mikhail-panzo commited on
Commit
d1b2ded
1 Parent(s): 0be76e0

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": "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-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 sha256:6b1c5cd8237c317573a8050e348c7184b1c2d802d8d072ed1456503e0216943e
3
+ size 577789320
checkpoint-7000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:836adbd639be0f69bc3ea493d6c7eccd6bc8bf76daa8c89a567abe993b1da68f
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 sha256:7b3b259180fd41b794ed20d44e18c60d0ce7c26053f290e5d51f7909e215e07a
3
+ size 14244
checkpoint-7000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:507e305dfe4256b26c68fb0c82345ea0f5163e277ea888ae062da4422c6fa9ea
3
+ size 1064
checkpoint-7000/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-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": 0.3966299593448639,
3
+ "best_model_checkpoint": "mikhail-panzo/ceb_b128_le4_s8000/checkpoint-2000",
4
+ "epoch": 554.4554455445544,
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": 3.9603960396039604,
13
+ "grad_norm": 2.95651912689209,
14
+ "learning_rate": 2.5e-06,
15
+ "loss": 0.7707,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 7.920792079207921,
20
+ "grad_norm": 1.1273679733276367,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.6934,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 11.881188118811881,
27
+ "grad_norm": 2.1830034255981445,
28
+ "learning_rate": 7.5e-06,
29
+ "loss": 0.5983,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 15.841584158415841,
34
+ "grad_norm": 1.0733364820480347,
35
+ "learning_rate": 1e-05,
36
+ "loss": 0.5094,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 19.801980198019802,
41
+ "grad_norm": 1.2824387550354004,
42
+ "learning_rate": 1.25e-05,
43
+ "loss": 0.4913,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 23.762376237623762,
48
+ "grad_norm": 0.9164418578147888,
49
+ "learning_rate": 1.5e-05,
50
+ "loss": 0.4764,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 27.722772277227723,
55
+ "grad_norm": 0.8938279747962952,
56
+ "learning_rate": 1.75e-05,
57
+ "loss": 0.4648,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 31.683168316831683,
62
+ "grad_norm": 0.7855221033096313,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.459,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 35.64356435643565,
69
+ "grad_norm": 0.9247761964797974,
70
+ "learning_rate": 2.25e-05,
71
+ "loss": 0.4519,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 39.603960396039604,
76
+ "grad_norm": 1.1086606979370117,
77
+ "learning_rate": 2.5e-05,
78
+ "loss": 0.4432,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 39.603960396039604,
83
+ "eval_loss": 0.4058004915714264,
84
+ "eval_runtime": 7.6054,
85
+ "eval_samples_per_second": 23.667,
86
+ "eval_steps_per_second": 3.024,
87
+ "step": 500
88
+ },
89
+ {
90
+ "epoch": 43.56435643564357,
91
+ "grad_norm": 1.3994853496551514,
92
+ "learning_rate": 2.7500000000000004e-05,
93
+ "loss": 0.4438,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 47.524752475247524,
98
+ "grad_norm": 1.1104933023452759,
99
+ "learning_rate": 3e-05,
100
+ "loss": 0.4367,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 51.48514851485149,
105
+ "grad_norm": 0.7456745505332947,
106
+ "learning_rate": 3.2500000000000004e-05,
107
+ "loss": 0.4315,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 55.445544554455445,
112
+ "grad_norm": 1.0349308252334595,
113
+ "learning_rate": 3.5e-05,
114
+ "loss": 0.4294,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 59.40594059405941,
