mikhail-panzo commited on
Commit
b6ef785
1 Parent(s): 0abae19

Training in progress, step 8500, checkpoint

Browse files
checkpoint-8500/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-8500/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
+ "torch_dtype": "float32",
88
+ "transformers_version": "4.41.0.dev0",
89
+ "use_cache": false,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
92
+ }
checkpoint-8500/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "eos_token_id": 2,
6
+ "max_length": 1876,
7
+ "pad_token_id": 1,
8
+ "transformers_version": "4.41.0.dev0"
9
+ }
checkpoint-8500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7e6d765d54fba24a261941c48db3f45d151cc3543e4620fc74841008e9d38c4
3
+ size 577789320
checkpoint-8500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4932bfcabceeaabb9f98d9c6304c48dc2f667000d0fff35bc9e76838ea3c811a
3
+ size 1155772233
checkpoint-8500/preprocessor_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": false,
3
+ "feature_extractor_type": "SpeechT5FeatureExtractor",
4
+ "feature_size": 1,
5
+ "fmax": 7600,
6
+ "fmin": 80,
7
+ "frame_signal_scale": 1.0,
8
+ "hop_length": 16,
9
+ "mel_floor": 1e-10,
10
+ "num_mel_bins": 80,
11
+ "padding_side": "right",
12
+ "padding_value": 0.0,
13
+ "processor_class": "SpeechT5Processor",
14
+ "reduction_factor": 2,
15
+ "return_attention_mask": true,
16
+ "sampling_rate": 16000,
17
+ "win_function": "hann_window",
18
+ "win_length": 64
19
+ }
checkpoint-8500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0ebb5d5ae294e81ab1a2faa6ce3a79a3d8cd9922b5cd8c1a74072b1685a262f
3
+ size 14244
checkpoint-8500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5efdf488bd79f4078dcc22ed21cb113945810568eae9749dc0cb12dc690695b7
3
+ size 1064
checkpoint-8500/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-8500/spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
checkpoint-8500/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": true,
54
+ "eos_token": "</s>",
55
+ "mask_token": "<mask>",
56
+ "model_max_length": 600,
57
+ "normalize": false,
58
+ "pad_token": "<pad>",
59
+ "processor_class": "SpeechT5Processor",
60
+ "sp_model_kwargs": {},
61
+ "tokenizer_class": "SpeechT5Tokenizer",
62
+ "unk_token": "<unk>"
63
+ }
checkpoint-8500/trainer_state.json ADDED
@@ -0,0 +1,1359 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.31820428371429443,
3
+ "best_model_checkpoint": "mikhail_panzo/zlm_b64_le4_s12000/checkpoint-8500",
4
+ "epoch": 7.11892797319933,
5
+ "eval_steps": 500,
6
+ "global_step": 8500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04187604690117253,
13
+ "grad_norm": 3.0184223651885986,
14
+ "learning_rate": 2.35e-06,
15
+ "loss": 1.1206,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.08375209380234507,
20
+ "grad_norm": 3.4442005157470703,
21
+ "learning_rate": 4.85e-06,
22
+ "loss": 0.8185,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.12562814070351758,
27
+ "grad_norm": 2.6812493801116943,
28
+ "learning_rate": 7.35e-06,
29
+ "loss": 0.7576,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.16750418760469013,
34
+ "grad_norm": 4.518482685089111,
35
+ "learning_rate": 9.85e-06,
36
+ "loss": 0.6938,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.20938023450586266,
41
+ "grad_norm": 4.683146953582764,
42
+ "learning_rate": 1.235e-05,
43
+ "loss": 0.6386,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.25125628140703515,
48
+ "grad_norm": 2.7392578125,
49
+ "learning_rate": 1.485e-05,
50
+ "loss": 0.6226,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.2931323283082077,
55
+ "grad_norm": 2.017056465148926,
56
+ "learning_rate": 1.7349999999999998e-05,
57
+ "loss": 0.6128,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.33500837520938026,
62
+ "grad_norm": 3.6669230461120605,
63
+ "learning_rate": 1.985e-05,
64
+ "loss": 0.5675,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.3768844221105528,
69
+ "grad_norm": 4.16896915435791,
70
+ "learning_rate": 2.235e-05,
71
+ "loss": 0.5655,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.4187604690117253,
76
+ "grad_norm": 3.691459894180298,
77
+ "learning_rate": 2.485e-05,
78
+ "loss": 0.5487,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.4187604690117253,
83
+ "eval_loss": 0.47458764910697937,
84
+ "eval_runtime": 214.7196,
85
+ "eval_samples_per_second": 39.535,
86
+ "eval_steps_per_second": 4.946,
87
+ "step": 500
88
+ },
89
+ {
90
+ "epoch": 0.46063651591289784,
91
+ "grad_norm": 3.3792996406555176,
92
