mikhail-panzo commited on
Commit
2cc0e7b
1 Parent(s): b95a9ba

Training in progress, step 7500, checkpoint

Browse files
checkpoint-7500/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-7500/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-7500/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-7500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbf51c633a114b828fa7da67f7778f64b0664e8d596980516137f579ac02c51b
3
+ size 577789320
checkpoint-7500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:921457aabfb1d1b1212159362ed706ccd507c154df718ca3fcfb42cd9579277c
3
+ size 1155772233
checkpoint-7500/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-7500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3bfd2a16b4e4f15db0cc3f93abde98aba5f2a4c591eaa1f8d640b28b586a74c
3
+ size 14244
checkpoint-7500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5cdfe5ed4f14bdadfee62402701e9c3c91a7e1b8246c7c7f0be536b67574fb3
3
+ size 1064
checkpoint-7500/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-7500/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-7500/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-7500/trainer_state.json ADDED
@@ -0,0 +1,1203 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.32830387353897095,
3
+ "best_model_checkpoint": "mikhail-panzo/zlm_b32_le4_s8000/checkpoint-7500",
4
+ "epoch": 3.141361256544503,
5
+ "eval_steps": 500,
6
+ "global_step": 7500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.020942408376963352,
13
+ "grad_norm": 3.1775436401367188,
14
+ "learning_rate": 2.35e-06,
15
+ "loss": 1.1228,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.041884816753926704,
20
+ "grad_norm": 4.18980073928833,
21
+ "learning_rate": 4.85e-06,
22
+ "loss": 0.8304,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.06282722513089005,
27
+ "grad_norm": 4.5355024337768555,
28
+ "learning_rate": 7.35e-06,
29
+ "loss": 0.7701,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.08376963350785341,
34
+ "grad_norm": 3.9590957164764404,
35
+ "learning_rate": 9.85e-06,
36
+ "loss": 0.7096,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.10471204188481675,
41
+ "grad_norm": 5.419675350189209,
42
+ "learning_rate": 1.235e-05,
43
+ "loss": 0.6518,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.1256544502617801,
48
+ "grad_norm": 3.1052777767181396,
49
+ "learning_rate": 1.485e-05,
50
+ "loss": 0.6326,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.14659685863874344,
55
+ "grad_norm": 3.4649453163146973,
56
+ "learning_rate": 1.7349999999999998e-05,
57
+ "loss": 0.6241,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.16753926701570682,
62
+ "grad_norm": 4.962624549865723,
63
+ "learning_rate": 1.985e-05,
64
+ "loss": 0.5723,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.18848167539267016,
69
+ "grad_norm": 2.7586095333099365,
70
+ "learning_rate": 2.235e-05,
71
+ "loss": 0.5764,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.2094240837696335,
76
+ "grad_norm": 3.5610299110412598,
77
+ "learning_rate": 2.485e-05,
78
+ "loss": 0.5552,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.2094240837696335,
83
+ "eval_loss": 0.48825809359550476,
84
+ "eval_runtime": 277.5892,
85
+ "eval_samples_per_second": 30.581,
86
+ "eval_steps_per_second": 3.826,
87
+ "step": 500
88
+ },
89
+ {
90
+ "epoch": 0.23036649214659685,
91
+ "grad_norm": 3.3351809978485107,
92
+ "learning_rate": 2.7350000000000004e-05,
93
+ "loss": 0.5322,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 0.2513089005235602,
98
+ "grad_norm": 3.7222373485565186,
99
+ "learning_rate": 2.985e-05,
100
+ "loss": 0.5186,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 0.27225130890052357,
105
+ "grad_norm": 3.1811156272888184,
106
+ "learning_rate": 3.235e-05,
107
+ "loss": 0.5157,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 0.2931937172774869,
112
+ "grad_norm": 2.2585642337799072,
113
+ "learning_rate": 3.485e-05,
114