119
+ "grad_norm": 0.5290626287460327,
120
+ "learning_rate": 3.7500000000000003e-05,
121
+ "loss": 0.4234,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 63.366336633663366,
126
+ "grad_norm": 1.054995059967041,
127
+ "learning_rate": 4e-05,
128
+ "loss": 0.4231,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 67.32673267326733,
133
+ "grad_norm": 1.1774096488952637,
134
+ "learning_rate": 4.25e-05,
135
+ "loss": 0.4198,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 71.2871287128713,
140
+ "grad_norm": 0.7513113617897034,
141
+ "learning_rate": 4.5e-05,
142
+ "loss": 0.4182,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 75.24752475247524,
147
+ "grad_norm": 0.8465940952301025,
148
+ "learning_rate": 4.75e-05,
149
+ "loss": 0.4179,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 79.20792079207921,
154
+ "grad_norm": 1.601589322090149,
155
+ "learning_rate": 5e-05,
156
+ "loss": 0.4129,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 79.20792079207921,
161
+ "eval_loss": 0.3995668292045593,
162
+ "eval_runtime": 8.0229,
163
+ "eval_samples_per_second": 22.436,
164
+ "eval_steps_per_second": 2.867,
165
+ "step": 1000
166
+ },
167
+ {
168
+ "epoch": 83.16831683168317,
169
+ "grad_norm": 2.049264669418335,
170
+ "learning_rate": 5.25e-05,
171
+ "loss": 0.4129,
172
+ "step": 1050
173
+ },
174
+ {
175
+ "epoch": 87.12871287128714,
176
+ "grad_norm": 1.0590511560440063,
177
+ "learning_rate": 5.500000000000001e-05,
178
+ "loss": 0.4118,
179
+ "step": 1100
180
+ },
181
+ {
182
+ "epoch": 91.08910891089108,
183
+ "grad_norm": 1.015751600265503,
184
+ "learning_rate": 5.7499999999999995e-05,
185
+ "loss": 0.4074,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 95.04950495049505,
190
+ "grad_norm": 1.295837640762329,
191
+ "learning_rate": 6e-05,
192
+ "loss": 0.4056,
193
+ "step": 1200
194
+ },
195
+ {
196
+ "epoch": 99.00990099009901,
197
+ "grad_norm": 1.49947988986969,
198
+ "learning_rate": 6.25e-05,
199
+ "loss": 0.4052,
200
+ "step": 1250
201
+ },
202
+ {
203
+ "epoch": 102.97029702970298,
204
+ "grad_norm": 1.3930383920669556,
205
+ "learning_rate": 6.500000000000001e-05,
206
+ "loss": 0.4072,
207
+ "step": 1300
208
+ },
209
+ {
210
+ "epoch": 106.93069306930693,
211
+ "grad_norm": 1.1130273342132568,
212
+ "learning_rate": 6.750000000000001e-05,
213
+ "loss": 0.4015,
214
+ "step": 1350
215
+ },
216
+ {
217
+ "epoch": 110.89108910891089,
218
+ "grad_norm": 2.1075382232666016,
219
+ "learning_rate": 7e-05,
220
+ "loss": 0.3994,
221
+ "step": 1400
222
+ },
223
+ {
224
+ "epoch": 114.85148514851485,
225
+ "grad_norm": 1.8470244407653809,
226
+ "learning_rate": 7.25e-05,
227
+ "loss": 0.397,
228
+ "step": 1450
229
+ },
230
+ {
231
+ "epoch": 118.81188118811882,
232
+ "grad_norm": 2.0715270042419434,
233
+ "learning_rate": 7.500000000000001e-05,
234
+ "loss": 0.3992,
235
+ "step": 1500
236
+ },
237
+ {
238
+ "epoch": 118.81188118811882,
239
+ "eval_loss": 0.39855822920799255,
240
+ "eval_runtime": 7.8631,
241
+ "eval_samples_per_second": 22.892,
242
+ "eval_steps_per_second": 2.925,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 122.77227722772277,
247
+ "grad_norm": 0.9546501636505127,
248
+ "learning_rate": 7.75e-05,
249
+ "loss": 0.3934,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 126.73267326732673,
254
+ "grad_norm": 1.1335992813110352,
255
+ "learning_rate": 8e-05,
256
+ "loss": 0.3927,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 130.69306930693068,
261
+ "grad_norm": 1.905771017074585,
262
+ "learning_rate": 8.25e-05,
263
