riken01 commited on
Commit
7fac7aa
1 Parent(s): 8b3a9ef

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ tags:
4
+ - autotrain
5
+ - text2text-generation
6
+ base_model: google-t5/t5-base
7
+ widget:
8
+ - text: "I love AutoTrain"
9
+ ---
10
+
11
+ # Model Trained Using AutoTrain
12
+
13
+ - Problem type: Seq2Seq
14
+
15
+ ## Validation Metrics
16
+ loss: 3.4647490978240967
17
+
18
+ rouge1: 34.5915
19
+
20
+ rouge2: 14.7399
21
+
22
+ rougeL: 33.5742
23
+
24
+ rougeLsum: 33.7167
25
+
26
+ gen_len: 17.5112
27
+
28
+ runtime: 117.889
29
+
30
+ samples_per_second: 9.475
31
+
32
+ steps_per_second: 2.375
33
+
34
+ : 3.0
checkpoint-13407/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google-t5/t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "relu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": false,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "n_positions": 512,
21
+ "num_decoder_layers": 12,
22
+ "num_heads": 12,
23
+ "num_layers": 12,
24
+ "output_past": true,
25
+ "pad_token_id": 0,
26
+ "relative_attention_max_distance": 128,
27
+ "relative_attention_num_buckets": 32,
28
+ "task_specific_params": {
29
+ "summarization": {
30
+ "early_stopping": true,
31
+ "length_penalty": 2.0,
32
+ "max_length": 200,
33
+ "min_length": 30,
34
+ "no_repeat_ngram_size": 3,
35
+ "num_beams": 4,
36
+ "prefix": "summarize: "
37
+ },
38
+ "translation_en_to_de": {
39
+ "early_stopping": true,
40
+ "max_length": 300,
41
+ "num_beams": 4,
42
+ "prefix": "translate English to German: "
43
+ },
44
+ "translation_en_to_fr": {
45
+ "early_stopping": true,
46
+ "max_length": 300,
47
+ "num_beams": 4,
48
+ "prefix": "translate English to French: "
49
+ },
50
+ "translation_en_to_ro": {
51
+ "early_stopping": true,
52
+ "max_length": 300,
53
+ "num_beams": 4,
54
+ "prefix": "translate English to Romanian: "
55
+ }
56
+ },
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.41.0",
59
+ "use_cache": false,
60
+ "vocab_size": 32128
61
+ }
checkpoint-13407/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "decoder_start_token_id": 0,
3
+ "eos_token_id": 1,
4
+ "pad_token_id": 0,
5
+ "transformers_version": "4.41.0"
6
+ }
checkpoint-13407/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1451374ea6f9efea6b2c3934a2252d4458f9545ad6ad102a223e0c2e12e8a029
3
+ size 891644712
checkpoint-13407/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97079dbfb597ac2b0e1e6df69f8daa4ddfcbe97cba48382c0520e9705312c4ae
3
+ size 1783444794
checkpoint-13407/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b38d05d859a5186f3e0dda8966ba0b98f9434cbf5f22333dde2f78c847bc3d4e
3
+ size 14244
checkpoint-13407/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6f612b8258c68ad9ecb53f347e07ca490bd57e97a73c6645dc72ef36d1e4bf0
3
+ size 1064
checkpoint-13407/special_tokens_map.json ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "eos_token": "</s>",
105
+ "pad_token": "<pad>",
106
+ "unk_token": "<unk>"
107
+ }
checkpoint-13407/spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
checkpoint-13407/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-13407/tokenizer_config.json ADDED
@@ -0,0 +1,937 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<pad>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "</s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<unk>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "32000": {
28
+ "content": "<extra_id_99>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "32001": {
36
+ "content": "<extra_id_98>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "32002": {
44
+ "content": "<extra_id_97>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "32003": {
52
+ "content": "<extra_id_96>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "32004": {
60
+ "content": "<extra_id_95>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "32005": {
68
+ "content": "<extra_id_94>",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "32006": {
76
+ "content": "<extra_id_93>",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "32007": {
84
+ "content": "<extra_id_92>",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "32008": {
92
+ "content": "<extra_id_91>",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "32009": {
100
+ "content": "<extra_id_90>",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "32010": {
108
+ "content": "<extra_id_89>",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "32011": {
116
+ "content": "<extra_id_88>",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "32012": {
124
+ "content": "<extra_id_87>",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "32013": {
132
+ "content": "<extra_id_86>",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "32014": {
140
+ "content": "<extra_id_85>",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "32015": {
148
+ "content": "<extra_id_84>",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "32016": {
156
+ "content": "<extra_id_83>",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "32017": {
164
+ "content": "<extra_id_82>",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "32018": {
172
+ "content": "<extra_id_81>",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "32019": {
180
+ "content": "<extra_id_80>",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "32020": {
188
+ "content": "<extra_id_79>",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "32021": {
196
+ "content": "<extra_id_78>",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "32022": {
204
+ "content": "<extra_id_77>",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "32023": {
212
