mikhail-panzo commited on
Commit
d725156
1 Parent(s): 8e2e3f7

Training in progress, step 7000, checkpoint

Browse files
checkpoint-7000/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-7000/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "mikhail-panzo/zlm-fil_b64_le5_s8000",
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-7000/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-7000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37d4c27c655947218d181f9ef807cafec2df765bf37ad1d841e68cd1a979b27a
3
+ size 577789320
checkpoint-7000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7066d3c3fc5a4edf9c08746bdb32ff9aa2e10236d9897c285e5b1005fca995ff
3
+ size 1155772233
checkpoint-7000/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-7000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4846a01c5dc5b38eede0301d449ad98c5b1eba671fa48afe5662a5636959841
3
+ size 14244
checkpoint-7000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79faef7859961c2a638e4312a796703fd0c86e5877b740a2ce8b47db225af025
3
+ size 1064
checkpoint-7000/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "mask_token": {
17
+ "content": "<mask>",
18
+ "lstrip": true,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "pad_token": {
24
+ "content": "<pad>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "<unk>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-7000/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-7000/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-7000/trainer_state.json ADDED
@@ -0,0 +1,1125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.4122372269630432,
3
+ "best_model_checkpoint": "mikhail-panzo/zlm-fil-ceb_b64_le5_s8000/checkpoint-5000",
4
+ "epoch": 304.3478260869565,
5
+ "eval_steps": 500,
6
+ "global_step": 7000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 2.1739130434782608,
13
+ "grad_norm": 0.7357823252677917,
14
+ "learning_rate": 2.5000000000000004e-07,
15
+ "loss": 0.4055,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 4.3478260869565215,
20
+ "grad_norm": 0.7944665551185608,
21
+ "learning_rate": 5.000000000000001e-07,
22
+ "loss": 0.3988,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 6.521739130434782,
27
+ "grad_norm": 0.7226569652557373,
28
+ "learning_rate": 7.5e-07,
29
+ "loss": 0.4072,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 8.695652173913043,
34
+ "grad_norm": 0.7845238447189331,
35
+ "learning_rate": 1.0000000000000002e-06,
36
+ "loss": 0.4028,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 10.869565217391305,
41
+ "grad_norm": 0.9769566655158997,
42
+ "learning_rate": 1.25e-06,
43
+ "loss": 0.4066,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 13.043478260869565,
48
+ "grad_norm": 0.8634074330329895,
49
+ "learning_rate": 1.5e-06,
50
+ "loss": 0.405,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 15.217391304347826,
55
+ "grad_norm": 0.9029327034950256,
56
+ "learning_rate": 1.75e-06,
57
+ "loss": 0.4074,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 17.391304347826086,
62
+ "grad_norm": 0.7624074220657349,
63
+ "learning_rate": 2.0000000000000003e-06,
64
+ "loss": 0.406,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 19.565217391304348,
69
+ "grad_norm": 1.005196213722229,
70
+ "learning_rate": 2.25e-06,
71
+ "loss": 0.4052,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 21.73913043478261,
76
+ "grad_norm": 1.0276380777359009,
77
+ "learning_rate": 2.5e-06,
78
+ "loss": 0.4019,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 21.73913043478261,
83
+ "eval_loss": 0.4143177568912506,
84
+ "eval_runtime": 6.4074,
85
+ "eval_samples_per_second": 24.815,
86
+ "eval_steps_per_second": 3.121,
87
+ "step": 500
88
+ },
89
+ {
90
+ "epoch": 23.91304347826087,