+ "learning_rate": 2.7350000000000004e-05,
93
+ "loss": 0.5232,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 0.5025125628140703,
98
+ "grad_norm": 3.6263363361358643,
99
+ "learning_rate": 2.985e-05,
100
+ "loss": 0.5132,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 0.5443886097152428,
105
+ "grad_norm": 2.652538776397705,
106
+ "learning_rate": 3.235e-05,
107
+ "loss": 0.5067,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 0.5862646566164154,
112
+ "grad_norm": 3.5355172157287598,
113
+ "learning_rate": 3.485e-05,
114
+ "loss": 0.4981,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 0.628140703517588,
119
+ "grad_norm": 2.5035619735717773,
120
+ "learning_rate": 3.735e-05,
121
+ "loss": 0.5094,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 0.6700167504187605,
126
+ "grad_norm": 2.9585518836975098,
127
+ "learning_rate": 3.9850000000000006e-05,
128
+ "loss": 0.5053,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 0.711892797319933,
133
+ "grad_norm": 9.997116088867188,
134
+ "learning_rate": 4.235e-05,
135
+ "loss": 0.502,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 0.7537688442211056,
140
+ "grad_norm": 1.8630377054214478,
141
+ "learning_rate": 4.4850000000000006e-05,
142
+ "loss": 0.483,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 0.7956448911222781,
147
+ "grad_norm": 1.5534099340438843,
148
+ "learning_rate": 4.735e-05,
149
+ "loss": 0.4743,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 0.8375209380234506,
154
+ "grad_norm": 2.8155503273010254,
155
+ "learning_rate": 4.9850000000000006e-05,
156
+ "loss": 0.483,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 0.8375209380234506,
161
+ "eval_loss": 0.4226582944393158,
162
+ "eval_runtime": 213.2339,
163
+ "eval_samples_per_second": 39.811,
164
+ "eval_steps_per_second": 4.98,
165
+ "step": 1000
166
+ },
167
+ {
168
+ "epoch": 0.8793969849246231,
169
+ "grad_norm": 4.2074055671691895,
170
+ "learning_rate": 5.235e-05,
171
+ "loss": 0.4625,
172
+ "step": 1050
173
+ },
174
+ {
175
+ "epoch": 0.9212730318257957,
176
+ "grad_norm": 2.215416431427002,
177
+ "learning_rate": 5.485e-05,
178
+ "loss": 0.461,
179
+ "step": 1100
180
+ },
181
+ {
182
+ "epoch": 0.9631490787269682,
183
+ "grad_norm": 3.3321831226348877,
184
+ "learning_rate": 5.7350000000000005e-05,
185
+ "loss": 0.4807,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 1.0050251256281406,
190
+ "grad_norm": 3.5796420574188232,
191
+ "learning_rate": 5.9850000000000005e-05,
192
+ "loss": 0.4456,
193
+ "step": 1200
194
+ },
195
+ {
196
+ "epoch": 1.0469011725293131,
197
+ "grad_norm": 2.3565518856048584,
198
+ "learning_rate": 6.235000000000001e-05,
199
+ "loss": 0.4664,
200
+ "step": 1250
201
+ },
202
+ {
203
+ "epoch": 1.0887772194304857,
204
+ "grad_norm": 3.147794485092163,
205
+ "learning_rate": 6.485e-05,
206
+ "loss": 0.4681,
207
+ "step": 1300
208
+ },
209
+ {
210
+ "epoch": 1.1306532663316582,
211
+ "grad_norm": 3.028409957885742,
212
+ "learning_rate": 6.735e-05,
213
+ "loss": 0.4663,
214
+ "step": 1350
215
+ },
216
+ {
217
+ "epoch": 1.1725293132328307,
218
+ "grad_norm": 4.820915222167969,
219
+ "learning_rate": 6.985e-05,
220
+ "loss": 0.4493,
221
+ "step": 1400
222
+ },
223
+ {
224
+ "epoch": 1.2144053601340032,
225
+ "grad_norm": 1.93254816532135,
226
+ "learning_rate": 7.235000000000001e-05,
227
+ "loss": 0.4413,
228
+ "step": 1450
229
+ },
230
+ {
231
+ "epoch": 1.2562814070351758,
232
+ "grad_norm": 2.237015724182129,
233
+ "learning_rate": 7.485e-05,
234
+ "loss": 0.432,
235
+ "step": 1500
236
+ },
237
+ {
238
+ "epoch": 1.2562814070351758,
239
+ "eval_loss": 0.39829039573669434,
240
+ "eval_runtime": 217.3782,
241
+ "eval_samples_per_second": 39.052,
242
+ "eval_steps_per_second": 4.885,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 1.2981574539363483,
247
+ "grad_norm": 1.709120273590088,
248
+ "learning_rate": 7.735e-05,
249
+ "loss": 0.4361,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 1.3400335008375208,
254
+ "grad_norm": 3.733328104019165,
255
+ "learning_rate": 7.985e-05,
256
+ "loss": 0.4357,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 1.3819095477386933,
261
+ "grad_norm": 4.131324768066406,
262
+ "learning_rate": 8.235000000000001e-05,
263
+ "loss": 0.4445,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 1.4237855946398659,
268
+ "grad_norm": 2.2893927097320557,
269
+ "learning_rate": 8.485e-05,
270
+ "loss": 0.4379,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 1.4656616415410384,
275
+ "grad_norm": 2.1464734077453613,
