Hugging Face's logo

aimonbc24
/
whisper-large-v3-jacob-tune

Safetensors
whisper
Model card Files Community
EC2 Default User commited on
Commit
d25ea70
·
1 Parent(s): 8494dcb

add ckpt for epoch 2

Browse files
Files changed (11) hide show
  1. checkpoint-26868/config.json +52 -0
  2. checkpoint-26868/generation_config.json +265 -0
  3. checkpoint-26868/model-00001-of-00002.safetensors +3 -0
  4. checkpoint-26868/model-00002-of-00002.safetensors +3 -0
  5. checkpoint-26868/model.safetensors.index.json +0 -0
  6. checkpoint-26868/optimizer.pt +3 -0
  7. checkpoint-26868/preprocessor_config.json +14 -0
  8. checkpoint-26868/rng_state.pth +3 -0
  9. checkpoint-26868/scheduler.pt +3 -0
  10. checkpoint-26868/trainer_state.json +2535 -0
  11. checkpoint-26868/training_args.bin +3 -0
checkpoint-26868/config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-large-v3",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "apply_spec_augment": false,
6
+ "architectures": [
7
+ "WhisperForConditionalGeneration"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "begin_suppress_tokens": [
11
+ 220,
12
+ 50257
13
+ ],
14
+ "bos_token_id": 50257,
15
+ "classifier_proj_size": 256,
16
+ "d_model": 1280,
17
+ "decoder_attention_heads": 20,
18
+ "decoder_ffn_dim": 5120,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 32,
21
+ "decoder_start_token_id": 50258,
22
+ "dropout": 0.0,
23
+ "encoder_attention_heads": 20,
24
+ "encoder_ffn_dim": 5120,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 32,
27
+ "eos_token_id": 50257,
28
+ "forced_decoder_ids": null,
29
+ "init_std": 0.02,
30
+ "is_encoder_decoder": true,
31
+ "mask_feature_length": 10,
32
+ "mask_feature_min_masks": 0,
33
+ "mask_feature_prob": 0.0,
34
+ "mask_time_length": 10,
35
+ "mask_time_min_masks": 2,
36
+ "mask_time_prob": 0.05,
37
+ "max_length": 448,
38
+ "max_source_positions": 1500,
39
+ "max_target_positions": 448,
40
+ "median_filter_width": 7,
41
+ "model_type": "whisper",
42
+ "num_hidden_layers": 32,
43
+ "num_mel_bins": 128,
44
+ "pad_token_id": 50256,
45
+ "scale_embedding": false,
46
+ "suppress_tokens": [],
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.37.2",
49
+ "use_cache": true,
50
+ "use_weighted_layer_sum": false,
51
+ "vocab_size": 51866
52
+ }
checkpoint-26868/generation_config.json ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 7,
5
+ 0
6
+ ],
7
+ [
8
+ 10,
9
+ 17
10
+ ],
11
+ [
12
+ 12,
13
+ 18
14
+ ],
15
+ [
16
+ 13,
17
+ 12
18
+ ],
19
+ [
20
+ 16,
21
+ 1
22
+ ],
23
+ [
24
+ 17,
25
+ 14
26
+ ],
27
+ [
28
+ 19,
29
+ 11
30
+ ],
31
+ [
32
+ 21,
33
+ 4
34
+ ],
35
+ [
36
+ 24,
37
+ 1
38
+ ],
39
+ [
40
+ 25,
41
+ 6
42
+ ]
43
+ ],
44
+ "begin_suppress_tokens": [
45
+ 220,
46
+ 50257
47
+ ],
48
+ "bos_token_id": 50257,
49
+ "decoder_start_token_id": 50258,
50
+ "eos_token_id": 50257,
51
+ "forced_decoder_ids": [
52
+ [
53
+ 1,
54
+ null
55
+ ],
56
+ [
57
+ 2,
58
+ 50360
59
+ ]
60
+ ],
61
+ "is_multilingual": true,
62
+ "lang_to_id": {
63
+ "<|af|>": 50327,
64
+ "<|am|>": 50334,
65
+ "<|ar|>": 50272,
66
+ "<|as|>": 50350,
67
+ "<|az|>": 50304,
68
+ "<|ba|>": 50355,
69
+ "<|be|>": 50330,
70
+ "<|bg|>": 50292,
71
+ "<|bn|>": 50302,
72
+ "<|bo|>": 50347,
73
+ "<|br|>": 50309,
74
+ "<|bs|>": 50315,
75
+ "<|ca|>": 50270,
76
+ "<|cs|>": 50283,
77
+ "<|cy|>": 50297,
78
+ "<|da|>": 50285,
79
+ "<|de|>": 50261,
80
+ "<|el|>": 50281,
81
+ "<|en|>": 50259,
82
+ "<|es|>": 50262,
83
+ "<|et|>": 50307,
84
+ "<|eu|>": 50310,
85
+ "<|fa|>": 50300,
86
+ "<|fi|>": 50277,
87
+ "<|fo|>": 50338,
88
+ "<|fr|>": 50265,
89
+ "<|gl|>": 50319,
90
+ "<|gu|>": 50333,
91
+ "<|haw|>": 50352,
92
+ "<|ha|>": 50354,
93
+ "<|he|>": 50279,
94
+ "<|hi|>": 50276,
95
+ "<|hr|>": 50291,
96
+ "<|ht|>": 50339,
97
+ "<|hu|>": 50286,
98
+ "<|hy|>": 50312,
99
+ "<|id|>": 50275,
100
+ "<|is|>": 50311,
101
+ "<|it|>": 50274,
102
+ "<|ja|>": 50266,
103
+ "<|jw|>": 50356,
104
+ "<|ka|>": 50329,
105
+ "<|kk|>": 50316,
106
+ "<|km|>": 50323,
107
+ "<|kn|>": 50306,
108
+ "<|ko|>": 50264,
109
+ "<|la|>": 50294,
110
+ "<|lb|>": 50345,
111
+ "<|ln|>": 50353,
112
+ "<|lo|>": 50336,
113
+ "<|lt|>": 50293,
114
+ "<|lv|>": 50301,
115
+ "<|mg|>": 50349,
116
+ "<|mi|>": 50295,
117
+ "<|mk|>": 50308,
118
+ "<|ml|>": 50296,
119
+ "<|mn|>": 50314,
120
+ "<|mr|>": 50320,
121
+ "<|ms|>": 50282,
122
+ "<|mt|>": 50343,
123
+ "<|my|>": 50346,
124
+ "<|ne|>": 50313,
125
+ "<|nl|>": 50271,
126
+ "<|nn|>": 50342,
127
+ "<|no|>": 50288,
128
+ "<|oc|>": 50328,
129
+ "<|pa|>": 50321,
130
+ "<|pl|>": 50269,
131
+ "<|ps|>": 50340,
132
+ "<|pt|>": 50267,
133
+ "<|ro|>": 50284,
134
+ "<|ru|>": 50263,
135
+ "<|sa|>": 50344,
136
+ "<|sd|>": 50332,
137
+ "<|si|>": 50322,
138
+ "<|sk|>": 50298,
139
+ "<|sl|>": 50305,
140
+ "<|sn|>": 50324,
141
+ "<|so|>": 50326,
142
+ "<|sq|>": 50317,
143
+ "<|sr|>": 50303,
144
+ "<|su|>": 50357,
145
+ "<|sv|>": 50273,
146
+ "<|sw|>": 50318,
147
+ "<|ta|>": 50287,
148
+ "<|te|>": 50299,
149
+ "<|tg|>": 50331,
150
+ "<|th|>": 50289,
151
+ "<|tk|>": 50341,
152
+ "<|tl|>": 50348,
153
+ "<|tr|>": 50268,
154
+ "<|tt|>": 50351,
155
+ "<|uk|>": 50280,
156
+ "<|ur|>": 50290,
157
+ "<|uz|>": 50337,
158
+ "<|vi|>": 50278,
159
+ "<|yi|>": 50335,
160
+ "<|yo|>": 50325,
161
+ "<|yue|>": 50358,
162
+ "<|zh|>": 50260
163
+ },
164
+ "max_initial_timestamp_index": 50,
165
+ "max_length": 448,
166
+ "no_timestamps_token_id": 50364,
167
+ "pad_token_id": 50257,
168
+ "prev_sot_token_id": 50362,
169
+ "return_timestamps": false,
170
+ "suppress_tokens": [
171
+ 1,
172
+ 2,
173
+ 7,
174
+ 8,
175
+ 9,
176
+ 10,
177
+ 14,
178
+ 25,
179
+ 26,
180
+ 27,
181
+ 28,
182
+ 29,
183
+ 31,
184
+ 58,
185
+ 59,
186
+ 60,
187
+ 61,
188
+ 62,
189
+ 63,
190
+ 90,
191
+ 91,
192
+ 92,
193
+ 93,
194
+ 359,
195
+ 503,
196
+ 522,
197
+ 542,
198
+ 873,
199
+ 893,
200
+ 902,
201
+ 918,
202
+ 922,
203
+ 931,
204
+ 1350,
205
+ 1853,
206
+ 1982,
207
+ 2460,
208
+ 2627,
209
+ 3246,
210
+ 3253,
211
+ 3268,
212
+ 3536,
213
+ 3846,
214
+ 3961,
215
+ 4183,
216
+ 4667,
217
+ 6585,
218
+ 6647,
219
+ 7273,
220
+ 9061,
221
+ 9383,
222
+ 10428,
223
+ 10929,
224
+ 11938,
225
+ 12033,
226
+ 12331,
227
+ 12562,
228
+ 13793,
229
+ 14157,
230
+ 14635,
231
+ 15265,
232
+ 15618,
233
+ 16553,
234
+ 16604,
235
+ 18362,
236
+ 18956,
237
+ 20075,
238
+ 21675,
239
+ 22520,
240
+ 26130,
241
+ 26161,
242
+ 26435,
243
+ 28279,
244
+ 29464,
245
+ 31650,
246
+ 32302,
247
+ 32470,
248
+ 36865,
249
+ 42863,
250
+ 47425,
251
+ 49870,
252
+ 50254,
253
+ 50258,
254
+ 50359,
255
+ 50360,
256
+ 50361,
257
+ 50362,
258
+ 50363
259
+ ],
260
+ "task_to_id": {
261
+ "transcribe": 50360,
262
+ "translate": 50359
263
+ },
264
+ "transformers_version": "4.37.2"
265
+ }
checkpoint-26868/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7d6f53db1a4e09acbcbaa1b924c20a29adf9e2a00d16a2bd4adee3ba8d34fbe
3
+ size 4993448880
checkpoint-26868/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6ee839665ed1abf49fd10092951f7012635c8f9c3402967f9f4deff0cd6823b
3
+ size 1180663192
checkpoint-26868/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-26868/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9653293516c6fbf28369be5b87c8cbcb3e7eecaed2d0271edf54083d8ba334bc
3
+ size 3095446256
checkpoint-26868/preprocessor_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "feature_extractor_type": "WhisperFeatureExtractor",
4
+ "feature_size": 128,
5
+ "hop_length": 160,
6
+ "n_fft": 400,
7
+ "n_samples": 480000,
8
+ "nb_max_frames": 3000,
9
+ "padding_side": "right",
10
+ "padding_value": 0.0,
11
+ "processor_class": "WhisperProcessor",
12
+ "return_attention_mask": false,
13
+ "sampling_rate": 16000
14
+ }
checkpoint-26868/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a5507cf4bac232d465d449fe7165a7d03a310cee070797b15ef47cd4dc0458b
3
+ size 14244