+ "loss": 0.3895,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 134.65346534653466,
268
+ "grad_norm": 1.129404067993164,
269
+ "learning_rate": 8.5e-05,
270
+ "loss": 0.3897,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 138.6138613861386,
275
+ "grad_norm": 0.9640969038009644,
276
+ "learning_rate": 8.75e-05,
277
+ "loss": 0.3894,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 142.5742574257426,
282
+ "grad_norm": 0.9840293526649475,
283
+ "learning_rate": 9e-05,
284
+ "loss": 0.3862,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 146.53465346534654,
289
+ "grad_norm": 2.770336627960205,
290
+ "learning_rate": 9.250000000000001e-05,
291
+ "loss": 0.3861,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 150.4950495049505,
296
+ "grad_norm": 1.7189514636993408,
297
+ "learning_rate": 9.5e-05,
298
+ "loss": 0.3914,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 154.45544554455446,
303
+ "grad_norm": 2.454707622528076,
304
+ "learning_rate": 9.75e-05,
305
+ "loss": 0.3848,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 158.41584158415841,
310
+ "grad_norm": 2.1555542945861816,
311
+ "learning_rate": 0.0001,
312
+ "loss": 0.3814,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 158.41584158415841,
317
+ "eval_loss": 0.3966299593448639,
318
+ "eval_runtime": 7.7159,
319
+ "eval_samples_per_second": 23.329,
320
+ "eval_steps_per_second": 2.981,
321
+ "step": 2000
322
+ },
323
+ {
324
+ "epoch": 162.37623762376236,
325
+ "grad_norm": 1.1228975057601929,
326
+ "learning_rate": 9.916666666666667e-05,
327
+ "loss": 0.3817,
328
+ "step": 2050
329
+ },
330
+ {
331
+ "epoch": 166.33663366336634,
332
+ "grad_norm": 1.629294991493225,
333
+ "learning_rate": 9.833333333333333e-05,
334
+ "loss": 0.3813,
335
+ "step": 2100
336
+ },
337
+ {
338
+ "epoch": 170.2970297029703,
339
+ "grad_norm": 0.9008976221084595,
340
+ "learning_rate": 9.75e-05,
341
+ "loss": 0.377,
342
+ "step": 2150
343
+ },
344
+ {
345
+ "epoch": 174.25742574257427,
346
+ "grad_norm": 2.174643039703369,
347
+ "learning_rate": 9.666666666666667e-05,
348
+ "loss": 0.3737,
349
+ "step": 2200
350
+ },
351
+ {
352
+ "epoch": 178.21782178217822,
353
+ "grad_norm": 1.617219090461731,
354
+ "learning_rate": 9.583333333333334e-05,
355
+ "loss": 0.3736,
356
+ "step": 2250
357
+ },
358
+ {
359
+ "epoch": 182.17821782178217,
360
+ "grad_norm": 1.292871117591858,
361
+ "learning_rate": 9.5e-05,
362
+ "loss": 0.3786,
363
+ "step": 2300
364
+ },
365
+ {
366
+ "epoch": 186.13861386138615,
367
+ "grad_norm": 1.5530415773391724,
368
+ "learning_rate": 9.416666666666667e-05,
369
+ "loss": 0.3726,
370
+ "step": 2350
371
+ },
372
+ {
373
+ "epoch": 190.0990099009901,
374
+ "grad_norm": 0.9837754964828491,
375
+ "learning_rate": 9.333333333333334e-05,
376
+ "loss": 0.3713,
377
+ "step": 2400
378
+ },
379
+ {
380
+ "epoch": 194.05940594059405,
381
+ "grad_norm": 0.8343392610549927,
382
+ "learning_rate": 9.250000000000001e-05,
383
+ "loss": 0.3727,
384
+ "step": 2450
385
+ },
386
+ {
387
+ "epoch": 198.01980198019803,
388
+ "grad_norm": 2.9330906867980957,
389
+ "learning_rate": 9.166666666666667e-05,
390
+ "loss": 0.3688,
391
+ "step": 2500
392
+ },
393
+ {
394
+ "epoch": 198.01980198019803,
395
+ "eval_loss": 0.40214845538139343,
396
+ "eval_runtime": 7.9273,
397
+ "eval_samples_per_second": 22.706,
398
+ "eval_steps_per_second": 2.901,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 201.98019801980197,
403
+ "grad_norm": 2.427091598510742,
404
+ "learning_rate": 9.083333333333334e-05,
405
+ "loss": 0.3699,
406
+ "step": 2550
407
+ },
408
+ {
409
+ "epoch": 205.94059405940595,
