+ "grad_norm": 6.311560153961182,
1359
+ "learning_rate": 1.1160000000000002e-05,
1360
+ "loss": 0.4642,
1361
+ "step": 4700
1362
+ },
1363
+ {
1364
+ "epoch": 7.797687861271676,
1365
+ "grad_norm": 7.201409816741943,
1366
+ "learning_rate": 1.1107368421052633e-05,
1367
+ "loss": 0.4545,
1368
+ "step": 4725
1369
+ },
1370
+ {
1371
+ "epoch": 7.838976052848885,
1372
+ "grad_norm": 6.123779773712158,
1373
+ "learning_rate": 1.1054736842105265e-05,
1374
+ "loss": 0.4287,
1375
+ "step": 4750
1376
+ },
1377
+ {
1378
+ "epoch": 7.880264244426094,
1379
+ "grad_norm": 7.237743377685547,
1380
+ "learning_rate": 1.1002105263157896e-05,
1381
+ "loss": 0.4993,
1382
+ "step": 4775
1383
+ },
1384
+ {
1385
+ "epoch": 7.921552436003303,
1386
+ "grad_norm": 6.445917129516602,
1387
+ "learning_rate": 1.0949473684210528e-05,
1388
+ "loss": 0.4843,
1389
+ "step": 4800
1390
+ },
1391
+ {
1392
+ "epoch": 7.962840627580512,
1393
+ "grad_norm": 6.234178066253662,
1394
+ "learning_rate": 1.089684210526316e-05,
1395
+ "loss": 0.4657,
1396
+ "step": 4825
1397
+ },
1398
+ {
1399
+ "epoch": 8.003303055326176,
1400
+ "grad_norm": 7.247951507568359,
1401
+ "learning_rate": 1.084421052631579e-05,
1402
+ "loss": 0.4923,
1403
+ "step": 4850
1404
+ },
1405
+ {
1406
+ "epoch": 8.044591246903385,
1407
+ "grad_norm": 5.9410223960876465,
1408
+ "learning_rate": 1.0791578947368421e-05,
1409
+ "loss": 0.3872,
1410
+ "step": 4875
1411
+ },
1412
+ {
1413
+ "epoch": 8.085879438480594,
1414
+ "grad_norm": 6.340057849884033,
1415
+ "learning_rate": 1.0738947368421053e-05,
1416
+ "loss": 0.3966,
1417
+ "step": 4900
1418
+ },
1419
+ {
1420
+ "epoch": 8.127167630057803,
1421
+ "grad_norm": 6.049722194671631,
1422
+ "learning_rate": 1.0686315789473685e-05,
1423
+ "loss": 0.3706,
1424
+ "step": 4925
1425
+ },
1426
+ {
1427
+ "epoch": 8.168455821635012,
1428
+ "grad_norm": 5.964404582977295,
1429
+ "learning_rate": 1.0633684210526316e-05,
1430
+ "loss": 0.3854,
1431
+ "step": 4950
1432
+ },
1433
+ {
1434
+ "epoch": 8.209744013212221,
1435
+ "grad_norm": 6.412349224090576,
1436
+ "learning_rate": 1.0581052631578948e-05,
1437
+ "loss": 0.3923,
1438
+ "step": 4975
1439
+ },
1440
+ {
1441
+ "epoch": 8.25103220478943,
1442
+ "grad_norm": 6.039672374725342,
1443
+ "learning_rate": 1.052842105263158e-05,
1444
+ "loss": 0.4191,
1445
+ "step": 5000
1446
+ },
1447
+ {
1448
+ "epoch": 8.25103220478943,
1449
+ "eval_loss": 1.0567598342895508,
1450
+ "eval_runtime": 694.1236,
1451
+ "eval_samples_per_second": 15.244,
1452
+ "eval_steps_per_second": 1.271,
1453
+ "eval_wer": Infinity,
1454
+ "step": 5000
1455
+ },
1456
+ {
1457
+ "epoch": 8.29232039636664,
1458
+ "grad_norm": 6.2119011878967285,
1459
+ "learning_rate": 1.047578947368421e-05,
1460
+ "loss": 0.3807,
1461
+ "step": 5025
1462
+ },
1463
+ {
1464
+ "epoch": 8.333608587943848,
1465
+ "grad_norm": 6.253497123718262,
1466
+ "learning_rate": 1.0423157894736844e-05,
1467
+ "loss": 0.3827,
1468
+ "step": 5050
1469
+ },
1470
+ {
1471
+ "epoch": 8.374896779521057,
1472
+ "grad_norm": 6.059364318847656,
1473
+ "learning_rate": 1.0370526315789476e-05,
1474
+ "loss": 0.389,
1475
+ "step": 5075
1476
+ },
1477
+ {
1478
+ "epoch": 8.416184971098266,
1479
+ "grad_norm": 5.26777982711792,
1480
+ "learning_rate": 1.0317894736842106e-05,
1481
+ "loss": 0.3998,
1482
+ "step": 5100
1483
+ },
1484
+ {
1485
+ "epoch": 8.457473162675475,
1486
+ "grad_norm": 5.529355525970459,
1487
+ "learning_rate": 1.0265263157894738e-05,
1488
+ "loss": 0.3954,
1489
+ "step": 5125
1490
+ },
1491
+ {
1492
+ "epoch": 8.498761354252684,
1493
+ "grad_norm": 6.751101016998291,
1494
+ "learning_rate": 1.021263157894737e-05,
1495
+ "loss": 0.3964,
1496
+ "step": 5150
1497
+ },
1498
+ {
1499
+ "epoch": 8.540049545829893,
1500
+ "grad_norm": 5.405825138092041,
1501
+ "learning_rate": 1.0160000000000001e-05,
1502
+ "loss": 0.389,
1503
+ "step": 5175
1504
+ },
1505
+ {
1506
+ "epoch": 8.581337737407102,
1507
+ "grad_norm": 7.287736415863037,
1508
+ "learning_rate": 1.0107368421052633e-05,
1509
+ "loss": 0.4196,
1510
+ "step": 5200
1511
+ },
1512
+ {
1513
+ "epoch": 8.622625928984311,
1514
+ "grad_norm": 6.384584903717041,
1515
+ "learning_rate": 1.0054736842105264e-05,
1516
+ "loss": 0.4133,
1517
+ "step": 5225
1518
+ },
1519
+ {
1520
+ "epoch": 8.66391412056152,
1521
+ "grad_norm": 6.451061725616455,
1522
+ "learning_rate": 1.0002105263157896e-05,
1523
+ "loss": 0.3837,
1524
+ "step": 5250
1525
+ },
1526
+ {
1527
+ "epoch": 8.705202312138729,
1528
+ "grad_norm": 5.920280933380127,
1529
+ "learning_rate": 9.949473684210526e-06,
1530
+ "loss": 0.4004,
1531
+ "step": 5275
1532
+ },
1533
+ {
1534
+ "epoch": 8.746490503715938,
1535
+ "grad_norm": 5.891169548034668,
1536
+ "learning_rate": 9.89684210526316e-06,
1537
+ "loss": 0.3932,
1538
+ "step": 5300
1539
+ },
1540
+ {
1541
+ "epoch": 8.787778695293147,
1542
+ "grad_norm": 6.570876598358154,
1543
+ "learning_rate": 9.84421052631579e-06,
1544
+ "loss": 0.4032,
1545
+ "step": 5325
1546
+ },
1547
+ {
1548
+ "epoch": 8.829066886870356,
1549
+ "grad_norm": 7.074426174163818,
1550
+ "learning_rate": 9.791578947368422e-06,
1551
+ "loss": 0.3951,
1552
+ "step": 5350
1553
+ },
1554
+ {
1555