checkpoint-26868/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57beb52ea991361080b12ea2f54875aaa868bb0fb2c485385647d815b28c9574
3
+ size 1064
checkpoint-26868/trainer_state.json ADDED
@@ -0,0 +1,2535 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 26868,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 2.5227460711331677e-08,
14
+ "loss": 3.3575,
15
+ "step": 64
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 5.128205128205128e-08,
20
+ "loss": 3.3023,
21
+ "step": 128
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 7.775020678246484e-08,
26
+ "loss": 2.9134,
27
+ "step": 192
28
+ },
29
+ {
30
+ "epoch": 0.02,
31
+ "learning_rate": 1.0421836228287841e-07,
32
+ "loss": 2.7766,
33
+ "step": 256
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 1.3068651778329198e-07,
38
+ "loss": 2.6203,
39
+ "step": 320
40
+ },
41
+ {
42
+ "epoch": 0.03,
43
+ "learning_rate": 1.5715467328370554e-07,
44
+ "loss": 2.221,
45
+ "step": 384
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 1.836228287841191e-07,
50
+ "loss": 1.6761,
51
+ "step": 448
52
+ },
53
+ {
54
+ "epoch": 0.04,
55
+ "learning_rate": 2.1009098428453268e-07,
56
+ "loss": 1.3371,
57
+ "step": 512
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 2.3655913978494625e-07,
62
+ "loss": 1.102,
63
+ "step": 576
64
+ },
65
+ {
66
+ "epoch": 0.05,
67
+ "learning_rate": 2.6302729528535976e-07,
68
+ "loss": 0.9863,
69
+ "step": 640
70
+ },
71
+ {
72
+ "epoch": 0.05,
73
+ "learning_rate": 2.8949545078577336e-07,
74
+ "loss": 0.8123,
75
+ "step": 704
76
+ },
77
+ {
78
+ "epoch": 0.06,
79
+ "learning_rate": 3.159636062861869e-07,
80
+ "loss": 0.6005,
81
+ "step": 768
82
+ },
83
+ {
84
+ "epoch": 0.06,
85
+ "learning_rate": 3.424317617866005e-07,
86
+ "loss": 0.4275,
87
+ "step": 832
88
+ },
89
+ {
90
+ "epoch": 0.07,
91
+ "learning_rate": 3.6889991728701403e-07,
92
+ "loss": 0.4868,
93
+ "step": 896
94
+ },
95
+ {
96
+ "epoch": 0.07,
97
+ "learning_rate": 3.9536807278742763e-07,
98
+ "loss": 0.4454,
99
+ "step": 960
100
+ },
101
+ {
102
+ "epoch": 0.08,
103
+ "learning_rate": 4.2183622828784117e-07,
104
+ "loss": 0.3431,
105
+ "step": 1024
106
+ },
107
+ {
108
+ "epoch": 0.08,
109
+ "learning_rate": 4.4830438378825477e-07,
110
+ "loss": 0.3373,
111
+ "step": 1088
112
+ },
113
+ {
114
+ "epoch": 0.09,
115
+ "learning_rate": 4.743589743589743e-07,
116
+ "loss": 0.351,
117
+ "step": 1152
118
+ },
119
+ {
120
+ "epoch": 0.09,
121
+ "learning_rate": 5.00827129859388e-07,
122
+ "loss": 0.3565,
123
+ "step": 1216
124
+ },
125
+ {
126
+ "epoch": 0.1,
127
+ "learning_rate": 5.272952853598015e-07,
128
+ "loss": 0.2179,
129
+ "step": 1280
130
+ },
131
+ {
132
+ "epoch": 0.1,
133
+ "learning_rate": 5.537634408602149e-07,
134
+ "loss": 0.3892,
135
+ "step": 1344
136
+ },
137
+ {
138
+ "epoch": 0.1,
139
+ "learning_rate": 5.802315963606285e-07,
140
+ "loss": 0.3388,
141
+ "step": 1408
142
+ },
143
+ {
144
+ "epoch": 0.11,
145
+ "learning_rate": 6.066997518610421e-07,
146
+ "loss": 0.4345,
147
+ "step": 1472
148
+ },
149
+ {
150
+ "epoch": 0.11,
151
+ "learning_rate": 6.331679073614557e-07,
152
+ "loss": 0.1948,
153
+ "step": 1536
154
+ },
155
+ {
156
+ "epoch": 0.12,
157
+ "learning_rate": 6.596360628618692e-07,
158
+ "loss": 0.2907,
159
+ "step": 1600
160
+ },
161
+ {
162
+ "epoch": 0.12,
163
+ "learning_rate": 6.861042183622828e-07,
164
+ "loss": 0.3393,
165
+ "step": 1664
166
+ },
167
+ {
168
+ "epoch": 0.13,
169
+ "learning_rate": 7.125723738626964e-07,
170
+ "loss": 0.2568,
171
+ "step": 1728
172
+ },
173
+ {
174
+ "epoch": 0.13,
175
+ "learning_rate": 7.3904052936311e-07,
176
+ "loss": 0.3041,
177
+ "step": 1792
178
+ },
179
+ {
180
+ "epoch": 0.14,
181
+ "learning_rate": 7.655086848635235e-07,
182
+ "loss": 0.3316,
183
+ "step": 1856
184
+ },
185
+ {
186
+ "epoch": 0.14,
187
+ "learning_rate": 7.919768403639371e-07,
188
+ "loss": 0.3193,
189
+ "step": 1920
190
+ },
191
+ {
192
+ "epoch": 0.15,
193
+ "learning_rate": 8.184449958643507e-07,
194
+ "loss": 0.3393,
195
+ "step": 1984
196
+ },
197
+ {
198
+ "epoch": 0.15,
199
+ "learning_rate": 8.449131513647643e-07,
200
+ "loss": 0.2933,
201
+ "step": 2048
202
+ },
203
+ {
204
+ "epoch": 0.16,
205
+ "learning_rate": 8.713813068651778e-07,
206
+ "loss": 0.236,
207
+ "step": 2112
208
+ },
209
+ {
210
+ "epoch": 0.16,
211
+ "learning_rate": 8.978494623655913e-07,
212
+ "loss": 0.305,
213
+ "step": 2176
214
+ },
215
+ {
216
+ "epoch": 0.17,
217
+ "learning_rate": 9.243176178660049e-07,
218
+ "loss": 0.2158,
219
+ "step": 2240
220
+ },
221
+ {
222
+ "epoch": 0.17,
223
+ "learning_rate": 9.507857733664185e-07,
224
+ "loss": 0.3261,
225
+ "step": 2304
226
+ },
227
+ {
228
+ "epoch": 0.18,
229
+ "learning_rate": 9.77253928866832e-07,
230
+ "loss": 0.2282,
231
+ "step": 2368
232
+ },
233
+ {
234
+ "epoch": 0.18,
235
+ "learning_rate": 9.997624326892619e-07,
236
+ "loss": 0.3758,
237
+ "step": 2432
238
+ },
239
+ {
240
+ "epoch": 0.19,
241
+ "learning_rate": 9.980730651462358e-07,
242
+ "loss": 0.2466,
243
+ "step": 2496
244
+ },
245
+ {
246
+ "epoch": 0.19,
247
+ "learning_rate": 9.963836976032098e-07,
248
+ "loss": 0.2283,
249
+ "step": 2560
250
+ },
251
+ {
252
+ "epoch": 0.2,
253
+ "learning_rate": 9.946943300601838e-07,
254
+ "loss": 0.2605,
255
+ "step": 2624
256
+ },
257
+ {
258
+ "epoch": 0.2,
259
+ "learning_rate": 9.930049625171575e-07,
260
+ "loss": 0.2818,
261
+ "step": 2688
262
+ },
263
+ {
264
+ "epoch": 0.2,
265
+ "learning_rate": 9.913155949741315e-07,
266
+ "loss": 0.3004,
267
+ "step": 2752
268
+ },
269
+ {
270
+ "epoch": 0.21,
271
+ "learning_rate": 9.896262274311054e-07,
272
+ "loss": 0.2703,
273
+ "step": 2816
274
+ },
275
+ {
276
+ "epoch": 0.21,
277
+ "learning_rate": 9.879368598880794e-07,
278
+ "loss": 0.2924,
279
+ "step": 2880
280
+ },
281
+ {
282
+ "epoch": 0.22,
283
+ "learning_rate": 9.862474923450532e-07,
284
+ "loss": 0.3661,
285
+ "step": 2944
286
+ },
287
+ {
288
+ "epoch": 0.22,
289
+ "learning_rate": 9.845581248020271e-07,
290
+ "loss": 0.3517,
291
+ "step": 3008
292
+ },
293
+ {
294
+ "epoch": 0.23,
295
+ "learning_rate": 9.82868757259001e-07,
296
+ "loss": 0.2773,
297
+ "step": 3072
298
+ },
299
+ {
300
+ "epoch": 0.23,
301
+ "learning_rate": 9.81179389715975e-07,
302
+ "loss": 0.3023,
303
+ "step": 3136
304
+ },
305
+ {
306
+ "epoch": 0.24,
307
+ "learning_rate": 9.79490022172949e-07,
308
+ "loss": 0.2831,
309
+ "step": 3200
310
+ },
311
+ {
312
+ "epoch": 0.24,
313
+ "learning_rate": 9.77800654629923e-07,
314
+ "loss": 0.2693,
315
+ "step": 3264
316
+ },
317
+ {
318
+ "epoch": 0.25,
319
+ "learning_rate": 9.76111287086897e-07,
320
+ "loss": 0.2534,
321
+ "step": 3328
322
+ },
323
+ {
324
+ "epoch": 0.25,
325
+ "learning_rate": 9.744219195438707e-07,
326
+ "loss": 0.183,
327
+ "step": 3392
328
+ },
329
+ {
330
+ "epoch": 0.26,
331
+ "learning_rate": 9.727325520008446e-07,
332
+ "loss": 0.1786,
333
+ "step": 3456
334
+ },
335
+ {
336
+ "epoch": 0.26,
337
+ "learning_rate": 9.710431844578186e-07,
338
+ "loss": 0.3985,
339
+ "step": 3520
340
+ },
341
+ {
342
+ "epoch": 0.27,
343
+ "learning_rate": 9.693538169147924e-07,
344
+ "loss": 0.2585,
345
+ "step": 3584
346
+ },
347
+ {
348
+ "epoch": 0.27,
349
+ "learning_rate": 9.676644493717663e-07,
350
+ "loss": 0.2088,
351
+ "step": 3648
352
+ },
353
+ {
354
+ "epoch": 0.28,
355
+ "learning_rate": 9.659750818287403e-07,
356
+ "loss": 0.1913,
357
+ "step": 3712
358
+ },
359
+ {
360
+ "epoch": 0.28,
361
+ "learning_rate": 9.642857142857142e-07,
362
+ "loss": 0.2747,
363
+ "step": 3776
364
+ },
365
+ {
366
+ "epoch": 0.29,
367
+ "learning_rate": 9.625963467426882e-07,
368
+ "loss": 0.2283,
369
+ "step": 3840
370
+ },
371
+ {
372
+ "epoch": 0.29,
373
+ "learning_rate": 9.609069791996622e-07,
374
+ "loss": 0.305,
375
+ "step": 3904
376
+ },
377
+ {
378
+ "epoch": 0.3,
379
+ "learning_rate": 9.59217611656636e-07,
380
+ "loss": 0.2648,