+ "loss": 0.5089,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 0.31413612565445026,
119
+ "grad_norm": 3.999460220336914,
120
+ "learning_rate": 3.735e-05,
121
+ "loss": 0.5117,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 0.33507853403141363,
126
+ "grad_norm": 3.739990472793579,
127
+ "learning_rate": 3.9850000000000006e-05,
128
+ "loss": 0.5031,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 0.35602094240837695,
133
+ "grad_norm": 4.251980781555176,
134
+ "learning_rate": 4.235e-05,
135
+ "loss": 0.5064,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 0.3769633507853403,
140
+ "grad_norm": 2.770602226257324,
141
+ "learning_rate": 4.4850000000000006e-05,
142
+ "loss": 0.4921,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 0.39790575916230364,
147
+ "grad_norm": 2.506974220275879,
148
+ "learning_rate": 4.735e-05,
149
+ "loss": 0.4839,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 0.418848167539267,
154
+ "grad_norm": 2.2666189670562744,
155
+ "learning_rate": 4.9850000000000006e-05,
156
+ "loss": 0.4913,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 0.418848167539267,
161
+ "eval_loss": 0.4265913963317871,
162
+ "eval_runtime": 265.1599,
163
+ "eval_samples_per_second": 32.015,
164
+ "eval_steps_per_second": 4.005,
165
+ "step": 1000
166
+ },
167
+ {
168
+ "epoch": 0.4397905759162304,
169
+ "grad_norm": 3.1451058387756348,
170
+ "learning_rate": 5.235e-05,
171
+ "loss": 0.4722,
172
+ "step": 1050
173
+ },
174
+ {
175
+ "epoch": 0.4607329842931937,
176
+ "grad_norm": 3.197997570037842,
177
+ "learning_rate": 5.485e-05,
178
+ "loss": 0.4692,
179
+ "step": 1100
180
+ },
181
+ {
182
+ "epoch": 0.4816753926701571,
183
+ "grad_norm": 2.9112601280212402,
184
+ "learning_rate": 5.7350000000000005e-05,
185
+ "loss": 0.4738,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 0.5026178010471204,
190
+ "grad_norm": 3.036731243133545,
191
+ "learning_rate": 5.9850000000000005e-05,
192
+ "loss": 0.4521,
193
+ "step": 1200
194
+ },
195
+ {
196
+ "epoch": 0.5235602094240838,
197
+ "grad_norm": 5.021958351135254,
198
+ "learning_rate": 6.235000000000001e-05,
199
+ "loss": 0.4666,
200
+ "step": 1250
201
+ },
202
+ {
203
+ "epoch": 0.5445026178010471,
204
+ "grad_norm": 3.302204132080078,
205
+ "learning_rate": 6.485e-05,
206
+ "loss": 0.4625,
207
+ "step": 1300
208
+ },
209
+ {
210
+ "epoch": 0.5654450261780105,
211
+ "grad_norm": 3.2643635272979736,
212
+ "learning_rate": 6.735e-05,
213
+ "loss": 0.4683,
214
+ "step": 1350
215
+ },
216
+ {
217
+ "epoch": 0.5863874345549738,
218
+ "grad_norm": 1.7499467134475708,
219
+ "learning_rate": 6.985e-05,
220
+ "loss": 0.449,
221
+ "step": 1400
222
+ },
223
+ {
224
+ "epoch": 0.6073298429319371,
225
+ "grad_norm": 1.3616622686386108,
226
+ "learning_rate": 7.235000000000001e-05,
227
+ "loss": 0.4523,
228
+ "step": 1450
229
+ },
230
+ {
231
+ "epoch": 0.6282722513089005,
232
+ "grad_norm": 2.5826191902160645,
233
+ "learning_rate": 7.485e-05,
234
+ "loss": 0.446,
235
+ "step": 1500
236
+ },
237
+ {
238
+ "epoch": 0.6282722513089005,
239
+ "eval_loss": 0.3975289463996887,
240
+ "eval_runtime": 266.4222,
241
+ "eval_samples_per_second": 31.863,
242
+ "eval_steps_per_second": 3.986,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 0.6492146596858639,
247
+ "grad_norm": 3.6052303314208984,
248
+ "learning_rate": 7.735e-05,
249
+ "loss": 0.4449,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 0.6701570680628273,
254
+ "grad_norm": 3.4120566844940186,
255
+ "learning_rate": 7.985e-05,
256
+ "loss": 0.4477,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 0.6910994764397905,
261
+ "grad_norm": 2.187040090560913,
262
+ "learning_rate": 8.235000000000001e-05,
263
+ "loss": 0.4522,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 0.7120418848167539,
268
+ "grad_norm": 1.718518853187561,
269
+ "learning_rate": 8.485e-05,
270
+ "loss": 0.4431,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 0.7329842931937173,
275
+ "grad_norm": 1.8248894214630127,