276
+ "learning_rate": 8.735000000000001e-05,
277
+ "loss": 0.4393,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 1.507537688442211,
282
+ "grad_norm": 2.01698637008667,
283
+ "learning_rate": 8.985e-05,
284
+ "loss": 0.4341,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 1.5494137353433834,
289
+ "grad_norm": 2.259115219116211,
290
+ "learning_rate": 9.235000000000001e-05,
291
+ "loss": 0.4324,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 1.591289782244556,
296
+ "grad_norm": 2.0072858333587646,
297
+ "learning_rate": 9.485e-05,
298
+ "loss": 0.4327,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 1.6331658291457285,
303
+ "grad_norm": 2.060303211212158,
304
+ "learning_rate": 9.735000000000001e-05,
305
+ "loss": 0.4279,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 1.675041876046901,
310
+ "grad_norm": 1.2846741676330566,
311
+ "learning_rate": 9.985000000000001e-05,
312
+ "loss": 0.429,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 1.675041876046901,
317
+ "eval_loss": 0.3953319191932678,
318
+ "eval_runtime": 217.8108,
319
+ "eval_samples_per_second": 38.974,
320
+ "eval_steps_per_second": 4.876,
321
+ "step": 2000
322
+ },
323
+ {
324
+ "epoch": 1.7169179229480735,
325
+ "grad_norm": 2.253840923309326,
326
+ "learning_rate": 9.953e-05,
327
+ "loss": 0.4264,
328
+ "step": 2050
329
+ },
330
+ {
331
+ "epoch": 1.758793969849246,
332
+ "grad_norm": 1.3114299774169922,
333
+ "learning_rate": 9.903e-05,
334
+ "loss": 0.4355,
335
+ "step": 2100
336
+ },
337
+ {
338
+ "epoch": 1.8006700167504186,
339
+ "grad_norm": 1.8178610801696777,
340
+ "learning_rate": 9.853e-05,
341
+ "loss": 0.4256,
342
+ "step": 2150
343
+ },
344
+ {
345
+ "epoch": 1.8425460636515911,
346
+ "grad_norm": 2.1193735599517822,
347
+ "learning_rate": 9.803e-05,
348
+ "loss": 0.4305,
349
+ "step": 2200
350
+ },
351
+ {
352
+ "epoch": 1.8844221105527639,
353
+ "grad_norm": 1.6871263980865479,
354
+ "learning_rate": 9.753e-05,
355
+ "loss": 0.4106,
356
+ "step": 2250
357
+ },
358
+ {
359
+ "epoch": 1.9262981574539364,
360
+ "grad_norm": 1.2699921131134033,
361
+ "learning_rate": 9.703000000000001e-05,
362
+ "loss": 0.4156,
363
+ "step": 2300
364
+ },
365
+ {
366
+ "epoch": 1.968174204355109,
367
+ "grad_norm": 1.7966769933700562,
368
+ "learning_rate": 9.653000000000001e-05,
369
+ "loss": 0.4257,
370
+ "step": 2350
371
+ },
372
+ {
373
+ "epoch": 2.0100502512562812,
374
+ "grad_norm": 1.7838683128356934,
375
+ "learning_rate": 9.603000000000001e-05,
376
+ "loss": 0.4156,
377
+ "step": 2400
378
+ },
379
+ {
380
+ "epoch": 2.0519262981574538,
381
+ "grad_norm": 1.4161527156829834,
382
+ "learning_rate": 9.553e-05,
383
+ "loss": 0.4096,
384
+ "step": 2450
385
+ },
386
+ {
387
+ "epoch": 2.0938023450586263,
388
+ "grad_norm": 1.5864896774291992,
389
+ "learning_rate": 9.503e-05,
390
+ "loss": 0.4168,
391
+ "step": 2500
392
+ },
393
+ {
394
+ "epoch": 2.0938023450586263,
395
+ "eval_loss": 0.3701198697090149,
396
+ "eval_runtime": 220.4399,
397
+ "eval_samples_per_second": 38.509,
398
+ "eval_steps_per_second": 4.818,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 2.135678391959799,
403
+ "grad_norm": 1.2460856437683105,
404
+ "learning_rate": 9.453e-05,
405
+ "loss": 0.415,
406
+ "step": 2550
407
+ },
408
+ {
409
+ "epoch": 2.1775544388609713,
410
+ "grad_norm": 1.994315266609192,
411
+ "learning_rate": 9.403e-05,
412
+ "loss": 0.4075,
413
+ "step": 2600
414
+ },
415
+ {
416
+ "epoch": 2.219430485762144,
417
+ "grad_norm": 2.148324966430664,
418
+ "learning_rate": 9.353000000000001e-05,
419
+ "loss": 0.4077,
420
+ "step": 2650
421
+ },
422
+ {
423
+ "epoch": 2.2613065326633164,
424
+ "grad_norm": 1.7242035865783691,
425
+ "learning_rate": 9.303000000000001e-05,
426
+ "loss": 0.4044,
427
+ "step": 2700
428
+ },
429
+ {
430
+ "epoch": 2.3031825795644894,
431
+ "grad_norm": 1.912110447883606,
432
+ "learning_rate": 9.253000000000001e-05,
433
+ "loss": 0.4075,
434
+ "step": 2750
435
+ },
436
+ {
437
+ "epoch": 2.3450586264656614,
438
+ "grad_norm": 1.3661189079284668,
439
+ "learning_rate": 9.203000000000001e-05,
440
+ "loss": 0.4041,
441
+ "step": 2800
442
+ },
443
+ {
444
+ "epoch": 2.3869346733668344,
445
+ "grad_norm": 1.7083531618118286,
446
+ "learning_rate": 9.153e-05,
447
+ "loss": 0.396,
448
+ "step": 2850
449
+ },
450
+ {
451
+ "epoch": 2.4288107202680065,
452
+ "grad_norm": 1.7829670906066895,
453
+ "learning_rate": 9.103e-05,
454
+ "loss": 0.4042,
455
+ "step": 2900
456
+ },
457
+ {
458
+ "epoch": 2.4706867671691795,
