Hugging Face's logo

s3nh
/
mamba-1.4b_dolly_instruction_polish

Transformers
PyTorch
TensorBoard
mamba
text-generation-inference
Model card Files Metrics Community
s3nh commited on
Commit
753f4fa
·
1 Parent(s): 6d7d022

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. config.json +16 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. runs/Dec23_10-26-57_nilpzfd9qz/events.out.tfevents.1703327218.nilpzfd9qz.1101.0 +3 -0
  6. runs/Dec23_19-31-17_n6e878fh75/events.out.tfevents.1703359877.n6e878fh75.696.0 +3 -0
  7. runs/Dec23_19-33-04_n6e878fh75/events.out.tfevents.1703360003.n6e878fh75.1086.0 +3 -0
  8. runs/Dec24_10-17-37_nqp3dbtvv8/events.out.tfevents.1703413076.nqp3dbtvv8.867.0 +3 -0
  9. scheduler.pt +3 -0
  10. special_tokens_map.json +30 -0
  11. tokenizer.json +0 -0
  12. tokenizer_config.json +220 -0
  13. trainer_state.json +2361 -0
  14. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "d_model": 2048,
4
+ "eos_token_id": 0,
5
+ "fused_add_norm": true,
6
+ "model_type": "mamba",
7
+ "n_layer": 48,
8
+ "pad_token_id": 50277,
9
+ "pad_vocab_size_multiple": 8,
10
+ "residual_in_fp32": true,
11
+ "rms_norm": true,
12
+ "tie_word_embeddings": false,
13
+ "transformers_version": "4.37.0.dev0",
14
+ "use_cache": false,
15
+ "vocab_size": 50280
16
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89d2162940de44abc7bc21fa8a50617ae040aed52857f6ba3c213b7e8d72db61
3
+ size 2751975898
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60662a7568e580c0038ccaef592cc9198797cd58fba14db20e677af89b2925cc
3
+ size 2957372562
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f14dccb5e034d9e607ba93b28001555145746872876643eb46ec30207534820
3
+ size 14244
runs/Dec23_10-26-57_nilpzfd9qz/events.out.tfevents.1703327218.nilpzfd9qz.1101.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdd2d7206603d8563b1f884142472a93bcce95157b7beffb23d4b4cdb5cef87e
3
+ size 27856
runs/Dec23_19-31-17_n6e878fh75/events.out.tfevents.1703359877.n6e878fh75.696.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7358c548ed25875bb673fe39f75cbde9133024eaba80ce03f53a14451354bdc
3
+ size 40
runs/Dec23_19-33-04_n6e878fh75/events.out.tfevents.1703360003.n6e878fh75.1086.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56038b04eaff6df24e0a384ef49bf42220fe4854e210057a75ee7ad77106a08f
3
+ size 26737
runs/Dec24_10-17-37_nqp3dbtvv8/events.out.tfevents.1703413076.nqp3dbtvv8.867.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca0275c2870c50044a25fb249e3b58d3eb630bd5ee86473894d6bb20dd839c26
3
+ size 29586
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3df0f03622770e8b51b49c8af4547307b56146e0972a11d43920a4b3109f047
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|endoftext|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<|padding|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "50254": {
21
+ "content": " ",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": false
27
+ },
28
+ "50255": {
29
+ "content": " ",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": false
35
+ },
36
+ "50256": {
37
+ "content": " ",
38
+ "lstrip": false,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": false
43
+ },
44
+ "50257": {
45
+ "content": " ",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": false
51
+ },
52
+ "50258": {
53
+ "content": " ",
54
+ "lstrip": false,
55
+ "normalized": true,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": false
59
+ },
60
+ "50259": {
61
+ "content": " ",
62
+ "lstrip": false,
63
+ "normalized": true,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": false
67
+ },
68
+ "50260": {
69
+ "content": " ",
70
+ "lstrip": false,
71
+ "normalized": true,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": false
75
+ },
76
+ "50261": {
77
+ "content": " ",
78
+ "lstrip": false,
79
+ "normalized": true,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": false
83
+ },
84
+ "50262": {
85
+ "content": " ",
86
+ "lstrip": false,
87
+ "normalized": true,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": false
91
+ },
92
+ "50263": {
93
+ "content": " ",
94
+ "lstrip": false,
95
+ "normalized": true,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": false
99
+ },
100
+ "50264": {
101
+ "content": " ",
102
+ "lstrip": false,
103
+ "normalized": true,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": false
107
+ },
108
+ "50265": {
109
+ "content": " ",
110
+ "lstrip": false,
111
+ "normalized": true,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": false
115
+ },
116
+ "50266": {
117
+ "content": " ",
118
+ "lstrip": false,
119
+ "normalized": true,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "50267": {
125
+ "content": " ",
126
+ "lstrip": false,
127
+ "normalized": true,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "50268": {
133
+ "content": " ",
134
+ "lstrip": false,
135
+ "normalized": true,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "50269": {
141
+ "content": " ",
142
+ "lstrip": false,
143
+ "normalized": true,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "50270": {
149
+ "content": " ",
150
+ "lstrip": false,
151
+ "normalized": true,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "50271": {
157
+ "content": " ",
158
+ "lstrip": false,
159
+ "normalized": true,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "50272": {
165
+ "content": " ",
166
+ "lstrip": false,
167
+ "normalized": true,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "50273": {
173
+ "content": " ",
174
+ "lstrip": false,
175
+ "normalized": true,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "50274": {
181
+ "content": " ",
182
+ "lstrip": false,
183
+ "normalized": true,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "50275": {
189
+ "content": " ",
190
+ "lstrip": false,
191
+ "normalized": true,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "50276": {
197
+ "content": " ",
198
+ "lstrip": false,
199
+ "normalized": true,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "50277": {
205
+ "content": "[PAD]",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": true
211
+ }
212
+ },
213
+ "bos_token": "<|endoftext|>",
214
+ "clean_up_tokenization_spaces": true,
215
+ "eos_token": "<|endoftext|>",
216
+ "model_max_length": 1000000000000000019884624838656,
217
+ "pad_token": "[PAD]",
218
+ "tokenizer_class": "GPTNeoXTokenizer",
219
+ "unk_token": "<|endoftext|>"
220
+ }
trainer_state.json ADDED
@@ -0,0 +1,2361 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 6.875501300074919,
5
+ "eval_steps": 500,
6
+ "global_step": 39000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "learning_rate": 4.9999514451748454e-05,
14
+ "loss": 1.2252,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.04,
19
+ "learning_rate": 4.999783603756215e-05,
20
+ "loss": 1.1874,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.05,
25
+ "learning_rate": 4.999495884348801e-05,
26
+ "loss": 1.1343,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.07,
31
+ "learning_rate": 4.9990883007502865e-05,
32
+ "loss": 1.1173,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 0.09,
37
+ "learning_rate": 4.998560872506484e-05,
38
+ "loss": 1.1058,
39
+ "step": 500