276
+ "learning_rate": 8.735000000000001e-05,
277
+ "loss": 0.4496,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 0.7539267015706806,
282
+ "grad_norm": 2.733355760574341,
283
+ "learning_rate": 8.985e-05,
284
+ "loss": 0.4297,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 0.774869109947644,
289
+ "grad_norm": 2.5899884700775146,
290
+ "learning_rate": 9.235000000000001e-05,
291
+ "loss": 0.4588,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 0.7958115183246073,
296
+ "grad_norm": 1.5704914331436157,
297
+ "learning_rate": 9.485e-05,
298
+ "loss": 0.4345,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 0.8167539267015707,
303
+ "grad_norm": 5.18487548828125,
304
+ "learning_rate": 9.735000000000001e-05,
305
+ "loss": 0.4325,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 0.837696335078534,
310
+ "grad_norm": 2.4392499923706055,
311
+ "learning_rate": 9.985000000000001e-05,
312
+ "loss": 0.4222,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 0.837696335078534,
317
+ "eval_loss": 0.39489272236824036,
318
+ "eval_runtime": 267.0393,
319
+ "eval_samples_per_second": 31.789,
320
+ "eval_steps_per_second": 3.977,
321
+ "step": 2000
322
+ },
323
+ {
324
+ "epoch": 0.8586387434554974,
325
+ "grad_norm": 2.1138336658477783,
326
+ "learning_rate": 9.921666666666668e-05,
327
+ "loss": 0.4438,
328
+ "step": 2050
329
+ },
330
+ {
331
+ "epoch": 0.8795811518324608,
332
+ "grad_norm": 2.3158669471740723,
333
+ "learning_rate": 9.838333333333334e-05,
334
+ "loss": 0.421,
335
+ "step": 2100
336
+ },
337
+ {
338
+ "epoch": 0.900523560209424,
339
+ "grad_norm": 2.1935179233551025,
340
+ "learning_rate": 9.755000000000001e-05,
341
+ "loss": 0.4247,
342
+ "step": 2150
343
+ },
344
+ {
345
+ "epoch": 0.9214659685863874,
346
+ "grad_norm": 1.7845231294631958,
347
+ "learning_rate": 9.671666666666667e-05,
348
+ "loss": 0.4203,
349
+ "step": 2200
350
+ },
351
+ {
352
+ "epoch": 0.9424083769633508,
353
+ "grad_norm": 1.7287964820861816,
354
+ "learning_rate": 9.588333333333334e-05,
355
+ "loss": 0.4306,
356
+ "step": 2250
357
+ },
358
+ {
359
+ "epoch": 0.9633507853403142,
360
+ "grad_norm": 2.2276289463043213,
361
+ "learning_rate": 9.505e-05,
362
+ "loss": 0.4392,
363
+ "step": 2300
364
+ },
365
+ {
366
+ "epoch": 0.9842931937172775,
367
+ "grad_norm": 2.248389720916748,
368
+ "learning_rate": 9.421666666666668e-05,
369
+ "loss": 0.4304,
370
+ "step": 2350
371
+ },
372
+ {
373
+ "epoch": 1.0052356020942408,
374
+ "grad_norm": 1.5984878540039062,
375
+ "learning_rate": 9.338333333333333e-05,
376
+ "loss": 0.4125,
377
+ "step": 2400
378
+ },
379
+ {
380
+ "epoch": 1.0261780104712042,
381
+ "grad_norm": 2.6061534881591797,
382
+ "learning_rate": 9.255e-05,
383
+ "loss": 0.4229,
384
+ "step": 2450
385
+ },
386
+ {
387
+ "epoch": 1.0471204188481675,
388
+ "grad_norm": 2.127110481262207,
389
+ "learning_rate": 9.171666666666667e-05,
390
+ "loss": 0.4273,
391
+ "step": 2500
392
+ },
393
+ {
394
+ "epoch": 1.0471204188481675,
395
+ "eval_loss": 0.38292551040649414,
396
+ "eval_runtime": 263.8923,
397
+ "eval_samples_per_second": 32.168,
398
+ "eval_steps_per_second": 4.024,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 1.068062827225131,
403
+ "grad_norm": 1.9867545366287231,
404
+ "learning_rate": 9.088333333333334e-05,
405
+ "loss": 0.4164,
406
+ "step": 2550
407
+ },
408
+ {
409
+ "epoch": 1.0890052356020943,
410
+ "grad_norm": 2.1518051624298096,
411
+ "learning_rate": 9.005000000000001e-05,
412
+ "loss": 0.4038,
413
+ "step": 2600
414
+ },
415
+ {
416
+ "epoch": 1.1099476439790577,
417
+ "grad_norm": 2.1266932487487793,
418
+ "learning_rate": 8.921666666666668e-05,
419
+ "loss": 0.418,
420
+ "step": 2650
421
+ },
422
+ {
423
+ "epoch": 1.130890052356021,
424
+ "grad_norm": 1.9351108074188232,
425
+ "learning_rate": 8.838333333333334e-05,
426
+ "loss": 0.4213,
427
+ "step": 2700
428
+ },
429
+ {
430
+ "epoch": 1.1518324607329844,
431
+ "grad_norm": 1.4189810752868652,
432