459
+ "grad_norm": 2.188753366470337,
460
+ "learning_rate": 9.053e-05,
461
+ "loss": 0.3963,
462
+ "step": 2950
463
+ },
464
+ {
465
+ "epoch": 2.5125628140703515,
466
+ "grad_norm": 1.2608036994934082,
467
+ "learning_rate": 9.003e-05,
468
+ "loss": 0.4021,
469
+ "step": 3000
470
+ },
471
+ {
472
+ "epoch": 2.5125628140703515,
473
+ "eval_loss": 0.3613213300704956,
474
+ "eval_runtime": 214.7226,
475
+ "eval_samples_per_second": 39.535,
476
+ "eval_steps_per_second": 4.946,
477
+ "step": 3000
478
+ },
479
+ {
480
+ "epoch": 2.5544388609715245,
481
+ "grad_norm": 1.501607060432434,
482
+ "learning_rate": 8.953e-05,
483
+ "loss": 0.3897,
484
+ "step": 3050
485
+ },
486
+ {
487
+ "epoch": 2.5963149078726966,
488
+ "grad_norm": 1.321588397026062,
489
+ "learning_rate": 8.903e-05,
490
+ "loss": 0.3948,
491
+ "step": 3100
492
+ },
493
+ {
494
+ "epoch": 2.6381909547738696,
495
+ "grad_norm": 1.5581566095352173,
496
+ "learning_rate": 8.853000000000001e-05,
497
+ "loss": 0.4005,
498
+ "step": 3150
499
+ },
500
+ {
501
+ "epoch": 2.6800670016750416,
502
+ "grad_norm": 1.3824856281280518,
503
+ "learning_rate": 8.803e-05,
504
+ "loss": 0.3981,
505
+ "step": 3200
506
+ },
507
+ {
508
+ "epoch": 2.7219430485762146,
509
+ "grad_norm": 1.7730051279067993,
510
+ "learning_rate": 8.753e-05,
511
+ "loss": 0.3982,
512
+ "step": 3250
513
+ },
514
+ {
515
+ "epoch": 2.7638190954773867,
516
+ "grad_norm": 1.516622543334961,
517
+ "learning_rate": 8.703e-05,
518
+ "loss": 0.3906,
519
+ "step": 3300
520
+ },
521
+ {
522
+ "epoch": 2.8056951423785597,
523
+ "grad_norm": 1.2814849615097046,
524
+ "learning_rate": 8.653e-05,
525
+ "loss": 0.3859,
526
+ "step": 3350
527
+ },
528
+ {
529
+ "epoch": 2.8475711892797317,
530
+ "grad_norm": 1.239619255065918,
531
+ "learning_rate": 8.603e-05,
532
+ "loss": 0.3942,
533
+ "step": 3400
534
+ },
535
+ {
536
+ "epoch": 2.8894472361809047,
537
+ "grad_norm": 2.24064040184021,
538
+ "learning_rate": 8.553e-05,
539
+ "loss": 0.3938,
540
+ "step": 3450
541
+ },
542
+ {
543
+ "epoch": 2.931323283082077,
544
+ "grad_norm": 1.1459012031555176,
545
+ "learning_rate": 8.503e-05,
546
+ "loss": 0.3925,
547
+ "step": 3500
548
+ },
549
+ {
550
+ "epoch": 2.931323283082077,
551
+ "eval_loss": 0.35087427496910095,
552
+ "eval_runtime": 217.1018,
553
+ "eval_samples_per_second": 39.101,
554
+ "eval_steps_per_second": 4.892,
555
+ "step": 3500
556
+ },
557
+ {
558
+ "epoch": 2.9731993299832498,
559
+ "grad_norm": 1.8844960927963257,
560
+ "learning_rate": 8.453e-05,
561
+ "loss": 0.3803,
562
+ "step": 3550
563
+ },
564
+ {
565
+ "epoch": 3.0150753768844223,
566
+ "grad_norm": 1.5738004446029663,
567
+ "learning_rate": 8.403000000000001e-05,
568
+ "loss": 0.3895,
569
+ "step": 3600
570
+ },
571
+ {
572
+ "epoch": 3.056951423785595,
573
+ "grad_norm": 1.19874906539917,
574
+ "learning_rate": 8.353000000000001e-05,
575
+ "loss": 0.3902,
576
+ "step": 3650
577
+ },
578
+ {
579
+ "epoch": 3.0988274706867673,
580
+ "grad_norm": 2.4810609817504883,
581
+ "learning_rate": 8.303000000000001e-05,
582
+ "loss": 0.3863,
583
+ "step": 3700
584
+ },
585
+ {
586
+ "epoch": 3.14070351758794,
587
+ "grad_norm": 1.4734585285186768,
588
+ "learning_rate": 8.253000000000001e-05,
589
+ "loss": 0.391,
590
+ "step": 3750
591
+ },
592
+ {
593
+ "epoch": 3.1825795644891124,
594
+ "grad_norm": 1.6449118852615356,
595
+ "learning_rate": 8.203000000000001e-05,
596
+ "loss": 0.3888,
597
+ "step": 3800
598
+ },
599
+ {
600
+ "epoch": 3.224455611390285,
601
+ "grad_norm": 2.5607736110687256,
602
+ "learning_rate": 8.153e-05,
603
+ "loss": 0.3893,
604
+ "step": 3850
605
+ },
606
+ {
607
+ "epoch": 3.2663316582914574,
608
+ "grad_norm": 1.0670469999313354,
609
+ "learning_rate": 8.103e-05,
610
+ "loss": 0.3853,
611
+ "step": 3900
612
+ },
613
+ {
614
+ "epoch": 3.30820770519263,
615
+ "grad_norm": 2.191486120223999,
616
+ "learning_rate": 8.053e-05,
617
+ "loss": 0.3832,
618
+ "step": 3950
619
+ },
620
+ {
621
+ "epoch": 3.3500837520938025,
622
+ "grad_norm": 1.0259532928466797,
623
+ "learning_rate": 8.003e-05,
624
+ "loss": 0.3839,
625
+ "step": 4000
626
+ },
627
+ {
628
+ "epoch": 3.3500837520938025,
629
+ "eval_loss": 0.35058465600013733,
630
+ "eval_runtime": 214.7925,
631
+ "eval_samples_per_second": 39.522,
632
+ "eval_steps_per_second": 4.944,
633
+ "step": 4000
634
+ },
635
+ {
636
+ "epoch": 3.391959798994975,
637
+ "grad_norm": 2.1106534004211426,
638
+ "learning_rate": 7.953e-05,
639
+ "loss": 0.4088,
640
+ "step": 4050
641
+ },
642
+ {