40
+ },
41
+ {
42
+ "epoch": 0.11,
43
+ "learning_rate": 4.997913624910397e-05,
44
+ "loss": 1.0854,
45
+ "step": 600
46
+ },
47
+ {
48
+ "epoch": 0.12,
49
+ "learning_rate": 4.9971465890010096e-05,
50
+ "loss": 1.0827,
51
+ "step": 700
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "learning_rate": 4.996259801561794e-05,
56
+ "loss": 1.075,
57
+ "step": 800
58
+ },
59
+ {
60
+ "epoch": 0.16,
61
+ "learning_rate": 4.995253305118949e-05,
62
+ "loss": 1.0566,
63
+ "step": 900
64
+ },
65
+ {
66
+ "epoch": 0.18,
67
+ "learning_rate": 4.994127147939361e-05,
68
+ "loss": 1.0469,
69
+ "step": 1000
70
+ },
71
+ {
72
+ "epoch": 0.19,
73
+ "learning_rate": 4.9928813840282865e-05,
74
+ "loss": 1.0765,
75
+ "step": 1100
76
+ },
77
+ {
78
+ "epoch": 0.21,
79
+ "learning_rate": 4.9915160731267663e-05,
80
+ "loss": 1.0569,
81
+ "step": 1200
82
+ },
83
+ {
84
+ "epoch": 0.23,
85
+ "learning_rate": 4.990031280708757e-05,
86
+ "loss": 1.0188,
87
+ "step": 1300
88
+ },
89
+ {
90
+ "epoch": 0.25,
91
+ "learning_rate": 4.988427077977994e-05,
92
+ "loss": 1.0402,
93
+ "step": 1400
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "learning_rate": 4.986703541864576e-05,
98
+ "loss": 1.0333,
99
+ "step": 1500
100
+ },
101
+ {
102
+ "epoch": 0.28,
103
+ "learning_rate": 4.9848607550212716e-05,
104
+ "loss": 1.0394,
105
+ "step": 1600
106
+ },
107
+ {
108
+ "epoch": 0.3,
109
+ "learning_rate": 4.982898805819565e-05,
110
+ "loss": 0.9886,
111
+ "step": 1700
112
+ },
113
+ {
114
+ "epoch": 0.32,
115
+ "learning_rate": 4.980817788345409e-05,
116
+ "loss": 0.9914,
117
+ "step": 1800
118
+ },
119
+ {
120
+ "epoch": 0.33,
121
+ "learning_rate": 4.9786178023947174e-05,
122
+ "loss": 1.0033,
123
+ "step": 1900
124
+ },
125
+ {
126
+ "epoch": 0.35,
127
+ "learning_rate": 4.976298953468578e-05,
128
+ "loss": 1.0169,
129
+ "step": 2000
130
+ },
131
+ {
132
+ "epoch": 0.37,
133
+ "learning_rate": 4.9738613527681964e-05,
134
+ "loss": 0.9913,
135
+ "step": 2100
136
+ },
137
+ {
138
+ "epoch": 0.39,
139
+ "learning_rate": 4.971305117189556e-05,
140
+ "loss": 1.0201,
141
+ "step": 2200
142
+ },
143
+ {
144
+ "epoch": 0.41,
145
+ "learning_rate": 4.968630369317823e-05,
146
+ "loss": 0.9812,
147
+ "step": 2300
148
+ },
149
+ {
150
+ "epoch": 0.42,
151
+ "learning_rate": 4.965837237421455e-05,
152
+ "loss": 0.9856,
153
+ "step": 2400
154
+ },
155
+ {
156
+ "epoch": 0.44,
157
+ "learning_rate": 4.962925855446062e-05,
158
+ "loss": 0.9902,
159
+ "step": 2500
160
+ },
161
+ {
162
+ "epoch": 0.46,
163
+ "learning_rate": 4.9598963630079745e-05,
164
+ "loss": 0.9933,
165
+ "step": 2600
166
+ },
167
+ {
168
+ "epoch": 0.48,
169
+ "learning_rate": 4.956748905387551e-05,
170
+ "loss": 0.9731,
171
+ "step": 2700
172
+ },
173
+ {
174
+ "epoch": 0.49,
175
+ "learning_rate": 4.9534836335222136e-05,
176
+ "loss": 0.9774,
177
+ "step": 2800
178
+ },
179
+ {
180
+ "epoch": 0.51,
181
+ "learning_rate": 4.9501007039992056e-05,
182
+ "loss": 0.9875,
183
+ "step": 2900
184
+ },
185
+ {
186
+ "epoch": 0.53,
187
+ "learning_rate": 4.9466002790480836e-05,
188
+ "loss": 0.9679,
189
+ "step": 3000
190
+ },
191
+ {
192
+ "epoch": 0.55,
193
+ "learning_rate": 4.942982526532939e-05,
194
+ "loss": 0.9698,
195
+ "step": 3100
196
+ },
197
+ {
198
+ "epoch": 0.56,
199
+ "learning_rate": 4.939247619944348e-05,
200
+ "loss": 0.9421,
201
+ "step": 3200
202
+ },
203
+ {
204
+ "epoch": 0.58,
205
+ "learning_rate": 4.93539573839105e-05,
206
+ "loss": 0.9811,
207
+ "step": 3300
208
+ },
209
+ {
210
+ "epoch": 0.6,
211
+ "learning_rate": 4.9314270665913574e-05,
212
+ "loss": 0.971,
213
+ "step": 3400
214
+ },
215
+ {
216
+ "epoch": 0.62,
217
+ "learning_rate": 4.927341794864304e-05,
218
+ "loss": 0.9824,
219
+ "step": 3500
220
+ },
221
+ {
222
+ "epoch": 0.63,
223
+ "learning_rate": 4.923140119120509e-05,
224
+ "loss": 0.9692,
225
+ "step": 3600
226
+ },
227
+ {
228
+ "epoch": 0.65,
229
+ "learning_rate": 4.91882224085279e-05,
230
+ "loss": 0.9309,
231
+ "step": 3700
232
+ },
233
+ {
234
+ "epoch": 0.67,
235
+ "learning_rate": 4.9143883671264926e-05,
236
+ "loss": 0.9732,
237
+ "step": 3800
238
+ },
239
+ {
240
+ "epoch": 0.69,
241
+ "learning_rate": 4.90983871056957e-05,
242
+ "loss": 0.9353,
243
+ "step": 3900
244
+ },
245
+ {
246
+ "epoch": 0.71,
247
+ "learning_rate": 4.905173489362378e-05,
248
+ "loss": 0.9745,
249
+ "step": 4000
250
+ },
251
+ {
252
+ "epoch": 0.72,
253
+ "learning_rate": 4.9003929272272154e-05,
254
+ "loss": 0.9583,
255
+ "step": 4100
256
+ },
257
+ {
258
+ "epoch": 0.74,
259
+ "learning_rate": 4.895497253417596e-05,
260
+ "loss": 0.9289,
261
+ "step": 4200
262
+ },
263
+ {
264
+ "epoch": 0.76,
265
+ "learning_rate": 4.890486702707253e-05,
266
+ "loss": 0.932,
267
+ "step": 4300
268
+ },
269
+ {
270
+ "epoch": 0.78,
271
+ "learning_rate": 4.8853615153788824e-05,
272
+ "loss": 0.974,
273
+ "step": 4400
274
+ },
275
+ {
276
+ "epoch": 0.79,
277
+ "learning_rate": 4.880121937212619e-05,
278
+ "loss": 0.9636,
279
+ "step": 4500
280
+ },
281
+ {
282
+ "epoch": 0.81,
283
+ "learning_rate": 4.874768219474246e-05,
284
+ "loss": 0.9213,
285
+ "step": 4600
286
+ },
287
+ {
288
+ "epoch": 0.83,
289
+ "learning_rate": 4.8693006189031556e-05,
290
+ "loss": 0.9523,
291
+ "step": 4700
292
+ },
293
+ {
294
+ "epoch": 0.85,
295
+ "learning_rate": 4.863719397700025e-05,
296
+ "loss": 0.9133,
297
+ "step": 4800
298
+ },
299
+ {
300
+ "epoch": 0.86,
301
+ "learning_rate": 4.858024823514253e-05,
302
+ "loss": 0.9305,
303
+ "step": 4900
304
+ },
305
+ {
306
+ "epoch": 0.88,
307
+ "learning_rate": 4.852217169431116e-05,
308
+ "loss": 0.9098,
309
+ "step": 5000
310
+ },
311
+ {
312
+ "epoch": 0.9,
313
+ "learning_rate": 4.846296713958678e-05,
314
+ "loss": 0.9354,
315
+ "step": 5100
316
+ },
317
+ {
318
+ "epoch": 0.92,
319
+ "learning_rate": 4.840263741014432e-05,
320
+ "loss": 0.921,
321
+ "step": 5200
322
+ },
323
+ {
324
+ "epoch": 0.93,
325
+ "learning_rate": 4.834118539911686e-05,
326
+ "loss": 0.9017,
327
+ "step": 5300
328
+ },
329
+ {
330
+ "epoch": 0.95,
331
+ "learning_rate": 4.827861405345689e-05,
332
+ "loss": 0.9236,
333
+ "step": 5400
334
+ },
335
+ {
336
+ "epoch": 0.97,
337
+ "learning_rate": 4.821492637379496e-05,
338
+ "loss": 0.8932,
339
+ "step": 5500
340
+ },
341
+ {
342
+ "epoch": 0.99,
343
+ "learning_rate": 4.81501254142958e-05,
344
+ "loss": 0.8994,
345
+ "step": 5600
346
+ },
347
+ {
348
+ "epoch": 1.0,
349
+ "learning_rate": 4.8084214282511896e-05,
350
+ "loss": 0.89,
351
+ "step": 5700
352
+ },
353
+ {
354
+ "epoch": 1.02,
355
+ "learning_rate": 4.80171961392344e-05,
356
+ "loss": 0.7001,
357
+ "step": 5800
358
+ },
359
+ {
360
+ "epoch": 1.04,
361
+ "learning_rate": 4.794907419834159e-05,
362
+ "loss": 0.6979,
363
+ "step": 5900
364
+ },
365
+ {
366
+ "epoch": 1.06,
367
+ "learning_rate": 4.787985172664474e-05,
368
+ "loss": 0.6947,
369
+ "step": 6000
370
+ },
371
+ {
372
+ "epoch": 1.08,
373
+ "learning_rate": 4.780953204373146e-05,
374
+ "loss": 0.6908,
375
+ "step": 6100
376
+ },
377
+ {
378
+ "epoch": 1.09,
379
+ "learning_rate": 4.773811852180652e-05,
380
+ "loss": 0.6758,
381
+ "step": 6200
382
+ },
383
+ {
384
+ "epoch": 1.11,
385
+ "learning_rate": 4.76656145855301e-05,