381
+ "step": 3968
382
+ },
383
+ {
384
+ "epoch": 0.3,
385
+ "learning_rate": 9.575282441136099e-07,
386
+ "loss": 0.2277,
387
+ "step": 4032
388
+ },
389
+ {
390
+ "epoch": 0.3,
391
+ "learning_rate": 9.558388765705839e-07,
392
+ "loss": 0.2445,
393
+ "step": 4096
394
+ },
395
+ {
396
+ "epoch": 0.31,
397
+ "learning_rate": 9.541495090275578e-07,
398
+ "loss": 0.3094,
399
+ "step": 4160
400
+ },
401
+ {
402
+ "epoch": 0.31,
403
+ "learning_rate": 9.524601414845317e-07,
404
+ "loss": 0.2558,
405
+ "step": 4224
406
+ },
407
+ {
408
+ "epoch": 0.32,
409
+ "learning_rate": 9.507707739415056e-07,
410
+ "loss": 0.2467,
411
+ "step": 4288
412
+ },
413
+ {
414
+ "epoch": 0.32,
415
+ "learning_rate": 9.490814063984796e-07,
416
+ "loss": 0.2412,
417
+ "step": 4352
418
+ },
419
+ {
420
+ "epoch": 0.33,
421
+ "learning_rate": 9.473920388554535e-07,
422
+ "loss": 0.2564,
423
+ "step": 4416
424
+ },
425
+ {
426
+ "epoch": 0.33,
427
+ "learning_rate": 9.457026713124273e-07,
428
+ "loss": 0.2373,
429
+ "step": 4480
430
+ },
431
+ {
432
+ "epoch": 0.34,
433
+ "learning_rate": 9.440133037694013e-07,
434
+ "loss": 0.34,
435
+ "step": 4544
436
+ },
437
+ {
438
+ "epoch": 0.34,
439
+ "learning_rate": 9.423239362263751e-07,
440
+ "loss": 0.2398,
441
+ "step": 4608
442
+ },
443
+ {
444
+ "epoch": 0.35,
445
+ "learning_rate": 9.406345686833491e-07,
446
+ "loss": 0.2493,
447
+ "step": 4672
448
+ },
449
+ {
450
+ "epoch": 0.35,
451
+ "learning_rate": 9.389452011403231e-07,
452
+ "loss": 0.2398,
453
+ "step": 4736
454
+ },
455
+ {
456
+ "epoch": 0.36,
457
+ "learning_rate": 9.37255833597297e-07,
458
+ "loss": 0.2536,
459
+ "step": 4800
460
+ },
461
+ {
462
+ "epoch": 0.36,
463
+ "learning_rate": 9.355664660542709e-07,
464
+ "loss": 0.3538,
465
+ "step": 4864
466
+ },
467
+ {
468
+ "epoch": 0.37,
469
+ "learning_rate": 9.338770985112448e-07,
470
+ "loss": 0.2481,
471
+ "step": 4928
472
+ },
473
+ {
474
+ "epoch": 0.37,
475
+ "learning_rate": 9.321877309682187e-07,
476
+ "loss": 0.2152,
477
+ "step": 4992
478
+ },
479
+ {
480
+ "epoch": 0.38,
481
+ "learning_rate": 9.304983634251927e-07,
482
+ "loss": 0.3056,
483
+ "step": 5056
484
+ },
485
+ {
486
+ "epoch": 0.38,
487
+ "learning_rate": 9.288089958821665e-07,
488
+ "loss": 0.2473,
489
+ "step": 5120
490
+ },
491
+ {
492
+ "epoch": 0.39,
493
+ "learning_rate": 9.271196283391405e-07,
494
+ "loss": 0.234,
495
+ "step": 5184
496
+ },
497
+ {
498
+ "epoch": 0.39,
499
+ "learning_rate": 9.254302607961144e-07,
500
+ "loss": 0.1578,
501
+ "step": 5248
502
+ },
503
+ {
504
+ "epoch": 0.4,
505
+ "learning_rate": 9.237672896209481e-07,
506
+ "loss": 0.2566,
507
+ "step": 5312
508
+ },
509
+ {
510
+ "epoch": 0.4,
511
+ "learning_rate": 9.22077922077922e-07,
512
+ "loss": 0.3794,
513
+ "step": 5376
514
+ },
515
+ {
516
+ "epoch": 0.4,
517
+ "learning_rate": 9.203885545348959e-07,
518
+ "loss": 0.2453,
519
+ "step": 5440
520
+ },
521
+ {
522
+ "epoch": 0.41,
523
+ "learning_rate": 9.186991869918699e-07,
524
+ "loss": 0.2361,
525
+ "step": 5504
526
+ },
527
+ {
528
+ "epoch": 0.41,
529
+ "learning_rate": 9.170098194488438e-07,
530
+ "loss": 0.1734,
531
+ "step": 5568
532
+ },
533
+ {
534
+ "epoch": 0.42,
535
+ "learning_rate": 9.153204519058178e-07,
536
+ "loss": 0.1762,
537
+ "step": 5632
538
+ },
539
+ {
540
+ "epoch": 0.42,
541
+ "learning_rate": 9.136310843627916e-07,
542
+ "loss": 0.2933,
543
+ "step": 5696
544
+ },
545
+ {
546
+ "epoch": 0.43,
547
+ "learning_rate": 9.119417168197655e-07,
548
+ "loss": 0.3152,
549
+ "step": 5760
550
+ },
551
+ {
552
+ "epoch": 0.43,
553
+ "learning_rate": 9.102523492767395e-07,
554
+ "loss": 0.1966,
555
+ "step": 5824
556
+ },
557
+ {
558
+ "epoch": 0.44,
559
+ "learning_rate": 9.085629817337133e-07,
560
+ "loss": 0.298,
561
+ "step": 5888
562
+ },
563
+ {
564
+ "epoch": 0.44,
565
+ "learning_rate": 9.068736141906873e-07,
566
+ "loss": 0.2437,
567
+ "step": 5952
568
+ },
569
+ {
570
+ "epoch": 0.45,
571
+ "learning_rate": 9.051842466476613e-07,
572
+ "loss": 0.305,
573
+ "step": 6016
574
+ },
575
+ {
576
+ "epoch": 0.45,
577
+ "learning_rate": 9.034948791046352e-07,
578
+ "loss": 0.1959,
579
+ "step": 6080
580
+ },
581
+ {
582
+ "epoch": 0.46,
583
+ "learning_rate": 9.018055115616091e-07,
584
+ "loss": 0.2465,
585
+ "step": 6144
586
+ },
587
+ {
588
+ "epoch": 0.46,
589
+ "learning_rate": 9.00116144018583e-07,
590
+ "loss": 0.2624,
591
+ "step": 6208
592
+ },
593
+ {
594
+ "epoch": 0.47,
595
+ "learning_rate": 8.984267764755569e-07,
596
+ "loss": 0.2772,
597
+ "step": 6272
598
+ },
599
+ {
600
+ "epoch": 0.47,
601
+ "learning_rate": 8.967374089325309e-07,
602
+ "loss": 0.2208,
603
+ "step": 6336
604
+ },
605
+ {
606
+ "epoch": 0.48,
607
+ "learning_rate": 8.950480413895047e-07,
608
+ "loss": 0.2695,
609
+ "step": 6400
610
+ },
611
+ {
612
+ "epoch": 0.48,
613
+ "learning_rate": 8.933586738464787e-07,
614
+ "loss": 0.3235,
615
+ "step": 6464
616
+ },
617
+ {
618
+ "epoch": 0.49,
619
+ "learning_rate": 8.916693063034526e-07,
620
+ "loss": 0.2802,
621
+ "step": 6528
622
+ },
623
+ {
624
+ "epoch": 0.49,
625
+ "learning_rate": 8.899799387604265e-07,
626
+ "loss": 0.2383,
627
+ "step": 6592
628
+ },
629
+ {
630
+ "epoch": 0.5,
631
+ "learning_rate": 8.882905712174005e-07,
632
+ "loss": 0.2238,
633
+ "step": 6656
634
+ },
635
+ {
636
+ "epoch": 0.5,
637
+ "learning_rate": 8.866012036743744e-07,
638
+ "loss": 0.2847,
639
+ "step": 6720
640
+ },
641
+ {
642
+ "epoch": 0.5,
643
+ "learning_rate": 8.849118361313483e-07,
644
+ "loss": 0.2713,
645
+ "step": 6784
646
+ },
647
+ {
648
+ "epoch": 0.51,
649
+ "learning_rate": 8.832224685883221e-07,
650
+ "loss": 0.2156,
651
+ "step": 6848
652
+ },
653
+ {
654
+ "epoch": 0.51,
655
+ "learning_rate": 8.815331010452961e-07,
656
+ "loss": 0.2832,
657
+ "step": 6912
658
+ },
659
+ {
660
+ "epoch": 0.52,
661
+ "learning_rate": 8.798437335022701e-07,
662
+ "loss": 0.2675,
663
+ "step": 6976
664
+ },
665
+ {
666
+ "epoch": 0.52,
667
+ "learning_rate": 8.781543659592439e-07,
668
+ "loss": 0.2506,
669
+ "step": 7040
670
+ },
671
+ {
672
+ "epoch": 0.53,
673
+ "learning_rate": 8.764649984162179e-07,
674
+ "loss": 0.2807,
675
+ "step": 7104
676
+ },
677
+ {
678
+ "epoch": 0.53,
679
+ "learning_rate": 8.747756308731918e-07,
680
+ "loss": 0.2802,
681
+ "step": 7168
682
+ },
683
+ {
684
+ "epoch": 0.54,
685
+ "learning_rate": 8.730862633301658e-07,
686
+ "loss": 0.2125,
687
+ "step": 7232
688
+ },
689
+ {
690
+ "epoch": 0.54,
691
+ "learning_rate": 8.713968957871396e-07,
692
+ "loss": 0.2739,
693
+ "step": 7296
694
+ },
695
+ {
696
+ "epoch": 0.55,
697
+ "learning_rate": 8.697075282441135e-07,
698
+ "loss": 0.2341,
699
+ "step": 7360
700
+ },
701
+ {
702
+ "epoch": 0.55,
703
+ "learning_rate": 8.680181607010875e-07,
704
+ "loss": 0.2267,
705
+ "step": 7424
706
+ },
707
+ {
708
+ "epoch": 0.56,
709
+ "learning_rate": 8.663287931580614e-07,
710
+ "loss": 0.3504,
711
+ "step": 7488
712
+ },
713
+ {
714
+ "epoch": 0.56,
715
+ "learning_rate": 8.646394256150353e-07,
716
+ "loss": 0.2675,
717
+ "step": 7552
718
+ },
719
+ {
720
+ "epoch": 0.57,
721
+ "learning_rate": 8.629500580720093e-07,
722
+ "loss": 0.1694,
723
+ "step": 7616
724
+ },
725
+ {
726
+ "epoch": 0.57,
727
+ "learning_rate": 8.612606905289832e-07,
728
+ "loss": 0.2865,
729
+ "step": 7680
730
+ },
731
+ {
732
+ "epoch": 0.58,
733
+ "learning_rate": 8.595713229859571e-07,
734
+ "loss": 0.2737,
735
+ "step": 7744
736
+ },
737
+ {
738
+ "epoch": 0.58,
739
+ "learning_rate": 8.578819554429309e-07,
740
+ "loss": 0.2201,
741
+ "step": 7808
742
+ },
743
+ {
744
+ "epoch": 0.59,
745
+ "learning_rate": 8.561925878999049e-07,
746
+ "loss": 0.1701,
747
+ "step": 7872
748
+ },
749
+ {
750
+ "epoch": 0.59,
751
+ "learning_rate": 8.545032203568789e-07,
752
+ "loss": 0.2566,
753
+ "step": 7936
754
+ },
755
+ {
756
+ "epoch": 0.6,
757