+ "epoch": 8.870355078447563,
1556
+ "grad_norm": 6.7733564376831055,
1557
+ "learning_rate": 9.738947368421054e-06,
1558
+ "loss": 0.3992,
1559
+ "step": 5375
1560
+ },
1561
+ {
1562
+ "epoch": 8.911643270024772,
1563
+ "grad_norm": 6.8463969230651855,
1564
+ "learning_rate": 9.686315789473684e-06,
1565
+ "loss": 0.4004,
1566
+ "step": 5400
1567
+ },
1568
+ {
1569
+ "epoch": 8.952931461601981,
1570
+ "grad_norm": 6.444080352783203,
1571
+ "learning_rate": 9.633684210526316e-06,
1572
+ "loss": 0.3844,
1573
+ "step": 5425
1574
+ },
1575
+ {
1576
+ "epoch": 8.99421965317919,
1577
+ "grad_norm": 6.648684024810791,
1578
+ "learning_rate": 9.581052631578947e-06,
1579
+ "loss": 0.3962,
1580
+ "step": 5450
1581
+ },
1582
+ {
1583
+ "epoch": 9.034682080924856,
1584
+ "grad_norm": 5.462916851043701,
1585
+ "learning_rate": 9.52842105263158e-06,
1586
+ "loss": 0.3186,
1587
+ "step": 5475
1588
+ },
1589
+ {
1590
+ "epoch": 9.075970272502065,
1591
+ "grad_norm": 4.91270112991333,
1592
+ "learning_rate": 9.475789473684212e-06,
1593
+ "loss": 0.3091,
1594
+ "step": 5500
1595
+ },
1596
+ {
1597
+ "epoch": 9.117258464079274,
1598
+ "grad_norm": 6.3385467529296875,
1599
+ "learning_rate": 9.423157894736842e-06,
1600
+ "loss": 0.3088,
1601
+ "step": 5525
1602
+ },
1603
+ {
1604
+ "epoch": 9.158546655656481,
1605
+ "grad_norm": 6.092479228973389,
1606
+ "learning_rate": 9.370526315789474e-06,
1607
+ "loss": 0.3366,
1608
+ "step": 5550
1609
+ },
1610
+ {
1611
+ "epoch": 9.19983484723369,
1612
+ "grad_norm": 6.2821364402771,
1613
+ "learning_rate": 9.317894736842105e-06,
1614
+ "loss": 0.328,
1615
+ "step": 5575
1616
+ },
1617
+ {
1618
+ "epoch": 9.2411230388109,
1619
+ "grad_norm": 6.607597827911377,
1620
+ "learning_rate": 9.265263157894737e-06,
1621
+ "loss": 0.3237,
1622
+ "step": 5600
1623
+ },
1624
+ {
1625
+ "epoch": 9.282411230388108,
1626
+ "grad_norm": 6.044376850128174,
1627
+ "learning_rate": 9.21263157894737e-06,
1628
+ "loss": 0.3405,
1629
+ "step": 5625
1630
+ },
1631
+ {
1632
+ "epoch": 9.323699421965317,
1633
+ "grad_norm": 5.687169551849365,
1634
+ "learning_rate": 9.16e-06,
1635
+ "loss": 0.3433,
1636
+ "step": 5650
1637
+ },
1638
+ {
1639
+ "epoch": 9.364987613542526,
1640
+ "grad_norm": 6.336760997772217,
1641
+ "learning_rate": 9.107368421052632e-06,
1642
+ "loss": 0.3681,
1643
+ "step": 5675
1644
+ },
1645
+ {
1646
+ "epoch": 9.406275805119735,
1647
+ "grad_norm": 5.349738597869873,
1648
+ "learning_rate": 9.054736842105264e-06,
1649
+ "loss": 0.3414,
1650
+ "step": 5700
1651
+ },
1652
+ {
1653
+ "epoch": 9.447563996696944,
1654
+ "grad_norm": 5.843466758728027,
1655
+ "learning_rate": 9.002105263157895e-06,
1656
+ "loss": 0.3494,
1657
+ "step": 5725
1658
+ },
1659
+ {
1660
+ "epoch": 9.488852188274153,
1661
+ "grad_norm": 5.646897315979004,
1662
+ "learning_rate": 8.949473684210527e-06,
1663
+ "loss": 0.3491,
1664
+ "step": 5750
1665
+ },
1666
+ {
1667
+ "epoch": 9.530140379851362,
1668
+ "grad_norm": 6.2915472984313965,
1669
+ "learning_rate": 8.896842105263159e-06,
1670
+ "loss": 0.3627,
1671
+ "step": 5775
1672
+ },
1673
+ {
1674
+ "epoch": 9.571428571428571,
1675
+ "grad_norm": 5.424402236938477,
1676
+ "learning_rate": 8.84421052631579e-06,
1677
+ "loss": 0.3203,
1678
+ "step": 5800
1679
+ },
1680
+ {
1681
+ "epoch": 9.61271676300578,
1682
+ "grad_norm": 6.5985260009765625,
1683
+ "learning_rate": 8.791578947368422e-06,
1684
+ "loss": 0.3449,
1685
+ "step": 5825
1686
+ },
1687
+ {
1688
+ "epoch": 9.65400495458299,
1689
+ "grad_norm": 5.538530349731445,
1690
+ "learning_rate": 8.738947368421053e-06,
1691
+ "loss": 0.3407,
1692
+ "step": 5850
1693
+ },
1694
+ {
1695
+ "epoch": 9.695293146160198,
1696
+ "grad_norm": 5.327653408050537,
1697
+ "learning_rate": 8.686315789473685e-06,
1698
+ "loss": 0.3235,
1699
+ "step": 5875
1700
+ },
1701
+ {
1702
+ "epoch": 9.736581337737407,
1703
+ "grad_norm": 6.340036392211914,
1704
+ "learning_rate": 8.633684210526317e-06,
1705
+ "loss": 0.3595,
1706
+ "step": 5900
1707
+ },
1708
+ {
1709
+ "epoch": 9.777869529314616,
1710
+ "grad_norm": 5.395780086517334,
1711
+ "learning_rate": 8.581052631578948e-06,
1712
+ "loss": 0.3368,
1713
+ "step": 5925
1714
+ },
1715
+ {
1716
+ "epoch": 9.819157720891825,
1717
+ "grad_norm": 5.803914546966553,
1718
+ "learning_rate": 8.528421052631578e-06,
1719
+ "loss": 0.3497,
1720
+ "step": 5950
1721
+ },
1722
+ {
1723
+ "epoch": 9.860445912469034,
1724
+ "grad_norm": 6.23885440826416,
1725
+ "learning_rate": 8.475789473684212e-06,
1726
+ "loss": 0.3504,
1727
+ "step": 5975
1728
+ },
1729
+ {
1730
+ "epoch": 9.901734104046243,
1731
+ "grad_norm": 6.402750015258789,
1732
+ "learning_rate": 8.425263157894737e-06,
1733
+ "loss": 0.3167,
1734
+ "step": 6000
1735
+ },
1736
+ {
1737
+ "epoch": 9.901734104046243,
1738
+ "eval_loss": 1.060942530632019,
1739
+ "eval_runtime": 695.1498,
1740
+ "eval_samples_per_second": 15.221,
1741
+ "eval_steps_per_second": 1.269,
1742
+ "eval_wer": Infinity,
1743
+ "step": 6000
1744
+ },
1745
+ {
1746
+ "epoch": 9.943022295623452,
1747
+ "grad_norm": 5.562705993652344,
1748
+ "learning_rate": 8.372631578947368e-06,
1749
+ "loss": 0.3252,
1750
+ "step": 6025
1751
+ },
1752
+ {
1753