386
+ "loss": 0.675,
387
+ "step": 6300
388
+ },
389
+ {
390
+ "epoch": 1.13,
391
+ "learning_rate": 4.759202371185358e-05,
392
+ "loss": 0.6911,
393
+ "step": 6400
394
+ },
395
+ {
396
+ "epoch": 1.15,
397
+ "learning_rate": 4.751734942985281e-05,
398
+ "loss": 0.6938,
399
+ "step": 6500
400
+ },
401
+ {
402
+ "epoch": 1.16,
403
+ "learning_rate": 4.744159532055884e-05,
404
+ "loss": 0.6903,
405
+ "step": 6600
406
+ },
407
+ {
408
+ "epoch": 1.18,
409
+ "learning_rate": 4.7364765016786237e-05,
410
+ "loss": 0.6888,
411
+ "step": 6700
412
+ },
413
+ {
414
+ "epoch": 1.2,
415
+ "learning_rate": 4.728686220295885e-05,
416
+ "loss": 0.6845,
417
+ "step": 6800
418
+ },
419
+ {
420
+ "epoch": 1.22,
421
+ "learning_rate": 4.7207890614933084e-05,
422
+ "loss": 0.6822,
423
+ "step": 6900
424
+ },
425
+ {
426
+ "epoch": 1.23,
427
+ "learning_rate": 4.71278540398188e-05,
428
+ "loss": 0.6844,
429
+ "step": 7000
430
+ },
431
+ {
432
+ "epoch": 1.25,
433
+ "learning_rate": 4.704675631579771e-05,
434
+ "loss": 0.6947,
435
+ "step": 7100
436
+ },
437
+ {
438
+ "epoch": 1.27,
439
+ "learning_rate": 4.696460133193926e-05,
440
+ "loss": 0.6961,
441
+ "step": 7200
442
+ },
443
+ {
444
+ "epoch": 1.29,
445
+ "learning_rate": 4.688139302801414e-05,
446
+ "loss": 0.6947,
447
+ "step": 7300
448
+ },
449
+ {
450
+ "epoch": 1.3,
451
+ "learning_rate": 4.6797135394305426e-05,
452
+ "loss": 0.7161,
453
+ "step": 7400
454
+ },
455
+ {
456
+ "epoch": 1.32,
457
+ "learning_rate": 4.671183247141712e-05,
458
+ "loss": 0.7073,
459
+ "step": 7500
460
+ },
461
+ {
462
+ "epoch": 1.34,
463
+ "learning_rate": 4.6625488350080435e-05,
464
+ "loss": 0.7095,
465
+ "step": 7600
466
+ },
467
+ {
468
+ "epoch": 1.36,
469
+ "learning_rate": 4.6538107170957636e-05,
470
+ "loss": 0.692,
471
+ "step": 7700
472
+ },
473
+ {
474
+ "epoch": 1.37,
475
+ "learning_rate": 4.6449693124443435e-05,
476
+ "loss": 0.6974,
477
+ "step": 7800
478
+ },
479
+ {
480
+ "epoch": 1.39,
481
+ "learning_rate": 4.636025045046407e-05,
482
+ "loss": 0.6813,
483
+ "step": 7900
484
+ },
485
+ {
486
+ "epoch": 1.41,
487
+ "learning_rate": 4.626978343827395e-05,
488
+ "loss": 0.6981,
489
+ "step": 8000
490
+ },
491
+ {
492
+ "epoch": 1.43,
493
+ "learning_rate": 4.617829642625e-05,
494
+ "loss": 0.709,
495
+ "step": 8100
496
+ },
497
+ {
498
+ "epoch": 1.45,
499
+ "learning_rate": 4.608579380168356e-05,
500
+ "loss": 0.6806,
501
+ "step": 8200
502
+ },
503
+ {
504
+ "epoch": 1.46,
505
+ "learning_rate": 4.599228000057006e-05,
506
+ "loss": 0.7207,
507
+ "step": 8300
508
+ },
509
+ {
510
+ "epoch": 1.48,
511
+ "learning_rate": 4.589775950739622e-05,
512
+ "loss": 0.6917,
513
+ "step": 8400
514
+ },
515
+ {
516
+ "epoch": 1.5,
517
+ "learning_rate": 4.580223685492504e-05,
518
+ "loss": 0.6964,
519
+ "step": 8500
520
+ },
521
+ {
522
+ "epoch": 1.52,
523
+ "learning_rate": 4.5705716623978404e-05,
524
+ "loss": 0.7057,
525
+ "step": 8600
526
+ },
527
+ {
528
+ "epoch": 1.53,
529
+ "learning_rate": 4.560820344321745e-05,
530
+ "loss": 0.6869,
531
+ "step": 8700
532
+ },
533
+ {
534
+ "epoch": 1.55,
535
+ "learning_rate": 4.550970198892052e-05,
536
+ "loss": 0.706,
537
+ "step": 8800
538
+ },
539
+ {
540
+ "epoch": 1.57,
541
+ "learning_rate": 4.541021698475899e-05,
542
+ "loss": 0.7045,
543
+ "step": 8900
544
+ },
545
+ {
546
+ "epoch": 1.59,
547
+ "learning_rate": 4.53097532015707e-05,
548
+ "loss": 0.7009,
549
+ "step": 9000
550
+ },
551
+ {
552
+ "epoch": 1.6,
553
+ "learning_rate": 4.52083154571312e-05,
554
+ "loss": 0.696,
555
+ "step": 9100
556
+ },
557
+ {
558
+ "epoch": 1.62,
559
+ "learning_rate": 4.510590861592264e-05,
560
+ "loss": 0.7188,
561
+ "step": 9200
562
+ },
563
+ {
564
+ "epoch": 1.64,
565
+ "learning_rate": 4.500253758890058e-05,
566
+ "loss": 0.7196,
567
+ "step": 9300
568
+ },
569
+ {
570
+ "epoch": 1.66,
571
+ "learning_rate": 4.489820733325845e-05,
572
+ "loss": 0.682,
573
+ "step": 9400
574
+ },
575
+ {
576
+ "epoch": 1.67,
577
+ "learning_rate": 4.4792922852189775e-05,
578
+ "loss": 0.6848,
579
+ "step": 9500
580
+ },
581
+ {
582
+ "epoch": 1.69,
583
+ "learning_rate": 4.4686689194648344e-05,
584
+ "loss": 0.6907,
585
+ "step": 9600
586
+ },
587
+ {
588
+ "epoch": 1.71,
589
+ "learning_rate": 4.4579511455106004e-05,
590
+ "loss": 0.7091,
591
+ "step": 9700
592
+ },
593
+ {
594
+ "epoch": 1.73,
595
+ "learning_rate": 4.447139477330838e-05,
596
+ "loss": 0.6947,
597
+ "step": 9800
598
+ },
599
+ {
600
+ "epoch": 1.75,
601
+ "learning_rate": 4.436234433402842e-05,
602
+ "loss": 0.6908,
603
+ "step": 9900
604
+ },
605
+ {
606
+ "epoch": 1.76,
607
+ "learning_rate": 4.42523653668177e-05,
608
+ "loss": 0.6968,
609
+ "step": 10000
610
+ },
611
+ {
612
+ "epoch": 1.78,
613
+ "learning_rate": 4.414146314575571e-05,
614
+ "loss": 0.7141,
615
+ "step": 10100
616
+ },
617
+ {
618
+ "epoch": 1.8,
619
+ "learning_rate": 4.402964298919688e-05,
620
+ "loss": 0.7172,
621
+ "step": 10200
622
+ },
623
+ {
624
+ "epoch": 1.82,
625
+ "learning_rate": 4.3916910259515574e-05,
626
+ "loss": 0.7058,
627
+ "step": 10300
628
+ },
629
+ {
630
+ "epoch": 1.83,
631
+ "learning_rate": 4.3803270362848876e-05,
632
+ "loss": 0.6794,
633
+ "step": 10400
634
+ },
635
+ {
636
+ "epoch": 1.85,
637
+ "learning_rate": 4.368872874883742e-05,
638
+ "loss": 0.6936,
639
+ "step": 10500
640
+ },
641
+ {
642
+ "epoch": 1.87,
643
+ "learning_rate": 4.357329091036397e-05,
644
+ "loss": 0.7033,
645
+ "step": 10600
646
+ },
647
+ {
648
+ "epoch": 1.89,
649
+ "learning_rate": 4.345696238329009e-05,
650
+ "loss": 0.6627,
651
+ "step": 10700
652
+ },
653
+ {
654
+ "epoch": 1.9,
655
+ "learning_rate": 4.333974874619058e-05,
656
+ "loss": 0.6819,
657
+ "step": 10800
658
+ },
659
+ {
660
+ "epoch": 1.92,
661
+ "learning_rate": 4.322165562008602e-05,
662
+ "loss": 0.6926,
663
+ "step": 10900
664
+ },
665
+ {
666
+ "epoch": 1.94,
667
+ "learning_rate": 4.310268866817318e-05,
668
+ "loss": 0.6857,
669
+ "step": 11000
670
+ },
671
+ {
672
+ "epoch": 1.96,
673
+ "learning_rate": 4.2982853595553476e-05,
674
+ "loss": 0.6833,
675
+ "step": 11100
676
+ },
677
+ {
678
+ "epoch": 1.97,
679
+ "learning_rate": 4.286215614895932e-05,
680
+ "loss": 0.6819,
681
+ "step": 11200
682
+ },
683
+ {
684
+ "epoch": 1.99,
685
+ "learning_rate": 4.274060211647858e-05,
686
+ "loss": 0.6981,
687
+ "step": 11300
688
+ },
689
+ {
690
+ "epoch": 2.01,
691
+ "learning_rate": 4.261819732727699e-05,
692
+ "loss": 0.5492,
693
+ "step": 11400
694
+ },
695
+ {
696
+ "epoch": 2.03,
697
+ "learning_rate": 4.249494765131864e-05,
698
+ "loss": 0.4547,
699
+ "step": 11500
700
+ },
701
+ {
702
+ "epoch": 2.04,
703
+ "learning_rate": 4.237085899908443e-05,
704
+ "loss": 0.4511,
705
+ "step": 11600
706
+ },
707
+ {
708
+ "epoch": 2.06,
709
+ "learning_rate": 4.224593732128867e-05,
710
+ "loss": 0.4488,
711
+ "step": 11700
712
+ },
713
+ {
714
+ "epoch": 2.08,
715
+ "learning_rate": 4.212018860859371e-05,
716
+ "loss": 0.4548,
717
+ "step": 11800
718
+ },
719
+ {
720
+ "epoch": 2.1,
721
+ "learning_rate": 4.199361889132262e-05,
722
+ "loss": 0.4568,
723
+ "step": 11900
724
+ },
725
+ {
726
+ "epoch": 2.12,
727
+ "learning_rate": 4.186623423917005e-05,
728
+ "loss": 0.4442,
729