+ "learning_rate": 8.528138528138527e-07,
758
+ "loss": 0.274,
759
+ "step": 8000
760
+ },
761
+ {
762
+ "epoch": 0.6,
763
+ "learning_rate": 8.511244852708267e-07,
764
+ "loss": 0.3413,
765
+ "step": 8064
766
+ },
767
+ {
768
+ "epoch": 0.61,
769
+ "learning_rate": 8.494351177278007e-07,
770
+ "loss": 0.2035,
771
+ "step": 8128
772
+ },
773
+ {
774
+ "epoch": 0.61,
775
+ "learning_rate": 8.477721465526343e-07,
776
+ "loss": 0.2612,
777
+ "step": 8192
778
+ },
779
+ {
780
+ "epoch": 0.61,
781
+ "learning_rate": 8.460827790096083e-07,
782
+ "loss": 0.2322,
783
+ "step": 8256
784
+ },
785
+ {
786
+ "epoch": 0.62,
787
+ "learning_rate": 8.443934114665822e-07,
788
+ "loss": 0.1929,
789
+ "step": 8320
790
+ },
791
+ {
792
+ "epoch": 0.62,
793
+ "learning_rate": 8.427040439235561e-07,
794
+ "loss": 0.2212,
795
+ "step": 8384
796
+ },
797
+ {
798
+ "epoch": 0.63,
799
+ "learning_rate": 8.4101467638053e-07,
800
+ "loss": 0.1912,
801
+ "step": 8448
802
+ },
803
+ {
804
+ "epoch": 0.63,
805
+ "learning_rate": 8.39325308837504e-07,
806
+ "loss": 0.3325,
807
+ "step": 8512
808
+ },
809
+ {
810
+ "epoch": 0.64,
811
+ "learning_rate": 8.376359412944777e-07,
812
+ "loss": 0.2341,
813
+ "step": 8576
814
+ },
815
+ {
816
+ "epoch": 0.64,
817
+ "learning_rate": 8.359465737514517e-07,
818
+ "loss": 0.248,
819
+ "step": 8640
820
+ },
821
+ {
822
+ "epoch": 0.65,
823
+ "learning_rate": 8.342572062084257e-07,
824
+ "loss": 0.268,
825
+ "step": 8704
826
+ },
827
+ {
828
+ "epoch": 0.65,
829
+ "learning_rate": 8.325678386653996e-07,
830
+ "loss": 0.3279,
831
+ "step": 8768
832
+ },
833
+ {
834
+ "epoch": 0.66,
835
+ "learning_rate": 8.308784711223735e-07,
836
+ "loss": 0.2175,
837
+ "step": 8832
838
+ },
839
+ {
840
+ "epoch": 0.66,
841
+ "learning_rate": 8.291891035793475e-07,
842
+ "loss": 0.2641,
843
+ "step": 8896
844
+ },
845
+ {
846
+ "epoch": 0.67,
847
+ "learning_rate": 8.274997360363214e-07,
848
+ "loss": 0.2435,
849
+ "step": 8960
850
+ },
851
+ {
852
+ "epoch": 0.67,
853
+ "learning_rate": 8.258103684932953e-07,
854
+ "loss": 0.208,
855
+ "step": 9024
856
+ },
857
+ {
858
+ "epoch": 0.68,
859
+ "learning_rate": 8.241210009502691e-07,
860
+ "loss": 0.2631,
861
+ "step": 9088
862
+ },
863
+ {
864
+ "epoch": 0.68,
865
+ "learning_rate": 8.224316334072431e-07,
866
+ "loss": 0.2774,
867
+ "step": 9152
868
+ },
869
+ {
870
+ "epoch": 0.69,
871
+ "learning_rate": 8.207422658642171e-07,
872
+ "loss": 0.2279,
873
+ "step": 9216
874
+ },
875
+ {
876
+ "epoch": 0.69,
877
+ "learning_rate": 8.190528983211909e-07,
878
+ "loss": 0.1885,
879
+ "step": 9280
880
+ },
881
+ {
882
+ "epoch": 0.7,
883
+ "learning_rate": 8.173635307781649e-07,
884
+ "loss": 0.2396,
885
+ "step": 9344
886
+ },
887
+ {
888
+ "epoch": 0.7,
889
+ "learning_rate": 8.156741632351388e-07,
890
+ "loss": 0.269,
891
+ "step": 9408
892
+ },
893
+ {
894
+ "epoch": 0.71,
895
+ "learning_rate": 8.139847956921128e-07,
896
+ "loss": 0.315,
897
+ "step": 9472
898
+ },
899
+ {
900
+ "epoch": 0.71,
901
+ "learning_rate": 8.122954281490867e-07,
902
+ "loss": 0.2165,
903
+ "step": 9536
904
+ },
905
+ {
906
+ "epoch": 0.71,
907
+ "learning_rate": 8.106060606060605e-07,
908
+ "loss": 0.2035,
909
+ "step": 9600
910
+ },
911
+ {
912
+ "epoch": 0.72,
913
+ "learning_rate": 8.089166930630345e-07,
914
+ "loss": 0.2472,
915
+ "step": 9664
916
+ },
917
+ {
918
+ "epoch": 0.72,
919
+ "learning_rate": 8.072273255200083e-07,
920
+ "loss": 0.2641,
921
+ "step": 9728
922
+ },
923
+ {
924
+ "epoch": 0.73,
925
+ "learning_rate": 8.055379579769823e-07,
926
+ "loss": 0.1937,
927
+ "step": 9792
928
+ },
929
+ {
930
+ "epoch": 0.73,
931
+ "learning_rate": 8.038485904339563e-07,
932
+ "loss": 0.2108,
933
+ "step": 9856
934
+ },
935
+ {
936
+ "epoch": 0.74,
937
+ "learning_rate": 8.021592228909302e-07,
938
+ "loss": 0.2866,
939
+ "step": 9920
940
+ },
941
+ {
942
+ "epoch": 0.74,
943
+ "learning_rate": 8.004698553479041e-07,
944
+ "loss": 0.2707,
945
+ "step": 9984
946
+ },
947
+ {
948
+ "epoch": 0.75,
949
+ "learning_rate": 7.98780487804878e-07,
950
+ "loss": 0.2249,
951
+ "step": 10048
952
+ },
953
+ {
954
+ "epoch": 0.75,
955
+ "learning_rate": 7.970911202618519e-07,
956
+ "loss": 0.2183,
957
+ "step": 10112
958
+ },
959
+ {
960
+ "epoch": 0.76,
961
+ "learning_rate": 7.954017527188258e-07,
962
+ "loss": 0.234,
963
+ "step": 10176
964
+ },
965
+ {
966
+ "epoch": 0.76,
967
+ "learning_rate": 7.937123851757997e-07,
968
+ "loss": 0.2504,
969
+ "step": 10240
970
+ },
971
+ {
972
+ "epoch": 0.77,
973
+ "learning_rate": 7.920230176327737e-07,
974
+ "loss": 0.1973,
975
+ "step": 10304
976
+ },
977
+ {
978
+ "epoch": 0.77,
979
+ "learning_rate": 7.903336500897477e-07,
980
+ "loss": 0.2003,
981
+ "step": 10368
982
+ },
983
+ {
984
+ "epoch": 0.78,
985
+ "learning_rate": 7.886442825467215e-07,
986
+ "loss": 0.2558,
987
+ "step": 10432
988
+ },
989
+ {
990
+ "epoch": 0.78,
991
+ "learning_rate": 7.869549150036955e-07,
992
+ "loss": 0.2694,
993
+ "step": 10496
994
+ },
995
+ {
996
+ "epoch": 0.79,
997
+ "learning_rate": 7.852655474606694e-07,
998
+ "loss": 0.2633,
999
+ "step": 10560
1000
+ },
1001
+ {
1002
+ "epoch": 0.79,
1003
+ "learning_rate": 7.835761799176432e-07,
1004
+ "loss": 0.2649,
1005
+ "step": 10624
1006
+ },
1007
+ {
1008
+ "epoch": 0.8,
1009
+ "learning_rate": 7.818868123746172e-07,
1010
+ "loss": 0.2874,
1011
+ "step": 10688
1012
+ },
1013
+ {
1014
+ "epoch": 0.8,
1015
+ "learning_rate": 7.801974448315911e-07,
1016
+ "loss": 0.2393,
1017
+ "step": 10752
1018
+ },
1019
+ {
1020
+ "epoch": 0.81,
1021
+ "learning_rate": 7.785080772885651e-07,
1022
+ "loss": 0.2303,
1023
+ "step": 10816
1024
+ },
1025
+ {
1026
+ "epoch": 0.81,
1027
+ "learning_rate": 7.768451061133987e-07,
1028
+ "loss": 0.2417,
1029
+ "step": 10880
1030
+ },
1031
+ {
1032
+ "epoch": 0.81,
1033
+ "learning_rate": 7.751557385703727e-07,
1034
+ "loss": 0.272,
1035
+ "step": 10944
1036
+ },
1037
+ {
1038
+ "epoch": 0.82,
1039
+ "learning_rate": 7.734663710273465e-07,
1040
+ "loss": 0.2005,
1041
+ "step": 11008
1042
+ },
1043
+ {
1044
+ "epoch": 0.82,
1045
+ "learning_rate": 7.717770034843205e-07,
1046
+ "loss": 0.1505,
1047
+ "step": 11072
1048
+ },
1049
+ {
1050
+ "epoch": 0.83,
1051
+ "learning_rate": 7.700876359412945e-07,
1052
+ "loss": 0.2839,
1053
+ "step": 11136
1054
+ },
1055
+ {
1056
+ "epoch": 0.83,
1057
+ "learning_rate": 7.683982683982684e-07,
1058
+ "loss": 0.2805,
1059
+ "step": 11200
1060
+ },
1061
+ {
1062
+ "epoch": 0.84,
1063
+ "learning_rate": 7.667089008552423e-07,
1064
+ "loss": 0.2407,
1065
+ "step": 11264
1066
+ },
1067
+ {
1068
+ "epoch": 0.84,
1069
+ "learning_rate": 7.650195333122162e-07,
1070
+ "loss": 0.1769,
1071
+ "step": 11328
1072
+ },
1073
+ {
1074
+ "epoch": 0.85,
1075
+ "learning_rate": 7.633301657691901e-07,
1076
+ "loss": 0.2223,
1077
+ "step": 11392
1078
+ },
1079
+ {
1080
+ "epoch": 0.85,
1081
+ "learning_rate": 7.61640798226164e-07,
1082
+ "loss": 0.2609,
1083
+ "step": 11456
1084
+ },
1085
+ {
1086
+ "epoch": 0.86,
1087
+ "learning_rate": 7.599514306831379e-07,
1088
+ "loss": 0.3837,
1089
+ "step": 11520
1090
+ },
1091
+ {
1092
+ "epoch": 0.86,
1093
+ "learning_rate": 7.582620631401119e-07,
1094
+ "loss": 0.2162,
1095
+ "step": 11584
1096
+ },
1097
+ {
1098
+ "epoch": 0.87,
1099
+ "learning_rate": 7.565726955970858e-07,
1100
+ "loss": 0.2076,
1101
+ "step": 11648
1102
+ },
1103
+ {
1104
+ "epoch": 0.87,
1105
+ "learning_rate": 7.548833280540597e-07,
1106
+ "loss": 0.266,
1107
+ "step": 11712
1108
+ },
1109
+ {
1110
+ "epoch": 0.88,
1111
+ "learning_rate": 7.531939605110337e-07,
1112
+ "loss": 0.1652,
1113
+ "step": 11776
1114
+ },
1115
+ {
1116
+ "epoch": 0.88,
1117
+ "learning_rate": 7.515045929680076e-07,
1118
+ "loss": 0.2416,
1119
+ "step": 11840
1120
+ },
1121
+ {
1122
+ "epoch": 0.89,
1123
+ "learning_rate": 7.498152254249815e-07,
1124