91
+ "grad_norm": 0.9067140817642212,
92
+ "learning_rate": 2.7500000000000004e-06,
93
+ "loss": 0.4021,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 26.08695652173913,
98
+ "grad_norm": 0.8799042701721191,
99
+ "learning_rate": 3e-06,
100
+ "loss": 0.4071,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 28.26086956521739,
105
+ "grad_norm": 1.0451533794403076,
106
+ "learning_rate": 3.2500000000000002e-06,
107
+ "loss": 0.4146,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 30.434782608695652,
112
+ "grad_norm": 0.7998800277709961,
113
+ "learning_rate": 3.5e-06,
114
+ "loss": 0.4023,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 32.608695652173914,
119
+ "grad_norm": 1.0301792621612549,
120
+ "learning_rate": 3.7500000000000005e-06,
121
+ "loss": 0.4045,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 34.78260869565217,
126
+ "grad_norm": 0.6388185620307922,
127
+ "learning_rate": 4.000000000000001e-06,
128
+ "loss": 0.4045,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 36.95652173913044,
133
+ "grad_norm": 0.9676840901374817,
134
+ "learning_rate": 4.25e-06,
135
+ "loss": 0.4116,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 39.130434782608695,
140
+ "grad_norm": 0.8355514407157898,
141
+ "learning_rate": 4.5e-06,
142
+ "loss": 0.4013,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 41.30434782608695,
147
+ "grad_norm": 0.7534067630767822,
148
+ "learning_rate": 4.75e-06,
149
+ "loss": 0.4019,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 43.47826086956522,
154
+ "grad_norm": 0.9366117119789124,
155
+ "learning_rate": 5e-06,
156
+ "loss": 0.4064,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 43.47826086956522,
161
+ "eval_loss": 0.41316837072372437,
162
+ "eval_runtime": 6.5616,
163
+ "eval_samples_per_second": 24.232,
164
+ "eval_steps_per_second": 3.048,
165
+ "step": 1000
166
+ },
167
+ {
168
+ "epoch": 45.65217391304348,
169
+ "grad_norm": 0.7025179266929626,
170
+ "learning_rate": 5.2500000000000006e-06,
171
+ "loss": 0.4061,
172
+ "step": 1050
173
+ },
174
+ {
175
+ "epoch": 47.82608695652174,
176
+ "grad_norm": 0.9240646958351135,
177
+ "learning_rate": 5.500000000000001e-06,
178
+ "loss": 0.4061,
179
+ "step": 1100
180
+ },
181
+ {
182
+ "epoch": 50.0,
183
+ "grad_norm": 0.9685391187667847,
184
+ "learning_rate": 5.75e-06,
185
+ "loss": 0.4063,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 52.17391304347826,
190
+ "grad_norm": 0.8591094017028809,
191
+ "learning_rate": 6e-06,
192
+ "loss": 0.4013,
193
+ "step": 1200
194
+ },
195
+ {
196
+ "epoch": 54.34782608695652,
197
+ "grad_norm": 0.9348228573799133,
198
+ "learning_rate": 6.25e-06,
199
+ "loss": 0.402,
200
+ "step": 1250
201
+ },
202
+ {
203
+ "epoch": 56.52173913043478,
204
+ "grad_norm": 0.7908409237861633,
205
+ "learning_rate": 6.5000000000000004e-06,
206
+ "loss": 0.4046,
207
+ "step": 1300
208
+ },
209
+ {
210
+ "epoch": 58.69565217391305,
211
+ "grad_norm": 1.109017014503479,
212
+ "learning_rate": 6.750000000000001e-06,
213
+ "loss": 0.3995,
214
+ "step": 1350
215
+ },
216
+ {
217
+ "epoch": 60.869565217391305,
218
+ "grad_norm": 0.8529478311538696,
219
+ "learning_rate": 7e-06,
220
+ "loss": 0.4018,
221
+ "step": 1400
222
+ },
223
+ {
224
+ "epoch": 63.04347826086956,
225
+ "grad_norm": 0.9947476387023926,
226
+ "learning_rate": 7.25e-06,
227
+ "loss": 0.401,
228
+ "step": 1450
229
+ },
230
+ {
231
+ "epoch": 65.21739130434783,
232
+ "grad_norm": 1.2476813793182373,
233
+ "learning_rate": 7.500000000000001e-06,
234
+ "loss": 0.4034,
235
+ "step": 1500
236
+ },
237
+ {
238
+ "epoch": 65.21739130434783,
239
+ "eval_loss": 0.41223829984664917,
240
+ "eval_runtime": 6.5846,
241
+ "eval_samples_per_second": 24.147,
242