+ "step": 12000
730
+ },
731
+ {
732
+ "epoch": 2.13,
733
+ "learning_rate": 4.1738040760911125e-05,
734
+ "loss": 0.4616,
735
+ "step": 12100
736
+ },
737
+ {
738
+ "epoch": 2.15,
739
+ "learning_rate": 4.1609044604108494e-05,
740
+ "loss": 0.4655,
741
+ "step": 12200
742
+ },
743
+ {
744
+ "epoch": 2.17,
745
+ "learning_rate": 4.147925195481758e-05,
746
+ "loss": 0.4497,
747
+ "step": 12300
748
+ },
749
+ {
750
+ "epoch": 2.19,
751
+ "learning_rate": 4.1348669037289814e-05,
752
+ "loss": 0.4613,
753
+ "step": 12400
754
+ },
755
+ {
756
+ "epoch": 2.2,
757
+ "learning_rate": 4.121730211367427e-05,
758
+ "loss": 0.4434,
759
+ "step": 12500
760
+ },
761
+ {
762
+ "epoch": 2.22,
763
+ "learning_rate": 4.1085157483717274e-05,
764
+ "loss": 0.472,
765
+ "step": 12600
766
+ },
767
+ {
768
+ "epoch": 2.24,
769
+ "learning_rate": 4.095224148446034e-05,
770
+ "loss": 0.4497,
771
+ "step": 12700
772
+ },
773
+ {
774
+ "epoch": 2.26,
775
+ "learning_rate": 4.0818560489936264e-05,
776
+ "loss": 0.4668,
777
+ "step": 12800
778
+ },
779
+ {
780
+ "epoch": 2.27,
781
+ "learning_rate": 4.0684120910863446e-05,
782
+ "loss": 0.4875,
783
+ "step": 12900
784
+ },
785
+ {
786
+ "epoch": 2.29,
787
+ "learning_rate": 4.054892919433849e-05,
788
+ "loss": 0.4858,
789
+ "step": 13000
790
+ },
791
+ {
792
+ "epoch": 2.31,
793
+ "learning_rate": 4.041299182352698e-05,
794
+ "loss": 0.4548,
795
+ "step": 13100
796
+ },
797
+ {
798
+ "epoch": 2.33,
799
+ "learning_rate": 4.027631531735265e-05,
800
+ "loss": 0.4834,
801
+ "step": 13200
802
+ },
803
+ {
804
+ "epoch": 2.34,
805
+ "learning_rate": 4.0138906230184706e-05,
806
+ "loss": 0.4618,
807
+ "step": 13300
808
+ },
809
+ {
810
+ "epoch": 2.36,
811
+ "learning_rate": 4.00007711515235e-05,
812
+ "loss": 0.4617,
813
+ "step": 13400
814
+ },
815
+ {
816
+ "epoch": 2.38,
817
+ "learning_rate": 3.986191670568462e-05,
818
+ "loss": 0.4596,
819
+ "step": 13500
820
+ },
821
+ {
822
+ "epoch": 2.4,
823
+ "learning_rate": 3.972234955148112e-05,
824
+ "loss": 0.4636,
825
+ "step": 13600
826
+ },
827
+ {
828
+ "epoch": 2.42,
829
+ "learning_rate": 3.9582076381904214e-05,
830
+ "loss": 0.4535,
831
+ "step": 13700
832
+ },
833
+ {
834
+ "epoch": 2.43,
835
+ "learning_rate": 3.944110392380236e-05,
836
+ "loss": 0.4651,
837
+ "step": 13800
838
+ },
839
+ {
840
+ "epoch": 2.45,
841
+ "learning_rate": 3.929943893755863e-05,
842
+ "loss": 0.4485,
843
+ "step": 13900
844
+ },
845
+ {
846
+ "epoch": 2.47,
847
+ "learning_rate": 3.915708821676651e-05,
848
+ "loss": 0.4611,
849
+ "step": 14000
850
+ },
851
+ {
852
+ "epoch": 2.49,
853
+ "learning_rate": 3.9014058587904144e-05,
854
+ "loss": 0.4708,
855
+ "step": 14100
856
+ },
857
+ {
858
+ "epoch": 2.5,
859
+ "learning_rate": 3.887035691000694e-05,
860
+ "loss": 0.4625,
861
+ "step": 14200
862
+ },
863
+ {
864
+ "epoch": 2.52,
865
+ "learning_rate": 3.8725990074338635e-05,
866
+ "loss": 0.4714,
867
+ "step": 14300
868
+ },
869
+ {
870
+ "epoch": 2.54,
871
+ "learning_rate": 3.858096500406085e-05,
872
+ "loss": 0.4645,
873
+ "step": 14400
874
+ },
875
+ {
876
+ "epoch": 2.56,
877
+ "learning_rate": 3.8435288653901095e-05,
878
+ "loss": 0.466,
879
+ "step": 14500
880
+ },
881
+ {
882
+ "epoch": 2.57,
883
+ "learning_rate": 3.828896800981919e-05,
884
+ "loss": 0.4658,
885
+ "step": 14600
886
+ },
887
+ {
888
+ "epoch": 2.59,
889
+ "learning_rate": 3.814201008867233e-05,
890
+ "loss": 0.4773,
891
+ "step": 14700
892
+ },
893
+ {
894
+ "epoch": 2.61,
895
+ "learning_rate": 3.799442193787853e-05,
896
+ "loss": 0.4791,
897
+ "step": 14800
898
+ },
899
+ {
900
+ "epoch": 2.63,
901
+ "learning_rate": 3.7846210635078705e-05,
902
+ "loss": 0.4665,
903
+ "step": 14900
904
+ },
905
+ {
906
+ "epoch": 2.64,
907
+ "learning_rate": 3.7697383287797214e-05,
908
+ "loss": 0.4676,
909
+ "step": 15000
910
+ },
911
+ {
912
+ "epoch": 2.66,
913
+ "learning_rate": 3.754794703310107e-05,
914
+ "loss": 0.4496,
915
+ "step": 15100
916
+ },
917
+ {
918
+ "epoch": 2.68,
919
+ "learning_rate": 3.739790903725763e-05,
920
+ "loss": 0.4662,
921
+ "step": 15200
922
+ },
923
+ {
924
+ "epoch": 2.7,
925
+ "learning_rate": 3.724727649539098e-05,
926
+ "loss": 0.4703,
927
+ "step": 15300
928
+ },
929
+ {
930
+ "epoch": 2.71,
931
+ "learning_rate": 3.7096056631136844e-05,
932
+ "loss": 0.458,
933
+ "step": 15400
934
+ },
935
+ {
936
+ "epoch": 2.73,
937
+ "learning_rate": 3.69442566962962e-05,
938
+ "loss": 0.47,
939
+ "step": 15500
940
+ },
941
+ {
942
+ "epoch": 2.75,
943
+ "learning_rate": 3.679188397048752e-05,
944
+ "loss": 0.4731,
945
+ "step": 15600
946
+ },
947
+ {
948
+ "epoch": 2.77,
949
+ "learning_rate": 3.663894576079767e-05,
950
+ "loss": 0.4733,
951
+ "step": 15700
952
+ },
953
+ {
954
+ "epoch": 2.79,
955
+ "learning_rate": 3.648544940143147e-05,
956
+ "loss": 0.4466,
957
+ "step": 15800
958
+ },
959
+ {
960
+ "epoch": 2.8,
961
+ "learning_rate": 3.633140225336001e-05,
962
+ "loss": 0.4748,
963
+ "step": 15900
964
+ },
965
+ {
966
+ "epoch": 2.82,
967
+ "learning_rate": 3.617681170396765e-05,
968
+ "loss": 0.4605,
969
+ "step": 16000
970
+ },
971
+ {
972
+ "epoch": 2.84,
973
+ "learning_rate": 3.602168516669776e-05,
974
+ "loss": 0.4617,
975
+ "step": 16100
976
+ },
977
+ {
978
+ "epoch": 2.86,
979
+ "learning_rate": 3.5866030080697155e-05,
980
+ "loss": 0.465,
981
+ "step": 16200
982
+ },
983
+ {
984
+ "epoch": 2.87,
985
+ "learning_rate": 3.5709853910459425e-05,
986
+ "loss": 0.459,
987
+ "step": 16300
988
+ },
989
+ {
990
+ "epoch": 2.89,
991
+ "learning_rate": 3.555316414546693e-05,
992
+ "loss": 0.4764,
993
+ "step": 16400
994
+ },
995
+ {
996
+ "epoch": 2.91,
997
+ "learning_rate": 3.5395968299831615e-05,
998
+ "loss": 0.4568,
999
+ "step": 16500
1000
+ },
1001
+ {
1002
+ "epoch": 2.93,
1003
+ "learning_rate": 3.5238273911934737e-05,
1004
+ "loss": 0.4746,
1005
+ "step": 16600
1006
+ },
1007
+ {
1008
+ "epoch": 2.94,
1009
+ "learning_rate": 3.508008854406529e-05,
1010
+ "loss": 0.4681,
1011
+ "step": 16700
1012
+ },
1013
+ {
1014
+ "epoch": 2.96,
1015
+ "learning_rate": 3.492141978205741e-05,
1016
+ "loss": 0.458,
1017
+ "step": 16800
1018
+ },
1019
+ {
1020
+ "epoch": 2.98,
1021
+ "learning_rate": 3.4762275234926535e-05,
1022
+ "loss": 0.4764,
1023
+ "step": 16900
1024
+ },
1025
+ {
1026
+ "epoch": 3.0,
1027
+ "learning_rate": 3.460266253450457e-05,
1028
+ "loss": 0.4684,
1029
+ "step": 17000
1030
+ },
1031
+ {
1032
+ "epoch": 3.01,
1033
+ "learning_rate": 3.444258933507386e-05,
1034
+ "loss": 0.3093,
1035
+ "step": 17100
1036
+ },
1037
+ {
1038
+ "epoch": 3.03,
1039
+ "learning_rate": 3.428206331300013e-05,
1040
+ "loss": 0.2806,
1041
+ "step": 17200
1042
+ },
1043
+ {
1044
+ "epoch": 3.05,
1045
+ "learning_rate": 3.4121092166364405e-05,
1046
+ "loss": 0.28,
1047
+ "step": 17300
1048
+ },
1049
+ {
1050
+ "epoch": 3.07,
1051
+ "learning_rate": 3.395968361459377e-05,
1052
+ "loss": 0.2668,
1053
+ "step": 17400
1054
+ },
1055
+ {
1056
+ "epoch": 3.09,
1057
+ "learning_rate": 3.379784539809127e-05,
1058
+ "loss": 0.2847,
1059
+ "step": 17500
1060
+ },
1061
+ {
1062
+ "epoch": 3.1,
1063
+ "learning_rate": 3.3635585277864624e-05,
1064
+ "loss": 0.2908,
1065
+ "step": 17600
1066
+ },
1067
+ {
1068