+ "loss": 0.2471,
1125
+ "step": 11904
1126
+ },
1127
+ {
1128
+ "epoch": 0.89,
1129
+ "learning_rate": 7.481258578819553e-07,
1130
+ "loss": 0.2285,
1131
+ "step": 11968
1132
+ },
1133
+ {
1134
+ "epoch": 0.9,
1135
+ "learning_rate": 7.464364903389293e-07,
1136
+ "loss": 0.2368,
1137
+ "step": 12032
1138
+ },
1139
+ {
1140
+ "epoch": 0.9,
1141
+ "learning_rate": 7.447471227959033e-07,
1142
+ "loss": 0.1847,
1143
+ "step": 12096
1144
+ },
1145
+ {
1146
+ "epoch": 0.91,
1147
+ "learning_rate": 7.430577552528771e-07,
1148
+ "loss": 0.2826,
1149
+ "step": 12160
1150
+ },
1151
+ {
1152
+ "epoch": 0.91,
1153
+ "learning_rate": 7.413683877098511e-07,
1154
+ "loss": 0.2931,
1155
+ "step": 12224
1156
+ },
1157
+ {
1158
+ "epoch": 0.91,
1159
+ "learning_rate": 7.396790201668251e-07,
1160
+ "loss": 0.1722,
1161
+ "step": 12288
1162
+ },
1163
+ {
1164
+ "epoch": 0.92,
1165
+ "learning_rate": 7.37989652623799e-07,
1166
+ "loss": 0.2673,
1167
+ "step": 12352
1168
+ },
1169
+ {
1170
+ "epoch": 0.92,
1171
+ "learning_rate": 7.363002850807728e-07,
1172
+ "loss": 0.2244,
1173
+ "step": 12416
1174
+ },
1175
+ {
1176
+ "epoch": 0.93,
1177
+ "learning_rate": 7.346109175377467e-07,
1178
+ "loss": 0.1532,
1179
+ "step": 12480
1180
+ },
1181
+ {
1182
+ "epoch": 0.93,
1183
+ "learning_rate": 7.329215499947207e-07,
1184
+ "loss": 0.2207,
1185
+ "step": 12544
1186
+ },
1187
+ {
1188
+ "epoch": 0.94,
1189
+ "learning_rate": 7.312321824516946e-07,
1190
+ "loss": 0.2362,
1191
+ "step": 12608
1192
+ },
1193
+ {
1194
+ "epoch": 0.94,
1195
+ "learning_rate": 7.295428149086685e-07,
1196
+ "loss": 0.2222,
1197
+ "step": 12672
1198
+ },
1199
+ {
1200
+ "epoch": 0.95,
1201
+ "learning_rate": 7.278534473656425e-07,
1202
+ "loss": 0.2664,
1203
+ "step": 12736
1204
+ },
1205
+ {
1206
+ "epoch": 0.95,
1207
+ "learning_rate": 7.261640798226164e-07,
1208
+ "loss": 0.1632,
1209
+ "step": 12800
1210
+ },
1211
+ {
1212
+ "epoch": 0.96,
1213
+ "learning_rate": 7.244747122795903e-07,
1214
+ "loss": 0.2212,
1215
+ "step": 12864
1216
+ },
1217
+ {
1218
+ "epoch": 0.96,
1219
+ "learning_rate": 7.227853447365642e-07,
1220
+ "loss": 0.2026,
1221
+ "step": 12928
1222
+ },
1223
+ {
1224
+ "epoch": 0.97,
1225
+ "learning_rate": 7.210959771935381e-07,
1226
+ "loss": 0.2562,
1227
+ "step": 12992
1228
+ },
1229
+ {
1230
+ "epoch": 0.97,
1231
+ "learning_rate": 7.194066096505121e-07,
1232
+ "loss": 0.2038,
1233
+ "step": 13056
1234
+ },
1235
+ {
1236
+ "epoch": 0.98,
1237
+ "learning_rate": 7.177436384753458e-07,
1238
+ "loss": 0.1829,
1239
+ "step": 13120
1240
+ },
1241
+ {
1242
+ "epoch": 0.98,
1243
+ "learning_rate": 7.160542709323197e-07,
1244
+ "loss": 0.217,
1245
+ "step": 13184
1246
+ },
1247
+ {
1248
+ "epoch": 0.99,
1249
+ "learning_rate": 7.143649033892935e-07,
1250
+ "loss": 0.2303,
1251
+ "step": 13248
1252
+ },
1253
+ {
1254
+ "epoch": 0.99,
1255
+ "learning_rate": 7.126755358462675e-07,
1256
+ "loss": 0.1996,
1257
+ "step": 13312
1258
+ },
1259
+ {
1260
+ "epoch": 1.0,
1261
+ "learning_rate": 7.109861683032415e-07,
1262
+ "loss": 0.206,
1263
+ "step": 13376
1264
+ },
1265
+ {
1266
+ "epoch": 1.0,
1267
+ "learning_rate": 7.092968007602153e-07,
1268
+ "loss": 0.2405,
1269
+ "step": 13440
1270
+ },
1271
+ {
1272
+ "epoch": 1.01,
1273
+ "learning_rate": 7.076074332171893e-07,
1274
+ "loss": 0.1609,
1275
+ "step": 13504
1276
+ },
1277
+ {
1278
+ "epoch": 1.01,
1279
+ "learning_rate": 7.059180656741632e-07,
1280
+ "loss": 0.1735,
1281
+ "step": 13568
1282
+ },
1283
+ {
1284
+ "epoch": 1.01,
1285
+ "learning_rate": 7.042286981311372e-07,
1286
+ "loss": 0.1477,
1287
+ "step": 13632
1288
+ },
1289
+ {
1290
+ "epoch": 1.02,
1291
+ "learning_rate": 7.02539330588111e-07,
1292
+ "loss": 0.2664,
1293
+ "step": 13696
1294
+ },
1295
+ {
1296
+ "epoch": 1.02,
1297
+ "learning_rate": 7.008499630450849e-07,
1298
+ "loss": 0.1506,
1299
+ "step": 13760
1300
+ },
1301
+ {
1302
+ "epoch": 1.03,
1303
+ "learning_rate": 6.991605955020589e-07,
1304
+ "loss": 0.1315,
1305
+ "step": 13824
1306
+ },
1307
+ {
1308
+ "epoch": 1.03,
1309
+ "learning_rate": 6.974712279590327e-07,
1310
+ "loss": 0.1494,
1311
+ "step": 13888
1312
+ },
1313
+ {
1314
+ "epoch": 1.04,
1315
+ "learning_rate": 6.957818604160067e-07,
1316
+ "loss": 0.1959,
1317
+ "step": 13952
1318
+ },
1319
+ {
1320
+ "epoch": 1.04,
1321
+ "learning_rate": 6.940924928729807e-07,
1322
+ "loss": 0.1266,
1323
+ "step": 14016
1324
+ },
1325
+ {
1326
+ "epoch": 1.05,
1327
+ "learning_rate": 6.924031253299546e-07,
1328
+ "loss": 0.1257,
1329
+ "step": 14080
1330
+ },
1331
+ {
1332
+ "epoch": 1.05,
1333
+ "learning_rate": 6.907137577869285e-07,
1334
+ "loss": 0.1429,
1335
+ "step": 14144
1336
+ },
1337
+ {
1338
+ "epoch": 1.06,
1339
+ "learning_rate": 6.890243902439023e-07,
1340
+ "loss": 0.1911,
1341
+ "step": 14208
1342
+ },
1343
+ {
1344
+ "epoch": 1.06,
1345
+ "learning_rate": 6.873350227008763e-07,
1346
+ "loss": 0.1307,
1347
+ "step": 14272
1348
+ },
1349
+ {
1350
+ "epoch": 1.07,
1351
+ "learning_rate": 6.856456551578503e-07,
1352
+ "loss": 0.1199,
1353
+ "step": 14336
1354
+ },
1355
+ {
1356
+ "epoch": 1.07,
1357
+ "learning_rate": 6.839562876148241e-07,
1358
+ "loss": 0.1367,
1359
+ "step": 14400
1360
+ },
1361
+ {
1362
+ "epoch": 1.08,
1363
+ "learning_rate": 6.822669200717981e-07,
1364
+ "loss": 0.102,
1365
+ "step": 14464
1366
+ },
1367
+ {
1368
+ "epoch": 1.08,
1369
+ "learning_rate": 6.805775525287721e-07,
1370
+ "loss": 0.2138,
1371
+ "step": 14528
1372
+ },
1373
+ {
1374
+ "epoch": 1.09,
1375
+ "learning_rate": 6.788881849857459e-07,
1376
+ "loss": 0.1363,
1377
+ "step": 14592
1378
+ },
1379
+ {
1380
+ "epoch": 1.09,
1381
+ "learning_rate": 6.771988174427199e-07,
1382
+ "loss": 0.1177,
1383
+ "step": 14656
1384
+ },
1385
+ {
1386
+ "epoch": 1.1,
1387
+ "learning_rate": 6.755094498996937e-07,
1388
+ "loss": 0.1916,
1389
+ "step": 14720
1390
+ },
1391
+ {
1392
+ "epoch": 1.1,
1393
+ "learning_rate": 6.738200823566677e-07,
1394
+ "loss": 0.1557,
1395
+ "step": 14784
1396
+ },
1397
+ {
1398
+ "epoch": 1.11,
1399
+ "learning_rate": 6.721307148136416e-07,
1400
+ "loss": 0.1847,
1401
+ "step": 14848
1402
+ },
1403
+ {
1404
+ "epoch": 1.11,
1405
+ "learning_rate": 6.704413472706155e-07,
1406
+ "loss": 0.1381,
1407
+ "step": 14912
1408
+ },
1409
+ {
1410
+ "epoch": 1.11,
1411
+ "learning_rate": 6.687519797275895e-07,
1412
+ "loss": 0.1897,
1413
+ "step": 14976
1414
+ },
1415
+ {
1416
+ "epoch": 1.12,
1417
+ "learning_rate": 6.670626121845634e-07,
1418
+ "loss": 0.1866,
1419
+ "step": 15040
1420
+ },
1421
+ {
1422
+ "epoch": 1.12,
1423
+ "learning_rate": 6.653732446415373e-07,
1424
+ "loss": 0.143,
1425
+ "step": 15104
1426
+ },
1427
+ {
1428
+ "epoch": 1.13,
1429
+ "learning_rate": 6.636838770985113e-07,
1430
+ "loss": 0.1729,
1431
+ "step": 15168
1432
+ },
1433
+ {
1434
+ "epoch": 1.13,
1435
+ "learning_rate": 6.619945095554851e-07,
1436
+ "loss": 0.2373,
1437
+ "step": 15232
1438
+ },
1439
+ {
1440
+ "epoch": 1.14,
1441
+ "learning_rate": 6.60305142012459e-07,
1442
+ "loss": 0.1464,
1443
+ "step": 15296
1444
+ },
1445
+ {
1446
+ "epoch": 1.14,
1447
+ "learning_rate": 6.586157744694329e-07,
1448
+ "loss": 0.1707,
1449
+ "step": 15360
1450
+ },
1451
+ {
1452
+ "epoch": 1.15,
1453
+ "learning_rate": 6.569264069264069e-07,
1454
+ "loss": 0.1757,
1455
+ "step": 15424
1456
+ },
1457
+ {
1458
+ "epoch": 1.15,
1459
+ "learning_rate": 6.552370393833809e-07,
1460
+ "loss": 0.2099,
1461
+ "step": 15488
1462
+ },
1463
+ {
1464
+ "epoch": 1.16,
1465
+ "learning_rate": 6.535476718403547e-07,
1466
+ "loss": 0.2021,
1467
+ "step": 15552
1468
+ },
1469
+ {
1470
+ "epoch": 1.16,
1471
+ "learning_rate": 6.518583042973287e-07,
1472
+ "loss": 0.1288,
1473
+ "step": 15616
1474
+ },
1475
+ {
1476
+ "epoch": 1.17,
1477
+ "learning_rate": 6.501689367543026e-07,
1478
+ "loss": 0.1461,
1479