410
+ "grad_norm": 0.680686891078949,
411
+ "learning_rate": 9e-05,
412
+ "loss": 0.3659,
413
+ "step": 2600
414
+ },
415
+ {
416
+ "epoch": 209.9009900990099,
417
+ "grad_norm": 2.0860631465911865,
418
+ "learning_rate": 8.916666666666667e-05,
419
+ "loss": 0.3631,
420
+ "step": 2650
421
+ },
422
+ {
423
+ "epoch": 213.86138613861385,
424
+ "grad_norm": 0.7303352355957031,
425
+ "learning_rate": 8.833333333333333e-05,
426
+ "loss": 0.3656,
427
+ "step": 2700
428
+ },
429
+ {
430
+ "epoch": 217.82178217821783,
431
+ "grad_norm": 1.2716140747070312,
432
+ "learning_rate": 8.75e-05,
433
+ "loss": 0.3627,
434
+ "step": 2750
435
+ },
436
+ {
437
+ "epoch": 221.78217821782178,
438
+ "grad_norm": 1.5935007333755493,
439
+ "learning_rate": 8.666666666666667e-05,
440
+ "loss": 0.3607,
441
+ "step": 2800
442
+ },
443
+ {
444
+ "epoch": 225.74257425742573,
445
+ "grad_norm": 1.156733512878418,
446
+ "learning_rate": 8.583333333333334e-05,
447
+ "loss": 0.3594,
448
+ "step": 2850
449
+ },
450
+ {
451
+ "epoch": 229.7029702970297,
452
+ "grad_norm": 1.1708168983459473,
453
+ "learning_rate": 8.5e-05,
454
+ "loss": 0.3596,
455
+ "step": 2900
456
+ },
457
+ {
458
+ "epoch": 233.66336633663366,
459
+ "grad_norm": 1.4882978200912476,
460
+ "learning_rate": 8.416666666666668e-05,
461
+ "loss": 0.3613,
462
+ "step": 2950
463
+ },
464
+ {
465
+ "epoch": 237.62376237623764,
466
+ "grad_norm": 1.0416059494018555,
467
+ "learning_rate": 8.333333333333334e-05,
468
+ "loss": 0.3574,
469
+ "step": 3000
470
+ },
471
+ {
472
+ "epoch": 237.62376237623764,
473
+ "eval_loss": 0.39999938011169434,
474
+ "eval_runtime": 7.4331,
475
+ "eval_samples_per_second": 24.216,
476
+ "eval_steps_per_second": 3.094,
477
+ "step": 3000
478
+ },
479
+ {
480
+ "epoch": 241.58415841584159,
481
+ "grad_norm": 0.9737104773521423,
482
+ "learning_rate": 8.25e-05,
483
+ "loss": 0.3555,
484
+ "step": 3050
485
+ },
486
+ {
487
+ "epoch": 245.54455445544554,
488
+ "grad_norm": 2.4858767986297607,
489
+ "learning_rate": 8.166666666666667e-05,
490
+ "loss": 0.3585,
491
+ "step": 3100
492
+ },
493
+ {
494
+ "epoch": 249.5049504950495,
495
+ "grad_norm": 0.7558372616767883,
496
+ "learning_rate": 8.083333333333334e-05,
497
+ "loss": 0.3562,
498
+ "step": 3150
499
+ },
500
+ {
501
+ "epoch": 253.46534653465346,
502
+ "grad_norm": 0.5738223195075989,
503
+ "learning_rate": 8e-05,
504
+ "loss": 0.3515,
505
+ "step": 3200
506
+ },
507
+ {
508
+ "epoch": 257.4257425742574,
509
+ "grad_norm": 0.8245187401771545,
510
+ "learning_rate": 7.916666666666666e-05,
511
+ "loss": 0.3557,
512
+ "step": 3250
513
+ },
514
+ {
515
+ "epoch": 261.38613861386136,
516
+ "grad_norm": 0.8837783336639404,
517
+ "learning_rate": 7.833333333333333e-05,
518
+ "loss": 0.3556,
519
+ "step": 3300
520
+ },
521
+ {
522
+ "epoch": 265.34653465346537,
523
+ "grad_norm": 1.0626522302627563,
524
+ "learning_rate": 7.75e-05,
525
+ "loss": 0.3531,
526
+ "step": 3350
527
+ },
528
+ {
529
+ "epoch": 269.3069306930693,
530
+ "grad_norm": 0.9158534407615662,
531
+ "learning_rate": 7.666666666666667e-05,
532
+ "loss": 0.3504,
533
+ "step": 3400
534
+ },
535
+ {
536
+ "epoch": 273.26732673267327,
537
+ "grad_norm": 1.1604119539260864,
538
+ "learning_rate": 7.583333333333334e-05,
539
+ "loss": 0.3502,
540
+ "step": 3450
541
+ },
542
+ {
543
+ "epoch": 277.2277227722772,
544
+ "grad_norm": 0.5547149181365967,
545
+ "learning_rate": 7.500000000000001e-05,
546
+ "loss": 0.3482,
547
+ "step": 3500
548
+ },
549
+ {
550
+ "epoch": 277.2277227722772,
551
+ "eval_loss": 0.39900416135787964,
552
+ "eval_runtime": 7.7665,
553