+ "epoch": 3.12,
1069
+ "learning_rate": 3.347291103515412e-05,
1070
+ "loss": 0.2768,
1071
+ "step": 17700
1072
+ },
1073
+ {
1074
+ "epoch": 3.14,
1075
+ "learning_rate": 3.330983047105944e-05,
1076
+ "loss": 0.2875,
1077
+ "step": 17800
1078
+ },
1079
+ {
1080
+ "epoch": 3.16,
1081
+ "learning_rate": 3.314635140616552e-05,
1082
+ "loss": 0.2817,
1083
+ "step": 17900
1084
+ },
1085
+ {
1086
+ "epoch": 3.17,
1087
+ "learning_rate": 3.2982481680167564e-05,
1088
+ "loss": 0.2799,
1089
+ "step": 18000
1090
+ },
1091
+ {
1092
+ "epoch": 3.19,
1093
+ "learning_rate": 3.281822915149505e-05,
1094
+ "loss": 0.2865,
1095
+ "step": 18100
1096
+ },
1097
+ {
1098
+ "epoch": 3.21,
1099
+ "learning_rate": 3.265360169693491e-05,
1100
+ "loss": 0.2738,
1101
+ "step": 18200
1102
+ },
1103
+ {
1104
+ "epoch": 3.23,
1105
+ "learning_rate": 3.248860721125374e-05,
1106
+ "loss": 0.2899,
1107
+ "step": 18300
1108
+ },
1109
+ {
1110
+ "epoch": 3.24,
1111
+ "learning_rate": 3.232325360681928e-05,
1112
+ "loss": 0.2819,
1113
+ "step": 18400
1114
+ },
1115
+ {
1116
+ "epoch": 3.26,
1117
+ "learning_rate": 3.2157548813220926e-05,
1118
+ "loss": 0.2857,
1119
+ "step": 18500
1120
+ },
1121
+ {
1122
+ "epoch": 3.28,
1123
+ "learning_rate": 3.199150077688945e-05,
1124
+ "loss": 0.2822,
1125
+ "step": 18600
1126
+ },
1127
+ {
1128
+ "epoch": 3.3,
1129
+ "learning_rate": 3.1825117460715966e-05,
1130
+ "loss": 0.2895,
1131
+ "step": 18700
1132
+ },
1133
+ {
1134
+ "epoch": 3.31,
1135
+ "learning_rate": 3.165840684367004e-05,
1136
+ "loss": 0.2972,
1137
+ "step": 18800
1138
+ },
1139
+ {
1140
+ "epoch": 3.33,
1141
+ "learning_rate": 3.149137692041709e-05,
1142
+ "loss": 0.2829,
1143
+ "step": 18900
1144
+ },
1145
+ {
1146
+ "epoch": 3.35,
1147
+ "learning_rate": 3.132403570093494e-05,
1148
+ "loss": 0.2919,
1149
+ "step": 19000
1150
+ },
1151
+ {
1152
+ "epoch": 3.37,
1153
+ "learning_rate": 3.1156391210129746e-05,
1154
+ "loss": 0.2921,
1155
+ "step": 19100
1156
+ },
1157
+ {
1158
+ "epoch": 3.38,
1159
+ "learning_rate": 3.0988451487451146e-05,
1160
+ "loss": 0.2789,
1161
+ "step": 19200
1162
+ },
1163
+ {
1164
+ "epoch": 3.4,
1165
+ "learning_rate": 3.0820224586506727e-05,
1166
+ "loss": 0.2832,
1167
+ "step": 19300
1168
+ },
1169
+ {
1170
+ "epoch": 3.42,
1171
+ "learning_rate": 3.065171857467581e-05,
1172
+ "loss": 0.2767,
1173
+ "step": 19400
1174
+ },
1175
+ {
1176
+ "epoch": 3.44,
1177
+ "learning_rate": 3.0482941532722576e-05,
1178
+ "loss": 0.2948,
1179
+ "step": 19500
1180
+ },
1181
+ {
1182
+ "epoch": 3.46,
1183
+ "learning_rate": 3.0313901554408537e-05,
1184
+ "loss": 0.2978,
1185
+ "step": 19600
1186
+ },
1187
+ {
1188
+ "epoch": 3.47,
1189
+ "learning_rate": 3.014460674610441e-05,
1190
+ "loss": 0.2934,
1191
+ "step": 19700
1192
+ },
1193
+ {
1194
+ "epoch": 3.49,
1195
+ "learning_rate": 2.9975065226401398e-05,
1196
+ "loss": 0.2891,
1197
+ "step": 19800
1198
+ },
1199
+ {
1200
+ "epoch": 3.51,
1201
+ "learning_rate": 2.980528512572181e-05,
1202
+ "loss": 0.2839,
1203
+ "step": 19900
1204
+ },
1205
+ {
1206
+ "epoch": 3.53,
1207
+ "learning_rate": 2.9635274585929203e-05,
1208
+ "loss": 0.2826,
1209
+ "step": 20000
1210
+ },
1211
+ {
1212
+ "epoch": 3.54,
1213
+ "learning_rate": 2.94650417599379e-05,
1214
+ "loss": 0.2968,
1215
+ "step": 20100
1216
+ },
1217
+ {
1218
+ "epoch": 3.56,
1219
+ "learning_rate": 2.9294594811322058e-05,
1220
+ "loss": 0.2837,
1221
+ "step": 20200
1222
+ },
1223
+ {
1224
+ "epoch": 3.58,
1225
+ "learning_rate": 2.9123941913924145e-05,
1226
+ "loss": 0.2818,
1227
+ "step": 20300
1228
+ },
1229
+ {
1230
+ "epoch": 3.6,
1231
+ "learning_rate": 2.895309125146299e-05,
1232
+ "loss": 0.2835,
1233
+ "step": 20400
1234
+ },
1235
+ {
1236
+ "epoch": 3.61,
1237
+ "learning_rate": 2.8782051017141287e-05,
1238
+ "loss": 0.2912,
1239
+ "step": 20500
1240
+ },
1241
+ {
1242
+ "epoch": 3.63,
1243
+ "learning_rate": 2.8610829413252748e-05,
1244
+ "loss": 0.2999,
1245
+ "step": 20600
1246
+ },
1247
+ {
1248
+ "epoch": 3.65,
1249
+ "learning_rate": 2.8439434650788706e-05,
1250
+ "loss": 0.293,
1251
+ "step": 20700
1252
+ },
1253
+ {
1254
+ "epoch": 3.67,
1255
+ "learning_rate": 2.826787494904437e-05,
1256
+ "loss": 0.2788,
1257
+ "step": 20800
1258
+ },
1259
+ {
1260
+ "epoch": 3.68,
1261
+ "learning_rate": 2.8096158535224665e-05,
1262
+ "loss": 0.2811,
1263
+ "step": 20900
1264
+ },
1265
+ {
1266
+ "epoch": 3.7,
1267
+ "learning_rate": 2.7924293644049725e-05,
1268
+ "loss": 0.2809,
1269
+ "step": 21000
1270
+ },
1271
+ {
1272
+ "epoch": 3.72,
1273
+ "learning_rate": 2.7752288517359947e-05,
1274
+ "loss": 0.2642,
1275
+ "step": 21100
1276
+ },
1277
+ {
1278
+ "epoch": 3.74,
1279
+ "learning_rate": 2.758015140372077e-05,
1280
+ "loss": 0.2848,
1281
+ "step": 21200
1282
+ },
1283
+ {
1284
+ "epoch": 3.76,
1285
+ "learning_rate": 2.7407890558027117e-05,
1286
+ "loss": 0.2837,
1287
+ "step": 21300
1288
+ },
1289
+ {
1290
+ "epoch": 3.77,
1291
+ "learning_rate": 2.7235514241107524e-05,
1292
+ "loss": 0.2796,
1293
+ "step": 21400
1294
+ },
1295
+ {
1296
+ "epoch": 3.79,
1297
+ "learning_rate": 2.7063030719327992e-05,
1298
+ "loss": 0.2857,
1299
+ "step": 21500
1300
+ },
1301
+ {
1302
+ "epoch": 3.81,
1303
+ "learning_rate": 2.689044826419556e-05,
1304
+ "loss": 0.2892,
1305
+ "step": 21600
1306
+ },
1307
+ {
1308
+ "epoch": 3.83,
1309
+ "learning_rate": 2.671777515196166e-05,
1310
+ "loss": 0.2683,
1311
+ "step": 21700
1312
+ },
1313
+ {
1314
+ "epoch": 3.84,
1315
+ "learning_rate": 2.6545019663225218e-05,
1316
+ "loss": 0.2908,
1317
+ "step": 21800
1318
+ },
1319
+ {
1320
+ "epoch": 3.86,
1321
+ "learning_rate": 2.637219008253555e-05,
1322
+ "loss": 0.2902,
1323
+ "step": 21900
1324
+ },
1325
+ {
1326
+ "epoch": 3.88,
1327
+ "learning_rate": 2.6199294697995053e-05,
1328
+ "loss": 0.2738,
1329
+ "step": 22000
1330
+ },
1331
+ {
1332
+ "epoch": 3.9,
1333
+ "learning_rate": 2.6026341800861826e-05,
1334
+ "loss": 0.2883,
1335
+ "step": 22100
1336
+ },
1337
+ {
1338
+ "epoch": 3.91,
1339
+ "learning_rate": 2.5853339685151946e-05,
1340
+ "loss": 0.2944,
1341
+ "step": 22200
1342
+ },
1343
+ {
1344
+ "epoch": 3.93,
1345
+ "learning_rate": 2.568029664724182e-05,
1346
+ "loss": 0.2803,
1347
+ "step": 22300
1348
+ },
1349
+ {
1350
+ "epoch": 3.95,
1351
+ "learning_rate": 2.550722098547029e-05,
1352
+ "loss": 0.2847,
1353
+ "step": 22400
1354
+ },
1355
+ {
1356
+ "epoch": 3.97,
1357
+ "learning_rate": 2.533412099974067e-05,
1358
+ "loss": 0.2904,
1359
+ "step": 22500
1360
+ },
1361
+ {
1362
+ "epoch": 3.98,
1363
+ "learning_rate": 2.5161004991122738e-05,
1364
+ "loss": 0.3068,
1365
+ "step": 22600
1366
+ },
1367
+ {
1368
+ "epoch": 4.0,
1369
+ "learning_rate": 2.4987881261454686e-05,
1370
+ "loss": 0.2774,
1371
+ "step": 22700
1372
+ },
1373
+ {
1374
+ "epoch": 4.02,
1375
+ "learning_rate": 2.481475811294494e-05,
1376
+ "loss": 0.178,
1377
+ "step": 22800
1378
+ },
1379
+ {
1380
+ "epoch": 4.04,
1381
+ "learning_rate": 2.464164384777406e-05,
1382
+ "loss": 0.1733,
1383
+ "step": 22900
1384
+ },
1385
+ {
1386
+ "epoch": 4.05,
1387
+ "learning_rate": 2.446854676769662e-05,
1388
+ "loss": 0.1769,
1389
+ "step": 23000
1390
+ },
1391
+ {
1392
+ "epoch": 4.07,
1393