+ "epoch": 9.984310487200661,
1754
+ "grad_norm": 7.461434364318848,
1755
+ "learning_rate": 8.32e-06,
1756
+ "loss": 0.344,
1757
+ "step": 6050
1758
+ },
1759
+ {
1760
+ "epoch": 10.024772914946325,
1761
+ "grad_norm": 5.992921352386475,
1762
+ "learning_rate": 8.267368421052632e-06,
1763
+ "loss": 0.2875,
1764
+ "step": 6075
1765
+ },
1766
+ {
1767
+ "epoch": 10.066061106523534,
1768
+ "grad_norm": 4.986021995544434,
1769
+ "learning_rate": 8.214736842105265e-06,
1770
+ "loss": 0.2773,
1771
+ "step": 6100
1772
+ },
1773
+ {
1774
+ "epoch": 10.107349298100743,
1775
+ "grad_norm": 4.8151092529296875,
1776
+ "learning_rate": 8.162105263157895e-06,
1777
+ "loss": 0.2886,
1778
+ "step": 6125
1779
+ },
1780
+ {
1781
+ "epoch": 10.148637489677952,
1782
+ "grad_norm": 4.7961883544921875,
1783
+ "learning_rate": 8.109473684210527e-06,
1784
+ "loss": 0.3002,
1785
+ "step": 6150
1786
+ },
1787
+ {
1788
+ "epoch": 10.189925681255161,
1789
+ "grad_norm": 5.7180399894714355,
1790
+ "learning_rate": 8.056842105263158e-06,
1791
+ "loss": 0.2995,
1792
+ "step": 6175
1793
+ },
1794
+ {
1795
+ "epoch": 10.23121387283237,
1796
+ "grad_norm": 5.287256717681885,
1797
+ "learning_rate": 8.00421052631579e-06,
1798
+ "loss": 0.2723,
1799
+ "step": 6200
1800
+ },
1801
+ {
1802
+ "epoch": 10.27250206440958,
1803
+ "grad_norm": 6.107348442077637,
1804
+ "learning_rate": 7.951578947368421e-06,
1805
+ "loss": 0.2909,
1806
+ "step": 6225
1807
+ },
1808
+ {
1809
+ "epoch": 10.313790255986788,
1810
+ "grad_norm": 5.607096195220947,
1811
+ "learning_rate": 7.898947368421053e-06,
1812
+ "loss": 0.3019,
1813
+ "step": 6250
1814
+ },
1815
+ {
1816
+ "epoch": 10.355078447563997,
1817
+ "grad_norm": 5.941197872161865,
1818
+ "learning_rate": 7.846315789473685e-06,
1819
+ "loss": 0.29,
1820
+ "step": 6275
1821
+ },
1822
+ {
1823
+ "epoch": 10.396366639141206,
1824
+ "grad_norm": 5.736889362335205,
1825
+ "learning_rate": 7.793684210526316e-06,
1826
+ "loss": 0.3138,
1827
+ "step": 6300
1828
+ },
1829
+ {
1830
+ "epoch": 10.437654830718415,
1831
+ "grad_norm": 5.2913126945495605,
1832
+ "learning_rate": 7.741052631578948e-06,
1833
+ "loss": 0.28,
1834
+ "step": 6325
1835
+ },
1836
+ {
1837
+ "epoch": 10.478943022295624,
1838
+ "grad_norm": 5.760861396789551,
1839
+ "learning_rate": 7.68842105263158e-06,
1840
+ "loss": 0.3069,
1841
+ "step": 6350
1842
+ },
1843
+ {
1844
+ "epoch": 10.520231213872833,
1845
+ "grad_norm": 4.886830806732178,
1846
+ "learning_rate": 7.635789473684211e-06,
1847
+ "loss": 0.2785,
1848
+ "step": 6375
1849
+ },
1850
+ {
1851
+ "epoch": 10.561519405450042,
1852
+ "grad_norm": 5.5294389724731445,
1853
+ "learning_rate": 7.583157894736842e-06,
1854
+ "loss": 0.2862,
1855
+ "step": 6400
1856
+ },
1857
+ {
1858
+ "epoch": 10.602807597027251,
1859
+ "grad_norm": 5.370204448699951,
1860
+ "learning_rate": 7.5305263157894745e-06,
1861
+ "loss": 0.2794,
1862
+ "step": 6425
1863
+ },
1864
+ {
1865
+ "epoch": 10.64409578860446,
1866
+ "grad_norm": 4.194797039031982,
1867
+ "learning_rate": 7.477894736842106e-06,
1868
+ "loss": 0.3064,
1869
+ "step": 6450
1870
+ },
1871
+ {
1872
+ "epoch": 10.685383980181667,
1873
+ "grad_norm": 6.214608669281006,
1874
+ "learning_rate": 7.425263157894738e-06,
1875
+ "loss": 0.3177,
1876
+ "step": 6475
1877
+ },
1878
+ {
1879
+ "epoch": 10.726672171758876,
1880
+ "grad_norm": 5.77274227142334,
1881
+ "learning_rate": 7.3726315789473694e-06,
1882
+ "loss": 0.31,
1883
+ "step": 6500
1884
+ },
1885
+ {
1886
+ "epoch": 10.767960363336085,
1887
+ "grad_norm": 6.4990105628967285,
1888
+ "learning_rate": 7.32e-06,
1889
+ "loss": 0.2716,
1890
+ "step": 6525
1891
+ },
1892
+ {
1893
+ "epoch": 10.809248554913294,
1894
+ "grad_norm": 5.706321716308594,
1895
+ "learning_rate": 7.267368421052632e-06,
1896
+ "loss": 0.2983,
1897
+ "step": 6550
1898
+ },
1899
+ {
1900
+ "epoch": 10.850536746490503,
1901
+ "grad_norm": 6.846646308898926,
1902
+ "learning_rate": 7.2147368421052635e-06,
1903
+ "loss": 0.2678,
1904
+ "step": 6575
1905
+ },
1906
+ {
1907
+ "epoch": 10.891824938067712,
1908
+ "grad_norm": 5.539628982543945,
1909
+ "learning_rate": 7.162105263157896e-06,
1910
+ "loss": 0.2809,
1911
+ "step": 6600
1912
+ },
1913
+ {
1914
+ "epoch": 10.933113129644921,
1915
+ "grad_norm": 5.272496700286865,
1916
+ "learning_rate": 7.109473684210528e-06,
1917
+ "loss": 0.2473,
1918
+ "step": 6625
1919
+ },
1920
+ {
1921
+ "epoch": 10.97440132122213,
1922
+ "grad_norm": 5.260607719421387,
1923
+ "learning_rate": 7.056842105263158e-06,
1924
+ "loss": 0.2931,
1925
+ "step": 6650
1926
+ },
1927
+ {
1928
+ "epoch": 11.014863748967795,
1929
+ "grad_norm": 5.241586685180664,
1930
+ "learning_rate": 7.00421052631579e-06,
1931
+ "loss": 0.2771,
1932
+ "step": 6675
1933
+ },
1934
+ {
1935
+ "epoch": 11.056151940545003,
1936
+ "grad_norm": 4.1120781898498535,
1937
+ "learning_rate": 6.951578947368422e-06,
1938
+ "loss": 0.2511,
1939
+ "step": 6700
1940
+ },
1941
+ {
1942
+ "epoch": 11.097440132122212,
1943
+ "grad_norm": 5.060145378112793,
1944
+ "learning_rate": 6.8989473684210525e-06,
1945
+ "loss": 0.2502,
1946
+ "step": 6725
1947
+ },
1948
+ {
1949
+ "epoch": 11.138728323699421,
1950
+ "grad_norm": 6.010138988494873,