+ "learning_rate": 8.755e-05,
433
+ "loss": 0.4201,
434
+ "step": 2750
435
+ },
436
+ {
437
+ "epoch": 1.1727748691099475,
438
+ "grad_norm": 3.590930223464966,
439
+ "learning_rate": 8.671666666666667e-05,
440
+ "loss": 0.4043,
441
+ "step": 2800
442
+ },
443
+ {
444
+ "epoch": 1.193717277486911,
445
+ "grad_norm": 3.049971103668213,
446
+ "learning_rate": 8.588333333333334e-05,
447
+ "loss": 0.4036,
448
+ "step": 2850
449
+ },
450
+ {
451
+ "epoch": 1.2146596858638743,
452
+ "grad_norm": 2.1481425762176514,
453
+ "learning_rate": 8.505000000000001e-05,
454
+ "loss": 0.4067,
455
+ "step": 2900
456
+ },
457
+ {
458
+ "epoch": 1.2356020942408377,
459
+ "grad_norm": 1.774927020072937,
460
+ "learning_rate": 8.421666666666666e-05,
461
+ "loss": 0.4126,
462
+ "step": 2950
463
+ },
464
+ {
465
+ "epoch": 1.256544502617801,
466
+ "grad_norm": 1.624089241027832,
467
+ "learning_rate": 8.338333333333333e-05,
468
+ "loss": 0.4028,
469
+ "step": 3000
470
+ },
471
+ {
472
+ "epoch": 1.256544502617801,
473
+ "eval_loss": 0.3674115538597107,
474
+ "eval_runtime": 263.8647,
475
+ "eval_samples_per_second": 32.172,
476
+ "eval_steps_per_second": 4.025,
477
+ "step": 3000
478
+ },
479
+ {
480
+ "epoch": 1.2774869109947644,
481
+ "grad_norm": 1.8440332412719727,
482
+ "learning_rate": 8.255e-05,
483
+ "loss": 0.4026,
484
+ "step": 3050
485
+ },
486
+ {
487
+ "epoch": 1.2984293193717278,
488
+ "grad_norm": 2.144713878631592,
489
+ "learning_rate": 8.171666666666667e-05,
490
+ "loss": 0.4007,
491
+ "step": 3100
492
+ },
493
+ {
494
+ "epoch": 1.3193717277486912,
495
+ "grad_norm": 1.9491198062896729,
496
+ "learning_rate": 8.088333333333334e-05,
497
+ "loss": 0.4087,
498
+ "step": 3150
499
+ },
500
+ {
501
+ "epoch": 1.3403141361256545,
502
+ "grad_norm": 2.0903196334838867,
503
+ "learning_rate": 8.005000000000001e-05,
504
+ "loss": 0.4034,
505
+ "step": 3200
506
+ },
507
+ {
508
+ "epoch": 1.3612565445026177,
509
+ "grad_norm": 2.226724863052368,
510
+ "learning_rate": 7.921666666666668e-05,
511
+ "loss": 0.4106,
512
+ "step": 3250
513
+ },
514
+ {
515
+ "epoch": 1.3821989528795813,
516
+ "grad_norm": 1.8056219816207886,
517
+ "learning_rate": 7.838333333333335e-05,
518
+ "loss": 0.3984,
519
+ "step": 3300
520
+ },
521
+ {
522
+ "epoch": 1.4031413612565444,
523
+ "grad_norm": 1.8196921348571777,
524
+ "learning_rate": 7.755e-05,
525
+ "loss": 0.4018,
526
+ "step": 3350
527
+ },
528
+ {
529
+ "epoch": 1.4240837696335078,
530
+ "grad_norm": 1.7930132150650024,
531
+ "learning_rate": 7.671666666666667e-05,
532
+ "loss": 0.395,
533
+ "step": 3400
534
+ },
535
+ {
536
+ "epoch": 1.4450261780104712,
537
+ "grad_norm": 2.525432586669922,
538
+ "learning_rate": 7.588333333333334e-05,
539
+ "loss": 0.399,
540
+ "step": 3450
541
+ },
542
+ {
543
+ "epoch": 1.4659685863874345,
544
+ "grad_norm": 1.2433278560638428,
545
+ "learning_rate": 7.505e-05,
546
+ "loss": 0.3941,
547
+ "step": 3500
548
+ },
549
+ {
550
+ "epoch": 1.4659685863874345,
551
+ "eval_loss": 0.3616171181201935,
552
+ "eval_runtime": 265.4304,
553
+ "eval_samples_per_second": 31.982,
554
+ "eval_steps_per_second": 4.001,
555
+ "step": 3500
556
+ },
557
+ {
558
+ "epoch": 1.486910994764398,
559
+ "grad_norm": 1.859680414199829,
560
+ "learning_rate": 7.421666666666666e-05,
561
+ "loss": 0.4023,
562
+ "step": 3550
563
+ },
564
+ {
565
+ "epoch": 1.5078534031413613,
566
+ "grad_norm": 1.9057739973068237,
567
+ "learning_rate": 7.338333333333333e-05,
568
+ "loss": 0.3938,
569
+ "step": 3600
570
+ },
571
+ {
572
+ "epoch": 1.5287958115183247,
573
+ "grad_norm": 1.6138437986373901,
574
+ "learning_rate": 7.255e-05,
575
+ "loss": 0.3995,
576
+ "step": 3650
577
+ },
578
+ {
579
+ "epoch": 1.5497382198952878,
580
+ "grad_norm": 1.9373960494995117,
581
+ "learning_rate": 7.171666666666667e-05,
582
+ "loss": 0.3934,
583
+ "step": 3700
584
+ },
585
+ {
586
+ "epoch": 1.5706806282722514,
587
+ "grad_norm": 1.7537634372711182,
588
+ "learning_rate": 7.088333333333334e-05,
589