+ "eval_steps_per_second": 3.037,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 67.3913043478261,
247
+ "grad_norm": 0.8658013343811035,
248
+ "learning_rate": 7.75e-06,
249
+ "loss": 0.4007,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 69.56521739130434,
254
+ "grad_norm": 0.7932788133621216,
255
+ "learning_rate": 8.000000000000001e-06,
256
+ "loss": 0.4054,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 71.73913043478261,
261
+ "grad_norm": 2.33347225189209,
262
+ "learning_rate": 8.25e-06,
263
+ "loss": 0.4075,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 73.91304347826087,
268
+ "grad_norm": 1.0095164775848389,
269
+ "learning_rate": 8.5e-06,
270
+ "loss": 0.3982,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 76.08695652173913,
275
+ "grad_norm": 0.8895902633666992,
276
+ "learning_rate": 8.750000000000001e-06,
277
+ "loss": 0.4018,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 78.26086956521739,
282
+ "grad_norm": 1.2252330780029297,
283
+ "learning_rate": 9e-06,
284
+ "loss": 0.4024,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 80.43478260869566,
289
+ "grad_norm": 0.7731898427009583,
290
+ "learning_rate": 9.250000000000001e-06,
291
+ "loss": 0.3994,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 82.6086956521739,
296
+ "grad_norm": 1.047607660293579,
297
+ "learning_rate": 9.5e-06,
298
+ "loss": 0.4013,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 84.78260869565217,
303
+ "grad_norm": 0.9284490346908569,
304
+ "learning_rate": 9.75e-06,
305
+ "loss": 0.3986,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 86.95652173913044,
310
+ "grad_norm": 4.725491046905518,
311
+ "learning_rate": 1e-05,
312
+ "loss": 0.3985,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 86.95652173913044,
317
+ "eval_loss": 0.4147759974002838,
318
+ "eval_runtime": 6.5475,
319
+ "eval_samples_per_second": 24.284,
320
+ "eval_steps_per_second": 3.055,
321
+ "step": 2000
322
+ },
323
+ {
324
+ "epoch": 89.1304347826087,
325
+ "grad_norm": 1.1678683757781982,
326
+ "learning_rate": 9.916666666666668e-06,
327
+ "loss": 0.4066,
328
+ "step": 2050
329
+ },
330
+ {
331
+ "epoch": 91.30434782608695,
332
+ "grad_norm": 1.988890528678894,
333
+ "learning_rate": 9.835000000000002e-06,
334
+ "loss": 0.3997,
335
+ "step": 2100
336
+ },
337
+ {
338
+ "epoch": 93.47826086956522,
339
+ "grad_norm": 1.0278427600860596,
340
+ "learning_rate": 9.751666666666667e-06,
341
+ "loss": 0.3948,
342
+ "step": 2150
343
+ },
344
+ {
345
+ "epoch": 95.65217391304348,
346
+ "grad_norm": 0.9606735110282898,
347
+ "learning_rate": 9.67e-06,
348
+ "loss": 0.42,
349
+ "step": 2200
350
+ },
351
+ {
352
+ "epoch": 97.82608695652173,
353
+ "grad_norm": 1.0861903429031372,
354
+ "learning_rate": 9.586666666666667e-06,
355
+ "loss": 0.398,
356
+ "step": 2250
357
+ },
358
+ {
359
+ "epoch": 100.0,
360
+ "grad_norm": 1.4872643947601318,
361
+ "learning_rate": 9.503333333333334e-06,
362
+ "loss": 0.4059,
363
+ "step": 2300
364
+ },
365
+ {
366
+ "epoch": 102.17391304347827,
367
+ "grad_norm": 1.0460981130599976,
368
+ "learning_rate": 9.42e-06,
369
+ "loss": 0.4014,
370
+ "step": 2350
371
+ },
372
+ {
373
+ "epoch": 104.34782608695652,
374
+ "grad_norm": 1.906476378440857,
375
+ "learning_rate": 9.336666666666666e-06,
376
+ "loss": 0.4015,
377
+ "step": 2400
378
+ },
379
+ {
380
+ "epoch": 106.52173913043478,
381
+ "grad_norm": 1.3965051174163818,
382
+ "learning_rate": 9.253333333333333e-06,
383
+ "loss": 0.4063,
384
+ "step": 2450
385
+ },
386
+ {
387
+ "epoch": 108.69565217391305,
388
+ "grad_norm": 1.7410107851028442,
389
+ "learning_rate": 9.17e-06,
390
+ "loss": 0.396,
391
+ "step": 2500
392
+ },
393
+ {
394
+ "epoch": 108.69565217391305,
395
+ "eval_loss": 0.41233959794044495,