1951
+ "learning_rate": 6.846315789473684e-06,
1952
+ "loss": 0.2442,
1953
+ "step": 6750
1954
+ },
1955
+ {
1956
+ "epoch": 11.18001651527663,
1957
+ "grad_norm": 5.282010078430176,
1958
+ "learning_rate": 6.793684210526317e-06,
1959
+ "loss": 0.2485,
1960
+ "step": 6775
1961
+ },
1962
+ {
1963
+ "epoch": 11.22130470685384,
1964
+ "grad_norm": 4.690225601196289,
1965
+ "learning_rate": 6.741052631578948e-06,
1966
+ "loss": 0.2506,
1967
+ "step": 6800
1968
+ },
1969
+ {
1970
+ "epoch": 11.262592898431048,
1971
+ "grad_norm": 3.9595465660095215,
1972
+ "learning_rate": 6.68842105263158e-06,
1973
+ "loss": 0.2325,
1974
+ "step": 6825
1975
+ },
1976
+ {
1977
+ "epoch": 11.303881090008257,
1978
+ "grad_norm": 5.889693737030029,
1979
+ "learning_rate": 6.635789473684211e-06,
1980
+ "loss": 0.2527,
1981
+ "step": 6850
1982
+ },
1983
+ {
1984
+ "epoch": 11.345169281585466,
1985
+ "grad_norm": 5.33756685256958,
1986
+ "learning_rate": 6.583157894736842e-06,
1987
+ "loss": 0.2488,
1988
+ "step": 6875
1989
+ },
1990
+ {
1991
+ "epoch": 11.386457473162675,
1992
+ "grad_norm": 5.049898147583008,
1993
+ "learning_rate": 6.530526315789474e-06,
1994
+ "loss": 0.2355,
1995
+ "step": 6900
1996
+ },
1997
+ {
1998
+ "epoch": 11.427745664739884,
1999
+ "grad_norm": 5.452274799346924,
2000
+ "learning_rate": 6.477894736842106e-06,
2001
+ "loss": 0.249,
2002
+ "step": 6925
2003
+ },
2004
+ {
2005
+ "epoch": 11.469033856317093,
2006
+ "grad_norm": 4.464880466461182,
2007
+ "learning_rate": 6.425263157894738e-06,
2008
+ "loss": 0.2523,
2009
+ "step": 6950
2010
+ },
2011
+ {
2012
+ "epoch": 11.510322047894302,
2013
+ "grad_norm": 5.923793792724609,
2014
+ "learning_rate": 6.372631578947369e-06,
2015
+ "loss": 0.2564,
2016
+ "step": 6975
2017
+ },
2018
+ {
2019
+ "epoch": 11.551610239471511,
2020
+ "grad_norm": 4.821681976318359,
2021
+ "learning_rate": 6.3200000000000005e-06,
2022
+ "loss": 0.2322,
2023
+ "step": 7000
2024
+ },
2025
+ {
2026
+ "epoch": 11.551610239471511,
2027
+ "eval_loss": 1.0933067798614502,
2028
+ "eval_runtime": 721.5402,
2029
+ "eval_samples_per_second": 14.664,
2030
+ "eval_steps_per_second": 1.222,
2031
+ "eval_wer": Infinity,
2032
+ "step": 7000
2033
+ },
2034
+ {
2035
+ "epoch": 11.59289843104872,
2036
+ "grad_norm": 6.9122748374938965,
2037
+ "learning_rate": 6.267368421052632e-06,
2038
+ "loss": 0.2701,
2039
+ "step": 7025
2040
+ },
2041
+ {
2042
+ "epoch": 11.63418662262593,
2043
+ "grad_norm": 4.586194038391113,
2044
+ "learning_rate": 6.214736842105264e-06,
2045
+ "loss": 0.2496,
2046
+ "step": 7050
2047
+ },
2048
+ {
2049
+ "epoch": 11.675474814203138,
2050
+ "grad_norm": 5.90902853012085,
2051
+ "learning_rate": 6.1621052631578945e-06,
2052
+ "loss": 0.2384,
2053
+ "step": 7075
2054
+ },
2055
+ {
2056
+ "epoch": 11.716763005780347,
2057
+ "grad_norm": 5.328728199005127,
2058
+ "learning_rate": 6.109473684210527e-06,
2059
+ "loss": 0.253,
2060
+ "step": 7100
2061
+ },
2062
+ {
2063
+ "epoch": 11.758051197357556,
2064
+ "grad_norm": 6.127770900726318,
2065
+ "learning_rate": 6.056842105263159e-06,
2066
+ "loss": 0.2359,
2067
+ "step": 7125
2068
+ },
2069
+ {
2070
+ "epoch": 11.799339388934765,
2071
+ "grad_norm": 5.675428867340088,
2072
+ "learning_rate": 6.00421052631579e-06,
2073
+ "loss": 0.2383,
2074
+ "step": 7150
2075
+ },
2076
+ {
2077
+ "epoch": 11.840627580511974,
2078
+ "grad_norm": 4.921765327453613,
2079
+ "learning_rate": 5.951578947368422e-06,
2080
+ "loss": 0.2648,
2081
+ "step": 7175
2082
+ },
2083
+ {
2084
+ "epoch": 11.881915772089183,
2085
+ "grad_norm": 4.198760986328125,
2086
+ "learning_rate": 5.898947368421053e-06,
2087
+ "loss": 0.2351,
2088
+ "step": 7200
2089
+ },
2090
+ {
2091
+ "epoch": 11.923203963666392,
2092
+ "grad_norm": 4.786856174468994,
2093
+ "learning_rate": 5.846315789473684e-06,
2094
+ "loss": 0.2468,
2095
+ "step": 7225
2096
+ },
2097
+ {
2098
+ "epoch": 11.964492155243601,
2099
+ "grad_norm": 4.955503463745117,
2100
+ "learning_rate": 5.793684210526316e-06,
2101
+ "loss": 0.2471,
2102
+ "step": 7250
2103
+ },
2104
+ {
2105
+ "epoch": 12.004954582989265,
2106
+ "grad_norm": 4.753754138946533,
2107
+ "learning_rate": 5.7410526315789485e-06,
2108
+ "loss": 0.2625,
2109
+ "step": 7275
2110
+ },
2111
+ {
2112
+ "epoch": 12.046242774566474,
2113
+ "grad_norm": 3.7155563831329346,
2114
+ "learning_rate": 5.68842105263158e-06,
2115
+ "loss": 0.2123,
2116
+ "step": 7300
2117
+ },
2118
+ {
2119
+ "epoch": 12.087530966143683,
2120
+ "grad_norm": 4.637264251708984,
2121
+ "learning_rate": 5.635789473684211e-06,
2122
+ "loss": 0.209,
2123
+ "step": 7325
2124
+ },
2125
+ {
2126
+ "epoch": 12.128819157720892,
2127
+ "grad_norm": 5.065235614776611,
2128
+ "learning_rate": 5.5831578947368425e-06,
2129
+ "loss": 0.2016,
2130
+ "step": 7350
2131
+ },
2132
+ {
2133
+ "epoch": 12.170107349298101,
2134
+ "grad_norm": 4.761934280395508,
2135
+ "learning_rate": 5.530526315789474e-06,
2136
+ "loss": 0.2081,
2137
+ "step": 7375
2138
+ },
2139
+ {
2140
+ "epoch": 12.21139554087531,
2141
+ "grad_norm": 5.17978048324585,
2142
+ "learning_rate": 5.477894736842105e-06,
2143
+ "loss": 0.2288,
2144
+ "step": 7400
2145
+ },
2146
+ {
2147
+ "epoch": 12.25268373245252,