+ "loss": 0.3969,
590
+ "step": 3750
591
+ },
592
+ {
593
+ "epoch": 1.5916230366492146,
594
+ "grad_norm": 2.310279607772827,
595
+ "learning_rate": 7.005000000000001e-05,
596
+ "loss": 0.3967,
597
+ "step": 3800
598
+ },
599
+ {
600
+ "epoch": 1.6125654450261782,
601
+ "grad_norm": 3.323341131210327,
602
+ "learning_rate": 6.921666666666668e-05,
603
+ "loss": 0.4058,
604
+ "step": 3850
605
+ },
606
+ {
607
+ "epoch": 1.6335078534031413,
608
+ "grad_norm": 1.784133791923523,
609
+ "learning_rate": 6.838333333333333e-05,
610
+ "loss": 0.3898,
611
+ "step": 3900
612
+ },
613
+ {
614
+ "epoch": 1.6544502617801047,
615
+ "grad_norm": 1.8042694330215454,
616
+ "learning_rate": 6.755e-05,
617
+ "loss": 0.3902,
618
+ "step": 3950
619
+ },
620
+ {
621
+ "epoch": 1.675392670157068,
622
+ "grad_norm": 1.8511914014816284,
623
+ "learning_rate": 6.671666666666667e-05,
624
+ "loss": 0.3871,
625
+ "step": 4000
626
+ },
627
+ {
628
+ "epoch": 1.675392670157068,
629
+ "eval_loss": 0.3518759310245514,
630
+ "eval_runtime": 268.3666,
631
+ "eval_samples_per_second": 31.632,
632
+ "eval_steps_per_second": 3.957,
633
+ "step": 4000
634
+ },
635
+ {
636
+ "epoch": 1.6963350785340314,
637
+ "grad_norm": 1.1159111261367798,
638
+ "learning_rate": 6.588333333333334e-05,
639
+ "loss": 0.3842,
640
+ "step": 4050
641
+ },
642
+ {
643
+ "epoch": 1.7172774869109948,
644
+ "grad_norm": 1.8107820749282837,
645
+ "learning_rate": 6.505e-05,
646
+ "loss": 0.3909,
647
+ "step": 4100
648
+ },
649
+ {
650
+ "epoch": 1.738219895287958,
651
+ "grad_norm": 1.5878946781158447,
652
+ "learning_rate": 6.421666666666666e-05,
653
+ "loss": 0.3974,
654
+ "step": 4150
655
+ },
656
+ {
657
+ "epoch": 1.7591623036649215,
658
+ "grad_norm": 3.3478894233703613,
659
+ "learning_rate": 6.338333333333333e-05,
660
+ "loss": 0.3874,
661
+ "step": 4200
662
+ },
663
+ {
664
+ "epoch": 1.7801047120418847,
665
+ "grad_norm": 2.9195618629455566,
666
+ "learning_rate": 6.255e-05,
667
+ "loss": 0.3893,
668
+ "step": 4250
669
+ },
670
+ {
671
+ "epoch": 1.8010471204188483,
672
+ "grad_norm": 1.6017109155654907,
673
+ "learning_rate": 6.171666666666667e-05,
674
+ "loss": 0.3827,
675
+ "step": 4300
676
+ },
677
+ {
678
+ "epoch": 1.8219895287958114,
679
+ "grad_norm": 2.468721628189087,
680
+ "learning_rate": 6.0883333333333334e-05,
681
+ "loss": 0.3835,
682
+ "step": 4350
683
+ },
684
+ {
685
+ "epoch": 1.8429319371727748,
686
+ "grad_norm": 1.5800402164459229,
687
+ "learning_rate": 6.005000000000001e-05,
688
+ "loss": 0.3767,
689
+ "step": 4400
690
+ },
691
+ {
692
+ "epoch": 1.8638743455497382,
693
+ "grad_norm": 2.512275218963623,
694
+ "learning_rate": 5.9216666666666665e-05,
695
+ "loss": 0.3835,
696
+ "step": 4450
697
+ },
698
+ {
699
+ "epoch": 1.8848167539267016,
700
+ "grad_norm": 1.4855787754058838,
701
+ "learning_rate": 5.8383333333333334e-05,
702
+ "loss": 0.3828,
703
+ "step": 4500
704
+ },
705
+ {
706
+ "epoch": 1.8848167539267016,
707
+ "eval_loss": 0.3493475914001465,
708
+ "eval_runtime": 265.6954,
709
+ "eval_samples_per_second": 31.95,
710
+ "eval_steps_per_second": 3.997,
711
+ "step": 4500
712
+ },
713
+ {
714
+ "epoch": 1.905759162303665,
715
+ "grad_norm": 1.5831927061080933,
716
+ "learning_rate": 5.755e-05,
717
+ "loss": 0.3798,
718
+ "step": 4550
719
+ },
720
+ {
721
+ "epoch": 1.9267015706806283,
722
+ "grad_norm": 1.8235031366348267,
723
+ "learning_rate": 5.671666666666667e-05,
724
+ "loss": 0.3772,
725
+ "step": 4600
726
+ },
727
+ {
728
+ "epoch": 1.9476439790575917,
729
+ "grad_norm": 2.0443477630615234,
730
+ "learning_rate": 5.5883333333333334e-05,
731
+ "loss": 0.3818,
732
+ "step": 4650
733
+ },
734
+ {
735
+ "epoch": 1.9685863874345548,
736
+ "grad_norm": 1.5491324663162231,
737
+ "learning_rate": 5.505e-05,
738
+ "loss": 0.376,
739
+ "step": 4700
740
+ },
741
+ {
742
+ "epoch": 1.9895287958115184,
743
+ "grad_norm": 1.934651494026184,
744
+ "learning_rate": 5.421666666666667e-05,
745
+ "loss": 0.3762,
746
+ "step": 4750