+ "step": 15680
1480
+ },
1481
+ {
1482
+ "epoch": 1.17,
1483
+ "learning_rate": 6.484795692112764e-07,
1484
+ "loss": 0.1831,
1485
+ "step": 15744
1486
+ },
1487
+ {
1488
+ "epoch": 1.18,
1489
+ "learning_rate": 6.467902016682504e-07,
1490
+ "loss": 0.0907,
1491
+ "step": 15808
1492
+ },
1493
+ {
1494
+ "epoch": 1.18,
1495
+ "learning_rate": 6.451008341252243e-07,
1496
+ "loss": 0.1144,
1497
+ "step": 15872
1498
+ },
1499
+ {
1500
+ "epoch": 1.19,
1501
+ "learning_rate": 6.434114665821983e-07,
1502
+ "loss": 0.1916,
1503
+ "step": 15936
1504
+ },
1505
+ {
1506
+ "epoch": 1.19,
1507
+ "learning_rate": 6.417220990391721e-07,
1508
+ "loss": 0.138,
1509
+ "step": 16000
1510
+ },
1511
+ {
1512
+ "epoch": 1.2,
1513
+ "learning_rate": 6.400327314961461e-07,
1514
+ "loss": 0.1923,
1515
+ "step": 16064
1516
+ },
1517
+ {
1518
+ "epoch": 1.2,
1519
+ "learning_rate": 6.383433639531201e-07,
1520
+ "loss": 0.1353,
1521
+ "step": 16128
1522
+ },
1523
+ {
1524
+ "epoch": 1.21,
1525
+ "learning_rate": 6.36653996410094e-07,
1526
+ "loss": 0.2265,
1527
+ "step": 16192
1528
+ },
1529
+ {
1530
+ "epoch": 1.21,
1531
+ "learning_rate": 6.349646288670678e-07,
1532
+ "loss": 0.1781,
1533
+ "step": 16256
1534
+ },
1535
+ {
1536
+ "epoch": 1.21,
1537
+ "learning_rate": 6.332752613240417e-07,
1538
+ "loss": 0.1241,
1539
+ "step": 16320
1540
+ },
1541
+ {
1542
+ "epoch": 1.22,
1543
+ "learning_rate": 6.315858937810157e-07,
1544
+ "loss": 0.1227,
1545
+ "step": 16384
1546
+ },
1547
+ {
1548
+ "epoch": 1.22,
1549
+ "learning_rate": 6.298965262379896e-07,
1550
+ "loss": 0.1769,
1551
+ "step": 16448
1552
+ },
1553
+ {
1554
+ "epoch": 1.23,
1555
+ "learning_rate": 6.282071586949635e-07,
1556
+ "loss": 0.1482,
1557
+ "step": 16512
1558
+ },
1559
+ {
1560
+ "epoch": 1.23,
1561
+ "learning_rate": 6.265177911519375e-07,
1562
+ "loss": 0.1909,
1563
+ "step": 16576
1564
+ },
1565
+ {
1566
+ "epoch": 1.24,
1567
+ "learning_rate": 6.248284236089115e-07,
1568
+ "loss": 0.1083,
1569
+ "step": 16640
1570
+ },
1571
+ {
1572
+ "epoch": 1.24,
1573
+ "learning_rate": 6.231390560658853e-07,
1574
+ "loss": 0.1837,
1575
+ "step": 16704
1576
+ },
1577
+ {
1578
+ "epoch": 1.25,
1579
+ "learning_rate": 6.214496885228592e-07,
1580
+ "loss": 0.1213,
1581
+ "step": 16768
1582
+ },
1583
+ {
1584
+ "epoch": 1.25,
1585
+ "learning_rate": 6.197603209798331e-07,
1586
+ "loss": 0.1744,
1587
+ "step": 16832
1588
+ },
1589
+ {
1590
+ "epoch": 1.26,
1591
+ "learning_rate": 6.18070953436807e-07,
1592
+ "loss": 0.1638,
1593
+ "step": 16896
1594
+ },
1595
+ {
1596
+ "epoch": 1.26,
1597
+ "learning_rate": 6.16381585893781e-07,
1598
+ "loss": 0.1712,
1599
+ "step": 16960
1600
+ },
1601
+ {
1602
+ "epoch": 1.27,
1603
+ "learning_rate": 6.146922183507549e-07,
1604
+ "loss": 0.1842,
1605
+ "step": 17024
1606
+ },
1607
+ {
1608
+ "epoch": 1.27,
1609
+ "learning_rate": 6.130028508077289e-07,
1610
+ "loss": 0.2725,
1611
+ "step": 17088
1612
+ },
1613
+ {
1614
+ "epoch": 1.28,
1615
+ "learning_rate": 6.113398796325625e-07,
1616
+ "loss": 0.1689,
1617
+ "step": 17152
1618
+ },
1619
+ {
1620
+ "epoch": 1.28,
1621
+ "learning_rate": 6.096505120895365e-07,
1622
+ "loss": 0.1041,
1623
+ "step": 17216
1624
+ },
1625
+ {
1626
+ "epoch": 1.29,
1627
+ "learning_rate": 6.079611445465103e-07,
1628
+ "loss": 0.1797,
1629
+ "step": 17280
1630
+ },
1631
+ {
1632
+ "epoch": 1.29,
1633
+ "learning_rate": 6.062717770034843e-07,
1634
+ "loss": 0.2498,
1635
+ "step": 17344
1636
+ },
1637
+ {
1638
+ "epoch": 1.3,
1639
+ "learning_rate": 6.045824094604583e-07,
1640
+ "loss": 0.1451,
1641
+ "step": 17408
1642
+ },
1643
+ {
1644
+ "epoch": 1.3,
1645
+ "learning_rate": 6.028930419174322e-07,
1646
+ "loss": 0.1862,
1647
+ "step": 17472
1648
+ },
1649
+ {
1650
+ "epoch": 1.31,
1651
+ "learning_rate": 6.01203674374406e-07,
1652
+ "loss": 0.1495,
1653
+ "step": 17536
1654
+ },
1655
+ {
1656
+ "epoch": 1.31,
1657
+ "learning_rate": 5.995143068313799e-07,
1658
+ "loss": 0.1119,
1659
+ "step": 17600
1660
+ },
1661
+ {
1662
+ "epoch": 1.31,
1663
+ "learning_rate": 5.978249392883539e-07,
1664
+ "loss": 0.2057,
1665
+ "step": 17664
1666
+ },
1667
+ {
1668
+ "epoch": 1.32,
1669
+ "learning_rate": 5.961355717453278e-07,
1670
+ "loss": 0.1437,
1671
+ "step": 17728
1672
+ },
1673
+ {
1674
+ "epoch": 1.32,
1675
+ "learning_rate": 5.944462042023017e-07,
1676
+ "loss": 0.1174,
1677
+ "step": 17792
1678
+ },
1679
+ {
1680
+ "epoch": 1.33,
1681
+ "learning_rate": 5.927568366592757e-07,
1682
+ "loss": 0.1908,
1683
+ "step": 17856
1684
+ },
1685
+ {
1686
+ "epoch": 1.33,
1687
+ "learning_rate": 5.910674691162496e-07,
1688
+ "loss": 0.1652,
1689
+ "step": 17920
1690
+ },
1691
+ {
1692
+ "epoch": 1.34,
1693
+ "learning_rate": 5.893781015732235e-07,
1694
+ "loss": 0.1413,
1695
+ "step": 17984
1696
+ },
1697
+ {
1698
+ "epoch": 1.34,
1699
+ "learning_rate": 5.876887340301974e-07,
1700
+ "loss": 0.1883,
1701
+ "step": 18048
1702
+ },
1703
+ {
1704
+ "epoch": 1.35,
1705
+ "learning_rate": 5.859993664871713e-07,
1706
+ "loss": 0.1673,
1707
+ "step": 18112
1708
+ },
1709
+ {
1710
+ "epoch": 1.35,
1711
+ "learning_rate": 5.843099989441452e-07,
1712
+ "loss": 0.1304,
1713
+ "step": 18176
1714
+ },
1715
+ {
1716
+ "epoch": 1.36,
1717
+ "learning_rate": 5.826206314011191e-07,
1718
+ "loss": 0.1971,
1719
+ "step": 18240
1720
+ },
1721
+ {
1722
+ "epoch": 1.36,
1723
+ "learning_rate": 5.809312638580931e-07,
1724
+ "loss": 0.1318,
1725
+ "step": 18304
1726
+ },
1727
+ {
1728
+ "epoch": 1.37,
1729
+ "learning_rate": 5.792418963150671e-07,
1730
+ "loss": 0.1817,
1731
+ "step": 18368
1732
+ },
1733
+ {
1734
+ "epoch": 1.37,
1735
+ "learning_rate": 5.775525287720409e-07,
1736
+ "loss": 0.1825,
1737
+ "step": 18432
1738
+ },
1739
+ {
1740
+ "epoch": 1.38,
1741
+ "learning_rate": 5.758895575968747e-07,
1742
+ "loss": 0.1423,
1743
+ "step": 18496
1744
+ },
1745
+ {
1746
+ "epoch": 1.38,
1747
+ "learning_rate": 5.742001900538485e-07,
1748
+ "loss": 0.1742,
1749
+ "step": 18560
1750
+ },
1751
+ {
1752
+ "epoch": 1.39,
1753
+ "learning_rate": 5.725108225108225e-07,
1754
+ "loss": 0.1447,
1755
+ "step": 18624
1756
+ },
1757
+ {
1758
+ "epoch": 1.39,
1759
+ "learning_rate": 5.708214549677965e-07,
1760
+ "loss": 0.1451,
1761
+ "step": 18688
1762
+ },
1763
+ {
1764
+ "epoch": 1.4,
1765
+ "learning_rate": 5.691320874247704e-07,
1766
+ "loss": 0.188,
1767
+ "step": 18752
1768
+ },
1769
+ {
1770
+ "epoch": 1.4,
1771
+ "learning_rate": 5.674427198817442e-07,
1772
+ "loss": 0.1554,
1773
+ "step": 18816
1774
+ },
1775
+ {
1776
+ "epoch": 1.41,
1777
+ "learning_rate": 5.657533523387181e-07,
1778
+ "loss": 0.0989,
1779
+ "step": 18880
1780
+ },
1781
+ {
1782
+ "epoch": 1.41,
1783
+ "learning_rate": 5.640639847956921e-07,
1784
+ "loss": 0.184,
1785
+ "step": 18944
1786
+ },
1787
+ {
1788
+ "epoch": 1.41,
1789
+ "learning_rate": 5.62374617252666e-07,
1790
+ "loss": 0.0934,
1791
+ "step": 19008
1792
+ },
1793
+ {
1794
+ "epoch": 1.42,
1795
+ "learning_rate": 5.606852497096399e-07,
1796
+ "loss": 0.173,
1797
+ "step": 19072
1798
+ },
1799
+ {
1800
+ "epoch": 1.42,
1801
+ "learning_rate": 5.589958821666139e-07,
1802
+ "loss": 0.1898,
1803
+ "step": 19136
1804
+ },
1805
+ {
1806
+ "epoch": 1.43,
1807
+ "learning_rate": 5.573065146235878e-07,
1808
+ "loss": 0.1446,
1809
+ "step": 19200
1810
+ },
1811
+ {
1812
+ "epoch": 1.43,
1813
+ "learning_rate": 5.556171470805617e-07,
1814
+ "loss": 0.1758,
1815
+ "step": 19264
1816
+ },
1817
+ {
1818
+ "epoch": 1.44,
1819
+ "learning_rate": 5.539277795375356e-07,
1820
+ "loss": 0.1509,
1821
+ "step": 19328
1822
+ },
1823
+ {
1824
+ "epoch": 1.44,
1825
+ "learning_rate": 5.522384119945095e-07,
1826
+ "loss": 0.301,
1827
+ "step": 19392
1828
+ },
1829
+ {
1830
+ "epoch": 1.45,
1831
+ "learning_rate": 5.505490444514834e-07,
1832
+ "loss": 0.107,
1833
+ "step": 19456
1834
+ },
1835
+ {
