Hugging Face's logo

imvladikon
/
wav2vec2-xls-r-1b-hebrew

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
Hebrew
wav2vec2
robust-speech-event
generated_from_trainer
hf-asr-leaderboard
Inference Endpoints
Model card Files Metrics Community
3
imvladikon commited on
Commit
e55479e
•
1 Parent(s): 0db3bbd

init

Browse files
Files changed (18) hide show
  1. checkpoint-4000/config.json +79 -0
  2. checkpoint-4000/optimizer.pt +3 -0
  3. checkpoint-4000/preprocessor_config.json +9 -0
  4. checkpoint-4000/pytorch_model.bin +3 -0
  5. checkpoint-4000/rng_state.pth +3 -0
  6. checkpoint-4000/scaler.pt +3 -0
  7. checkpoint-4000/scheduler.pt +3 -0
  8. checkpoint-4000/trainer_state.json +166 -0
  9. checkpoint-4000/training_args.bin +3 -0
  10. checkpoint-4400/config.json +79 -0
  11. checkpoint-4400/optimizer.pt +3 -0
  12. checkpoint-4400/preprocessor_config.json +9 -0
  13. checkpoint-4400/pytorch_model.bin +3 -0
  14. checkpoint-4400/rng_state.pth +3 -0
  15. checkpoint-4400/scaler.pt +3 -0
  16. checkpoint-4400/scheduler.pt +3 -0
  17. checkpoint-4400/trainer_state.json +181 -0
  18. checkpoint-4400/training_args.bin +3 -0
checkpoint-4000/config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-xls-r-300m",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 768,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": true,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": true,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_dropout": 0.0,
48
+ "feat_extract_norm": "layer",
49
+ "feat_proj_dropout": 0.0,
50
+ "feat_quantizer_dropout": 0.0,
51
+ "final_dropout": 0.0,
52
+ "gradient_checkpointing": false,
53
+ "hidden_act": "gelu",
54
+ "hidden_dropout": 0.0,
55
+ "hidden_size": 1024,
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 4096,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_prob": 0.0,
62
+ "mask_time_length": 10,
63
+ "mask_time_prob": 0.05,
64
+ "model_type": "wav2vec2",
65
+ "num_attention_heads": 16,
66
+ "num_codevector_groups": 2,
67
+ "num_codevectors_per_group": 320,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 24,
72
+ "num_negatives": 100,
73
+ "pad_token_id": 29,
74
+ "proj_codevector_dim": 768,
75
+ "torch_dtype": "float32",
76
+ "transformers_version": "4.11.3",
77
+ "use_weighted_layer_sum": false,
78
+ "vocab_size": 32
79
+ }
checkpoint-4000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:052d9d1f443d1b9f67c2f7a6c2808fe922f1eb7ec26dba77c36c332395ec93d6
3
+ size 2490321361
checkpoint-4000/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-4000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:092b750aa2825af7fb89c41d0967aa326857a67ed0c101103b080ef3668b8ca5
3
+ size 1262054897
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bf4dfa62752c868edada56217703db94c074fedf843e434f438dfbdfd0009a0
3
+ size 14631
checkpoint-4000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93d4c53e2e4d886d0eac00fb8b6e3a12d9b4941d302b762706e463005cd1e4ad
3
+ size 559
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a20530e63990facdbe1c575c37ff4cc0d4637f57452c1d8f3b74093a7afbf98
3
+ size 623
checkpoint-4000/trainer_state.json ADDED
@@ -0,0 +1,166 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.5759306677171558,
5
+ "global_step": 4000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.16,
12
+ "learning_rate": 0.0002388,
13
+ "loss": 4.5334,
14
+ "step": 400
15
+ },
16
+ {
17
+ "epoch": 0.16,
18
+ "eval_loss": 2.026416063308716,