+ "content": "<extra_id_76>",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "32024": {
220
+ "content": "<extra_id_75>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "32025": {
228
+ "content": "<extra_id_74>",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "32026": {
236
+ "content": "<extra_id_73>",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "32027": {
244
+ "content": "<extra_id_72>",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "32028": {
252
+ "content": "<extra_id_71>",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "32029": {
260
+ "content": "<extra_id_70>",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "32030": {
268
+ "content": "<extra_id_69>",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "32031": {
276
+ "content": "<extra_id_68>",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "32032": {
284
+ "content": "<extra_id_67>",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "32033": {
292
+ "content": "<extra_id_66>",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "32034": {
300
+ "content": "<extra_id_65>",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "32035": {
308
+ "content": "<extra_id_64>",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "32036": {
316
+ "content": "<extra_id_63>",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "32037": {
324
+ "content": "<extra_id_62>",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "32038": {
332
+ "content": "<extra_id_61>",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "32039": {
340
+ "content": "<extra_id_60>",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "32040": {
348
+ "content": "<extra_id_59>",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "32041": {
356
+ "content": "<extra_id_58>",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "32042": {
364
+ "content": "<extra_id_57>",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "32043": {
372
+ "content": "<extra_id_56>",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "32044": {
380
+ "content": "<extra_id_55>",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "32045": {
388
+ "content": "<extra_id_54>",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "32046": {
396
+ "content": "<extra_id_53>",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "32047": {
404
+ "content": "<extra_id_52>",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "32048": {
412
+ "content": "<extra_id_51>",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "32049": {
420
+ "content": "<extra_id_50>",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "32050": {
428
+ "content": "<extra_id_49>",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "32051": {
436
+ "content": "<extra_id_48>",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "32052": {
444
+ "content": "<extra_id_47>",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "32053": {
452
+ "content": "<extra_id_46>",
453
+ "lstrip": false,
454
+ "normalized": false,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ },
459
+ "32054": {
460
+ "content": "<extra_id_45>",
461
+ "lstrip": false,
462
+ "normalized": false,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": true
466
+ },
467
+ "32055": {
468
+ "content": "<extra_id_44>",
469
+ "lstrip": false,
470
+ "normalized": false,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": true
474
+ },
475
+ "32056": {
476
+ "content": "<extra_id_43>",
477
+ "lstrip": false,
478
+ "normalized": false,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": true
482
+ },
483
+ "32057": {
484
+ "content": "<extra_id_42>",
485
+ "lstrip": false,
486
+ "normalized": false,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": true
490
+ },
491
+ "32058": {
492
+ "content": "<extra_id_41>",
493
+ "lstrip": false,
494
+ "normalized": false,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": true
498
+ },
499
+ "32059": {
500
+ "content": "<extra_id_40>",
501
+ "lstrip": false,
502
+ "normalized": false,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": true
506
+ },
507
+ "32060": {
508
+ "content": "<extra_id_39>",
509
+ "lstrip": false,
510
+ "normalized": false,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": true
514
+ },
515
+ "32061": {
516
+ "content": "<extra_id_38>",
517
+ "lstrip": false,
518
+ "normalized": false,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": true
522
+ },
523
+ "32062": {
524
+ "content": "<extra_id_37>",
525
+ "lstrip": false,
526
+ "normalized": false,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": true
530
+ },
531
+ "32063": {
532
+ "content": "<extra_id_36>",
533
+ "lstrip": false,
534
+ "normalized": false,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": true
538
+ },
539
+ "32064": {
540
+ "content": "<extra_id_35>",
541
+ "lstrip": false,
542
+ "normalized": false,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": true
546
+ },
547
+ "32065": {
548
+ "content": "<extra_id_34>",
549
+ "lstrip": false,
550
+ "normalized": false,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": true
554
+ },
555
+ "32066": {
556
+ "content": "<extra_id_33>",
557
+ "lstrip": false,
558
+ "normalized": false,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": true
562
+ },
563
+ "32067": {
564
+ "content": "<extra_id_32>",
565
+ "lstrip": false,
566
+ "normalized": false,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": true
570
+ },
571
+ "32068": {
572
+ "content": "<extra_id_31>",
573
+ "lstrip": false,
574
+ "normalized": false,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": true
578
+ },
579