396
+ "eval_runtime": 6.6031,
397
+ "eval_samples_per_second": 24.08,
398
+ "eval_steps_per_second": 3.029,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 110.8695652173913,
403
+ "grad_norm": 1.4562053680419922,
404
+ "learning_rate": 9.088333333333334e-06,
405
+ "loss": 0.4042,
406
+ "step": 2550
407
+ },
408
+ {
409
+ "epoch": 113.04347826086956,
410
+ "grad_norm": 0.8741047978401184,
411
+ "learning_rate": 9.005000000000001e-06,
412
+ "loss": 0.3985,
413
+ "step": 2600
414
+ },
415
+ {
416
+ "epoch": 115.21739130434783,
417
+ "grad_norm": 1.1817028522491455,
418
+ "learning_rate": 8.921666666666668e-06,
419
+ "loss": 0.4036,
420
+ "step": 2650
421
+ },
422
+ {
423
+ "epoch": 117.3913043478261,
424
+ "grad_norm": 0.8515534996986389,
425
+ "learning_rate": 8.838333333333335e-06,
426
+ "loss": 0.3986,
427
+ "step": 2700
428
+ },
429
+ {
430
+ "epoch": 119.56521739130434,
431
+ "grad_norm": 0.9545428156852722,
432
+ "learning_rate": 8.755e-06,
433
+ "loss": 0.3946,
434
+ "step": 2750
435
+ },
436
+ {
437
+ "epoch": 121.73913043478261,
438
+ "grad_norm": 0.6653321385383606,
439
+ "learning_rate": 8.671666666666667e-06,
440
+ "loss": 0.3957,
441
+ "step": 2800
442
+ },
443
+ {
444
+ "epoch": 123.91304347826087,
445
+ "grad_norm": 0.8100114464759827,
446
+ "learning_rate": 8.588333333333334e-06,
447
+ "loss": 0.3953,
448
+ "step": 2850
449
+ },
450
+ {
451
+ "epoch": 126.08695652173913,
452
+ "grad_norm": 1.4174237251281738,
453
+ "learning_rate": 8.505e-06,
454
+ "loss": 0.3933,
455
+ "step": 2900
456
+ },
457
+ {
458
+ "epoch": 128.2608695652174,
459
+ "grad_norm": 1.6101123094558716,
460
+ "learning_rate": 8.421666666666668e-06,
461
+ "loss": 0.4027,
462
+ "step": 2950
463
+ },
464
+ {
465
+ "epoch": 130.43478260869566,
466
+ "grad_norm": 0.7453967332839966,
467
+ "learning_rate": 8.338333333333335e-06,
468
+ "loss": 0.3958,
469
+ "step": 3000
470
+ },
471
+ {
472
+ "epoch": 130.43478260869566,
473
+ "eval_loss": 0.4138609766960144,
474
+ "eval_runtime": 6.6093,
475
+ "eval_samples_per_second": 24.057,
476
+ "eval_steps_per_second": 3.026,
477
+ "step": 3000
478
+ },
479
+ {
480
+ "epoch": 132.6086956521739,
481
+ "grad_norm": 1.061684012413025,
482
+ "learning_rate": 8.255000000000001e-06,
483
+ "loss": 0.3966,
484
+ "step": 3050
485
+ },
486
+ {
487
+ "epoch": 134.7826086956522,
488
+ "grad_norm": 1.0820105075836182,
489
+ "learning_rate": 8.171666666666668e-06,
490
+ "loss": 0.3937,
491
+ "step": 3100
492
+ },
493
+ {
494
+ "epoch": 136.95652173913044,
495
+ "grad_norm": 0.8147062063217163,
496
+ "learning_rate": 8.088333333333334e-06,
497
+ "loss": 0.3961,
498
+ "step": 3150
499
+ },
500
+ {
501
+ "epoch": 139.1304347826087,
502
+ "grad_norm": 0.7450351715087891,
503
+ "learning_rate": 8.005e-06,
504
+ "loss": 0.3923,
505
+ "step": 3200
506
+ },
507
+ {
508
+ "epoch": 141.30434782608697,
509
+ "grad_norm": 1.05226731300354,
510
+ "learning_rate": 7.921666666666667e-06,
511
+ "loss": 0.3951,
512
+ "step": 3250
513
+ },
514
+ {
515
+ "epoch": 143.47826086956522,
516
+ "grad_norm": 1.5119163990020752,
517
+ "learning_rate": 7.838333333333334e-06,
518
+ "loss": 0.3957,
519
+ "step": 3300
520
+ },
521
+ {
522
+ "epoch": 145.65217391304347,
523
+ "grad_norm": 1.0422383546829224,
524
+ "learning_rate": 7.755000000000001e-06,
525
+ "loss": 0.3942,
526
+ "step": 3350
527
+ },
528
+ {
529
+ "epoch": 147.82608695652175,
530
+ "grad_norm": 1.2120935916900635,
531
+ "learning_rate": 7.671666666666668e-06,
532
+ "loss": 0.3946,
533
+ "step": 3400
534
+ },
535
+ {
536
+ "epoch": 150.0,
537
+ "grad_norm": 1.3127976655960083,
538
+ "learning_rate": 7.588333333333334e-06,
539
+ "loss": 0.3933,
540
+ "step": 3450
541
+ },
542
+ {
543
+ "epoch": 152.17391304347825,
544