+ "eval_samples_per_second": 23.177,
554
+ "eval_steps_per_second": 2.961,
555
+ "step": 3500
556
+ },
557
+ {
558
+ "epoch": 281.18811881188117,
559
+ "grad_norm": 1.213918924331665,
560
+ "learning_rate": 7.416666666666668e-05,
561
+ "loss": 0.3465,
562
+ "step": 3550
563
+ },
564
+ {
565
+ "epoch": 285.1485148514852,
566
+ "grad_norm": 1.0824240446090698,
567
+ "learning_rate": 7.333333333333333e-05,
568
+ "loss": 0.3467,
569
+ "step": 3600
570
+ },
571
+ {
572
+ "epoch": 289.1089108910891,
573
+ "grad_norm": 0.8079837560653687,
574
+ "learning_rate": 7.25e-05,
575
+ "loss": 0.3457,
576
+ "step": 3650
577
+ },
578
+ {
579
+ "epoch": 293.0693069306931,
580
+ "grad_norm": 1.1617168188095093,
581
+ "learning_rate": 7.166666666666667e-05,
582
+ "loss": 0.3457,
583
+ "step": 3700
584
+ },
585
+ {
586
+ "epoch": 297.029702970297,
587
+ "grad_norm": 1.0282728672027588,
588
+ "learning_rate": 7.083333333333334e-05,
589
+ "loss": 0.3491,
590
+ "step": 3750
591
+ },
592
+ {
593
+ "epoch": 300.990099009901,
594
+ "grad_norm": 0.8772862553596497,
595
+ "learning_rate": 7e-05,
596
+ "loss": 0.3494,
597
+ "step": 3800
598
+ },
599
+ {
600
+ "epoch": 304.9504950495049,
601
+ "grad_norm": 1.0405845642089844,
602
+ "learning_rate": 6.916666666666666e-05,
603
+ "loss": 0.3469,
604
+ "step": 3850
605
+ },
606
+ {
607
+ "epoch": 308.91089108910893,
608
+ "grad_norm": 0.7217592000961304,
609
+ "learning_rate": 6.833333333333333e-05,
610
+ "loss": 0.3449,
611
+ "step": 3900
612
+ },
613
+ {
614
+ "epoch": 312.8712871287129,
615
+ "grad_norm": 0.7694669365882874,
616
+ "learning_rate": 6.750000000000001e-05,
617
+ "loss": 0.341,
618
+ "step": 3950
619
+ },
620
+ {
621
+ "epoch": 316.83168316831683,
622
+ "grad_norm": 0.6203835010528564,
623
+ "learning_rate": 6.666666666666667e-05,
624
+ "loss": 0.3421,
625
+ "step": 4000
626
+ },
627
+ {
628
+ "epoch": 316.83168316831683,
629
+ "eval_loss": 0.4013901650905609,
630
+ "eval_runtime": 8.6905,
631
+ "eval_samples_per_second": 20.712,
632
+ "eval_steps_per_second": 2.647,
633
+ "step": 4000
634
+ },
635
+ {
636
+ "epoch": 320.7920792079208,
637
+ "grad_norm": 0.7539447546005249,
638
+ "learning_rate": 6.583333333333334e-05,
639
+ "loss": 0.3445,
640
+ "step": 4050
641
+ },
642
+ {
643
+ "epoch": 324.7524752475247,
644
+ "grad_norm": 0.9108504056930542,
645
+ "learning_rate": 6.500000000000001e-05,
646
+ "loss": 0.341,
647
+ "step": 4100
648
+ },
649
+ {
650
+ "epoch": 328.71287128712873,
651
+ "grad_norm": 0.5291554927825928,
652
+ "learning_rate": 6.416666666666668e-05,
653
+ "loss": 0.3412,
654
+ "step": 4150
655
+ },
656
+ {
657
+ "epoch": 332.6732673267327,
658
+ "grad_norm": 1.2332825660705566,
659
+ "learning_rate": 6.333333333333333e-05,
660
+ "loss": 0.3408,
661
+ "step": 4200
662
+ },
663
+ {
664
+ "epoch": 336.63366336633663,
665
+ "grad_norm": 0.5890567302703857,
666
+ "learning_rate": 6.25e-05,
667
+ "loss": 0.339,
668
+ "step": 4250
669
+ },
670
+ {
671
+ "epoch": 340.5940594059406,
672
+ "grad_norm": 1.05361807346344,
673
+ "learning_rate": 6.168333333333333e-05,
674
+ "loss": 0.3388,
675
+ "step": 4300
676
+ },
677
+ {
678
+ "epoch": 344.55445544554453,
679
+ "grad_norm": 1.6529746055603027,
680
+ "learning_rate": 6.085000000000001e-05,
681
+ "loss": 0.3423,
682
+ "step": 4350
683
+ },
684
+ {
685
+ "epoch": 348.51485148514854,
686
+ "grad_norm": 0.6428557634353638,
687
+ "learning_rate": 6.0016666666666664e-05,
688
+ "loss": 0.3406,
689
+ "step": 4400
690
+ },
691
+ {
692
+ "epoch": 352.4752475247525,
693
+ "grad_norm": 0.832099437713623,
694
+ "learning_rate": 5.918333333333333e-05,
695