2148
+ "grad_norm": 3.554265260696411,
2149
+ "learning_rate": 5.425263157894737e-06,
2150
+ "loss": 0.2143,
2151
+ "step": 7425
2152
+ },
2153
+ {
2154
+ "epoch": 12.293971924029728,
2155
+ "grad_norm": 5.236491680145264,
2156
+ "learning_rate": 5.372631578947369e-06,
2157
+ "loss": 0.2025,
2158
+ "step": 7450
2159
+ },
2160
+ {
2161
+ "epoch": 12.335260115606937,
2162
+ "grad_norm": 5.198639392852783,
2163
+ "learning_rate": 5.320000000000001e-06,
2164
+ "loss": 0.2199,
2165
+ "step": 7475
2166
+ },
2167
+ {
2168
+ "epoch": 12.376548307184144,
2169
+ "grad_norm": 5.509646892547607,
2170
+ "learning_rate": 5.267368421052632e-06,
2171
+ "loss": 0.1932,
2172
+ "step": 7500
2173
+ },
2174
+ {
2175
+ "epoch": 12.417836498761353,
2176
+ "grad_norm": 4.569394588470459,
2177
+ "learning_rate": 5.214736842105263e-06,
2178
+ "loss": 0.2018,
2179
+ "step": 7525
2180
+ },
2181
+ {
2182
+ "epoch": 12.459124690338562,
2183
+ "grad_norm": 4.686092376708984,
2184
+ "learning_rate": 5.162105263157895e-06,
2185
+ "loss": 0.2148,
2186
+ "step": 7550
2187
+ },
2188
+ {
2189
+ "epoch": 12.500412881915771,
2190
+ "grad_norm": 5.779174327850342,
2191
+ "learning_rate": 5.1094736842105264e-06,
2192
+ "loss": 0.2362,
2193
+ "step": 7575
2194
+ },
2195
+ {
2196
+ "epoch": 12.54170107349298,
2197
+ "grad_norm": 7.2701873779296875,
2198
+ "learning_rate": 5.056842105263158e-06,
2199
+ "loss": 0.2421,
2200
+ "step": 7600
2201
+ },
2202
+ {
2203
+ "epoch": 12.58298926507019,
2204
+ "grad_norm": 4.832854270935059,
2205
+ "learning_rate": 5.0042105263157906e-06,
2206
+ "loss": 0.219,
2207
+ "step": 7625
2208
+ },
2209
+ {
2210
+ "epoch": 12.624277456647398,
2211
+ "grad_norm": 5.803878307342529,
2212
+ "learning_rate": 4.951578947368421e-06,
2213
+ "loss": 0.2172,
2214
+ "step": 7650
2215
+ },
2216
+ {
2217
+ "epoch": 12.665565648224607,
2218
+ "grad_norm": 4.021119594573975,
2219
+ "learning_rate": 4.898947368421053e-06,
2220
+ "loss": 0.227,
2221
+ "step": 7675
2222
+ },
2223
+ {
2224
+ "epoch": 12.706853839801816,
2225
+ "grad_norm": 5.169255256652832,
2226
+ "learning_rate": 4.846315789473685e-06,
2227
+ "loss": 0.21,
2228
+ "step": 7700
2229
+ },
2230
+ {
2231
+ "epoch": 12.748142031379025,
2232
+ "grad_norm": 4.798264503479004,
2233
+ "learning_rate": 4.793684210526316e-06,
2234
+ "loss": 0.1984,
2235
+ "step": 7725
2236
+ },
2237
+ {
2238
+ "epoch": 12.789430222956234,
2239
+ "grad_norm": 5.631106853485107,
2240
+ "learning_rate": 4.741052631578948e-06,
2241
+ "loss": 0.225,
2242
+ "step": 7750
2243
+ },
2244
+ {
2245
+ "epoch": 12.830718414533443,
2246
+ "grad_norm": 4.223946571350098,
2247
+ "learning_rate": 4.6884210526315795e-06,
2248
+ "loss": 0.2163,
2249
+ "step": 7775
2250
+ },
2251
+ {
2252
+ "epoch": 12.872006606110652,
2253
+ "grad_norm": 4.958042621612549,
2254
+ "learning_rate": 4.63578947368421e-06,
2255
+ "loss": 0.2131,
2256
+ "step": 7800
2257
+ },
2258
+ {
2259
+ "epoch": 12.913294797687861,
2260
+ "grad_norm": 6.063140392303467,
2261
+ "learning_rate": 4.583157894736843e-06,
2262
+ "loss": 0.2443,
2263
+ "step": 7825
2264
+ },
2265
+ {
2266
+ "epoch": 12.95458298926507,
2267
+ "grad_norm": 4.037540435791016,
2268
+ "learning_rate": 4.5305263157894744e-06,
2269
+ "loss": 0.2033,
2270
+ "step": 7850
2271
+ },
2272
+ {
2273
+ "epoch": 12.995871180842279,
2274
+ "grad_norm": 4.495209217071533,
2275
+ "learning_rate": 4.477894736842105e-06,
2276
+ "loss": 0.2307,
2277
+ "step": 7875
2278
+ },
2279
+ {
2280
+ "epoch": 13.036333608587944,
2281
+ "grad_norm": 4.494256496429443,
2282
+ "learning_rate": 4.425263157894737e-06,
2283
+ "loss": 0.1894,
2284
+ "step": 7900
2285
+ },
2286
+ {
2287
+ "epoch": 13.077621800165153,
2288
+ "grad_norm": 3.877058744430542,
2289
+ "learning_rate": 4.3726315789473685e-06,
2290
+ "loss": 0.195,
2291
+ "step": 7925
2292
+ },
2293
+ {
2294
+ "epoch": 13.118909991742361,
2295
+ "grad_norm": 4.717246055603027,
2296
+ "learning_rate": 4.32e-06,
2297
+ "loss": 0.1745,
2298
+ "step": 7950
2299
+ },
2300
+ {
2301
+ "epoch": 13.16019818331957,
2302
+ "grad_norm": 4.590612411499023,
2303
+ "learning_rate": 4.267368421052632e-06,
2304
+ "loss": 0.195,
2305
+ "step": 7975
2306
+ },
2307
+ {
2308
+ "epoch": 13.20148637489678,
2309
+ "grad_norm": 6.558470249176025,
2310
+ "learning_rate": 4.214736842105263e-06,
2311
+ "loss": 0.1913,
2312
+ "step": 8000
2313
+ },
2314
+ {
2315
+ "epoch": 13.20148637489678,
2316
+ "eval_loss": 1.1226582527160645,
2317
+ "eval_runtime": 703.5225,
2318
+ "eval_samples_per_second": 15.04,
2319
+ "eval_steps_per_second": 1.254,
2320
+ "eval_wer": Infinity,
2321
+ "step": 8000
2322
+ },
2323
+ {
2324
+ "epoch": 13.242774566473988,
2325
+ "grad_norm": 4.316694736480713,
2326
+ "learning_rate": 4.162105263157895e-06,
2327
+ "loss": 0.2075,
2328
+ "step": 8025
2329
+ },
2330
+ {
2331
+ "epoch": 13.284062758051197,
2332
+ "grad_norm": 4.108456611633301,
2333
+ "learning_rate": 4.109473684210527e-06,
2334
+ "loss": 0.1654,
2335
+ "step": 8050
2336
+ },
2337
+ {
2338
+ "epoch": 13.325350949628406,
2339
+ "grad_norm": 3.923902750015259,
2340
+ "learning_rate": 4.056842105263158e-06,
2341
+ "loss": 0.19,
2342
+ "step": 8075
2343
+ },
2344
+ {