+ "grad_norm": 0.7704339027404785,
545
+ "learning_rate": 7.505e-06,
546
+ "loss": 0.3928,
547
+ "step": 3500
548
+ },
549
+ {
550
+ "epoch": 152.17391304347825,
551
+ "eval_loss": 0.4132215976715088,
552
+ "eval_runtime": 6.6162,
553
+ "eval_samples_per_second": 24.032,
554
+ "eval_steps_per_second": 3.023,
555
+ "step": 3500
556
+ },
557
+ {
558
+ "epoch": 154.34782608695653,
559
+ "grad_norm": 1.4827122688293457,
560
+ "learning_rate": 7.421666666666667e-06,
561
+ "loss": 0.3908,
562
+ "step": 3550
563
+ },
564
+ {
565
+ "epoch": 156.52173913043478,
566
+ "grad_norm": 1.0464049577713013,
567
+ "learning_rate": 7.338333333333334e-06,
568
+ "loss": 0.3993,
569
+ "step": 3600
570
+ },
571
+ {
572
+ "epoch": 158.69565217391303,
573
+ "grad_norm": 0.8279913663864136,
574
+ "learning_rate": 7.255000000000001e-06,
575
+ "loss": 0.3967,
576
+ "step": 3650
577
+ },
578
+ {
579
+ "epoch": 160.8695652173913,
580
+ "grad_norm": 0.9175034165382385,
581
+ "learning_rate": 7.171666666666667e-06,
582
+ "loss": 0.3911,
583
+ "step": 3700
584
+ },
585
+ {
586
+ "epoch": 163.04347826086956,
587
+ "grad_norm": 1.0080233812332153,
588
+ "learning_rate": 7.088333333333334e-06,
589
+ "loss": 0.3898,
590
+ "step": 3750
591
+ },
592
+ {
593
+ "epoch": 165.2173913043478,
594
+ "grad_norm": 0.9944823980331421,
595
+ "learning_rate": 7.005000000000001e-06,
596
+ "loss": 0.3901,
597
+ "step": 3800
598
+ },
599
+ {
600
+ "epoch": 167.3913043478261,
601
+ "grad_norm": 1.8222392797470093,
602
+ "learning_rate": 6.921666666666668e-06,
603
+ "loss": 0.3901,
604
+ "step": 3850
605
+ },
606
+ {
607
+ "epoch": 169.56521739130434,
608
+ "grad_norm": 0.9506675004959106,
609
+ "learning_rate": 6.838333333333334e-06,
610
+ "loss": 0.3945,
611
+ "step": 3900
612
+ },
613
+ {
614
+ "epoch": 171.7391304347826,
615
+ "grad_norm": 4.512270450592041,
616
+ "learning_rate": 6.7550000000000005e-06,
617
+ "loss": 0.3913,
618
+ "step": 3950
619
+ },
620
+ {
621
+ "epoch": 173.91304347826087,
622
+ "grad_norm": 0.74826580286026,
623
+ "learning_rate": 6.6716666666666674e-06,
624
+ "loss": 0.3914,
625
+ "step": 4000
626
+ },
627
+ {
628
+ "epoch": 173.91304347826087,
629
+ "eval_loss": 0.4138501286506653,
630
+ "eval_runtime": 6.5399,
631
+ "eval_samples_per_second": 24.312,
632
+ "eval_steps_per_second": 3.058,
633
+ "step": 4000
634
+ },
635
+ {
636
+ "epoch": 176.08695652173913,
637
+ "grad_norm": 1.4560046195983887,
638
+ "learning_rate": 6.588333333333334e-06,
639
+ "loss": 0.3903,
640
+ "step": 4050
641
+ },
642
+ {
643
+ "epoch": 178.2608695652174,
644
+ "grad_norm": 1.183271050453186,
645
+ "learning_rate": 6.505e-06,
646
+ "loss": 0.3901,
647
+ "step": 4100
648
+ },
649
+ {
650
+ "epoch": 180.43478260869566,
651
+ "grad_norm": 0.6697027683258057,
652
+ "learning_rate": 6.421666666666667e-06,
653
+ "loss": 0.399,
654
+ "step": 4150
655
+ },
656
+ {
657
+ "epoch": 182.6086956521739,
658
+ "grad_norm": 0.96327143907547,
659
+ "learning_rate": 6.338333333333334e-06,
660
+ "loss": 0.3895,
661
+ "step": 4200
662
+ },
663
+ {
664
+ "epoch": 184.7826086956522,
665
+ "grad_norm": 0.9833754301071167,
666
+ "learning_rate": 6.255e-06,
667
+ "loss": 0.3903,
668
+ "step": 4250
669
+ },
670
+ {
671
+ "epoch": 186.95652173913044,
672
+ "grad_norm": 0.857953667640686,
673
+ "learning_rate": 6.171666666666667e-06,
674
+ "loss": 0.39,
675
+ "step": 4300
676
+ },
677
+ {
678
+ "epoch": 189.1304347826087,
679
+ "grad_norm": 0.7506987452507019,
680
+ "learning_rate": 6.088333333333334e-06,
681
+ "loss": 0.394,
682
+ "step": 4350
683
+ },
684
+ {
685
+ "epoch": 191.30434782608697,
686
+ "grad_norm": 0.8283782601356506,
687
+ "learning_rate": 6.005000000000001e-06,
688
+ "loss": 0.3869,
689
+ "step": 4400
690
+ },
691
+ {
692
+ "epoch": 193.47826086956522,