19
+ "eval_runtime": 471.8471,
20
+ "eval_samples_per_second": 10.758,
21
+ "eval_steps_per_second": 1.346,
22
+ "eval_wer": 0.9800369528744665,
23
+ "step": 400
24
+ },
25
+ {
26
+ "epoch": 0.32,
27
+ "learning_rate": 0.00029881808566895816,
28
+ "loss": 1.0808,
29
+ "step": 800
30
+ },
31
+ {
32
+ "epoch": 0.32,
33
+ "eval_loss": 0.7069017887115479,
34
+ "eval_runtime": 472.368,
35
+ "eval_samples_per_second": 10.746,
36
+ "eval_steps_per_second": 1.344,
37
+ "eval_wer": 0.593412194448574,
38
+ "step": 800
39
+ },
40
+ {
41
+ "epoch": 0.47,
42
+ "learning_rate": 0.0002972316234796404,
43
+ "loss": 0.7499,
44
+ "step": 1200
45
+ },
46
+ {
47
+ "epoch": 0.47,
48
+ "eval_loss": 0.5792935490608215,
49
+ "eval_runtime": 471.2762,
50
+ "eval_samples_per_second": 10.771,
51
+ "eval_steps_per_second": 1.347,
52
+ "eval_wer": 0.49153694225582434,
53
+ "step": 1200
54
+ },
55
+ {
56
+ "epoch": 0.63,
57
+ "learning_rate": 0.00029564516129032255,
58
+ "loss": 0.6397,
59
+ "step": 1600
60
+ },
61
+ {
62
+ "epoch": 0.63,
63
+ "eval_loss": 0.5181192755699158,
64
+ "eval_runtime": 470.1504,
65
+ "eval_samples_per_second": 10.797,
66
+ "eval_steps_per_second": 1.351,
67
+ "eval_wer": 0.41618705799902306,
68
+ "step": 1600
69
+ },
70
+ {
71
+ "epoch": 0.79,
72
+ "learning_rate": 0.0002940586991010047,
73
+ "loss": 0.597,
74
+ "step": 2000
75
+ },
76
+ {
77
+ "epoch": 0.79,
78
+ "eval_loss": 0.47279468178749084,
79
+ "eval_runtime": 470.9283,
80
+ "eval_samples_per_second": 10.779,
81
+ "eval_steps_per_second": 1.348,
82
+ "eval_wer": 0.37248072716461017,
83
+ "step": 2000
84
+ },
85
+ {
86
+ "epoch": 0.95,
87
+ "learning_rate": 0.00029247223691168694,
88
+ "loss": 0.5666,
89
+ "step": 2400
90
+ },
91
+ {
92
+ "epoch": 0.95,
93
+ "eval_loss": 0.4448830783367157,
94
+ "eval_runtime": 471.0174,
95
+ "eval_samples_per_second": 10.777,
96
+ "eval_steps_per_second": 1.348,
97
+ "eval_wer": 0.3520292225030263,
98
+ "step": 2400
99
+ },
100
+ {
101
+ "epoch": 1.1,
102
+ "learning_rate": 0.0002908857747223691,
103
+ "loss": 0.5067,
104
+ "step": 2800
105
+ },
106
+ {
107
+ "epoch": 1.1,
108
+ "eval_loss": 0.421601802110672,
109
+ "eval_runtime": 470.0998,
110
+ "eval_samples_per_second": 10.798,
111
+ "eval_steps_per_second": 1.351,
112
+ "eval_wer": 0.3330855650179455,
113
+ "step": 2800
114
+ },
115
+ {
116
+ "epoch": 1.26,
117
+ "learning_rate": 0.0002892993125330513,
118
+ "loss": 0.477,
119
+ "step": 3200
120
+ },
121
+ {
122
+ "epoch": 1.26,
123
+ "eval_loss": 0.39891988039016724,
124
+ "eval_runtime": 471.7321,
125
+ "eval_samples_per_second": 10.76,
126
+ "eval_steps_per_second": 1.346,
127
+ "eval_wer": 0.30331089260305394,
128
+ "step": 3200
129
+ },
130
+ {
131
+ "epoch": 1.42,
132
+ "learning_rate": 0.00028771285034373344,
133
+ "loss": 0.4633,
134
+ "step": 3600
135
+ },
136
+ {
137
+ "epoch": 1.42,
138
+ "eval_loss": 0.41667959094047546,
139
+ "eval_runtime": 473.0088,
140
+ "eval_samples_per_second": 10.731,
141
+ "eval_steps_per_second": 1.342,
142
+ "eval_wer": 0.3046276042219721,
143
+ "step": 3600
144
+ },
145
+ {
146
+ "epoch": 1.58,
147
+ "learning_rate": 0.0002861263881544156,
148
+ "loss": 0.4606,
149
+ "step": 4000
150
+ },
151
+ {
152
+ "epoch": 1.58,
153
+ "eval_loss": 0.3892616033554077,
154
+ "eval_runtime": 473.5469,
155
+ "eval_samples_per_second": 10.719,
156
+ "eval_steps_per_second": 1.341,
157
+ "eval_wer": 0.29747063945462654,
158
+ "step": 4000
159
+ }
160
+ ],
161
+ "max_steps": 76140,