2345
+ "epoch": 13.366639141205615,
2346
+ "grad_norm": 3.792008638381958,
2347
+ "learning_rate": 4.00421052631579e-06,
2348
+ "loss": 0.1924,
2349
+ "step": 8100
2350
+ },
2351
+ {
2352
+ "epoch": 13.407927332782824,
2353
+ "grad_norm": 4.64762020111084,
2354
+ "learning_rate": 3.951578947368422e-06,
2355
+ "loss": 0.1988,
2356
+ "step": 8125
2357
+ },
2358
+ {
2359
+ "epoch": 13.449215524360033,
2360
+ "grad_norm": 4.5705246925354,
2361
+ "learning_rate": 3.898947368421052e-06,
2362
+ "loss": 0.1857,
2363
+ "step": 8150
2364
+ },
2365
+ {
2366
+ "epoch": 13.490503715937242,
2367
+ "grad_norm": 4.8831610679626465,
2368
+ "learning_rate": 3.846315789473685e-06,
2369
+ "loss": 0.1813,
2370
+ "step": 8175
2371
+ },
2372
+ {
2373
+ "epoch": 13.531791907514451,
2374
+ "grad_norm": 3.7012646198272705,
2375
+ "learning_rate": 3.793684210526316e-06,
2376
+ "loss": 0.212,
2377
+ "step": 8200
2378
+ },
2379
+ {
2380
+ "epoch": 13.57308009909166,
2381
+ "grad_norm": 4.698606967926025,
2382
+ "learning_rate": 3.7410526315789473e-06,
2383
+ "loss": 0.1999,
2384
+ "step": 8225
2385
+ },
2386
+ {
2387
+ "epoch": 13.61436829066887,
2388
+ "grad_norm": 5.355463027954102,
2389
+ "learning_rate": 3.6884210526315794e-06,
2390
+ "loss": 0.188,
2391
+ "step": 8250
2392
+ },
2393
+ {
2394
+ "epoch": 13.655656482246078,
2395
+ "grad_norm": 4.198258876800537,
2396
+ "learning_rate": 3.635789473684211e-06,
2397
+ "loss": 0.187,
2398
+ "step": 8275
2399
+ },
2400
+ {
2401
+ "epoch": 13.696944673823287,
2402
+ "grad_norm": 4.937903881072998,
2403
+ "learning_rate": 3.5831578947368422e-06,
2404
+ "loss": 0.1942,
2405
+ "step": 8300
2406
+ },
2407
+ {
2408
+ "epoch": 13.738232865400496,
2409
+ "grad_norm": 5.5391645431518555,
2410
+ "learning_rate": 3.5305263157894743e-06,
2411
+ "loss": 0.1897,
2412
+ "step": 8325
2413
+ },
2414
+ {
2415
+ "epoch": 13.779521056977705,
2416
+ "grad_norm": 3.7583065032958984,
2417
+ "learning_rate": 3.4778947368421055e-06,
2418
+ "loss": 0.1932,
2419
+ "step": 8350
2420
+ },
2421
+ {
2422
+ "epoch": 13.820809248554914,
2423
+ "grad_norm": 5.243954181671143,
2424
+ "learning_rate": 3.425263157894737e-06,
2425
+ "loss": 0.198,
2426
+ "step": 8375
2427
+ },
2428
+ {
2429
+ "epoch": 13.862097440132121,
2430
+ "grad_norm": 5.0445556640625,
2431
+ "learning_rate": 3.3726315789473683e-06,
2432
+ "loss": 0.1773,
2433
+ "step": 8400
2434
+ },
2435
+ {
2436
+ "epoch": 13.90338563170933,
2437
+ "grad_norm": 4.348180770874023,
2438
+ "learning_rate": 3.3200000000000004e-06,
2439
+ "loss": 0.1834,
2440
+ "step": 8425
2441
+ },
2442
+ {
2443
+ "epoch": 13.94467382328654,
2444
+ "grad_norm": 5.649644374847412,
2445
+ "learning_rate": 3.267368421052632e-06,
2446
+ "loss": 0.1772,
2447
+ "step": 8450
2448
+ },
2449
+ {
2450
+ "epoch": 13.985962014863748,
2451
+ "grad_norm": 5.276495933532715,
2452
+ "learning_rate": 3.2147368421052633e-06,
2453
+ "loss": 0.211,
2454
+ "step": 8475
2455
+ },
2456
+ {
2457
+ "epoch": 14.026424442609414,
2458
+ "grad_norm": 4.379940032958984,
2459
+ "learning_rate": 3.1621052631578953e-06,
2460
+ "loss": 0.1697,
2461
+ "step": 8500
2462
+ },
2463
+ {
2464
+ "epoch": 14.067712634186623,
2465
+ "grad_norm": 4.398263454437256,
2466
+ "learning_rate": 3.1094736842105265e-06,
2467
+ "loss": 0.1656,
2468
+ "step": 8525
2469
+ },
2470
+ {
2471
+ "epoch": 14.109000825763832,
2472
+ "grad_norm": 5.636984348297119,
2473
+ "learning_rate": 3.056842105263158e-06,
2474
+ "loss": 0.1613,
2475
+ "step": 8550
2476
+ },
2477
+ {
2478
+ "epoch": 14.15028901734104,
2479
+ "grad_norm": 5.673081398010254,
2480
+ "learning_rate": 3.0042105263157894e-06,
2481
+ "loss": 0.1814,
2482
+ "step": 8575
2483
+ },
2484
+ {
2485
+ "epoch": 14.191577208918249,
2486
+ "grad_norm": 4.340714454650879,
2487
+ "learning_rate": 2.9515789473684214e-06,
2488
+ "loss": 0.1482,
2489
+ "step": 8600
2490
+ },
2491
+ {
2492
+ "epoch": 14.232865400495458,
2493
+ "grad_norm": 4.788120269775391,
2494
+ "learning_rate": 2.8989473684210526e-06,
2495
+ "loss": 0.1688,
2496
+ "step": 8625
2497
+ },
2498
+ {
2499
+ "epoch": 14.274153592072667,
2500
+ "grad_norm": 4.258752822875977,
2501
+ "learning_rate": 2.8463157894736843e-06,
2502
+ "loss": 0.1623,
2503
+ "step": 8650
2504
+ },
2505
+ {
2506
+ "epoch": 14.315441783649876,
2507
+ "grad_norm": 5.6167988777160645,
2508
+ "learning_rate": 2.7936842105263163e-06,
2509
+ "loss": 0.1736,
2510
+ "step": 8675
2511
+ },
2512
+ {
2513
+ "epoch": 14.356729975227084,
2514
+ "grad_norm": 4.973904609680176,
2515
+ "learning_rate": 2.7410526315789476e-06,
2516
+ "loss": 0.163,
2517
+ "step": 8700
2518
+ },
2519
+ {
2520
+ "epoch": 14.398018166804293,
2521
+ "grad_norm": 5.393108367919922,
2522
+ "learning_rate": 2.688421052631579e-06,
2523
+ "loss": 0.1721,
2524
+ "step": 8725
2525
+ },
2526
+ {
2527
+ "epoch": 14.439306358381502,
2528
+ "grad_norm": 3.455310583114624,
2529
+ "learning_rate": 2.635789473684211e-06,
2530
+ "loss": 0.1762,
2531
+ "step": 8750
2532
+ },
2533
+ {
2534
+ "epoch": 14.480594549958711,
2535
+ "grad_norm": 4.485291957855225,
2536
+ "learning_rate": 2.5831578947368425e-06,
2537
+ "loss": 0.167,
2538
+ "step": 8775
2539
+ },
2540
+ {
2541