+ "loss": 0.3399,
696
+ "step": 4450
697
+ },
698
+ {
699
+ "epoch": 356.43564356435644,
700
+ "grad_norm": 1.4958369731903076,
701
+ "learning_rate": 5.835e-05,
702
+ "loss": 0.3414,
703
+ "step": 4500
704
+ },
705
+ {
706
+ "epoch": 356.43564356435644,
707
+ "eval_loss": 0.4059315621852875,
708
+ "eval_runtime": 7.3212,
709
+ "eval_samples_per_second": 24.586,
710
+ "eval_steps_per_second": 3.142,
711
+ "step": 4500
712
+ },
713
+ {
714
+ "epoch": 360.3960396039604,
715
+ "grad_norm": 0.7380837798118591,
716
+ "learning_rate": 5.751666666666667e-05,
717
+ "loss": 0.3363,
718
+ "step": 4550
719
+ },
720
+ {
721
+ "epoch": 364.35643564356434,
722
+ "grad_norm": 1.1721248626708984,
723
+ "learning_rate": 5.668333333333333e-05,
724
+ "loss": 0.3385,
725
+ "step": 4600
726
+ },
727
+ {
728
+ "epoch": 368.3168316831683,
729
+ "grad_norm": 0.6880260109901428,
730
+ "learning_rate": 5.585e-05,
731
+ "loss": 0.3355,
732
+ "step": 4650
733
+ },
734
+ {
735
+ "epoch": 372.2772277227723,
736
+ "grad_norm": 0.614639401435852,
737
+ "learning_rate": 5.501666666666667e-05,
738
+ "loss": 0.3378,
739
+ "step": 4700
740
+ },
741
+ {
742
+ "epoch": 376.23762376237624,
743
+ "grad_norm": 0.6579806208610535,
744
+ "learning_rate": 5.4183333333333334e-05,
745
+ "loss": 0.3342,
746
+ "step": 4750
747
+ },
748
+ {
749
+ "epoch": 380.1980198019802,
750
+ "grad_norm": 0.7675396203994751,
751
+ "learning_rate": 5.335e-05,
752
+ "loss": 0.3333,
753
+ "step": 4800
754
+ },
755
+ {
756
+ "epoch": 384.15841584158414,
757
+ "grad_norm": 0.6529810428619385,
758
+ "learning_rate": 5.251666666666667e-05,
759
+ "loss": 0.3348,
760
+ "step": 4850
761
+ },
762
+ {
763
+ "epoch": 388.1188118811881,
764
+ "grad_norm": 0.6606137752532959,
765
+ "learning_rate": 5.168333333333334e-05,
766
+ "loss": 0.3351,
767
+ "step": 4900
768
+ },
769
+ {
770
+ "epoch": 392.0792079207921,
771
+ "grad_norm": 0.5794258713722229,
772
+ "learning_rate": 5.0849999999999996e-05,
773
+ "loss": 0.3319,
774
+ "step": 4950
775
+ },
776
+ {
777
+ "epoch": 396.03960396039605,
778
+ "grad_norm": 0.8365175127983093,
779
+ "learning_rate": 5.0016666666666665e-05,
780
+ "loss": 0.3328,
781
+ "step": 5000
782
+ },
783
+ {
784
+ "epoch": 396.03960396039605,
785
+ "eval_loss": 0.40650251507759094,
786
+ "eval_runtime": 7.4758,
787
+ "eval_samples_per_second": 24.078,
788
+ "eval_steps_per_second": 3.077,
789
+ "step": 5000
790
+ },
791
+ {
792
+ "epoch": 400.0,
793
+ "grad_norm": 0.7176100611686707,
794
+ "learning_rate": 4.9183333333333334e-05,
795
+ "loss": 0.3346,
796
+ "step": 5050
797
+ },
798
+ {
799
+ "epoch": 403.96039603960395,
800
+ "grad_norm": 0.9245086312294006,
801
+ "learning_rate": 4.835e-05,
802
+ "loss": 0.3319,
803
+ "step": 5100
804
+ },
805
+ {
806
+ "epoch": 407.9207920792079,
807
+ "grad_norm": 0.7508360743522644,
808
+ "learning_rate": 4.751666666666667e-05,
809
+ "loss": 0.3344,
810
+ "step": 5150
811
+ },
812
+ {
813
+ "epoch": 411.8811881188119,
814
+ "grad_norm": 0.8295847177505493,
815
+ "learning_rate": 4.6683333333333334e-05,
816
+ "loss": 0.3311,
817
+ "step": 5200
818
+ },
819
+ {
820
+ "epoch": 415.84158415841586,
821
+ "grad_norm": 0.9669463634490967,
822
+ "learning_rate": 4.585e-05,
823
+ "loss": 0.3307,
824
+ "step": 5250
825
+ },
826
+ {
827
+ "epoch": 419.8019801980198,
828
+ "grad_norm": 0.6602340340614319,
829
+ "learning_rate": 4.5016666666666665e-05,
830
+ "loss": 0.3302,
831
+ "step": 5300
832
+ },
833
+ {
834
+ "epoch": 423.76237623762376,
835
+ "grad_norm": 0.7802284955978394,
836
+ "learning_rate": 4.4183333333333334e-05,
837
+ "loss": 0.332,
838