747
+ },
748
+ {
749
+ "epoch": 2.0104712041884816,
750
+ "grad_norm": 1.5579476356506348,
751
+ "learning_rate": 5.338333333333334e-05,
752
+ "loss": 0.3899,
753
+ "step": 4800
754
+ },
755
+ {
756
+ "epoch": 2.031413612565445,
757
+ "grad_norm": 3.049567461013794,
758
+ "learning_rate": 5.255e-05,
759
+ "loss": 0.379,
760
+ "step": 4850
761
+ },
762
+ {
763
+ "epoch": 2.0523560209424083,
764
+ "grad_norm": 2.302692174911499,
765
+ "learning_rate": 5.1716666666666666e-05,
766
+ "loss": 0.376,
767
+ "step": 4900
768
+ },
769
+ {
770
+ "epoch": 2.073298429319372,
771
+ "grad_norm": 1.2674555778503418,
772
+ "learning_rate": 5.088333333333334e-05,
773
+ "loss": 0.3804,
774
+ "step": 4950
775
+ },
776
+ {
777
+ "epoch": 2.094240837696335,
778
+ "grad_norm": 4.3979902267456055,
779
+ "learning_rate": 5.005e-05,
780
+ "loss": 0.3954,
781
+ "step": 5000
782
+ },
783
+ {
784
+ "epoch": 2.094240837696335,
785
+ "eval_loss": 0.3490452766418457,
786
+ "eval_runtime": 267.7969,
787
+ "eval_samples_per_second": 31.699,
788
+ "eval_steps_per_second": 3.966,
789
+ "step": 5000
790
+ },
791
+ {
792
+ "epoch": 2.115183246073298,
793
+ "grad_norm": 1.4126474857330322,
794
+ "learning_rate": 4.9216666666666666e-05,
795
+ "loss": 0.3814,
796
+ "step": 5050
797
+ },
798
+ {
799
+ "epoch": 2.136125654450262,
800
+ "grad_norm": 2.161151885986328,
801
+ "learning_rate": 4.8383333333333335e-05,
802
+ "loss": 0.3796,
803
+ "step": 5100
804
+ },
805
+ {
806
+ "epoch": 2.157068062827225,
807
+ "grad_norm": 1.7521085739135742,
808
+ "learning_rate": 4.755e-05,
809
+ "loss": 0.3796,
810
+ "step": 5150
811
+ },
812
+ {
813
+ "epoch": 2.1780104712041886,
814
+ "grad_norm": 1.831732988357544,
815
+ "learning_rate": 4.671666666666667e-05,
816
+ "loss": 0.3745,
817
+ "step": 5200
818
+ },
819
+ {
820
+ "epoch": 2.1989528795811517,
821
+ "grad_norm": 1.640143871307373,
822
+ "learning_rate": 4.5883333333333335e-05,
823
+ "loss": 0.3687,
824
+ "step": 5250
825
+ },
826
+ {
827
+ "epoch": 2.2198952879581153,
828
+ "grad_norm": 1.7855136394500732,
829
+ "learning_rate": 4.5050000000000004e-05,
830
+ "loss": 0.3737,
831
+ "step": 5300
832
+ },
833
+ {
834
+ "epoch": 2.2408376963350785,
835
+ "grad_norm": 2.1165611743927,
836
+ "learning_rate": 4.4216666666666666e-05,
837
+ "loss": 0.3681,
838
+ "step": 5350
839
+ },
840
+ {
841
+ "epoch": 2.261780104712042,
842
+ "grad_norm": 2.91493558883667,
843
+ "learning_rate": 4.3383333333333335e-05,
844
+ "loss": 0.3775,
845
+ "step": 5400
846
+ },
847
+ {
848
+ "epoch": 2.282722513089005,
849
+ "grad_norm": 3.0871639251708984,
850
+ "learning_rate": 4.2550000000000004e-05,
851
+ "loss": 0.3714,
852
+ "step": 5450
853
+ },
854
+ {
855
+ "epoch": 2.303664921465969,
856
+ "grad_norm": 1.933838129043579,
857
+ "learning_rate": 4.171666666666667e-05,
858
+ "loss": 0.381,
859
+ "step": 5500
860
+ },
861
+ {
862
+ "epoch": 2.303664921465969,
863
+ "eval_loss": 0.33975139260292053,
864
+ "eval_runtime": 264.665,
865
+ "eval_samples_per_second": 32.075,
866
+ "eval_steps_per_second": 4.013,
867
+ "step": 5500
868
+ },
869
+ {
870
+ "epoch": 2.324607329842932,
871
+ "grad_norm": 1.6925002336502075,
872
+ "learning_rate": 4.0883333333333335e-05,
873
+ "loss": 0.372,
874
+ "step": 5550
875
+ },
876
+ {
877
+ "epoch": 2.345549738219895,
878
+ "grad_norm": 1.2782198190689087,
879
+ "learning_rate": 4.0050000000000004e-05,
880
+ "loss": 0.3705,
881
+ "step": 5600
882
+ },
883
+ {
884
+ "epoch": 2.3664921465968587,
885
+ "grad_norm": 1.279897689819336,
886
+ "learning_rate": 3.921666666666667e-05,
887
+ "loss": 0.3739,
888
+ "step": 5650
889
+ },
890
+ {
891
+ "epoch": 2.387434554973822,
892
+ "grad_norm": 2.082263231277466,
893
+ "learning_rate": 3.8383333333333336e-05,
894
+ "loss": 0.3696,
895
+ "step": 5700
896
+ },
897
+ {
898
+ "epoch": 2.4083769633507854,
899
+ "grad_norm": 1.2950184345245361,
900
+ "learning_rate": 3.7550000000000005e-05,
901
+ "loss": 0.3729,
902
+ "step": 5750
903
+ },
904
+ {
905