1836
+ "epoch": 1.45,
1837
+ "learning_rate": 5.488596769084573e-07,
1838
+ "loss": 0.1417,
1839
+ "step": 19520
1840
+ },
1841
+ {
1842
+ "epoch": 1.46,
1843
+ "learning_rate": 5.471703093654313e-07,
1844
+ "loss": 0.152,
1845
+ "step": 19584
1846
+ },
1847
+ {
1848
+ "epoch": 1.46,
1849
+ "learning_rate": 5.454809418224053e-07,
1850
+ "loss": 0.1762,
1851
+ "step": 19648
1852
+ },
1853
+ {
1854
+ "epoch": 1.47,
1855
+ "learning_rate": 5.437915742793791e-07,
1856
+ "loss": 0.1559,
1857
+ "step": 19712
1858
+ },
1859
+ {
1860
+ "epoch": 1.47,
1861
+ "learning_rate": 5.421022067363531e-07,
1862
+ "loss": 0.1965,
1863
+ "step": 19776
1864
+ },
1865
+ {
1866
+ "epoch": 1.48,
1867
+ "learning_rate": 5.404128391933269e-07,
1868
+ "loss": 0.1924,
1869
+ "step": 19840
1870
+ },
1871
+ {
1872
+ "epoch": 1.48,
1873
+ "learning_rate": 5.387234716503009e-07,
1874
+ "loss": 0.1208,
1875
+ "step": 19904
1876
+ },
1877
+ {
1878
+ "epoch": 1.49,
1879
+ "learning_rate": 5.370341041072748e-07,
1880
+ "loss": 0.1676,
1881
+ "step": 19968
1882
+ },
1883
+ {
1884
+ "epoch": 1.49,
1885
+ "learning_rate": 5.353447365642487e-07,
1886
+ "loss": 0.1461,
1887
+ "step": 20032
1888
+ },
1889
+ {
1890
+ "epoch": 1.5,
1891
+ "learning_rate": 5.336553690212227e-07,
1892
+ "loss": 0.2259,
1893
+ "step": 20096
1894
+ },
1895
+ {
1896
+ "epoch": 1.5,
1897
+ "learning_rate": 5.319660014781965e-07,
1898
+ "loss": 0.1377,
1899
+ "step": 20160
1900
+ },
1901
+ {
1902
+ "epoch": 1.51,
1903
+ "learning_rate": 5.302766339351705e-07,
1904
+ "loss": 0.1578,
1905
+ "step": 20224
1906
+ },
1907
+ {
1908
+ "epoch": 1.51,
1909
+ "learning_rate": 5.285872663921445e-07,
1910
+ "loss": 0.1253,
1911
+ "step": 20288
1912
+ },
1913
+ {
1914
+ "epoch": 1.51,
1915
+ "learning_rate": 5.268978988491183e-07,
1916
+ "loss": 0.2099,
1917
+ "step": 20352
1918
+ },
1919
+ {
1920
+ "epoch": 1.52,
1921
+ "learning_rate": 5.252085313060922e-07,
1922
+ "loss": 0.1226,
1923
+ "step": 20416
1924
+ },
1925
+ {
1926
+ "epoch": 1.52,
1927
+ "learning_rate": 5.235191637630661e-07,
1928
+ "loss": 0.1869,
1929
+ "step": 20480
1930
+ },
1931
+ {
1932
+ "epoch": 1.53,
1933
+ "learning_rate": 5.218297962200401e-07,
1934
+ "loss": 0.1255,
1935
+ "step": 20544
1936
+ },
1937
+ {
1938
+ "epoch": 1.53,
1939
+ "learning_rate": 5.201668250448737e-07,
1940
+ "loss": 0.1905,
1941
+ "step": 20608
1942
+ },
1943
+ {
1944
+ "epoch": 1.54,
1945
+ "learning_rate": 5.184774575018477e-07,
1946
+ "loss": 0.1708,
1947
+ "step": 20672
1948
+ },
1949
+ {
1950
+ "epoch": 1.54,
1951
+ "learning_rate": 5.167880899588216e-07,
1952
+ "loss": 0.163,
1953
+ "step": 20736
1954
+ },
1955
+ {
1956
+ "epoch": 1.55,
1957
+ "learning_rate": 5.150987224157955e-07,
1958
+ "loss": 0.1716,
1959
+ "step": 20800
1960
+ },
1961
+ {
1962
+ "epoch": 1.55,
1963
+ "learning_rate": 5.134093548727695e-07,
1964
+ "loss": 0.1274,
1965
+ "step": 20864
1966
+ },
1967
+ {
1968
+ "epoch": 1.56,
1969
+ "learning_rate": 5.117199873297435e-07,
1970
+ "loss": 0.1403,
1971
+ "step": 20928
1972
+ },
1973
+ {
1974
+ "epoch": 1.56,
1975
+ "learning_rate": 5.100306197867173e-07,
1976
+ "loss": 0.1995,
1977
+ "step": 20992
1978
+ },
1979
+ {
1980
+ "epoch": 1.57,
1981
+ "learning_rate": 5.083412522436913e-07,
1982
+ "loss": 0.1767,
1983
+ "step": 21056
1984
+ },
1985
+ {
1986
+ "epoch": 1.57,
1987
+ "learning_rate": 5.066518847006651e-07,
1988
+ "loss": 0.1903,
1989
+ "step": 21120
1990
+ },
1991
+ {
1992
+ "epoch": 1.58,
1993
+ "learning_rate": 5.049625171576391e-07,
1994
+ "loss": 0.142,
1995
+ "step": 21184
1996
+ },
1997
+ {
1998
+ "epoch": 1.58,
1999
+ "learning_rate": 5.03273149614613e-07,
2000
+ "loss": 0.1716,
2001
+ "step": 21248
2002
+ },
2003
+ {
2004
+ "epoch": 1.59,
2005
+ "learning_rate": 5.015837820715869e-07,
2006
+ "loss": 0.1786,
2007
+ "step": 21312
2008
+ },
2009
+ {
2010
+ "epoch": 1.59,
2011
+ "learning_rate": 4.998944145285609e-07,
2012
+ "loss": 0.2228,
2013
+ "step": 21376
2014
+ },
2015
+ {
2016
+ "epoch": 1.6,
2017
+ "learning_rate": 4.982050469855347e-07,
2018
+ "loss": 0.1619,
2019
+ "step": 21440
2020
+ },
2021
+ {
2022
+ "epoch": 1.6,
2023
+ "learning_rate": 4.965156794425087e-07,
2024
+ "loss": 0.1718,
2025
+ "step": 21504
2026
+ },
2027
+ {
2028
+ "epoch": 1.61,
2029
+ "learning_rate": 4.948263118994826e-07,
2030
+ "loss": 0.1151,
2031
+ "step": 21568
2032
+ },
2033
+ {
2034
+ "epoch": 1.61,
2035
+ "learning_rate": 4.931369443564565e-07,
2036
+ "loss": 0.1673,
2037
+ "step": 21632
2038
+ },
2039
+ {
2040
+ "epoch": 1.62,
2041
+ "learning_rate": 4.914475768134305e-07,
2042
+ "loss": 0.2063,
2043
+ "step": 21696
2044
+ },
2045
+ {
2046
+ "epoch": 1.62,
2047
+ "learning_rate": 4.897582092704043e-07,
2048
+ "loss": 0.1335,
2049
+ "step": 21760
2050
+ },
2051
+ {
2052
+ "epoch": 1.62,
2053
+ "learning_rate": 4.880688417273783e-07,
2054
+ "loss": 0.1217,
2055
+ "step": 21824
2056
+ },
2057
+ {
2058
+ "epoch": 1.63,
2059
+ "learning_rate": 4.863794741843523e-07,
2060
+ "loss": 0.2009,
2061
+ "step": 21888
2062
+ },
2063
+ {
2064
+ "epoch": 1.63,
2065
+ "learning_rate": 4.846901066413261e-07,
2066
+ "loss": 0.2075,
2067
+ "step": 21952
2068
+ },
2069
+ {
2070
+ "epoch": 1.64,
2071
+ "learning_rate": 4.830007390983e-07,
2072
+ "loss": 0.2211,
2073
+ "step": 22016
2074
+ },
2075
+ {
2076
+ "epoch": 1.64,
2077
+ "learning_rate": 4.813113715552739e-07,
2078
+ "loss": 0.1547,
2079
+ "step": 22080
2080
+ },
2081
+ {
2082
+ "epoch": 1.65,
2083
+ "learning_rate": 4.796220040122479e-07,
2084
+ "loss": 0.2471,
2085
+ "step": 22144
2086
+ },
2087
+ {
2088
+ "epoch": 1.65,
2089
+ "learning_rate": 4.779326364692219e-07,
2090
+ "loss": 0.131,
2091
+ "step": 22208
2092
+ },
2093
+ {
2094
+ "epoch": 1.66,
2095
+ "learning_rate": 4.762432689261957e-07,
2096
+ "loss": 0.1717,
2097
+ "step": 22272
2098
+ },
2099
+ {
2100
+ "epoch": 1.66,
2101
+ "learning_rate": 4.7455390138316963e-07,
2102
+ "loss": 0.1274,
2103
+ "step": 22336
2104
+ },
2105
+ {
2106
+ "epoch": 1.67,
2107
+ "learning_rate": 4.728645338401436e-07,
2108
+ "loss": 0.1374,
2109
+ "step": 22400
2110
+ },
2111
+ {
2112
+ "epoch": 1.67,
2113
+ "learning_rate": 4.711751662971175e-07,
2114
+ "loss": 0.1467,
2115
+ "step": 22464
2116
+ },
2117
+ {
2118
+ "epoch": 1.68,
2119
+ "learning_rate": 4.6948579875409136e-07,
2120
+ "loss": 0.2168,
2121
+ "step": 22528
2122
+ },
2123
+ {
2124
+ "epoch": 1.68,
2125
+ "learning_rate": 4.6779643121106533e-07,
2126
+ "loss": 0.1264,
2127
+ "step": 22592
2128
+ },
2129
+ {
2130
+ "epoch": 1.69,
2131
+ "learning_rate": 4.6610706366803924e-07,
2132
+ "loss": 0.1222,
2133
+ "step": 22656
2134
+ },
2135
+ {
2136
+ "epoch": 1.69,
2137
+ "learning_rate": 4.64444092492873e-07,
2138
+ "loss": 0.1715,
2139
+ "step": 22720
2140
+ },
2141
+ {
2142
+ "epoch": 1.7,
2143
+ "learning_rate": 4.6275472494984683e-07,
2144
+ "loss": 0.1394,
2145
+ "step": 22784
2146
+ },
2147
+ {
2148
+ "epoch": 1.7,
2149
+ "learning_rate": 4.610653574068208e-07,
2150
+ "loss": 0.1648,
2151
+ "step": 22848
2152
+ },
2153
+ {
2154
+ "epoch": 1.71,
2155
+ "learning_rate": 4.593759898637947e-07,
2156
+ "loss": 0.1854,
2157
+ "step": 22912
2158
+ },
2159
+ {
2160
+ "epoch": 1.71,
2161
+ "learning_rate": 4.5768662232076867e-07,
2162
+ "loss": 0.1098,
2163
+ "step": 22976
2164
+ },
2165
+ {
2166
+ "epoch": 1.72,
2167
+ "learning_rate": 4.5599725477774253e-07,
2168
+ "loss": 0.1099,
2169
+ "step": 23040
2170
+ },
2171
+ {
2172
+ "epoch": 1.72,
2173
+ "learning_rate": 4.543078872347165e-07,
2174
+ "loss": 0.1653,
2175
+ "step": 23104
2176
+ },
2177
+ {
2178
+ "epoch": 1.72,
2179
+ "learning_rate": 4.526185196916904e-07,
2180
+ "loss": 0.1917,
2181
+ "step": 23168
2182
+ },
2183
+ {
2184
+ "epoch": 1.73,
2185
+ "learning_rate": 4.5092915214866436e-07,
2186
+ "loss": 0.2046,
2187
+ "step": 23232
2188
+ },
2189
+ {
2190
+ "epoch": 1.73,