+ "learning_rate": 2.4295475173643055e-05,
1394
+ "loss": 0.1688,
1395
+ "step": 23100
1396
+ },
1397
+ {
1398
+ "epoch": 4.09,
1399
+ "learning_rate": 2.4122437365321623e-05,
1400
+ "loss": 0.1757,
1401
+ "step": 23200
1402
+ },
1403
+ {
1404
+ "epoch": 4.11,
1405
+ "learning_rate": 2.3949441640820388e-05,
1406
+ "loss": 0.1739,
1407
+ "step": 23300
1408
+ },
1409
+ {
1410
+ "epoch": 4.13,
1411
+ "learning_rate": 2.3776496296209225e-05,
1412
+ "loss": 0.1652,
1413
+ "step": 23400
1414
+ },
1415
+ {
1416
+ "epoch": 4.14,
1417
+ "learning_rate": 2.3603609625142086e-05,
1418
+ "loss": 0.1774,
1419
+ "step": 23500
1420
+ },
1421
+ {
1422
+ "epoch": 4.16,
1423
+ "learning_rate": 2.3430789918459142e-05,
1424
+ "loss": 0.1788,
1425
+ "step": 23600
1426
+ },
1427
+ {
1428
+ "epoch": 4.18,
1429
+ "learning_rate": 2.325804546378933e-05,
1430
+ "loss": 0.1745,
1431
+ "step": 23700
1432
+ },
1433
+ {
1434
+ "epoch": 4.2,
1435
+ "learning_rate": 2.3085384545152818e-05,
1436
+ "loss": 0.1823,
1437
+ "step": 23800
1438
+ },
1439
+ {
1440
+ "epoch": 4.21,
1441
+ "learning_rate": 2.2912815442563764e-05,
1442
+ "loss": 0.1806,
1443
+ "step": 23900
1444
+ },
1445
+ {
1446
+ "epoch": 4.23,
1447
+ "learning_rate": 2.2740346431633284e-05,
1448
+ "loss": 0.1758,
1449
+ "step": 24000
1450
+ },
1451
+ {
1452
+ "epoch": 4.25,
1453
+ "learning_rate": 2.2567985783172555e-05,
1454
+ "loss": 0.1787,
1455
+ "step": 24100
1456
+ },
1457
+ {
1458
+ "epoch": 4.27,
1459
+ "learning_rate": 2.239574176279617e-05,
1460
+ "loss": 0.175,
1461
+ "step": 24200
1462
+ },
1463
+ {
1464
+ "epoch": 4.28,
1465
+ "learning_rate": 2.2223622630525822e-05,
1466
+ "loss": 0.1785,
1467
+ "step": 24300
1468
+ },
1469
+ {
1470
+ "epoch": 4.3,
1471
+ "learning_rate": 2.2051636640394116e-05,
1472
+ "loss": 0.1812,
1473
+ "step": 24400
1474
+ },
1475
+ {
1476
+ "epoch": 4.32,
1477
+ "learning_rate": 2.1879792040048797e-05,
1478
+ "loss": 0.1867,
1479
+ "step": 24500
1480
+ },
1481
+ {
1482
+ "epoch": 4.34,
1483
+ "learning_rate": 2.1708097070357218e-05,
1484
+ "loss": 0.1709,
1485
+ "step": 24600
1486
+ },
1487
+ {
1488
+ "epoch": 4.35,
1489
+ "learning_rate": 2.153655996501113e-05,
1490
+ "loss": 0.1777,
1491
+ "step": 24700
1492
+ },
1493
+ {
1494
+ "epoch": 4.37,
1495
+ "learning_rate": 2.1365188950131853e-05,
1496
+ "loss": 0.175,
1497
+ "step": 24800
1498
+ },
1499
+ {
1500
+ "epoch": 4.39,
1501
+ "learning_rate": 2.119399224387577e-05,
1502
+ "loss": 0.1817,
1503
+ "step": 24900
1504
+ },
1505
+ {
1506
+ "epoch": 4.41,
1507
+ "learning_rate": 2.102297805604025e-05,
1508
+ "loss": 0.1686,
1509
+ "step": 25000
1510
+ },
1511
+ {
1512
+ "epoch": 4.43,
1513
+ "learning_rate": 2.0852154587669925e-05,
1514
+ "loss": 0.1759,
1515
+ "step": 25100
1516
+ },
1517
+ {
1518
+ "epoch": 4.44,
1519
+ "learning_rate": 2.0681530030663386e-05,
1520
+ "loss": 0.178,
1521
+ "step": 25200
1522
+ },
1523
+ {
1524
+ "epoch": 4.46,
1525
+ "learning_rate": 2.051111256738039e-05,
1526
+ "loss": 0.1726,
1527
+ "step": 25300
1528
+ },
1529
+ {
1530
+ "epoch": 4.48,
1531
+ "learning_rate": 2.034091037024944e-05,
1532
+ "loss": 0.1815,
1533
+ "step": 25400
1534
+ },
1535
+ {
1536
+ "epoch": 4.5,
1537
+ "learning_rate": 2.017093160137584e-05,
1538
+ "loss": 0.1781,
1539
+ "step": 25500
1540
+ },
1541
+ {
1542
+ "epoch": 4.51,
1543
+ "learning_rate": 2.000118441215038e-05,
1544
+ "loss": 0.1788,
1545
+ "step": 25600
1546
+ },
1547
+ {
1548
+ "epoch": 4.53,
1549
+ "learning_rate": 1.9831676942858286e-05,
1550
+ "loss": 0.1733,
1551
+ "step": 25700
1552
+ },
1553
+ {
1554
+ "epoch": 4.55,
1555
+ "learning_rate": 1.9662417322289e-05,
1556
+ "loss": 0.1778,
1557
+ "step": 25800
1558
+ },
1559
+ {
1560
+ "epoch": 4.57,
1561
+ "learning_rate": 1.9493413667346272e-05,
1562
+ "loss": 0.176,
1563
+ "step": 25900
1564
+ },
1565
+ {
1566
+ "epoch": 4.58,
1567
+ "learning_rate": 1.9324674082658913e-05,
1568
+ "loss": 0.1753,
1569
+ "step": 26000
1570
+ },
1571
+ {
1572
+ "epoch": 4.6,
1573
+ "learning_rate": 1.9156206660192175e-05,
1574
+ "loss": 0.176,
1575
+ "step": 26100
1576
+ },
1577
+ {
1578
+ "epoch": 4.62,
1579
+ "learning_rate": 1.8988019478859672e-05,
1580
+ "loss": 0.1794,
1581
+ "step": 26200
1582
+ },
1583
+ {
1584
+ "epoch": 4.64,
1585
+ "learning_rate": 1.8820120604135944e-05,
1586
+ "loss": 0.1701,
1587
+ "step": 26300
1588
+ },
1589
+ {
1590
+ "epoch": 4.65,
1591
+ "learning_rate": 1.865251808766971e-05,
1592
+ "loss": 0.1797,
1593
+ "step": 26400
1594
+ },
1595
+ {
1596
+ "epoch": 4.67,
1597
+ "learning_rate": 1.84852199668977e-05,
1598
+ "loss": 0.1816,
1599
+ "step": 26500
1600
+ },
1601
+ {
1602
+ "epoch": 4.69,
1603
+ "learning_rate": 1.8318234264659266e-05,
1604
+ "loss": 0.182,
1605
+ "step": 26600
1606
+ },
1607
+ {
1608
+ "epoch": 4.71,
1609
+ "learning_rate": 1.8151568988811622e-05,
1610
+ "loss": 0.1829,
1611
+ "step": 26700
1612
+ },
1613
+ {
1614
+ "epoch": 4.72,
1615
+ "learning_rate": 1.798523213184579e-05,
1616
+ "loss": 0.1704,
1617
+ "step": 26800
1618
+ },
1619
+ {
1620
+ "epoch": 4.74,
1621
+ "learning_rate": 1.781923167050339e-05,
1622
+ "loss": 0.1798,
1623
+ "step": 26900
1624
+ },
1625
+ {
1626
+ "epoch": 4.76,
1627
+ "learning_rate": 1.7653575565394063e-05,
1628
+ "loss": 0.1836,
1629
+ "step": 27000
1630
+ },
1631
+ {
1632
+ "epoch": 4.78,
1633
+ "learning_rate": 1.74882717606137e-05,
1634
+ "loss": 0.1823,
1635
+ "step": 27100
1636
+ },
1637
+ {
1638
+ "epoch": 4.8,
1639
+ "learning_rate": 1.7323328183363562e-05,
1640
+ "loss": 0.1798,
1641
+ "step": 27200
1642
+ },
1643
+ {
1644
+ "epoch": 4.81,
1645
+ "learning_rate": 1.7158752743570005e-05,
1646
+ "loss": 0.1726,
1647
+ "step": 27300
1648
+ },
1649
+ {
1650
+ "epoch": 4.83,
1651
+ "learning_rate": 1.6994553333505287e-05,
1652
+ "loss": 0.1733,
1653
+ "step": 27400
1654
+ },
1655
+ {
1656
+ "epoch": 4.85,
1657
+ "learning_rate": 1.6830737827409006e-05,
1658
+ "loss": 0.1716,
1659
+ "step": 27500
1660
+ },
1661
+ {
1662
+ "epoch": 4.87,
1663
+ "learning_rate": 1.66673140811105e-05,
1664
+ "loss": 0.1734,
1665
+ "step": 27600
1666
+ },
1667
+ {
1668
+ "epoch": 4.88,
1669
+ "learning_rate": 1.650428993165215e-05,
1670
+ "loss": 0.182,
1671
+ "step": 27700
1672
+ },
1673
+ {
1674
+ "epoch": 4.9,
1675
+ "learning_rate": 1.634167319691352e-05,
1676
+ "loss": 0.17,
1677
+ "step": 27800
1678
+ },
1679
+ {
1680
+ "epoch": 4.92,
1681
+ "learning_rate": 1.617947167523646e-05,
1682
+ "loss": 0.1812,
1683
+ "step": 27900
1684
+ },
1685
+ {
1686
+ "epoch": 4.94,
1687
+ "learning_rate": 1.6017693145051155e-05,
1688
+ "loss": 0.1748,
1689
+ "step": 28000
1690
+ },
1691
+ {
1692
+ "epoch": 4.95,
1693
+ "learning_rate": 1.585634536450305e-05,
1694
+ "loss": 0.1724,
1695
+ "step": 28100
1696
+ },
1697
+ {
1698
+ "epoch": 4.97,
1699
+ "learning_rate": 1.5695436071080864e-05,
1700
+ "loss": 0.1794,
1701
+ "step": 28200
1702
+ },
1703
+ {
1704
+ "epoch": 4.99,
1705
+ "learning_rate": 1.5534972981245527e-05,
1706
+ "loss": 0.1858,
1707
+ "step": 28300
1708
+ },
1709
+ {
1710
+ "epoch": 5.01,
1711
+ "learning_rate": 1.5374963790060088e-05,
1712
+ "loss": 0.1436,
1713
+ "step": 28400
1714
+ },
1715
+ {
1716
+ "epoch": 5.02,
1717
+ "learning_rate": 1.5215416170820757e-05,