+ "epoch": 14.52188274153592,
2542
+ "grad_norm": 5.023384094238281,
2543
+ "learning_rate": 2.5305263157894737e-06,
2544
+ "loss": 0.1797,
2545
+ "step": 8800
2546
+ },
2547
+ {
2548
+ "epoch": 14.56317093311313,
2549
+ "grad_norm": 3.6772608757019043,
2550
+ "learning_rate": 2.4778947368421053e-06,
2551
+ "loss": 0.1688,
2552
+ "step": 8825
2553
+ },
2554
+ {
2555
+ "epoch": 14.604459124690338,
2556
+ "grad_norm": 4.417513847351074,
2557
+ "learning_rate": 2.425263157894737e-06,
2558
+ "loss": 0.176,
2559
+ "step": 8850
2560
+ },
2561
+ {
2562
+ "epoch": 14.645747316267547,
2563
+ "grad_norm": 5.2892231941223145,
2564
+ "learning_rate": 2.3726315789473686e-06,
2565
+ "loss": 0.1802,
2566
+ "step": 8875
2567
+ },
2568
+ {
2569
+ "epoch": 14.687035507844756,
2570
+ "grad_norm": 5.4643025398254395,
2571
+ "learning_rate": 2.3200000000000002e-06,
2572
+ "loss": 0.1914,
2573
+ "step": 8900
2574
+ },
2575
+ {
2576
+ "epoch": 14.728323699421965,
2577
+ "grad_norm": 3.4271011352539062,
2578
+ "learning_rate": 2.267368421052632e-06,
2579
+ "loss": 0.1668,
2580
+ "step": 8925
2581
+ },
2582
+ {
2583
+ "epoch": 14.769611890999174,
2584
+ "grad_norm": 4.431731224060059,
2585
+ "learning_rate": 2.2147368421052635e-06,
2586
+ "loss": 0.1725,
2587
+ "step": 8950
2588
+ },
2589
+ {
2590
+ "epoch": 14.810900082576383,
2591
+ "grad_norm": 3.5097200870513916,
2592
+ "learning_rate": 2.1621052631578947e-06,
2593
+ "loss": 0.169,
2594
+ "step": 8975
2595
+ },
2596
+ {
2597
+ "epoch": 14.852188274153592,
2598
+ "grad_norm": 4.471765995025635,
2599
+ "learning_rate": 2.1094736842105264e-06,
2600
+ "loss": 0.1424,
2601
+ "step": 9000
2602
+ },
2603
+ {
2604
+ "epoch": 14.852188274153592,
2605
+ "eval_loss": 1.1418445110321045,
2606
+ "eval_runtime": 697.6196,
2607
+ "eval_samples_per_second": 15.167,
2608
+ "eval_steps_per_second": 1.264,
2609
+ "eval_wer": Infinity,
2610
+ "step": 9000
2611
+ },
2612
+ {
2613
+ "epoch": 14.893476465730801,
2614
+ "grad_norm": 4.103869915008545,
2615
+ "learning_rate": 2.056842105263158e-06,
2616
+ "loss": 0.1855,
2617
+ "step": 9025
2618
+ },
2619
+ {
2620
+ "epoch": 14.93476465730801,
2621
+ "grad_norm": 4.589334487915039,
2622
+ "learning_rate": 2.0042105263157896e-06,
2623
+ "loss": 0.1723,
2624
+ "step": 9050
2625
+ },
2626
+ {
2627
+ "epoch": 14.97605284888522,
2628
+ "grad_norm": 5.06859016418457,
2629
+ "learning_rate": 1.9515789473684213e-06,
2630
+ "loss": 0.1568,
2631
+ "step": 9075
2632
+ },
2633
+ {
2634
+ "epoch": 15.016515276630884,
2635
+ "grad_norm": 3.8599202632904053,
2636
+ "learning_rate": 1.8989473684210527e-06,
2637
+ "loss": 0.1802,
2638
+ "step": 9100
2639
+ },
2640
+ {
2641
+ "epoch": 15.057803468208093,
2642
+ "grad_norm": 3.51918363571167,
2643
+ "learning_rate": 1.8463157894736843e-06,
2644
+ "loss": 0.163,
2645
+ "step": 9125
2646
+ },
2647
+ {
2648
+ "epoch": 15.099091659785302,
2649
+ "grad_norm": 3.9094550609588623,
2650
+ "learning_rate": 1.7936842105263158e-06,
2651
+ "loss": 0.1508,
2652
+ "step": 9150
2653
+ },
2654
+ {
2655
+ "epoch": 15.14037985136251,
2656
+ "grad_norm": 4.491118907928467,
2657
+ "learning_rate": 1.7410526315789474e-06,
2658
+ "loss": 0.1683,
2659
+ "step": 9175
2660
+ },
2661
+ {
2662
+ "epoch": 15.18166804293972,
2663
+ "grad_norm": 5.513247966766357,
2664
+ "learning_rate": 1.6884210526315792e-06,
2665
+ "loss": 0.1413,
2666
+ "step": 9200
2667
+ },
2668
+ {
2669
+ "epoch": 15.222956234516928,
2670
+ "grad_norm": 4.3536577224731445,
2671
+ "learning_rate": 1.6357894736842107e-06,
2672
+ "loss": 0.1749,
2673
+ "step": 9225
2674
+ },
2675
+ {
2676
+ "epoch": 15.264244426094137,
2677
+ "grad_norm": 3.8934166431427,
2678
+ "learning_rate": 1.5831578947368423e-06,
2679
+ "loss": 0.1727,
2680
+ "step": 9250
2681
+ },
2682
+ {
2683
+ "epoch": 15.305532617671346,
2684
+ "grad_norm": 4.610677242279053,
2685
+ "learning_rate": 1.5305263157894737e-06,
2686
+ "loss": 0.1518,
2687
+ "step": 9275
2688
+ },
2689
+ {
2690
+ "epoch": 15.346820809248555,
2691
+ "grad_norm": 3.230677366256714,
2692
+ "learning_rate": 1.4778947368421054e-06,
2693
+ "loss": 0.1525,
2694
+ "step": 9300
2695
+ },
2696
+ {
2697
+ "epoch": 15.388109000825764,
2698
+ "grad_norm": 3.854262590408325,
2699
+ "learning_rate": 1.425263157894737e-06,
2700
+ "loss": 0.1501,
2701
+ "step": 9325
2702
+ },
2703
+ {
2704
+ "epoch": 15.429397192402973,
2705
+ "grad_norm": 3.96171498298645,
2706
+ "learning_rate": 1.3726315789473684e-06,
2707
+ "loss": 0.1655,
2708
+ "step": 9350
2709
+ },
2710
+ {
2711
+ "epoch": 15.470685383980182,
2712
+ "grad_norm": 3.8165667057037354,
2713
+ "learning_rate": 1.32e-06,
2714
+ "loss": 0.1469,
2715
+ "step": 9375
2716
+ },
2717
+ {
2718
+ "epoch": 15.511973575557391,
2719
+ "grad_norm": 3.6944658756256104,
2720
+ "learning_rate": 1.2673684210526315e-06,
2721
+ "loss": 0.15,
2722
+ "step": 9400
2723
+ },
2724
+ {
2725
+ "epoch": 15.5532617671346,
2726
+ "grad_norm": 3.9097495079040527,
2727
+ "learning_rate": 1.2147368421052633e-06,
2728
+ "loss": 0.1492,
2729
+ "step": 9425
2730
+ },
2731
+ {
2732
+ "epoch": 15.59454995871181,
2733
+ "grad_norm": 5.010981559753418,
2734
+ "learning_rate": 1.1621052631578948e-06,
2735
+ "loss": 0.1551,
2736