643
+ "epoch": 3.4338358458961475,
644
+ "grad_norm": 1.7350833415985107,
645
+ "learning_rate": 7.903000000000001e-05,
646
+ "loss": 0.3923,
647
+ "step": 4100
648
+ },
649
+ {
650
+ "epoch": 3.47571189279732,
651
+ "grad_norm": 1.1461760997772217,
652
+ "learning_rate": 7.853000000000001e-05,
653
+ "loss": 0.3924,
654
+ "step": 4150
655
+ },
656
+ {
657
+ "epoch": 3.5175879396984926,
658
+ "grad_norm": 1.3420217037200928,
659
+ "learning_rate": 7.803e-05,
660
+ "loss": 0.3833,
661
+ "step": 4200
662
+ },
663
+ {
664
+ "epoch": 3.559463986599665,
665
+ "grad_norm": 1.1494131088256836,
666
+ "learning_rate": 7.753e-05,
667
+ "loss": 0.3821,
668
+ "step": 4250
669
+ },
670
+ {
671
+ "epoch": 3.6013400335008376,
672
+ "grad_norm": 2.058499813079834,
673
+ "learning_rate": 7.703e-05,
674
+ "loss": 0.3784,
675
+ "step": 4300
676
+ },
677
+ {
678
+ "epoch": 3.64321608040201,
679
+ "grad_norm": 1.8158782720565796,
680
+ "learning_rate": 7.653e-05,
681
+ "loss": 0.3833,
682
+ "step": 4350
683
+ },
684
+ {
685
+ "epoch": 3.6850921273031827,
686
+ "grad_norm": 1.8433794975280762,
687
+ "learning_rate": 7.603e-05,
688
+ "loss": 0.3768,
689
+ "step": 4400
690
+ },
691
+ {
692
+ "epoch": 3.726968174204355,
693
+ "grad_norm": 1.6182619333267212,
694
+ "learning_rate": 7.553e-05,
695
+ "loss": 0.3772,
696
+ "step": 4450
697
+ },
698
+ {
699
+ "epoch": 3.7688442211055277,
700
+ "grad_norm": 1.5222620964050293,
701
+ "learning_rate": 7.503e-05,
702
+ "loss": 0.3798,
703
+ "step": 4500
704
+ },
705
+ {
706
+ "epoch": 3.7688442211055277,
707
+ "eval_loss": 0.3422691822052002,
708
+ "eval_runtime": 214.805,
709
+ "eval_samples_per_second": 39.52,
710
+ "eval_steps_per_second": 4.944,
711
+ "step": 4500
712
+ },
713
+ {
714
+ "epoch": 3.8107202680067003,
715
+ "grad_norm": 1.981819748878479,
716
+ "learning_rate": 7.453e-05,
717
+ "loss": 0.3739,
718
+ "step": 4550
719
+ },
720
+ {
721
+ "epoch": 3.852596314907873,
722
+ "grad_norm": 1.1871320009231567,
723
+ "learning_rate": 7.403e-05,
724
+ "loss": 0.3769,
725
+ "step": 4600
726
+ },
727
+ {
728
+ "epoch": 3.8944723618090453,
729
+ "grad_norm": 1.5213547945022583,
730
+ "learning_rate": 7.353e-05,
731
+ "loss": 0.3734,
732
+ "step": 4650
733
+ },
734
+ {
735
+ "epoch": 3.936348408710218,
736
+ "grad_norm": 1.6221413612365723,
737
+ "learning_rate": 7.303e-05,
738
+ "loss": 0.3704,
739
+ "step": 4700
740
+ },
741
+ {
742
+ "epoch": 3.9782244556113904,
743
+ "grad_norm": 2.075160264968872,
744
+ "learning_rate": 7.253e-05,
745
+ "loss": 0.3701,
746
+ "step": 4750
747
+ },
748
+ {
749
+ "epoch": 4.0201005025125625,
750
+ "grad_norm": 1.217353105545044,
751
+ "learning_rate": 7.203000000000001e-05,
752
+ "loss": 0.3747,
753
+ "step": 4800
754
+ },
755
+ {
756
+ "epoch": 4.061976549413735,
757
+ "grad_norm": 1.9273918867111206,
758
+ "learning_rate": 7.153000000000001e-05,
759
+ "loss": 0.3718,
760
+ "step": 4850
761
+ },
762
+ {
763
+ "epoch": 4.1038525963149075,
764
+ "grad_norm": 1.2843226194381714,
765
+ "learning_rate": 7.103e-05,
766
+ "loss": 0.3689,
767
+ "step": 4900
768
+ },
769
+ {
770
+ "epoch": 4.1457286432160805,
771
+ "grad_norm": 2.8989481925964355,
772
+ "learning_rate": 7.053e-05,
773
+ "loss": 0.3785,
774
+ "step": 4950
775
+ },
776
+ {
777
+ "epoch": 4.187604690117253,
778
+ "grad_norm": 1.6432912349700928,
779
+ "learning_rate": 7.003e-05,
780
+ "loss": 0.3693,
781
+ "step": 5000
782
+ },
783
+ {
784
+ "epoch": 4.187604690117253,
785
+ "eval_loss": 0.3375495374202728,
786
+ "eval_runtime": 215.1803,
787
+ "eval_samples_per_second": 39.451,
788
+ "eval_steps_per_second": 4.935,
789
+ "step": 5000
790
+ },
791
+ {
792
+ "epoch": 4.2294807370184255,
793
+ "grad_norm": 1.5097157955169678,
794
+ "learning_rate": 6.953000000000001e-05,
795
+ "loss": 0.3731,
796
+ "step": 5050
797
+ },
798
+ {
799
+ "epoch": 4.271356783919598,
800
+ "grad_norm": 1.1934735774993896,
801
+ "learning_rate": 6.903000000000001e-05,
802
+ "loss": 0.3683,
803
+ "step": 5100
804
+ },
805
+ {
806
+ "epoch": 4.313232830820771,
807
+ "grad_norm": 1.1494152545928955,
808
+ "learning_rate": 6.853000000000001e-05,
809
+ "loss": 0.3752,
810
+ "step": 5150
811
+ },
812
+ {
813
+ "epoch": 4.355108877721943,
814
+ "grad_norm": 1.404390811920166,
815
+ "learning_rate": 6.803000000000001e-05,
816
+ "loss": 0.3697,
817
+ "step": 5200
818
+ },
819
+ {
820
+ "epoch": 4.396984924623116,
821
+ "grad_norm": 1.0370264053344727,
822
+ "learning_rate": 6.753e-05,
823
+ "loss": 0.3688,
824
+ "step": 5250
825
+ },
826
+ {
827