+ "step": 5350
839
+ },
840
+ {
841
+ "epoch": 427.7227722772277,
842
+ "grad_norm": 0.6156259179115295,
843
+ "learning_rate": 4.335e-05,
844
+ "loss": 0.3305,
845
+ "step": 5400
846
+ },
847
+ {
848
+ "epoch": 431.6831683168317,
849
+ "grad_norm": 0.636411726474762,
850
+ "learning_rate": 4.251666666666667e-05,
851
+ "loss": 0.3309,
852
+ "step": 5450
853
+ },
854
+ {
855
+ "epoch": 435.64356435643566,
856
+ "grad_norm": 0.7911040186882019,
857
+ "learning_rate": 4.1683333333333335e-05,
858
+ "loss": 0.3277,
859
+ "step": 5500
860
+ },
861
+ {
862
+ "epoch": 435.64356435643566,
863
+ "eval_loss": 0.40623539686203003,
864
+ "eval_runtime": 7.0665,
865
+ "eval_samples_per_second": 25.472,
866
+ "eval_steps_per_second": 3.255,
867
+ "step": 5500
868
+ },
869
+ {
870
+ "epoch": 439.6039603960396,
871
+ "grad_norm": 0.4689953625202179,
872
+ "learning_rate": 4.085e-05,
873
+ "loss": 0.3281,
874
+ "step": 5550
875
+ },
876
+ {
877
+ "epoch": 443.56435643564356,
878
+ "grad_norm": 0.57826828956604,
879
+ "learning_rate": 4.0016666666666666e-05,
880
+ "loss": 0.3316,
881
+ "step": 5600
882
+ },
883
+ {
884
+ "epoch": 447.5247524752475,
885
+ "grad_norm": 0.5948452949523926,
886
+ "learning_rate": 3.9183333333333335e-05,
887
+ "loss": 0.3287,
888
+ "step": 5650
889
+ },
890
+ {
891
+ "epoch": 451.48514851485146,
892
+ "grad_norm": 0.6122576594352722,
893
+ "learning_rate": 3.8350000000000004e-05,
894
+ "loss": 0.3267,
895
+ "step": 5700
896
+ },
897
+ {
898
+ "epoch": 455.44554455445547,
899
+ "grad_norm": 0.48124071955680847,
900
+ "learning_rate": 3.7516666666666666e-05,
901
+ "loss": 0.3279,
902
+ "step": 5750
903
+ },
904
+ {
905
+ "epoch": 459.4059405940594,
906
+ "grad_norm": 0.5144623517990112,
907
+ "learning_rate": 3.6683333333333335e-05,
908
+ "loss": 0.3257,
909
+ "step": 5800
910
+ },
911
+ {
912
+ "epoch": 463.36633663366337,
913
+ "grad_norm": 1.026330828666687,
914
+ "learning_rate": 3.585e-05,
915
+ "loss": 0.3272,
916
+ "step": 5850
917
+ },
918
+ {
919
+ "epoch": 467.3267326732673,
920
+ "grad_norm": 0.6319882869720459,
921
+ "learning_rate": 3.501666666666667e-05,
922
+ "loss": 0.3247,
923
+ "step": 5900
924
+ },
925
+ {
926
+ "epoch": 471.28712871287127,
927
+ "grad_norm": 0.7657246589660645,
928
+ "learning_rate": 3.4183333333333335e-05,
929
+ "loss": 0.3257,
930
+ "step": 5950
931
+ },
932
+ {
933
+ "epoch": 475.2475247524753,
934
+ "grad_norm": 0.67906254529953,
935
+ "learning_rate": 3.3350000000000004e-05,
936
+ "loss": 0.3248,
937
+ "step": 6000
938
+ },
939
+ {
940
+ "epoch": 475.2475247524753,
941
+ "eval_loss": 0.40691491961479187,
942
+ "eval_runtime": 8.6221,
943
+ "eval_samples_per_second": 20.876,
944
+ "eval_steps_per_second": 2.668,
945
+ "step": 6000
946
+ },
947
+ {
948
+ "epoch": 479.2079207920792,
949
+ "grad_norm": 0.5330384969711304,
950
+ "learning_rate": 3.2516666666666666e-05,
951
+ "loss": 0.3289,
952
+ "step": 6050
953
+ },
954
+ {
955
+ "epoch": 483.16831683168317,
956
+ "grad_norm": 0.7305939793586731,
957
+ "learning_rate": 3.1683333333333335e-05,
958
+ "loss": 0.3267,
959
+ "step": 6100
960
+ },
961
+ {
962
+ "epoch": 487.1287128712871,
963
+ "grad_norm": 1.1130585670471191,
964
+ "learning_rate": 3.0850000000000004e-05,
965
+ "loss": 0.3247,
966
+ "step": 6150
967
+ },
968
+ {
969
+ "epoch": 491.08910891089107,
970
+ "grad_norm": 0.5422760844230652,
971
+ "learning_rate": 3.001666666666667e-05,
972
+ "loss": 0.3252,
973
+ "step": 6200
974
+ },
975
+ {
976
+ "epoch": 495.0495049504951,
977
+ "grad_norm": 0.5144603252410889,
978
+ "learning_rate": 2.9183333333333336e-05,
979
+ "loss": 0.3276,