162
+ "num_train_epochs": 30,
163
+ "total_flos": 5.025867257304392e+18,
164
+ "trial_name": null,
165
+ "trial_params": null
166
+ }
checkpoint-4000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4c1c368c4db40de7b0ac3ee4ff79af2d311395b8ce794d1869af91457b7e050
3
+ size 2863
checkpoint-4400/config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-xls-r-300m",
3
+ "activation_dropout": 0.0,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 768,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": true,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": true,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_dropout": 0.0,
48
+ "feat_extract_norm": "layer",
49
+ "feat_proj_dropout": 0.0,
50
+ "feat_quantizer_dropout": 0.0,
51
+ "final_dropout": 0.0,
52
+ "gradient_checkpointing": false,
53
+ "hidden_act": "gelu",
54
+ "hidden_dropout": 0.0,
55
+ "hidden_size": 1024,
56
+ "initializer_range": 0.02,
57
+ "intermediate_size": 4096,
58
+ "layer_norm_eps": 1e-05,
59
+ "layerdrop": 0.0,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_prob": 0.0,
62
+ "mask_time_length": 10,
63
+ "mask_time_prob": 0.05,
64
+ "model_type": "wav2vec2",
65
+ "num_attention_heads": 16,
66
+ "num_codevector_groups": 2,
67
+ "num_codevectors_per_group": 320,
68
+ "num_conv_pos_embedding_groups": 16,
69
+ "num_conv_pos_embeddings": 128,
70
+ "num_feat_extract_layers": 7,
71
+ "num_hidden_layers": 24,
72
+ "num_negatives": 100,
73
+ "pad_token_id": 29,
74
+ "proj_codevector_dim": 768,
75
+ "torch_dtype": "float32",
76
+ "transformers_version": "4.11.3",
77
+ "use_weighted_layer_sum": false,
78
+ "vocab_size": 32
79
+ }
checkpoint-4400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10430c88cb27d80d4909d60b1f5b4649806024e17f499620af7b33e594972686
3
+ size 2490321361
checkpoint-4400/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-4400/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f6b19c2693a520d464d36c22f9e97cab05ee230f04af2c84cb79a18b55c8190
3
+ size 1262054897
checkpoint-4400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f850e9eef43b3d8d7daa339e72e185c9b03f2fbdfe48df592252aedbe1875965
3
+ size 14567
checkpoint-4400/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4affccbe68b9032483a9a189417afa898f74547fed0e8eff9cfb5b00deb58885
3
+ size 559
checkpoint-4400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e11bf57746f56037dc02c0f1b9ef426c313cafb919b3099b8b7f85426ae9e2d5
3
+ size 623
checkpoint-4400/trainer_state.json ADDED
@@ -0,0 +1,181 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.733504037817609,
5
+ "global_step": 4400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.16,
12
+ "learning_rate": 0.0002388,
13
+ "loss": 4.5334,
14
+ "step": 400
15
+ },
16
+ {
17
+ "epoch": 0.16,
18
+ "eval_loss": 2.026416063308716,
19
+ "eval_runtime": 471.8471,
20
+ "eval_samples_per_second": 10.758,
21
+ "eval_steps_per_second": 1.346,
22
+ "eval_wer": 0.9800369528744665,
23
+ "step": 400
24
+ },
25
+ {
26
+ "epoch": 0.32,
27
+ "learning_rate": 0.00029881808566895816,
28
+ "loss": 1.0808,
29
+ "step": 800
30
+ },
31
+ {
32
+ "epoch": 0.32,
33
+ "eval_loss": 0.7069017887115479,
34
+ "eval_runtime": 472.368,
35
+ "eval_samples_per_second": 10.746,
36
+ "eval_steps_per_second": 1.344,
37
+ "eval_wer": 0.593412194448574,
38
+ "step": 800
39
+ },
40
+ {
41
+ "epoch": 0.47,
42
+ "learning_rate": 0.0002972316234796404,
43
+ "loss": 0.7499,
44
+ "step": 1200
45
+ },
46
+ {
47