693
+ "grad_norm": 1.0252714157104492,
694
+ "learning_rate": 5.921666666666667e-06,
695
+ "loss": 0.3967,
696
+ "step": 4450
697
+ },
698
+ {
699
+ "epoch": 195.65217391304347,
700
+ "grad_norm": 0.9400416612625122,
701
+ "learning_rate": 5.838333333333334e-06,
702
+ "loss": 0.3916,
703
+ "step": 4500
704
+ },
705
+ {
706
+ "epoch": 195.65217391304347,
707
+ "eval_loss": 0.41232919692993164,
708
+ "eval_runtime": 6.3912,
709
+ "eval_samples_per_second": 24.878,
710
+ "eval_steps_per_second": 3.129,
711
+ "step": 4500
712
+ },
713
+ {
714
+ "epoch": 197.82608695652175,
715
+ "grad_norm": 0.9093385934829712,
716
+ "learning_rate": 5.755000000000001e-06,
717
+ "loss": 0.3894,
718
+ "step": 4550
719
+ },
720
+ {
721
+ "epoch": 200.0,
722
+ "grad_norm": 1.078837513923645,
723
+ "learning_rate": 5.671666666666668e-06,
724
+ "loss": 0.3899,
725
+ "step": 4600
726
+ },
727
+ {
728
+ "epoch": 202.17391304347825,
729
+ "grad_norm": 0.7748771905899048,
730
+ "learning_rate": 5.588333333333334e-06,
731
+ "loss": 0.3943,
732
+ "step": 4650
733
+ },
734
+ {
735
+ "epoch": 204.34782608695653,
736
+ "grad_norm": 1.1522585153579712,
737
+ "learning_rate": 5.505000000000001e-06,
738
+ "loss": 0.389,
739
+ "step": 4700
740
+ },
741
+ {
742
+ "epoch": 206.52173913043478,
743
+ "grad_norm": 1.6818900108337402,
744
+ "learning_rate": 5.4216666666666676e-06,
745
+ "loss": 0.3961,
746
+ "step": 4750
747
+ },
748
+ {
749
+ "epoch": 208.69565217391303,
750
+ "grad_norm": 0.9630489349365234,
751
+ "learning_rate": 5.3383333333333345e-06,
752
+ "loss": 0.3926,
753
+ "step": 4800
754
+ },
755
+ {
756
+ "epoch": 210.8695652173913,
757
+ "grad_norm": 0.915107250213623,
758
+ "learning_rate": 5.2550000000000005e-06,
759
+ "loss": 0.3843,
760
+ "step": 4850
761
+ },
762
+ {
763
+ "epoch": 213.04347826086956,
764
+ "grad_norm": 0.7301034331321716,
765
+ "learning_rate": 5.171666666666667e-06,
766
+ "loss": 0.3929,
767
+ "step": 4900
768
+ },
769
+ {
770
+ "epoch": 215.2173913043478,
771
+ "grad_norm": 0.8523257970809937,
772
+ "learning_rate": 5.088333333333334e-06,
773
+ "loss": 0.3873,
774
+ "step": 4950
775
+ },
776
+ {
777
+ "epoch": 217.3913043478261,
778
+ "grad_norm": 0.7538474798202515,
779
+ "learning_rate": 5.0049999999999995e-06,
780
+ "loss": 0.3919,
781
+ "step": 5000
782
+ },
783
+ {
784
+ "epoch": 217.3913043478261,
785
+ "eval_loss": 0.4122372269630432,
786
+ "eval_runtime": 6.5201,
787
+ "eval_samples_per_second": 24.386,
788
+ "eval_steps_per_second": 3.067,
789
+ "step": 5000
790
+ },
791
+ {
792
+ "epoch": 219.56521739130434,
793
+ "grad_norm": 1.006597638130188,
794
+ "learning_rate": 4.921666666666666e-06,
795
+ "loss": 0.3844,
796
+ "step": 5050
797
+ },
798
+ {
799
+ "epoch": 221.7391304347826,
800
+ "grad_norm": 0.6486532092094421,
801
+ "learning_rate": 4.838333333333334e-06,
802
+ "loss": 0.3853,
803
+ "step": 5100
804
+ },
805
+ {
806
+ "epoch": 223.91304347826087,
807
+ "grad_norm": 0.7955175638198853,
808
+ "learning_rate": 4.755e-06,
809
+ "loss": 0.3853,
810
+ "step": 5150
811
+ },
812
+ {
813
+ "epoch": 226.08695652173913,
814
+ "grad_norm": 0.8011429309844971,
815
+ "learning_rate": 4.671666666666667e-06,
816
+ "loss": 0.3888,
817
+ "step": 5200
818
+ },
819
+ {
820
+ "epoch": 228.2608695652174,
821
+ "grad_norm": 0.7421734929084778,
822
+ "learning_rate": 4.588333333333333e-06,
823
+ "loss": 0.3945,
824
+ "step": 5250
825
+ },
826
+ {
827
+ "epoch": 230.43478260869566,
828
+ "grad_norm": 1.5567547082901,
829
+ "learning_rate": 4.505e-06,
830
+ "loss": 0.3916,
831
+ "step": 5300
832
+ },
833
+ {
834
+ "epoch": 232.6086956521739,
835
+ "grad_norm": 0.8348387479782104,
836
+ "learning_rate": 4.421666666666667e-06,
837
+ "loss": 0.3887,
838
+ "step": 5350
839
+ },