1718
+ "loss": 0.1271,
1719
+ "step": 28500
1720
+ },
1721
+ {
1722
+ "epoch": 5.04,
1723
+ "learning_rate": 1.5056337774688881e-05,
1724
+ "loss": 0.1268,
1725
+ "step": 28600
1726
+ },
1727
+ {
1728
+ "epoch": 5.06,
1729
+ "learning_rate": 1.4897736230324067e-05,
1730
+ "loss": 0.1253,
1731
+ "step": 28700
1732
+ },
1733
+ {
1734
+ "epoch": 5.08,
1735
+ "learning_rate": 1.4739619143518327e-05,
1736
+ "loss": 0.1244,
1737
+ "step": 28800
1738
+ },
1739
+ {
1740
+ "epoch": 5.1,
1741
+ "learning_rate": 1.458199409683133e-05,
1742
+ "loss": 0.122,
1743
+ "step": 28900
1744
+ },
1745
+ {
1746
+ "epoch": 5.11,
1747
+ "learning_rate": 1.4424868649226802e-05,
1748
+ "loss": 0.1351,
1749
+ "step": 29000
1750
+ },
1751
+ {
1752
+ "epoch": 5.13,
1753
+ "learning_rate": 1.4268250335710026e-05,
1754
+ "loss": 0.12,
1755
+ "step": 29100
1756
+ },
1757
+ {
1758
+ "epoch": 5.15,
1759
+ "learning_rate": 1.4112146666966486e-05,
1760
+ "loss": 0.1204,
1761
+ "step": 29200
1762
+ },
1763
+ {
1764
+ "epoch": 5.17,
1765
+ "learning_rate": 1.3956565129001709e-05,
1766
+ "loss": 0.1219,
1767
+ "step": 29300
1768
+ },
1769
+ {
1770
+ "epoch": 5.18,
1771
+ "learning_rate": 1.380151318278223e-05,
1772
+ "loss": 0.1268,
1773
+ "step": 29400
1774
+ },
1775
+ {
1776
+ "epoch": 5.2,
1777
+ "learning_rate": 1.3646998263877857e-05,
1778
+ "loss": 0.1262,
1779
+ "step": 29500
1780
+ },
1781
+ {
1782
+ "epoch": 5.22,
1783
+ "learning_rate": 1.3493027782105063e-05,
1784
+ "loss": 0.1295,
1785
+ "step": 29600
1786
+ },
1787
+ {
1788
+ "epoch": 5.24,
1789
+ "learning_rate": 1.3339609121171642e-05,
1790
+ "loss": 0.1232,
1791
+ "step": 29700
1792
+ },
1793
+ {
1794
+ "epoch": 5.25,
1795
+ "learning_rate": 1.3186749638322643e-05,
1796
+ "loss": 0.1204,
1797
+ "step": 29800
1798
+ },
1799
+ {
1800
+ "epoch": 5.27,
1801
+ "learning_rate": 1.3034456663987508e-05,
1802
+ "loss": 0.1248,
1803
+ "step": 29900
1804
+ },
1805
+ {
1806
+ "epoch": 5.29,
1807
+ "learning_rate": 1.2882737501428579e-05,
1808
+ "loss": 0.1317,
1809
+ "step": 30000
1810
+ },
1811
+ {
1812
+ "epoch": 5.31,
1813
+ "learning_rate": 1.2731599426390894e-05,
1814
+ "loss": 0.1269,
1815
+ "step": 30100
1816
+ },
1817
+ {
1818
+ "epoch": 5.32,
1819
+ "learning_rate": 1.2581049686753171e-05,
1820
+ "loss": 0.1301,
1821
+ "step": 30200
1822
+ },
1823
+ {
1824
+ "epoch": 5.34,
1825
+ "learning_rate": 1.2431095502180349e-05,
1826
+ "loss": 0.1202,
1827
+ "step": 30300
1828
+ },
1829
+ {
1830
+ "epoch": 5.36,
1831
+ "learning_rate": 1.2281744063777295e-05,
1832
+ "loss": 0.1216,
1833
+ "step": 30400
1834
+ },
1835
+ {
1836
+ "epoch": 5.38,
1837
+ "learning_rate": 1.2133002533743987e-05,
1838
+ "loss": 0.122,
1839
+ "step": 30500
1840
+ },
1841
+ {
1842
+ "epoch": 5.39,
1843
+ "learning_rate": 1.1984878045032037e-05,
1844
+ "loss": 0.1323,
1845
+ "step": 30600
1846
+ },
1847
+ {
1848
+ "epoch": 5.41,
1849
+ "learning_rate": 1.18373777010026e-05,
1850
+ "loss": 0.1263,
1851
+ "step": 30700
1852
+ },
1853
+ {
1854
+ "epoch": 5.43,
1855
+ "learning_rate": 1.169050857508579e-05,
1856
+ "loss": 0.1254,
1857
+ "step": 30800
1858
+ },
1859
+ {
1860
+ "epoch": 5.45,
1861
+ "learning_rate": 1.1544277710441418e-05,
1862
+ "loss": 0.1259,
1863
+ "step": 30900
1864
+ },
1865
+ {
1866
+ "epoch": 5.47,
1867
+ "learning_rate": 1.1398692119621265e-05,
1868
+ "loss": 0.1245,
1869
+ "step": 31000
1870
+ },
1871
+ {
1872
+ "epoch": 5.48,
1873
+ "learning_rate": 1.1253758784232778e-05,
1874
+ "loss": 0.1255,
1875
+ "step": 31100
1876
+ },
1877
+ {
1878
+ "epoch": 5.5,
1879
+ "learning_rate": 1.1109484654604277e-05,
1880
+ "loss": 0.125,
1881
+ "step": 31200
1882
+ },
1883
+ {
1884
+ "epoch": 5.52,
1885
+ "learning_rate": 1.0965876649451612e-05,
1886
+ "loss": 0.1213,
1887
+ "step": 31300
1888
+ },
1889
+ {
1890
+ "epoch": 5.54,
1891
+ "learning_rate": 1.0822941655546436e-05,
1892
+ "loss": 0.1291,
1893
+ "step": 31400
1894
+ },
1895
+ {
1896
+ "epoch": 5.55,
1897
+ "learning_rate": 1.0680686527385902e-05,
1898
+ "loss": 0.125,
1899
+ "step": 31500
1900
+ },
1901
+ {
1902
+ "epoch": 5.57,
1903
+ "learning_rate": 1.0539118086863969e-05,
1904
+ "loss": 0.1186,
1905
+ "step": 31600
1906
+ },
1907
+ {
1908
+ "epoch": 5.59,
1909
+ "learning_rate": 1.0398243122944251e-05,
1910
+ "loss": 0.1305,
1911
+ "step": 31700
1912
+ },
1913
+ {
1914
+ "epoch": 5.61,
1915
+ "learning_rate": 1.025806839133443e-05,
1916
+ "loss": 0.1205,
1917
+ "step": 31800
1918
+ },
1919
+ {
1920
+ "epoch": 5.62,
1921
+ "learning_rate": 1.011860061416235e-05,
1922
+ "loss": 0.1225,
1923
+ "step": 31900
1924
+ },
1925
+ {
1926
+ "epoch": 5.64,
1927
+ "learning_rate": 9.979846479653585e-06,
1928
+ "loss": 0.1291,
1929
+ "step": 32000
1930
+ },
1931
+ {
1932
+ "epoch": 5.66,
1933
+ "learning_rate": 9.841812641810705e-06,
1934
+ "loss": 0.1261,
1935
+ "step": 32100
1936
+ },
1937
+ {
1938
+ "epoch": 5.68,
1939
+ "learning_rate": 9.70450572009425e-06,
1940
+ "loss": 0.1305,
1941
+ "step": 32200
1942
+ },
1943
+ {
1944
+ "epoch": 5.69,
1945
+ "learning_rate": 9.567932299105192e-06,
1946
+ "loss": 0.1303,
1947
+ "step": 32300
1948
+ },
1949
+ {
1950
+ "epoch": 5.71,
1951
+ "learning_rate": 9.43209892826928e-06,
1952
+ "loss": 0.1262,
1953
+ "step": 32400
1954
+ },
1955
+ {
1956
+ "epoch": 5.73,
1957
+ "learning_rate": 9.297012121522874e-06,
1958
+ "loss": 0.1309,
1959
+ "step": 32500
1960
+ },
1961
+ {
1962
+ "epoch": 5.75,
1963
+ "learning_rate": 9.16267835700057e-06,
1964
+ "loss": 0.1207,
1965
+ "step": 32600
1966
+ },
1967
+ {
1968
+ "epoch": 5.76,
1969
+ "learning_rate": 9.0291040767246e-06,
1970
+ "loss": 0.1274,
1971
+ "step": 32700
1972
+ },
1973
+ {
1974
+ "epoch": 5.78,
1975
+ "learning_rate": 8.896295686295853e-06,
1976
+ "loss": 0.1299,
1977
+ "step": 32800
1978
+ },
1979
+ {
1980
+ "epoch": 5.8,
1981
+ "learning_rate": 8.764259554586699e-06,
1982
+ "loss": 0.1247,
1983
+ "step": 32900
1984
+ },
1985
+ {
1986
+ "epoch": 5.82,
1987
+ "learning_rate": 8.633002013435593e-06,
1988
+ "loss": 0.1275,
1989
+ "step": 33000
1990
+ },
1991
+ {
1992
+ "epoch": 5.84,
1993
+ "learning_rate": 8.502529357343374e-06,
1994
+ "loss": 0.1245,
1995
+ "step": 33100
1996
+ },
1997
+ {
1998
+ "epoch": 5.85,
1999
+ "learning_rate": 8.372847843171482e-06,
2000
+ "loss": 0.128,
2001
+ "step": 33200
2002
+ },
2003
+ {
2004
+ "epoch": 5.87,
2005
+ "learning_rate": 8.243963689841855e-06,
2006
+ "loss": 0.1281,
2007
+ "step": 33300
2008
+ },
2009
+ {
2010
+ "epoch": 5.89,
2011
+ "learning_rate": 8.115883078038719e-06,
2012
+ "loss": 0.1297,
2013
+ "step": 33400
2014
+ },
2015
+ {
2016
+ "epoch": 5.91,
2017
+ "learning_rate": 7.988612149912195e-06,
2018
+ "loss": 0.1284,
2019
+ "step": 33500
2020
+ },
2021
+ {
2022
+ "epoch": 5.92,
2023
+ "learning_rate": 7.86215700878372e-06,
2024
+ "loss": 0.1217,
2025
+ "step": 33600
2026
+ },
2027
+ {
2028
+ "epoch": 5.94,
2029
+ "learning_rate": 7.736523718853395e-06,
2030
+ "loss": 0.1279,
2031
+ "step": 33700
2032
+ },
2033
+ {
2034
+ "epoch": 5.96,
2035
+ "learning_rate": 7.611718304909193e-06,
2036
+ "loss": 0.1277,
2037
+ "step": 33800
2038
+ },
2039
+ {
2040
+ "epoch": 5.98,
2041
+ "learning_rate": 7.487746752037955e-06,
2042
+ "loss": 0.1232,