2191
+ "learning_rate": 4.492397846056382e-07,
2192
+ "loss": 0.1353,
2193
+ "step": 23296
2194
+ },
2195
+ {
2196
+ "epoch": 1.74,
2197
+ "learning_rate": 4.4755041706261213e-07,
2198
+ "loss": 0.152,
2199
+ "step": 23360
2200
+ },
2201
+ {
2202
+ "epoch": 1.74,
2203
+ "learning_rate": 4.458610495195861e-07,
2204
+ "loss": 0.178,
2205
+ "step": 23424
2206
+ },
2207
+ {
2208
+ "epoch": 1.75,
2209
+ "learning_rate": 4.4417168197656e-07,
2210
+ "loss": 0.1666,
2211
+ "step": 23488
2212
+ },
2213
+ {
2214
+ "epoch": 1.75,
2215
+ "learning_rate": 4.424823144335339e-07,
2216
+ "loss": 0.1452,
2217
+ "step": 23552
2218
+ },
2219
+ {
2220
+ "epoch": 1.76,
2221
+ "learning_rate": 4.407929468905078e-07,
2222
+ "loss": 0.1384,
2223
+ "step": 23616
2224
+ },
2225
+ {
2226
+ "epoch": 1.76,
2227
+ "learning_rate": 4.391035793474818e-07,
2228
+ "loss": 0.171,
2229
+ "step": 23680
2230
+ },
2231
+ {
2232
+ "epoch": 1.77,
2233
+ "learning_rate": 4.374142118044557e-07,
2234
+ "loss": 0.2005,
2235
+ "step": 23744
2236
+ },
2237
+ {
2238
+ "epoch": 1.77,
2239
+ "learning_rate": 4.3572484426142955e-07,
2240
+ "loss": 0.1804,
2241
+ "step": 23808
2242
+ },
2243
+ {
2244
+ "epoch": 1.78,
2245
+ "learning_rate": 4.340354767184035e-07,
2246
+ "loss": 0.1528,
2247
+ "step": 23872
2248
+ },
2249
+ {
2250
+ "epoch": 1.78,
2251
+ "learning_rate": 4.3234610917537743e-07,
2252
+ "loss": 0.1233,
2253
+ "step": 23936
2254
+ },
2255
+ {
2256
+ "epoch": 1.79,
2257
+ "learning_rate": 4.306567416323514e-07,
2258
+ "loss": 0.1556,
2259
+ "step": 24000
2260
+ },
2261
+ {
2262
+ "epoch": 1.79,
2263
+ "learning_rate": 4.2896737408932525e-07,
2264
+ "loss": 0.1198,
2265
+ "step": 24064
2266
+ },
2267
+ {
2268
+ "epoch": 1.8,
2269
+ "learning_rate": 4.272780065462992e-07,
2270
+ "loss": 0.2046,
2271
+ "step": 24128
2272
+ },
2273
+ {
2274
+ "epoch": 1.8,
2275
+ "learning_rate": 4.255886390032731e-07,
2276
+ "loss": 0.1274,
2277
+ "step": 24192
2278
+ },
2279
+ {
2280
+ "epoch": 1.81,
2281
+ "learning_rate": 4.238992714602471e-07,
2282
+ "loss": 0.2104,
2283
+ "step": 24256
2284
+ },
2285
+ {
2286
+ "epoch": 1.81,
2287
+ "learning_rate": 4.2220990391722094e-07,
2288
+ "loss": 0.1954,
2289
+ "step": 24320
2290
+ },
2291
+ {
2292
+ "epoch": 1.82,
2293
+ "learning_rate": 4.205205363741949e-07,
2294
+ "loss": 0.1826,
2295
+ "step": 24384
2296
+ },
2297
+ {
2298
+ "epoch": 1.82,
2299
+ "learning_rate": 4.188311688311688e-07,
2300
+ "loss": 0.1573,
2301
+ "step": 24448
2302
+ },
2303
+ {
2304
+ "epoch": 1.82,
2305
+ "learning_rate": 4.171418012881427e-07,
2306
+ "loss": 0.0717,
2307
+ "step": 24512
2308
+ },
2309
+ {
2310
+ "epoch": 1.83,
2311
+ "learning_rate": 4.1545243374511664e-07,
2312
+ "loss": 0.1276,
2313
+ "step": 24576
2314
+ },
2315
+ {
2316
+ "epoch": 1.83,
2317
+ "learning_rate": 4.1376306620209055e-07,
2318
+ "loss": 0.1869,
2319
+ "step": 24640
2320
+ },
2321
+ {
2322
+ "epoch": 1.84,
2323
+ "learning_rate": 4.120736986590645e-07,
2324
+ "loss": 0.1593,
2325
+ "step": 24704
2326
+ },
2327
+ {
2328
+ "epoch": 1.84,
2329
+ "learning_rate": 4.103843311160384e-07,
2330
+ "loss": 0.159,
2331
+ "step": 24768
2332
+ },
2333
+ {
2334
+ "epoch": 1.85,
2335
+ "learning_rate": 4.0869496357301233e-07,
2336
+ "loss": 0.1226,
2337
+ "step": 24832
2338
+ },
2339
+ {
2340
+ "epoch": 1.85,
2341
+ "learning_rate": 4.0700559602998624e-07,
2342
+ "loss": 0.1372,
2343
+ "step": 24896
2344
+ },
2345
+ {
2346
+ "epoch": 1.86,
2347
+ "learning_rate": 4.053162284869602e-07,
2348
+ "loss": 0.1391,
2349
+ "step": 24960
2350
+ },
2351
+ {
2352
+ "epoch": 1.86,
2353
+ "learning_rate": 4.036268609439341e-07,
2354
+ "loss": 0.1583,
2355
+ "step": 25024
2356
+ },
2357
+ {
2358
+ "epoch": 1.87,
2359
+ "learning_rate": 4.0193749340090797e-07,
2360
+ "loss": 0.1139,
2361
+ "step": 25088
2362
+ },
2363
+ {
2364
+ "epoch": 1.87,
2365
+ "learning_rate": 4.0024812585788193e-07,
2366
+ "loss": 0.151,
2367
+ "step": 25152
2368
+ },
2369
+ {
2370
+ "epoch": 1.88,
2371
+ "learning_rate": 3.9855875831485584e-07,
2372
+ "loss": 0.0999,
2373
+ "step": 25216
2374
+ },
2375
+ {
2376
+ "epoch": 1.88,
2377
+ "learning_rate": 3.968693907718298e-07,
2378
+ "loss": 0.1652,
2379
+ "step": 25280
2380
+ },
2381
+ {
2382
+ "epoch": 1.89,
2383
+ "learning_rate": 3.9518002322880366e-07,
2384
+ "loss": 0.2249,
2385
+ "step": 25344
2386
+ },
2387
+ {
2388
+ "epoch": 1.89,
2389
+ "learning_rate": 3.934906556857776e-07,
2390
+ "loss": 0.1898,
2391
+ "step": 25408
2392
+ },
2393
+ {
2394
+ "epoch": 1.9,
2395
+ "learning_rate": 3.9180128814275154e-07,
2396
+ "loss": 0.1548,
2397
+ "step": 25472
2398
+ },
2399
+ {
2400
+ "epoch": 1.9,
2401
+ "learning_rate": 3.901383169675853e-07,
2402
+ "loss": 0.1457,
2403
+ "step": 25536
2404
+ },
2405
+ {
2406
+ "epoch": 1.91,
2407
+ "learning_rate": 3.8844894942455913e-07,
2408
+ "loss": 0.1548,
2409
+ "step": 25600
2410
+ },
2411
+ {
2412
+ "epoch": 1.91,
2413
+ "learning_rate": 3.867595818815331e-07,
2414
+ "loss": 0.1829,
2415
+ "step": 25664
2416
+ },
2417
+ {
2418
+ "epoch": 1.92,
2419
+ "learning_rate": 3.85070214338507e-07,
2420
+ "loss": 0.1119,
2421
+ "step": 25728
2422
+ },
2423
+ {
2424
+ "epoch": 1.92,
2425
+ "learning_rate": 3.833808467954809e-07,
2426
+ "loss": 0.1695,
2427
+ "step": 25792
2428
+ },
2429
+ {
2430
+ "epoch": 1.92,
2431
+ "learning_rate": 3.816914792524548e-07,
2432
+ "loss": 0.1507,
2433
+ "step": 25856
2434
+ },
2435
+ {
2436
+ "epoch": 1.93,
2437
+ "learning_rate": 3.8000211170942874e-07,
2438
+ "loss": 0.1416,
2439
+ "step": 25920
2440
+ },
2441
+ {
2442
+ "epoch": 1.93,
2443
+ "learning_rate": 3.783127441664027e-07,
2444
+ "loss": 0.0902,
2445
+ "step": 25984
2446
+ },
2447
+ {
2448
+ "epoch": 1.94,
2449
+ "learning_rate": 3.766233766233766e-07,
2450
+ "loss": 0.1472,
2451
+ "step": 26048
2452
+ },
2453
+ {
2454
+ "epoch": 1.94,
2455
+ "learning_rate": 3.749340090803505e-07,
2456
+ "loss": 0.1624,
2457
+ "step": 26112
2458
+ },
2459
+ {
2460
+ "epoch": 1.95,
2461
+ "learning_rate": 3.7324464153732443e-07,
2462
+ "loss": 0.1903,
2463
+ "step": 26176
2464
+ },
2465
+ {
2466
+ "epoch": 1.95,
2467
+ "learning_rate": 3.715552739942984e-07,
2468
+ "loss": 0.1224,
2469
+ "step": 26240
2470
+ },
2471
+ {
2472
+ "epoch": 1.96,
2473
+ "learning_rate": 3.698659064512723e-07,
2474
+ "loss": 0.16,
2475
+ "step": 26304
2476
+ },
2477
+ {
2478
+ "epoch": 1.96,
2479
+ "learning_rate": 3.6817653890824616e-07,
2480
+ "loss": 0.2316,
2481
+ "step": 26368
2482
+ },
2483
+ {
2484
+ "epoch": 1.97,
2485
+ "learning_rate": 3.664871713652201e-07,
2486
+ "loss": 0.2141,
2487
+ "step": 26432
2488
+ },
2489
+ {
2490
+ "epoch": 1.97,
2491
+ "learning_rate": 3.6479780382219403e-07,
2492
+ "loss": 0.1727,
2493
+ "step": 26496
2494
+ },
2495
+ {
2496
+ "epoch": 1.98,
2497
+ "learning_rate": 3.63108436279168e-07,
2498
+ "loss": 0.232,
2499
+ "step": 26560
2500
+ },
2501
+ {
2502
+ "epoch": 1.98,
2503
+ "learning_rate": 3.6141906873614185e-07,
2504
+ "loss": 0.1381,
2505
+ "step": 26624
2506
+ },
2507
+ {
2508
+ "epoch": 1.99,
2509
+ "learning_rate": 3.597297011931158e-07,
2510
+ "loss": 0.142,
2511
+ "step": 26688
2512
+ },
2513
+ {
2514
+ "epoch": 1.99,
2515
+ "learning_rate": 3.580403336500897e-07,
2516
+ "loss": 0.1338,
2517
+ "step": 26752
2518
+ },
2519
+ {
2520
+ "epoch": 2.0,
2521
+ "learning_rate": 3.563509661070637e-07,
2522
+ "loss": 0.1643,
2523
+ "step": 26816
2524
+ }
2525
+ ],
2526
+ "logging_steps": 64,
2527
+ "max_steps": 40302,
2528
+ "num_input_tokens_seen": 0,
2529
+ "num_train_epochs": 3,
2530
+ "save_steps": 500,
2531
+ "total_flos": 9.128398450065408e+19,
2532
+ "train_batch_size": 1,
2533
+ "trial_name": null,
2534
+ "trial_params": null
2535
+ }
checkpoint-26868/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78386e842565101e214de6fd490f11cca7ee20779a9fedba99952f8538bae8ab
3
+ size 4920