840
+ {
841
+ "epoch": 234.7826086956522,
842
+ "grad_norm": 0.965840756893158,
843
+ "learning_rate": 4.338333333333334e-06,
844
+ "loss": 0.3907,
845
+ "step": 5400
846
+ },
847
+ {
848
+ "epoch": 236.95652173913044,
849
+ "grad_norm": 1.0057460069656372,
850
+ "learning_rate": 4.255e-06,
851
+ "loss": 0.4004,
852
+ "step": 5450
853
+ },
854
+ {
855
+ "epoch": 239.1304347826087,
856
+ "grad_norm": 1.2100751399993896,
857
+ "learning_rate": 4.171666666666667e-06,
858
+ "loss": 0.3844,
859
+ "step": 5500
860
+ },
861
+ {
862
+ "epoch": 239.1304347826087,
863
+ "eval_loss": 0.41257867217063904,
864
+ "eval_runtime": 6.47,
865
+ "eval_samples_per_second": 24.575,
866
+ "eval_steps_per_second": 3.091,
867
+ "step": 5500
868
+ },
869
+ {
870
+ "epoch": 241.30434782608697,
871
+ "grad_norm": 0.7117742896080017,
872
+ "learning_rate": 4.088333333333334e-06,
873
+ "loss": 0.3845,
874
+ "step": 5550
875
+ },
876
+ {
877
+ "epoch": 243.47826086956522,
878
+ "grad_norm": 0.6881281733512878,
879
+ "learning_rate": 4.005000000000001e-06,
880
+ "loss": 0.3858,
881
+ "step": 5600
882
+ },
883
+ {
884
+ "epoch": 245.65217391304347,
885
+ "grad_norm": 1.3413081169128418,
886
+ "learning_rate": 3.921666666666667e-06,
887
+ "loss": 0.3842,
888
+ "step": 5650
889
+ },
890
+ {
891
+ "epoch": 247.82608695652175,
892
+ "grad_norm": 0.7762352228164673,
893
+ "learning_rate": 3.8383333333333336e-06,
894
+ "loss": 0.3839,
895
+ "step": 5700
896
+ },
897
+ {
898
+ "epoch": 250.0,
899
+ "grad_norm": 1.2861838340759277,
900
+ "learning_rate": 3.7550000000000005e-06,
901
+ "loss": 0.3836,
902
+ "step": 5750
903
+ },
904
+ {
905
+ "epoch": 252.17391304347825,
906
+ "grad_norm": 1.12875497341156,
907
+ "learning_rate": 3.6716666666666665e-06,
908
+ "loss": 0.3828,
909
+ "step": 5800
910
+ },
911
+ {
912
+ "epoch": 254.34782608695653,
913
+ "grad_norm": 0.9641019105911255,
914
+ "learning_rate": 3.588333333333334e-06,
915
+ "loss": 0.3815,
916
+ "step": 5850
917
+ },
918
+ {
919
+ "epoch": 256.5217391304348,
920
+ "grad_norm": 0.7063067555427551,
921
+ "learning_rate": 3.505e-06,
922
+ "loss": 0.3838,
923
+ "step": 5900
924
+ },
925
+ {
926
+ "epoch": 258.69565217391306,
927
+ "grad_norm": 1.0992182493209839,
928
+ "learning_rate": 3.4216666666666672e-06,
929
+ "loss": 0.3824,
930
+ "step": 5950
931
+ },
932
+ {
933
+ "epoch": 260.8695652173913,
934
+ "grad_norm": 0.6153731942176819,
935
+ "learning_rate": 3.3383333333333333e-06,
936
+ "loss": 0.3908,
937
+ "step": 6000
938
+ },
939
+ {
940
+ "epoch": 260.8695652173913,
941
+ "eval_loss": 0.4136714041233063,
942
+ "eval_runtime": 6.4416,
943
+ "eval_samples_per_second": 24.683,
944
+ "eval_steps_per_second": 3.105,
945
+ "step": 6000
946
+ },
947
+ {
948
+ "epoch": 263.04347826086956,
949
+ "grad_norm": 0.8322598934173584,
950
+ "learning_rate": 3.255e-06,
951
+ "loss": 0.3918,
952
+ "step": 6050
953
+ },
954
+ {
955
+ "epoch": 265.2173913043478,
956
+ "grad_norm": 0.7336969971656799,
957
+ "learning_rate": 3.1716666666666667e-06,
958
+ "loss": 0.3908,
959
+ "step": 6100
960
+ },
961
+ {
962
+ "epoch": 267.39130434782606,
963
+ "grad_norm": 0.7218737006187439,
964
+ "learning_rate": 3.0883333333333336e-06,
965
+ "loss": 0.3848,
966
+ "step": 6150
967
+ },
968
+ {
969
+ "epoch": 269.5652173913044,
970
+ "grad_norm": 1.2083243131637573,
971
+ "learning_rate": 3.005e-06,
972
+ "loss": 0.3868,
973
+ "step": 6200
974
+ },
975
+ {
976
+ "epoch": 271.7391304347826,
977
+ "grad_norm": 0.9182942509651184,
978
+ "learning_rate": 2.921666666666667e-06,
979
+ "loss": 0.3895,
980
+ "step": 6250
981
+ },
982
+ {
983
+ "epoch": 273.9130434782609,
984
+ "grad_norm": 0.8747413158416748,
985
+ "learning_rate": 2.8383333333333334e-06,
986
+ "loss": 0.3842,
987
+ "step": 6300