+ "epoch": 4.438860971524288,
828
+ "grad_norm": 1.6091570854187012,
829
+ "learning_rate": 6.703e-05,
830
+ "loss": 0.3692,
831
+ "step": 5300
832
+ },
833
+ {
834
+ "epoch": 4.480737018425461,
835
+ "grad_norm": 2.325129270553589,
836
+ "learning_rate": 6.653e-05,
837
+ "loss": 0.3711,
838
+ "step": 5350
839
+ },
840
+ {
841
+ "epoch": 4.522613065326633,
842
+ "grad_norm": 1.2980619668960571,
843
+ "learning_rate": 6.603e-05,
844
+ "loss": 0.379,
845
+ "step": 5400
846
+ },
847
+ {
848
+ "epoch": 4.564489112227806,
849
+ "grad_norm": 1.407894253730774,
850
+ "learning_rate": 6.553e-05,
851
+ "loss": 0.3678,
852
+ "step": 5450
853
+ },
854
+ {
855
+ "epoch": 4.606365159128979,
856
+ "grad_norm": 1.9543007612228394,
857
+ "learning_rate": 6.503e-05,
858
+ "loss": 0.3712,
859
+ "step": 5500
860
+ },
861
+ {
862
+ "epoch": 4.606365159128979,
863
+ "eval_loss": 0.33667415380477905,
864
+ "eval_runtime": 214.7408,
865
+ "eval_samples_per_second": 39.531,
866
+ "eval_steps_per_second": 4.945,
867
+ "step": 5500
868
+ },
869
+ {
870
+ "epoch": 4.648241206030151,
871
+ "grad_norm": 1.7211672067642212,
872
+ "learning_rate": 6.453000000000001e-05,
873
+ "loss": 0.369,
874
+ "step": 5550
875
+ },
876
+ {
877
+ "epoch": 4.690117252931323,
878
+ "grad_norm": 1.2072224617004395,
879
+ "learning_rate": 6.403e-05,
880
+ "loss": 0.3661,
881
+ "step": 5600
882
+ },
883
+ {
884
+ "epoch": 4.731993299832496,
885
+ "grad_norm": 1.994110345840454,
886
+ "learning_rate": 6.353e-05,
887
+ "loss": 0.3685,
888
+ "step": 5650
889
+ },
890
+ {
891
+ "epoch": 4.773869346733669,
892
+ "grad_norm": 1.9967550039291382,
893
+ "learning_rate": 6.303e-05,
894
+ "loss": 0.3647,
895
+ "step": 5700
896
+ },
897
+ {
898
+ "epoch": 4.815745393634841,
899
+ "grad_norm": 1.83970046043396,
900
+ "learning_rate": 6.253e-05,
901
+ "loss": 0.369,
902
+ "step": 5750
903
+ },
904
+ {
905
+ "epoch": 4.857621440536013,
906
+ "grad_norm": 1.1974071264266968,
907
+ "learning_rate": 6.203e-05,
908
+ "loss": 0.3662,
909
+ "step": 5800
910
+ },
911
+ {
912
+ "epoch": 4.899497487437186,
913
+ "grad_norm": 1.50179123878479,
914
+ "learning_rate": 6.153e-05,
915
+ "loss": 0.3663,
916
+ "step": 5850
917
+ },
918
+ {
919
+ "epoch": 4.941373534338359,
920
+ "grad_norm": 1.105136513710022,
921
+ "learning_rate": 6.103e-05,
922
+ "loss": 0.3654,
923
+ "step": 5900
924
+ },
925
+ {
926
+ "epoch": 4.983249581239531,
927
+ "grad_norm": 1.079147219657898,
928
+ "learning_rate": 6.053e-05,
929
+ "loss": 0.3613,
930
+ "step": 5950
931
+ },
932
+ {
933
+ "epoch": 5.025125628140704,
934
+ "grad_norm": 1.25212824344635,
935
+ "learning_rate": 6.003e-05,
936
+ "loss": 0.3668,
937
+ "step": 6000
938
+ },
939
+ {
940
+ "epoch": 5.025125628140704,
941
+ "eval_loss": 0.3315671682357788,
942
+ "eval_runtime": 216.0413,
943
+ "eval_samples_per_second": 39.293,
944
+ "eval_steps_per_second": 4.916,
945
+ "step": 6000
946
+ },
947
+ {
948
+ "epoch": 5.067001675041876,
949
+ "grad_norm": 1.2671592235565186,
950
+ "learning_rate": 5.953000000000001e-05,
951
+ "loss": 0.3589,
952
+ "step": 6050
953
+ },
954
+ {
955
+ "epoch": 5.108877721943049,
956
+ "grad_norm": 2.1920173168182373,
957
+ "learning_rate": 5.903000000000001e-05,
958
+ "loss": 0.3647,
959
+ "step": 6100
960
+ },
961
+ {
962
+ "epoch": 5.150753768844221,
963
+ "grad_norm": 0.9611026048660278,
964
+ "learning_rate": 5.8530000000000004e-05,
965
+ "loss": 0.3639,
966
+ "step": 6150
967
+ },
968
+ {
969
+ "epoch": 5.192629815745394,
970
+ "grad_norm": 1.5600687265396118,
971
+ "learning_rate": 5.8030000000000005e-05,
972
+ "loss": 0.3608,
973
+ "step": 6200
974
+ },
975
+ {
976
+ "epoch": 5.234505862646566,
977
+ "grad_norm": 1.410090446472168,
978
+ "learning_rate": 5.7530000000000007e-05,
979
+ "loss": 0.3602,
980
+ "step": 6250
981
+ },
982
+ {
983
+ "epoch": 5.276381909547739,
984
+ "grad_norm": 1.2535935640335083,
985
+ "learning_rate": 5.703000000000001e-05,
986
+ "loss": 0.3611,
987
+ "step": 6300
988
+ },
989
+ {
990
+ "epoch": 5.318257956448911,
991
+ "grad_norm": 1.2604366540908813,
992
+ "learning_rate": 5.653e-05,
993
+ "loss": 0.3609,
994
+ "step": 6350
995
+ },
996
+ {
997
+ "epoch": 5.360134003350084,
998
+ "grad_norm": 1.4509762525558472,
999
+ "learning_rate": 5.6030000000000004e-05,
1000
+ "loss": 0.366,
1001
+ "step": 6400
1002
+ },
1003
+ {
1004
+ "epoch": 5.402010050251256,
1005
+ "grad_norm": 2.065939426422119,
1006
+ "learning_rate": 5.5530000000000005e-05,
1007
+ "loss": 0.3622,
1008
+ "step": 6450
1009
+ },
1010
+ {
1011
+ "epoch": 5.443886097152429,