980
+ "step": 6250
981
+ },
982
+ {
983
+ "epoch": 499.009900990099,
984
+ "grad_norm": 0.5953711867332458,
985
+ "learning_rate": 2.8349999999999998e-05,
986
+ "loss": 0.3259,
987
+ "step": 6300
988
+ },
989
+ {
990
+ "epoch": 502.970297029703,
991
+ "grad_norm": 0.48666295409202576,
992
+ "learning_rate": 2.7516666666666667e-05,
993
+ "loss": 0.3233,
994
+ "step": 6350
995
+ },
996
+ {
997
+ "epoch": 506.9306930693069,
998
+ "grad_norm": 0.44972455501556396,
999
+ "learning_rate": 2.6683333333333333e-05,
1000
+ "loss": 0.3223,
1001
+ "step": 6400
1002
+ },
1003
+ {
1004
+ "epoch": 510.8910891089109,
1005
+ "grad_norm": 0.5804650783538818,
1006
+ "learning_rate": 2.585e-05,
1007
+ "loss": 0.3246,
1008
+ "step": 6450
1009
+ },
1010
+ {
1011
+ "epoch": 514.8514851485148,
1012
+ "grad_norm": 0.41495367884635925,
1013
+ "learning_rate": 2.5016666666666667e-05,
1014
+ "loss": 0.3245,
1015
+ "step": 6500
1016
+ },
1017
+ {
1018
+ "epoch": 514.8514851485148,
1019
+ "eval_loss": 0.4071478545665741,
1020
+ "eval_runtime": 8.0659,
1021
+ "eval_samples_per_second": 22.316,
1022
+ "eval_steps_per_second": 2.852,
1023
+ "step": 6500
1024
+ },
1025
+ {
1026
+ "epoch": 518.8118811881188,
1027
+ "grad_norm": 0.3836776912212372,
1028
+ "learning_rate": 2.4183333333333336e-05,
1029
+ "loss": 0.3236,
1030
+ "step": 6550
1031
+ },
1032
+ {
1033
+ "epoch": 522.7722772277227,
1034
+ "grad_norm": 0.45108506083488464,
1035
+ "learning_rate": 2.3350000000000002e-05,
1036
+ "loss": 0.3225,
1037
+ "step": 6600
1038
+ },
1039
+ {
1040
+ "epoch": 526.7326732673267,
1041
+ "grad_norm": 0.36493930220603943,
1042
+ "learning_rate": 2.2516666666666667e-05,
1043
+ "loss": 0.324,
1044
+ "step": 6650
1045
+ },
1046
+ {
1047
+ "epoch": 530.6930693069307,
1048
+ "grad_norm": 0.4647804796695709,
1049
+ "learning_rate": 2.1683333333333333e-05,
1050
+ "loss": 0.3236,
1051
+ "step": 6700
1052
+ },
1053
+ {
1054
+ "epoch": 534.6534653465346,
1055
+ "grad_norm": 0.3885102868080139,
1056
+ "learning_rate": 2.085e-05,
1057
+ "loss": 0.3207,
1058
+ "step": 6750
1059
+ },
1060
+ {
1061
+ "epoch": 538.6138613861386,
1062
+ "grad_norm": 0.6492519378662109,
1063
+ "learning_rate": 2.0016666666666668e-05,
1064
+ "loss": 0.3226,
1065
+ "step": 6800
1066
+ },
1067
+ {
1068
+ "epoch": 542.5742574257425,
1069
+ "grad_norm": 0.3782261610031128,
1070
+ "learning_rate": 1.9183333333333333e-05,
1071
+ "loss": 0.3211,
1072
+ "step": 6850
1073
+ },
1074
+ {
1075
+ "epoch": 546.5346534653465,
1076
+ "grad_norm": 0.5947457551956177,
1077
+ "learning_rate": 1.8350000000000002e-05,
1078
+ "loss": 0.3228,
1079
+ "step": 6900
1080
+ },
1081
+ {
1082
+ "epoch": 550.4950495049505,
1083
+ "grad_norm": 0.3643256425857544,
1084
+ "learning_rate": 1.7516666666666668e-05,
1085
+ "loss": 0.3206,
1086
+ "step": 6950
1087
+ },
1088
+ {
1089
+ "epoch": 554.4554455445544,
1090
+ "grad_norm": 0.4753130078315735,
1091
+ "learning_rate": 1.6683333333333333e-05,
1092
+ "loss": 0.3219,
1093
+ "step": 7000
1094
+ },
1095
+ {
1096
+ "epoch": 554.4554455445544,
1097
+ "eval_loss": 0.4096488654613495,
1098
+ "eval_runtime": 9.31,
1099
+ "eval_samples_per_second": 19.334,
1100
+ "eval_steps_per_second": 2.47,
1101
+ "step": 7000
1102
+ }
1103
+ ],
1104
+ "logging_steps": 50,
1105
+ "max_steps": 8000,
1106
+ "num_input_tokens_seen": 0,
1107
+ "num_train_epochs": 667,
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": 1.5126342636671082e+17,
1122
+ "train_batch_size": 16,
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 sha256:85df14de5cd709ebb874ec61066a56bc489b0cf6ac7193ea95d4fbf830394508
3
+ size 5304