+ "epoch": 2.4293193717277486,
906
+ "grad_norm": 2.0218422412872314,
907
+ "learning_rate": 3.671666666666667e-05,
908
+ "loss": 0.374,
909
+ "step": 5800
910
+ },
911
+ {
912
+ "epoch": 2.450261780104712,
913
+ "grad_norm": 1.9062658548355103,
914
+ "learning_rate": 3.5883333333333336e-05,
915
+ "loss": 0.3701,
916
+ "step": 5850
917
+ },
918
+ {
919
+ "epoch": 2.4712041884816753,
920
+ "grad_norm": 1.6323776245117188,
921
+ "learning_rate": 3.505e-05,
922
+ "loss": 0.3683,
923
+ "step": 5900
924
+ },
925
+ {
926
+ "epoch": 2.492146596858639,
927
+ "grad_norm": 1.4120862483978271,
928
+ "learning_rate": 3.421666666666667e-05,
929
+ "loss": 0.3674,
930
+ "step": 5950
931
+ },
932
+ {
933
+ "epoch": 2.513089005235602,
934
+ "grad_norm": 2.368260383605957,
935
+ "learning_rate": 3.3383333333333336e-05,
936
+ "loss": 0.372,
937
+ "step": 6000
938
+ },
939
+ {
940
+ "epoch": 2.513089005235602,
941
+ "eval_loss": 0.3372337222099304,
942
+ "eval_runtime": 267.436,
943
+ "eval_samples_per_second": 31.742,
944
+ "eval_steps_per_second": 3.971,
945
+ "step": 6000
946
+ },
947
+ {
948
+ "epoch": 2.5340314136125652,
949
+ "grad_norm": 1.5971177816390991,
950
+ "learning_rate": 3.2550000000000005e-05,
951
+ "loss": 0.3665,
952
+ "step": 6050
953
+ },
954
+ {
955
+ "epoch": 2.554973821989529,
956
+ "grad_norm": 1.4700311422348022,
957
+ "learning_rate": 3.171666666666667e-05,
958
+ "loss": 0.3681,
959
+ "step": 6100
960
+ },
961
+ {
962
+ "epoch": 2.5759162303664924,
963
+ "grad_norm": 1.6570911407470703,
964
+ "learning_rate": 3.0883333333333336e-05,
965
+ "loss": 0.3641,
966
+ "step": 6150
967
+ },
968
+ {
969
+ "epoch": 2.5968586387434556,
970
+ "grad_norm": 1.9697566032409668,
971
+ "learning_rate": 3.0050000000000002e-05,
972
+ "loss": 0.3736,
973
+ "step": 6200
974
+ },
975
+ {
976
+ "epoch": 2.6178010471204187,
977
+ "grad_norm": 3.0906572341918945,
978
+ "learning_rate": 2.921666666666667e-05,
979
+ "loss": 0.3606,
980
+ "step": 6250
981
+ },
982
+ {
983
+ "epoch": 2.6387434554973823,
984
+ "grad_norm": 1.426192283630371,
985
+ "learning_rate": 2.8383333333333333e-05,
986
+ "loss": 0.3746,
987
+ "step": 6300
988
+ },
989
+ {
990
+ "epoch": 2.6596858638743455,
991
+ "grad_norm": 1.8590399026870728,
992
+ "learning_rate": 2.7550000000000002e-05,
993
+ "loss": 0.3675,
994
+ "step": 6350
995
+ },
996
+ {
997
+ "epoch": 2.680628272251309,
998
+ "grad_norm": 1.3725066184997559,
999
+ "learning_rate": 2.6716666666666668e-05,
1000
+ "loss": 0.3669,
1001
+ "step": 6400
1002
+ },
1003
+ {
1004
+ "epoch": 2.701570680628272,
1005
+ "grad_norm": 1.1929622888565063,
1006
+ "learning_rate": 2.5883333333333337e-05,
1007
+ "loss": 0.363,
1008
+ "step": 6450
1009
+ },
1010
+ {
1011
+ "epoch": 2.7225130890052354,
1012
+ "grad_norm": 1.2838363647460938,
1013
+ "learning_rate": 2.5050000000000002e-05,
1014
+ "loss": 0.3642,
1015
+ "step": 6500
1016
+ },
1017
+ {
1018
+ "epoch": 2.7225130890052354,
1019
+ "eval_loss": 0.3314475417137146,
1020
+ "eval_runtime": 271.3873,
1021
+ "eval_samples_per_second": 31.28,
1022
+ "eval_steps_per_second": 3.913,
1023
+ "step": 6500
1024
+ },
1025
+ {
1026
+ "epoch": 2.743455497382199,
1027
+ "grad_norm": 1.5131884813308716,
1028
+ "learning_rate": 2.4216666666666668e-05,
1029
+ "loss": 0.3631,
1030
+ "step": 6550
1031
+ },
1032
+ {
1033
+ "epoch": 2.7643979057591626,
1034
+ "grad_norm": 2.1872663497924805,
1035
+ "learning_rate": 2.3383333333333334e-05,
1036
+ "loss": 0.3611,
1037
+ "step": 6600
1038
+ },
1039
+ {
1040
+ "epoch": 2.7853403141361257,
1041
+ "grad_norm": 1.3774406909942627,
1042
+ "learning_rate": 2.2550000000000003e-05,
1043
+ "loss": 0.3674,
1044
+ "step": 6650
1045
+ },
1046
+ {
1047
+ "epoch": 2.806282722513089,
1048
+ "grad_norm": 1.6061275005340576,
1049
+ "learning_rate": 2.1716666666666668e-05,
1050
+ "loss": 0.3654,
1051
+ "step": 6700
1052
+ },
1053
+ {
1054
+ "epoch": 2.8272251308900525,
1055
+ "grad_norm": 1.909685492515564,
1056
+ "learning_rate": 2.0883333333333334e-05,