988
+ },
989
+ {
990
+ "epoch": 276.0869565217391,
991
+ "grad_norm": 0.8834450840950012,
992
+ "learning_rate": 2.7550000000000003e-06,
993
+ "loss": 0.3879,
994
+ "step": 6350
995
+ },
996
+ {
997
+ "epoch": 278.2608695652174,
998
+ "grad_norm": 1.0217564105987549,
999
+ "learning_rate": 2.6716666666666668e-06,
1000
+ "loss": 0.3892,
1001
+ "step": 6400
1002
+ },
1003
+ {
1004
+ "epoch": 280.4347826086956,
1005
+ "grad_norm": 0.6895825862884521,
1006
+ "learning_rate": 2.5883333333333337e-06,
1007
+ "loss": 0.389,
1008
+ "step": 6450
1009
+ },
1010
+ {
1011
+ "epoch": 282.60869565217394,
1012
+ "grad_norm": 0.6623283624649048,
1013
+ "learning_rate": 2.505e-06,
1014
+ "loss": 0.3854,
1015
+ "step": 6500
1016
+ },
1017
+ {
1018
+ "epoch": 282.60869565217394,
1019
+ "eval_loss": 0.413311630487442,
1020
+ "eval_runtime": 6.7085,
1021
+ "eval_samples_per_second": 23.701,
1022
+ "eval_steps_per_second": 2.981,
1023
+ "step": 6500
1024
+ },
1025
+ {
1026
+ "epoch": 284.7826086956522,
1027
+ "grad_norm": 0.7647244930267334,
1028
+ "learning_rate": 2.421666666666667e-06,
1029
+ "loss": 0.384,
1030
+ "step": 6550
1031
+ },
1032
+ {
1033
+ "epoch": 286.95652173913044,
1034
+ "grad_norm": 0.7481832504272461,
1035
+ "learning_rate": 2.3383333333333335e-06,
1036
+ "loss": 0.3897,
1037
+ "step": 6600
1038
+ },
1039
+ {
1040
+ "epoch": 289.1304347826087,
1041
+ "grad_norm": 0.9159369468688965,
1042
+ "learning_rate": 2.2550000000000004e-06,
1043
+ "loss": 0.3852,
1044
+ "step": 6650
1045
+ },
1046
+ {
1047
+ "epoch": 291.30434782608694,
1048
+ "grad_norm": 0.5772305727005005,
1049
+ "learning_rate": 2.171666666666667e-06,
1050
+ "loss": 0.3864,
1051
+ "step": 6700
1052
+ },
1053
+ {
1054
+ "epoch": 293.4782608695652,
1055
+ "grad_norm": 0.8706581592559814,
1056
+ "learning_rate": 2.088333333333334e-06,
1057
+ "loss": 0.3818,
1058
+ "step": 6750
1059
+ },
1060
+ {
1061
+ "epoch": 295.6521739130435,
1062
+ "grad_norm": 0.6359920501708984,
1063
+ "learning_rate": 2.0050000000000003e-06,
1064
+ "loss": 0.3866,
1065
+ "step": 6800
1066
+ },
1067
+ {
1068
+ "epoch": 297.82608695652175,
1069
+ "grad_norm": 0.6971527934074402,
1070
+ "learning_rate": 1.9216666666666668e-06,
1071
+ "loss": 0.3835,
1072
+ "step": 6850
1073
+ },
1074
+ {
1075
+ "epoch": 300.0,
1076
+ "grad_norm": 1.396098017692566,
1077
+ "learning_rate": 1.8383333333333334e-06,
1078
+ "loss": 0.3878,
1079
+ "step": 6900
1080
+ },
1081
+ {
1082
+ "epoch": 302.17391304347825,
1083
+ "grad_norm": 0.6345895528793335,
1084
+ "learning_rate": 1.7550000000000001e-06,
1085
+ "loss": 0.382,
1086
+ "step": 6950
1087
+ },
1088
+ {
1089
+ "epoch": 304.3478260869565,
1090
+ "grad_norm": 1.3194807767868042,
1091
+ "learning_rate": 1.6716666666666666e-06,
1092
+ "loss": 0.3885,
1093
+ "step": 7000
1094
+ },
1095
+ {
1096
+ "epoch": 304.3478260869565,
1097
+ "eval_loss": 0.4140428602695465,
1098
+ "eval_runtime": 6.4587,
1099
+ "eval_samples_per_second": 24.618,
1100
+ "eval_steps_per_second": 3.097,
1101
+ "step": 7000
1102
+ }
1103
+ ],
1104
+ "logging_steps": 50,
1105
+ "max_steps": 8000,
1106
+ "num_input_tokens_seen": 0,
1107
+ "num_train_epochs": 348,
1108
+ "save_steps": 500,
1109
+ "stateful_callbacks": {
1110
+ "TrainerControl": {
1111
+ "args": {
1112
+ "should_epoch_stop": false,
1113
+ "should_evaluate": false,
1114
+ "should_log": false,
1115
+ "should_save": true,
1116
+ "should_training_stop": false
1117
+ },
1118
+ "attributes": {}
1119
+ }
1120
+ },
1121
+ "total_flos": 1.0358217888890573e+17,
1122
+ "train_batch_size": 64,
1123
+ "trial_name": null,
1124
+ "trial_params": null
1125
+ }
checkpoint-7000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f95de2b194fe7f16ec10fbf1b93d83e04c25498d479793b9a5748c1d338bfbd0
3
+ size 5304