+ "step": 9450
2737
+ },
2738
+ {
2739
+ "epoch": 15.635838150289018,
2740
+ "grad_norm": 4.151124000549316,
2741
+ "learning_rate": 1.1094736842105264e-06,
2742
+ "loss": 0.1569,
2743
+ "step": 9475
2744
+ },
2745
+ {
2746
+ "epoch": 15.677126341866225,
2747
+ "grad_norm": 4.455909729003906,
2748
+ "learning_rate": 1.0568421052631578e-06,
2749
+ "loss": 0.1551,
2750
+ "step": 9500
2751
+ },
2752
+ {
2753
+ "epoch": 15.718414533443434,
2754
+ "grad_norm": 4.5624895095825195,
2755
+ "learning_rate": 1.0042105263157897e-06,
2756
+ "loss": 0.1526,
2757
+ "step": 9525
2758
+ },
2759
+ {
2760
+ "epoch": 15.759702725020643,
2761
+ "grad_norm": 3.476386070251465,
2762
+ "learning_rate": 9.515789473684212e-07,
2763
+ "loss": 0.151,
2764
+ "step": 9550
2765
+ },
2766
+ {
2767
+ "epoch": 15.800990916597852,
2768
+ "grad_norm": 3.7594966888427734,
2769
+ "learning_rate": 8.989473684210527e-07,
2770
+ "loss": 0.1614,
2771
+ "step": 9575
2772
+ },
2773
+ {
2774
+ "epoch": 15.842279108175061,
2775
+ "grad_norm": 3.677940607070923,
2776
+ "learning_rate": 8.463157894736843e-07,
2777
+ "loss": 0.1607,
2778
+ "step": 9600
2779
+ },
2780
+ {
2781
+ "epoch": 15.88356729975227,
2782
+ "grad_norm": 2.814276695251465,
2783
+ "learning_rate": 7.936842105263158e-07,
2784
+ "loss": 0.1584,
2785
+ "step": 9625
2786
+ },
2787
+ {
2788
+ "epoch": 15.92485549132948,
2789
+ "grad_norm": 4.551584243774414,
2790
+ "learning_rate": 7.410526315789475e-07,
2791
+ "loss": 0.1746,
2792
+ "step": 9650
2793
+ },
2794
+ {
2795
+ "epoch": 15.966143682906688,
2796
+ "grad_norm": 3.54571533203125,
2797
+ "learning_rate": 6.884210526315791e-07,
2798
+ "loss": 0.1283,
2799
+ "step": 9675
2800
+ },
2801
+ {
2802
+ "epoch": 16.006606110652353,
2803
+ "grad_norm": 3.034546136856079,
2804
+ "learning_rate": 6.357894736842106e-07,
2805
+ "loss": 0.1445,
2806
+ "step": 9700
2807
+ },
2808
+ {
2809
+ "epoch": 16.04789430222956,
2810
+ "grad_norm": 4.239781379699707,
2811
+ "learning_rate": 5.831578947368421e-07,
2812
+ "loss": 0.1526,
2813
+ "step": 9725
2814
+ },
2815
+ {
2816
+ "epoch": 16.08918249380677,
2817
+ "grad_norm": 3.5759620666503906,
2818
+ "learning_rate": 5.305263157894737e-07,
2819
+ "loss": 0.1435,
2820
+ "step": 9750
2821
+ },
2822
+ {
2823
+ "epoch": 16.13047068538398,
2824
+ "grad_norm": 3.807652473449707,
2825
+ "learning_rate": 4.778947368421053e-07,
2826
+ "loss": 0.1616,
2827
+ "step": 9775
2828
+ },
2829
+ {
2830
+ "epoch": 16.17175887696119,
2831
+ "grad_norm": 2.9478118419647217,
2832
+ "learning_rate": 4.2526315789473684e-07,
2833
+ "loss": 0.1344,
2834
+ "step": 9800
2835
+ },
2836
+ {
2837
+ "epoch": 16.213047068538398,
2838
+ "grad_norm": 3.351928472518921,
2839
+ "learning_rate": 3.726315789473685e-07,
2840
+ "loss": 0.1616,
2841
+ "step": 9825
2842
+ },
2843
+ {
2844
+ "epoch": 16.254335260115607,
2845
+ "grad_norm": 4.71775484085083,
2846
+ "learning_rate": 3.2e-07,
2847
+ "loss": 0.138,
2848
+ "step": 9850
2849
+ },
2850
+ {
2851
+ "epoch": 16.295623451692816,
2852
+ "grad_norm": 4.4986138343811035,
2853
+ "learning_rate": 2.6736842105263164e-07,
2854
+ "loss": 0.1592,
2855
+ "step": 9875
2856
+ },
2857
+ {
2858
+ "epoch": 16.336911643270025,
2859
+ "grad_norm": 4.285621643066406,
2860
+ "learning_rate": 2.1473684210526317e-07,
2861
+ "loss": 0.1423,
2862
+ "step": 9900
2863
+ },
2864
+ {
2865
+ "epoch": 16.378199834847234,
2866
+ "grad_norm": 4.092608451843262,
2867
+ "learning_rate": 1.6210526315789476e-07,
2868
+ "loss": 0.1619,
2869
+ "step": 9925
2870
+ },
2871
+ {
2872
+ "epoch": 16.419488026424442,
2873
+ "grad_norm": 4.384539604187012,
2874
+ "learning_rate": 1.0947368421052632e-07,
2875
+ "loss": 0.1606,
2876
+ "step": 9950
2877
+ },
2878
+ {
2879
+ "epoch": 16.46077621800165,
2880
+ "grad_norm": 5.559605121612549,
2881
+ "learning_rate": 5.68421052631579e-08,
2882
+ "loss": 0.164,
2883
+ "step": 9975
2884
+ },
2885
+ {
2886
+ "epoch": 16.50206440957886,
2887
+ "grad_norm": 4.197207450866699,
2888
+ "learning_rate": 4.210526315789474e-09,
2889
+ "loss": 0.1615,
2890
+ "step": 10000
2891
+ },
2892
+ {
2893
+ "epoch": 16.50206440957886,
2894
+ "eval_loss": 1.1515085697174072,
2895
+ "eval_runtime": 702.9423,
2896
+ "eval_samples_per_second": 15.052,
2897
+ "eval_steps_per_second": 1.255,
2898
+ "eval_wer": Infinity,
2899
+ "step": 10000
2900
+ },
2901
+ {
2902
+ "epoch": 16.50206440957886,
2903
+ "step": 10000,
2904
+ "total_flos": 1.949150849531904e+19,
2905
+ "train_loss": 0.8644906110763549,
2906
+ "train_runtime": 22776.5193,
2907
+ "train_samples_per_second": 21.074,
2908
+ "train_steps_per_second": 0.439
2909
+ }
2910
+ ],
2911
+ "logging_steps": 25,
2912
+ "max_steps": 10000,
2913
+ "num_input_tokens_seen": 0,
2914
+ "num_train_epochs": 17,
2915
+ "save_steps": 1000,
2916
+ "stateful_callbacks": {
2917
+ "TrainerControl": {
2918
+ "args": {
2919
+ "should_epoch_stop": false,
2920
+ "should_evaluate": false,
2921
+ "should_log": false,
2922
+ "should_save": true,
2923
+ "should_training_stop": true
2924
+ },
2925
+ "attributes": {}
2926
+ }
2927
+ },
2928
+ "total_flos": 1.949150849531904e+19,
2929
+ "train_batch_size": 12,
2930
+ "trial_name": null,
2931
+ "trial_params": null
2932
+ }