1012
+ "grad_norm": 1.6851164102554321,
1013
+ "learning_rate": 5.503000000000001e-05,
1014
+ "loss": 0.3635,
1015
+ "step": 6500
1016
+ },
1017
+ {
1018
+ "epoch": 5.443886097152429,
1019
+ "eval_loss": 0.3290642201900482,
1020
+ "eval_runtime": 216.182,
1021
+ "eval_samples_per_second": 39.268,
1022
+ "eval_steps_per_second": 4.913,
1023
+ "step": 6500
1024
+ },
1025
+ {
1026
+ "epoch": 5.485762144053601,
1027
+ "grad_norm": 1.2512905597686768,
1028
+ "learning_rate": 5.453e-05,
1029
+ "loss": 0.3595,
1030
+ "step": 6550
1031
+ },
1032
+ {
1033
+ "epoch": 5.527638190954773,
1034
+ "grad_norm": 1.0366383790969849,
1035
+ "learning_rate": 5.403e-05,
1036
+ "loss": 0.3654,
1037
+ "step": 6600
1038
+ },
1039
+ {
1040
+ "epoch": 5.569514237855946,
1041
+ "grad_norm": 1.6947450637817383,
1042
+ "learning_rate": 5.3530000000000004e-05,
1043
+ "loss": 0.3573,
1044
+ "step": 6650
1045
+ },
1046
+ {
1047
+ "epoch": 5.611390284757119,
1048
+ "grad_norm": 1.5837823152542114,
1049
+ "learning_rate": 5.303e-05,
1050
+ "loss": 0.3575,
1051
+ "step": 6700
1052
+ },
1053
+ {
1054
+ "epoch": 5.653266331658291,
1055
+ "grad_norm": 1.6993337869644165,
1056
+ "learning_rate": 5.253e-05,
1057
+ "loss": 0.3606,
1058
+ "step": 6750
1059
+ },
1060
+ {
1061
+ "epoch": 5.695142378559464,
1062
+ "grad_norm": 1.208640456199646,
1063
+ "learning_rate": 5.203e-05,
1064
+ "loss": 0.3608,
1065
+ "step": 6800
1066
+ },
1067
+ {
1068
+ "epoch": 5.7370184254606365,
1069
+ "grad_norm": 1.5352587699890137,
1070
+ "learning_rate": 5.153e-05,
1071
+ "loss": 0.3613,
1072
+ "step": 6850
1073
+ },
1074
+ {
1075
+ "epoch": 5.778894472361809,
1076
+ "grad_norm": 1.9643837213516235,
1077
+ "learning_rate": 5.103e-05,
1078
+ "loss": 0.3542,
1079
+ "step": 6900
1080
+ },
1081
+ {
1082
+ "epoch": 5.8207705192629815,
1083
+ "grad_norm": 0.9766552448272705,
1084
+ "learning_rate": 5.053e-05,
1085
+ "loss": 0.3579,
1086
+ "step": 6950
1087
+ },
1088
+ {
1089
+ "epoch": 5.8626465661641545,
1090
+ "grad_norm": 1.5910046100616455,
1091
+ "learning_rate": 5.003e-05,
1092
+ "loss": 0.3543,
1093
+ "step": 7000
1094
+ },
1095
+ {
1096
+ "epoch": 5.8626465661641545,
1097
+ "eval_loss": 0.3250139653682709,
1098
+ "eval_runtime": 213.3758,
1099
+ "eval_samples_per_second": 39.784,
1100
+ "eval_steps_per_second": 4.977,
1101
+ "step": 7000
1102
+ },
1103
+ {
1104
+ "epoch": 5.9045226130653266,
1105
+ "grad_norm": 2.036315679550171,
1106
+ "learning_rate": 4.953e-05,
1107
+ "loss": 0.3615,
1108
+ "step": 7050
1109
+ },
1110
+ {
1111
+ "epoch": 5.9463986599664995,
1112
+ "grad_norm": 1.7725523710250854,
1113
+ "learning_rate": 4.903e-05,
1114
+ "loss": 0.3569,
1115
+ "step": 7100
1116
+ },
1117
+ {
1118
+ "epoch": 5.988274706867672,
1119
+ "grad_norm": 0.9128499031066895,
1120
+ "learning_rate": 4.8530000000000005e-05,
1121
+ "loss": 0.3607,
1122
+ "step": 7150
1123
+ },
1124
+ {
1125
+ "epoch": 6.030150753768845,
1126
+ "grad_norm": 1.2767729759216309,
1127
+ "learning_rate": 4.8030000000000006e-05,
1128
+ "loss": 0.3548,
1129
+ "step": 7200
1130
+ },
1131
+ {
1132
+ "epoch": 6.072026800670017,
1133
+ "grad_norm": 1.359195351600647,
1134
+ "learning_rate": 4.753e-05,
1135
+ "loss": 0.3595,
1136
+ "step": 7250
1137
+ },
1138
+ {
1139
+ "epoch": 6.11390284757119,
1140
+ "grad_norm": 1.290307641029358,
1141
+ "learning_rate": 4.703e-05,
1142
+ "loss": 0.3515,
1143
+ "step": 7300
1144
+ },
1145
+ {
1146
+ "epoch": 6.155778894472362,
1147
+ "grad_norm": 1.5797748565673828,
1148
+ "learning_rate": 4.6530000000000003e-05,
1149
+ "loss": 0.3594,
1150
+ "step": 7350
1151
+ },
1152
+ {
1153
+ "epoch": 6.197654941373535,
1154
+ "grad_norm": 3.1323928833007812,
1155
+ "learning_rate": 4.603e-05,
1156
+ "loss": 0.3531,
1157
+ "step": 7400
1158
+ },
1159
+ {
1160
+ "epoch": 6.239530988274707,
1161
+ "grad_norm": 1.1986955404281616,
1162
+ "learning_rate": 4.553e-05,
1163
+ "loss": 0.3503,
1164
+ "step": 7450
1165
+ },
1166
+ {
1167
+ "epoch": 6.28140703517588,
1168
+ "grad_norm": 2.56941294670105,
1169
+ "learning_rate": 4.503e-05,
1170
+ "loss": 0.3526,
1171
+ "step": 7500
1172
+ },
1173
+ {
1174
+ "epoch": 6.28140703517588,
1175
+ "eval_loss": 0.32207000255584717,
1176
+ "eval_runtime": 213.5905,
1177
+ "eval_samples_per_second": 39.744,
1178
+ "eval_steps_per_second": 4.972,
1179
+ "step": 7500
1180
+ },
1181
+ {
1182
+ "epoch": 6.323283082077052,
1183
+ "grad_norm": 1.4174665212631226,
1184
+ "learning_rate": 4.453e-05,
1185
+ "loss": 0.3597,
1186
+ "step": 7550
1187
+ },
1188
+ {
1189