+ "epoch": 0.47,
48
+ "eval_loss": 0.5792935490608215,
49
+ "eval_runtime": 471.2762,
50
+ "eval_samples_per_second": 10.771,
51
+ "eval_steps_per_second": 1.347,
52
+ "eval_wer": 0.49153694225582434,
53
+ "step": 1200
54
+ },
55
+ {
56
+ "epoch": 0.63,
57
+ "learning_rate": 0.00029564516129032255,
58
+ "loss": 0.6397,
59
+ "step": 1600
60
+ },
61
+ {
62
+ "epoch": 0.63,
63
+ "eval_loss": 0.5181192755699158,
64
+ "eval_runtime": 470.1504,
65
+ "eval_samples_per_second": 10.797,
66
+ "eval_steps_per_second": 1.351,
67
+ "eval_wer": 0.41618705799902306,
68
+ "step": 1600
69
+ },
70
+ {
71
+ "epoch": 0.79,
72
+ "learning_rate": 0.0002940586991010047,
73
+ "loss": 0.597,
74
+ "step": 2000
75
+ },
76
+ {
77
+ "epoch": 0.79,
78
+ "eval_loss": 0.47279468178749084,
79
+ "eval_runtime": 470.9283,
80
+ "eval_samples_per_second": 10.779,
81
+ "eval_steps_per_second": 1.348,
82
+ "eval_wer": 0.37248072716461017,
83
+ "step": 2000
84
+ },
85
+ {
86
+ "epoch": 0.95,
87
+ "learning_rate": 0.00029247223691168694,
88
+ "loss": 0.5666,
89
+ "step": 2400
90
+ },
91
+ {
92
+ "epoch": 0.95,
93
+ "eval_loss": 0.4448830783367157,
94
+ "eval_runtime": 471.0174,
95
+ "eval_samples_per_second": 10.777,
96
+ "eval_steps_per_second": 1.348,
97
+ "eval_wer": 0.3520292225030263,
98
+ "step": 2400
99
+ },
100
+ {
101
+ "epoch": 1.1,
102
+ "learning_rate": 0.0002908857747223691,
103
+ "loss": 0.5067,
104
+ "step": 2800
105
+ },
106
+ {
107
+ "epoch": 1.1,
108
+ "eval_loss": 0.421601802110672,
109
+ "eval_runtime": 470.0998,
110
+ "eval_samples_per_second": 10.798,
111
+ "eval_steps_per_second": 1.351,
112
+ "eval_wer": 0.3330855650179455,
113
+ "step": 2800
114
+ },
115
+ {
116
+ "epoch": 1.26,
117
+ "learning_rate": 0.0002892993125330513,
118
+ "loss": 0.477,
119
+ "step": 3200
120
+ },
121
+ {
122
+ "epoch": 1.26,
123
+ "eval_loss": 0.39891988039016724,
124
+ "eval_runtime": 471.7321,
125
+ "eval_samples_per_second": 10.76,
126
+ "eval_steps_per_second": 1.346,
127
+ "eval_wer": 0.30331089260305394,
128
+ "step": 3200
129
+ },
130
+ {
131
+ "epoch": 1.42,
132
+ "learning_rate": 0.00028771285034373344,
133
+ "loss": 0.4633,
134
+ "step": 3600
135
+ },
136
+ {
137
+ "epoch": 1.42,
138
+ "eval_loss": 0.41667959094047546,
139
+ "eval_runtime": 473.0088,
140
+ "eval_samples_per_second": 10.731,
141
+ "eval_steps_per_second": 1.342,
142
+ "eval_wer": 0.3046276042219721,
143
+ "step": 3600
144
+ },
145
+ {
146
+ "epoch": 1.58,
147
+ "learning_rate": 0.0002861263881544156,
148
+ "loss": 0.4606,
149
+ "step": 4000
150
+ },
151
+ {
152
+ "epoch": 1.58,
153
+ "eval_loss": 0.3892616033554077,
154
+ "eval_runtime": 473.5469,
155
+ "eval_samples_per_second": 10.719,
156
+ "eval_steps_per_second": 1.341,
157
+ "eval_wer": 0.29747063945462654,
158
+ "step": 4000
159
+ },
160
+ {
161
+ "epoch": 1.73,
162
+ "learning_rate": 0.0002845399259650978,
163
+ "loss": 0.4385,
164
+ "step": 4400
165
+ },
166
+ {
167
+ "epoch": 1.73,
168
+ "eval_loss": 0.3867528438568115,
169
+ "eval_runtime": 471.6975,
170
+ "eval_samples_per_second": 10.761,
171
+ "eval_steps_per_second": 1.346,
172
+ "eval_wer": 0.3090237220464247,
173
+ "step": 4400
174
+ }
175
+ ],
176
+ "max_steps": 76140,
177
+ "num_train_epochs": 30,
178
+ "total_flos": 5.530258185352643e+18,
179
+ "trial_name": null,
180
+ "trial_params": null
181
+ }
checkpoint-4400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4c1c368c4db40de7b0ac3ee4ff79af2d311395b8ce794d1869af91457b7e050
3
+ size 2863