1057
+ "loss": 0.3611,
1058
+ "step": 6750
1059
+ },
1060
+ {
1061
+ "epoch": 2.8481675392670156,
1062
+ "grad_norm": 1.6721019744873047,
1063
+ "learning_rate": 2.0050000000000003e-05,
1064
+ "loss": 0.3663,
1065
+ "step": 6800
1066
+ },
1067
+ {
1068
+ "epoch": 2.869109947643979,
1069
+ "grad_norm": 1.6825881004333496,
1070
+ "learning_rate": 1.921666666666667e-05,
1071
+ "loss": 0.3638,
1072
+ "step": 6850
1073
+ },
1074
+ {
1075
+ "epoch": 2.8900523560209423,
1076
+ "grad_norm": 1.6210919618606567,
1077
+ "learning_rate": 1.8383333333333334e-05,
1078
+ "loss": 0.3643,
1079
+ "step": 6900
1080
+ },
1081
+ {
1082
+ "epoch": 2.9109947643979055,
1083
+ "grad_norm": 1.5710177421569824,
1084
+ "learning_rate": 1.755e-05,
1085
+ "loss": 0.3657,
1086
+ "step": 6950
1087
+ },
1088
+ {
1089
+ "epoch": 2.931937172774869,
1090
+ "grad_norm": 1.158400535583496,
1091
+ "learning_rate": 1.6716666666666665e-05,
1092
+ "loss": 0.3692,
1093
+ "step": 7000
1094
+ },
1095
+ {
1096
+ "epoch": 2.931937172774869,
1097
+ "eval_loss": 0.3307726979255676,
1098
+ "eval_runtime": 273.8683,
1099
+ "eval_samples_per_second": 30.997,
1100
+ "eval_steps_per_second": 3.878,
1101
+ "step": 7000
1102
+ },
1103
+ {
1104
+ "epoch": 2.9528795811518327,
1105
+ "grad_norm": 1.70427668094635,
1106
+ "learning_rate": 1.5883333333333334e-05,
1107
+ "loss": 0.3604,
1108
+ "step": 7050
1109
+ },
1110
+ {
1111
+ "epoch": 2.973821989528796,
1112
+ "grad_norm": 1.7018851041793823,
1113
+ "learning_rate": 1.505e-05,
1114
+ "loss": 0.3603,
1115
+ "step": 7100
1116
+ },
1117
+ {
1118
+ "epoch": 2.994764397905759,
1119
+ "grad_norm": 1.4028674364089966,
1120
+ "learning_rate": 1.4216666666666667e-05,
1121
+ "loss": 0.3729,
1122
+ "step": 7150
1123
+ },
1124
+ {
1125
+ "epoch": 3.0157068062827226,
1126
+ "grad_norm": 1.1933259963989258,
1127
+ "learning_rate": 1.3383333333333335e-05,
1128
+ "loss": 0.3626,
1129
+ "step": 7200
1130
+ },
1131
+ {
1132
+ "epoch": 3.0366492146596857,
1133
+ "grad_norm": 1.202587366104126,
1134
+ "learning_rate": 1.255e-05,
1135
+ "loss": 0.3565,
1136
+ "step": 7250
1137
+ },
1138
+ {
1139
+ "epoch": 3.0575916230366493,
1140
+ "grad_norm": 1.1894952058792114,
1141
+ "learning_rate": 1.1716666666666667e-05,
1142
+ "loss": 0.3615,
1143
+ "step": 7300
1144
+ },
1145
+ {
1146
+ "epoch": 3.0785340314136125,
1147
+ "grad_norm": 1.1371740102767944,
1148
+ "learning_rate": 1.0883333333333335e-05,
1149
+ "loss": 0.3578,
1150
+ "step": 7350
1151
+ },
1152
+ {
1153
+ "epoch": 3.099476439790576,
1154
+ "grad_norm": 1.69568932056427,
1155
+ "learning_rate": 1.005e-05,
1156
+ "loss": 0.3593,
1157
+ "step": 7400
1158
+ },
1159
+ {
1160
+ "epoch": 3.1204188481675392,
1161
+ "grad_norm": 1.6758700609207153,
1162
+ "learning_rate": 9.216666666666666e-06,
1163
+ "loss": 0.3562,
1164
+ "step": 7450
1165
+ },
1166
+ {
1167
+ "epoch": 3.141361256544503,
1168
+ "grad_norm": 1.5202099084854126,
1169
+ "learning_rate": 8.383333333333333e-06,
1170
+ "loss": 0.3555,
1171
+ "step": 7500
1172
+ },
1173
+ {
1174
+ "epoch": 3.141361256544503,
1175
+ "eval_loss": 0.32830387353897095,
1176
+ "eval_runtime": 278.2777,
1177
+ "eval_samples_per_second": 30.506,
1178
+ "eval_steps_per_second": 3.816,
1179
+ "step": 7500
1180
+ }
1181
+ ],
1182
+ "logging_steps": 50,
1183
+ "max_steps": 8000,
1184
+ "num_input_tokens_seen": 0,
1185
+ "num_train_epochs": 4,
1186
+ "save_steps": 500,
1187
+ "stateful_callbacks": {
1188
+ "TrainerControl": {
1189
+ "args": {
1190
+ "should_epoch_stop": false,
1191
+ "should_evaluate": false,
1192
+ "should_log": false,
1193
+ "should_save": true,
1194
+ "should_training_stop": false
1195
+ },
1196
+ "attributes": {}
1197
+ }
1198
+ },
1199
+ "total_flos": 3.357131990640595e+16,
1200
+ "train_batch_size": 16,
1201
+ "trial_name": null,
1202
+ "trial_params": null
1203
+ }
checkpoint-7500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f8ba8e957961bfac2918c828bd8fb1b1b5848bc9b1402a9fd5cd5021a38c66f
3
+ size 5304