+ "epoch": 6.365159128978225,
1190
+ "grad_norm": 1.4059972763061523,
1191
+ "learning_rate": 4.4030000000000004e-05,
1192
+ "loss": 0.3495,
1193
+ "step": 7600
1194
+ },
1195
+ {
1196
+ "epoch": 6.407035175879397,
1197
+ "grad_norm": 1.2972431182861328,
1198
+ "learning_rate": 4.3530000000000005e-05,
1199
+ "loss": 0.3534,
1200
+ "step": 7650
1201
+ },
1202
+ {
1203
+ "epoch": 6.44891122278057,
1204
+ "grad_norm": 1.3449029922485352,
1205
+ "learning_rate": 4.3030000000000006e-05,
1206
+ "loss": 0.3505,
1207
+ "step": 7700
1208
+ },
1209
+ {
1210
+ "epoch": 6.490787269681742,
1211
+ "grad_norm": 1.2508536577224731,
1212
+ "learning_rate": 4.253e-05,
1213
+ "loss": 0.3583,
1214
+ "step": 7750
1215
+ },
1216
+ {
1217
+ "epoch": 6.532663316582915,
1218
+ "grad_norm": 1.881706714630127,
1219
+ "learning_rate": 4.203e-05,
1220
+ "loss": 0.3519,
1221
+ "step": 7800
1222
+ },
1223
+ {
1224
+ "epoch": 6.574539363484087,
1225
+ "grad_norm": 1.0013474225997925,
1226
+ "learning_rate": 4.1530000000000004e-05,
1227
+ "loss": 0.3512,
1228
+ "step": 7850
1229
+ },
1230
+ {
1231
+ "epoch": 6.61641541038526,
1232
+ "grad_norm": 1.5754601955413818,
1233
+ "learning_rate": 4.103e-05,
1234
+ "loss": 0.3517,
1235
+ "step": 7900
1236
+ },
1237
+ {
1238
+ "epoch": 6.658291457286432,
1239
+ "grad_norm": 1.0650445222854614,
1240
+ "learning_rate": 4.053e-05,
1241
+ "loss": 0.352,
1242
+ "step": 7950
1243
+ },
1244
+ {
1245
+ "epoch": 6.700167504187605,
1246
+ "grad_norm": 0.8827164769172668,
1247
+ "learning_rate": 4.003e-05,
1248
+ "loss": 0.3525,
1249
+ "step": 8000
1250
+ },
1251
+ {
1252
+ "epoch": 6.700167504187605,
1253
+ "eval_loss": 0.32177916169166565,
1254
+ "eval_runtime": 212.2379,
1255
+ "eval_samples_per_second": 39.998,
1256
+ "eval_steps_per_second": 5.004,
1257
+ "step": 8000
1258
+ },
1259
+ {
1260
+ "epoch": 6.742043551088777,
1261
+ "grad_norm": 1.0491342544555664,
1262
+ "learning_rate": 3.953e-05,
1263
+ "loss": 0.3508,
1264
+ "step": 8050
1265
+ },
1266
+ {
1267
+ "epoch": 6.78391959798995,
1268
+ "grad_norm": 1.8081963062286377,
1269
+ "learning_rate": 3.903e-05,
1270
+ "loss": 0.3598,
1271
+ "step": 8100
1272
+ },
1273
+ {
1274
+ "epoch": 6.825795644891122,
1275
+ "grad_norm": 1.4452039003372192,
1276
+ "learning_rate": 3.853e-05,
1277
+ "loss": 0.3475,
1278
+ "step": 8150
1279
+ },
1280
+ {
1281
+ "epoch": 6.867671691792295,
1282
+ "grad_norm": 1.9516081809997559,
1283
+ "learning_rate": 3.803000000000001e-05,
1284
+ "loss": 0.3576,
1285
+ "step": 8200
1286
+ },
1287
+ {
1288
+ "epoch": 6.909547738693467,
1289
+ "grad_norm": 2.1795125007629395,
1290
+ "learning_rate": 3.753e-05,
1291
+ "loss": 0.3523,
1292
+ "step": 8250
1293
+ },
1294
+ {
1295
+ "epoch": 6.95142378559464,
1296
+ "grad_norm": 1.8348519802093506,
1297
+ "learning_rate": 3.703e-05,
1298
+ "loss": 0.3554,
1299
+ "step": 8300
1300
+ },
1301
+ {
1302
+ "epoch": 6.993299832495812,
1303
+ "grad_norm": 1.1464494466781616,
1304
+ "learning_rate": 3.6530000000000004e-05,
1305
+ "loss": 0.3558,
1306
+ "step": 8350
1307
+ },
1308
+ {
1309
+ "epoch": 7.035175879396985,
1310
+ "grad_norm": 1.2294760942459106,
1311
+ "learning_rate": 3.6030000000000006e-05,
1312
+ "loss": 0.3502,
1313
+ "step": 8400
1314
+ },
1315
+ {
1316
+ "epoch": 7.077051926298157,
1317
+ "grad_norm": 4.840690612792969,
1318
+ "learning_rate": 3.553e-05,
1319
+ "loss": 0.3534,
1320
+ "step": 8450
1321
+ },
1322
+ {
1323
+ "epoch": 7.11892797319933,
1324
+ "grad_norm": 1.1307451725006104,
1325
+ "learning_rate": 3.503e-05,
1326
+ "loss": 0.3513,
1327
+ "step": 8500
1328
+ },
1329
+ {
1330
+ "epoch": 7.11892797319933,
1331
+ "eval_loss": 0.31820428371429443,
1332
+ "eval_runtime": 211.821,
1333
+ "eval_samples_per_second": 40.076,
1334
+ "eval_steps_per_second": 5.014,
1335
+ "step": 8500
1336
+ }
1337
+ ],
1338
+ "logging_steps": 50,
1339
+ "max_steps": 12000,
1340
+ "num_input_tokens_seen": 0,
1341
+ "num_train_epochs": 11,
1342
+ "save_steps": 500,
1343
+ "stateful_callbacks": {
1344
+ "TrainerControl": {
1345
+ "args": {
1346
+ "should_epoch_stop": false,
1347
+ "should_evaluate": false,
1348
+ "should_log": false,
1349
+ "should_save": true,
1350
+ "should_training_stop": false
1351
+ },
1352
+ "attributes": {}
1353
+ }
1354
+ },
1355
+ "total_flos": 8.263422718396243e+16,
1356
+ "train_batch_size": 32,
1357
+ "trial_name": null,
1358
+ "trial_params": null
1359
+ }
checkpoint-8500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb8203a5324b57ae1a64f032a5185016d65cbc7738fd0978239bc5b2518d051b
3
+ size 5304