2043
+ "step": 33900
2044
+ },
2045
+ {
2046
+ "epoch": 5.99,
2047
+ "learning_rate": 7.364615005338455e-06,
2048
+ "loss": 0.1269,
2049
+ "step": 34000
2050
+ },
2051
+ {
2052
+ "epoch": 6.01,
2053
+ "learning_rate": 7.242328969636267e-06,
2054
+ "loss": 0.1191,
2055
+ "step": 34100
2056
+ },
2057
+ {
2058
+ "epoch": 6.03,
2059
+ "learning_rate": 7.120894509200601e-06,
2060
+ "loss": 0.1115,
2061
+ "step": 34200
2062
+ },
2063
+ {
2064
+ "epoch": 6.05,
2065
+ "learning_rate": 7.0003174474630865e-06,
2066
+ "loss": 0.1094,
2067
+ "step": 34300
2068
+ },
2069
+ {
2070
+ "epoch": 6.06,
2071
+ "learning_rate": 6.8806035667384775e-06,
2072
+ "loss": 0.1062,
2073
+ "step": 34400
2074
+ },
2075
+ {
2076
+ "epoch": 6.08,
2077
+ "learning_rate": 6.761758607947408e-06,
2078
+ "loss": 0.1031,
2079
+ "step": 34500
2080
+ },
2081
+ {
2082
+ "epoch": 6.1,
2083
+ "learning_rate": 6.64378827034105e-06,
2084
+ "loss": 0.1115,
2085
+ "step": 34600
2086
+ },
2087
+ {
2088
+ "epoch": 6.12,
2089
+ "learning_rate": 6.526698211227819e-06,
2090
+ "loss": 0.1079,
2091
+ "step": 34700
2092
+ },
2093
+ {
2094
+ "epoch": 6.14,
2095
+ "learning_rate": 6.410494045702062e-06,
2096
+ "loss": 0.1049,
2097
+ "step": 34800
2098
+ },
2099
+ {
2100
+ "epoch": 6.15,
2101
+ "learning_rate": 6.295181346374806e-06,
2102
+ "loss": 0.1082,
2103
+ "step": 34900
2104
+ },
2105
+ {
2106
+ "epoch": 6.17,
2107
+ "learning_rate": 6.1807656431064815e-06,
2108
+ "loss": 0.1046,
2109
+ "step": 35000
2110
+ },
2111
+ {
2112
+ "epoch": 6.19,
2113
+ "learning_rate": 6.067252422741787e-06,
2114
+ "loss": 0.102,
2115
+ "step": 35100
2116
+ },
2117
+ {
2118
+ "epoch": 6.21,
2119
+ "learning_rate": 5.954647128846535e-06,
2120
+ "loss": 0.109,
2121
+ "step": 35200
2122
+ },
2123
+ {
2124
+ "epoch": 6.22,
2125
+ "learning_rate": 5.842955161446606e-06,
2126
+ "loss": 0.1073,
2127
+ "step": 35300
2128
+ },
2129
+ {
2130
+ "epoch": 6.24,
2131
+ "learning_rate": 5.7321818767689965e-06,
2132
+ "loss": 0.1111,
2133
+ "step": 35400
2134
+ },
2135
+ {
2136
+ "epoch": 6.26,
2137
+ "learning_rate": 5.622332586984932e-06,
2138
+ "loss": 0.1027,
2139
+ "step": 35500
2140
+ },
2141
+ {
2142
+ "epoch": 6.28,
2143
+ "learning_rate": 5.513412559955195e-06,
2144
+ "loss": 0.1076,
2145
+ "step": 35600
2146
+ },
2147
+ {
2148
+ "epoch": 6.29,
2149
+ "learning_rate": 5.405427018977391e-06,
2150
+ "loss": 0.1064,
2151
+ "step": 35700
2152
+ },
2153
+ {
2154
+ "epoch": 6.31,
2155
+ "learning_rate": 5.298381142535558e-06,
2156
+ "loss": 0.102,
2157
+ "step": 35800
2158
+ },
2159
+ {
2160
+ "epoch": 6.33,
2161
+ "learning_rate": 5.1922800640517934e-06,
2162
+ "loss": 0.1098,
2163
+ "step": 35900
2164
+ },
2165
+ {
2166
+ "epoch": 6.35,
2167
+ "learning_rate": 5.08712887164006e-06,
2168
+ "loss": 0.1134,
2169
+ "step": 36000
2170
+ },
2171
+ {
2172
+ "epoch": 6.36,
2173
+ "learning_rate": 4.982932607862234e-06,
2174
+ "loss": 0.1152,
2175
+ "step": 36100
2176
+ },
2177
+ {
2178
+ "epoch": 6.38,
2179
+ "learning_rate": 4.87969626948625e-06,
2180
+ "loss": 0.1079,
2181
+ "step": 36200
2182
+ },
2183
+ {
2184
+ "epoch": 6.4,
2185
+ "learning_rate": 4.777424807246464e-06,
2186
+ "loss": 0.1102,
2187
+ "step": 36300
2188
+ },
2189
+ {
2190
+ "epoch": 6.42,
2191
+ "learning_rate": 4.676123125606288e-06,
2192
+ "loss": 0.1052,
2193
+ "step": 36400
2194
+ },
2195
+ {
2196
+ "epoch": 6.43,
2197
+ "learning_rate": 4.575796082522965e-06,
2198
+ "loss": 0.1137,
2199
+ "step": 36500
2200
+ },
2201
+ {
2202
+ "epoch": 6.45,
2203
+ "learning_rate": 4.476448489214599e-06,
2204
+ "loss": 0.1103,
2205
+ "step": 36600
2206
+ },
2207
+ {
2208
+ "epoch": 6.47,
2209
+ "learning_rate": 4.378085109929458e-06,
2210
+ "loss": 0.105,
2211
+ "step": 36700
2212
+ },
2213
+ {
2214
+ "epoch": 6.49,
2215
+ "learning_rate": 4.280710661717463e-06,
2216
+ "loss": 0.1079,
2217
+ "step": 36800
2218
+ },
2219
+ {
2220
+ "epoch": 6.51,
2221
+ "learning_rate": 4.184329814204014e-06,
2222
+ "loss": 0.1107,
2223
+ "step": 36900
2224
+ },
2225
+ {
2226
+ "epoch": 6.52,
2227
+ "learning_rate": 4.0889471893660494e-06,
2228
+ "loss": 0.1094,
2229
+ "step": 37000
2230
+ },
2231
+ {
2232
+ "epoch": 6.54,
2233
+ "learning_rate": 3.994567361310386e-06,
2234
+ "loss": 0.1085,
2235
+ "step": 37100
2236
+ },
2237
+ {
2238
+ "epoch": 6.56,
2239
+ "learning_rate": 3.901194856054388e-06,
2240
+ "loss": 0.0993,
2241
+ "step": 37200
2242
+ },
2243
+ {
2244
+ "epoch": 6.58,
2245
+ "learning_rate": 3.808834151308879e-06,
2246
+ "loss": 0.1096,
2247
+ "step": 37300
2248
+ },
2249
+ {
2250
+ "epoch": 6.59,
2251
+ "learning_rate": 3.7174896762634554e-06,
2252
+ "loss": 0.1008,
2253
+ "step": 37400
2254
+ },
2255
+ {
2256
+ "epoch": 6.61,
2257
+ "learning_rate": 3.627165811374081e-06,
2258
+ "loss": 0.1093,
2259
+ "step": 37500
2260
+ },
2261
+ {
2262
+ "epoch": 6.63,
2263
+ "learning_rate": 3.5378668881529744e-06,
2264
+ "loss": 0.1064,
2265
+ "step": 37600
2266
+ },
2267
+ {
2268
+ "epoch": 6.65,
2269
+ "learning_rate": 3.44959718896094e-06,
2270
+ "loss": 0.1059,
2271
+ "step": 37700
2272
+ },
2273
+ {
2274
+ "epoch": 6.66,
2275
+ "learning_rate": 3.3623609468019845e-06,
2276
+ "loss": 0.11,
2277
+ "step": 37800
2278
+ },
2279
+ {
2280
+ "epoch": 6.68,
2281
+ "learning_rate": 3.276162345120329e-06,
2282
+ "loss": 0.1067,
2283
+ "step": 37900
2284
+ },
2285
+ {
2286
+ "epoch": 6.7,
2287
+ "learning_rate": 3.1910055175997794e-06,
2288
+ "loss": 0.1048,
2289
+ "step": 38000
2290
+ },
2291
+ {
2292
+ "epoch": 6.72,
2293
+ "learning_rate": 3.106894547965497e-06,
2294
+ "loss": 0.1098,
2295
+ "step": 38100
2296
+ },
2297
+ {
2298
+ "epoch": 6.73,
2299
+ "learning_rate": 3.023833469788176e-06,
2300
+ "loss": 0.1084,
2301
+ "step": 38200
2302
+ },
2303
+ {
2304
+ "epoch": 6.75,
2305
+ "learning_rate": 2.9418262662905955e-06,
2306
+ "loss": 0.1084,
2307
+ "step": 38300
2308
+ },
2309
+ {
2310
+ "epoch": 6.77,
2311
+ "learning_rate": 2.860876870156612e-06,
2312
+ "loss": 0.1119,
2313
+ "step": 38400
2314
+ },
2315
+ {
2316
+ "epoch": 6.79,
2317
+ "learning_rate": 2.7809891633425676e-06,
2318
+ "loss": 0.1092,
2319
+ "step": 38500
2320
+ },
2321
+ {
2322
+ "epoch": 6.8,
2323
+ "learning_rate": 2.70216697689111e-06,
2324
+ "loss": 0.1076,
2325
+ "step": 38600
2326
+ },
2327
+ {
2328
+ "epoch": 6.82,
2329
+ "learning_rate": 2.624414090747504e-06,
2330
+ "loss": 0.1113,
2331
+ "step": 38700
2332
+ },
2333
+ {
2334
+ "epoch": 6.84,
2335
+ "learning_rate": 2.5477342335783445e-06,
2336
+ "loss": 0.114,
2337
+ "step": 38800
2338
+ },
2339
+ {
2340
+ "epoch": 6.86,
2341
+ "learning_rate": 2.4721310825927464e-06,
2342
+ "loss": 0.1026,
2343
+ "step": 38900
2344
+ },
2345
+ {
2346
+ "epoch": 6.88,
2347
+ "learning_rate": 2.397608263366008e-06,
2348
+ "loss": 0.1115,
2349
+ "step": 39000
2350
+ }
2351
+ ],
2352
+ "logging_steps": 100,
2353
+ "max_steps": 45376,
2354
+ "num_input_tokens_seen": 0,
2355
+ "num_train_epochs": 8,
2356
+ "save_steps": 3000,
2357
+ "total_flos": 0.0,
2358
+ "train_batch_size": 1,
2359
+ "trial_name": null,
2360
+ "trial_params": null
2361
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd7640c7005dc98ac1372e8db15dadabf5f37abce0887d18e682d1997b396643
3
+ size 5176