+ "32069": {
580
+ "content": "<extra_id_30>",
581
+ "lstrip": false,
582
+ "normalized": false,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": true
586
+ },
587
+ "32070": {
588
+ "content": "<extra_id_29>",
589
+ "lstrip": false,
590
+ "normalized": false,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": true
594
+ },
595
+ "32071": {
596
+ "content": "<extra_id_28>",
597
+ "lstrip": false,
598
+ "normalized": false,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": true
602
+ },
603
+ "32072": {
604
+ "content": "<extra_id_27>",
605
+ "lstrip": false,
606
+ "normalized": false,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": true
610
+ },
611
+ "32073": {
612
+ "content": "<extra_id_26>",
613
+ "lstrip": false,
614
+ "normalized": false,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": true
618
+ },
619
+ "32074": {
620
+ "content": "<extra_id_25>",
621
+ "lstrip": false,
622
+ "normalized": false,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": true
626
+ },
627
+ "32075": {
628
+ "content": "<extra_id_24>",
629
+ "lstrip": false,
630
+ "normalized": false,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": true
634
+ },
635
+ "32076": {
636
+ "content": "<extra_id_23>",
637
+ "lstrip": false,
638
+ "normalized": false,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": true
642
+ },
643
+ "32077": {
644
+ "content": "<extra_id_22>",
645
+ "lstrip": false,
646
+ "normalized": false,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": true
650
+ },
651
+ "32078": {
652
+ "content": "<extra_id_21>",
653
+ "lstrip": false,
654
+ "normalized": false,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": true
658
+ },
659
+ "32079": {
660
+ "content": "<extra_id_20>",
661
+ "lstrip": false,
662
+ "normalized": false,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": true
666
+ },
667
+ "32080": {
668
+ "content": "<extra_id_19>",
669
+ "lstrip": false,
670
+ "normalized": false,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": true
674
+ },
675
+ "32081": {
676
+ "content": "<extra_id_18>",
677
+ "lstrip": false,
678
+ "normalized": false,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": true
682
+ },
683
+ "32082": {
684
+ "content": "<extra_id_17>",
685
+ "lstrip": false,
686
+ "normalized": false,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": true
690
+ },
691
+ "32083": {
692
+ "content": "<extra_id_16>",
693
+ "lstrip": false,
694
+ "normalized": false,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": true
698
+ },
699
+ "32084": {
700
+ "content": "<extra_id_15>",
701
+ "lstrip": false,
702
+ "normalized": false,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": true
706
+ },
707
+ "32085": {
708
+ "content": "<extra_id_14>",
709
+ "lstrip": false,
710
+ "normalized": false,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": true
714
+ },
715
+ "32086": {
716
+ "content": "<extra_id_13>",
717
+ "lstrip": false,
718
+ "normalized": false,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": true
722
+ },
723
+ "32087": {
724
+ "content": "<extra_id_12>",
725
+ "lstrip": false,
726
+ "normalized": false,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": true
730
+ },
731
+ "32088": {
732
+ "content": "<extra_id_11>",
733
+ "lstrip": false,
734
+ "normalized": false,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": true
738
+ },
739
+ "32089": {
740
+ "content": "<extra_id_10>",
741
+ "lstrip": false,
742
+ "normalized": false,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": true
746
+ },
747
+ "32090": {
748
+ "content": "<extra_id_9>",
749
+ "lstrip": false,
750
+ "normalized": false,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": true
754
+ },
755
+ "32091": {
756
+ "content": "<extra_id_8>",
757
+ "lstrip": false,
758
+ "normalized": false,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": true
762
+ },
763
+ "32092": {
764
+ "content": "<extra_id_7>",
765
+ "lstrip": false,
766
+ "normalized": false,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": true
770
+ },
771
+ "32093": {
772
+ "content": "<extra_id_6>",
773
+ "lstrip": false,
774
+ "normalized": false,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": true
778
+ },
779
+ "32094": {
780
+ "content": "<extra_id_5>",
781
+ "lstrip": false,
782
+ "normalized": false,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": true
786
+ },
787
+ "32095": {
788
+ "content": "<extra_id_4>",
789
+ "lstrip": false,
790
+ "normalized": false,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": true
794
+ },
795
+ "32096": {
796
+ "content": "<extra_id_3>",
797
+ "lstrip": false,
798
+ "normalized": false,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": true
802
+ },
803
+ "32097": {
804
+ "content": "<extra_id_2>",
805
+ "lstrip": false,
806
+ "normalized": false,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": true
810
+ },
811
+ "32098": {
812
+ "content": "<extra_id_1>",
813
+ "lstrip": false,
814
+ "normalized": false,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": true
818
+ },
819
+ "32099": {
820
+ "content": "<extra_id_0>",
821
+ "lstrip": false,
822
+ "normalized": false,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": true
826
+ }
827
+ },
828
+ "additional_special_tokens": [
829
+ "<extra_id_0>",
830
+ "<extra_id_1>",
831
+ "<extra_id_2>",
832
+ "<extra_id_3>",
833
+ "<extra_id_4>",
834
+ "<extra_id_5>",
835
+ "<extra_id_6>",
836
+ "<extra_id_7>",
837
+ "<extra_id_8>",
838
+ "<extra_id_9>",
839
+ "<extra_id_10>",
840
+ "<extra_id_11>",
841
+ "<extra_id_12>",
842
+ "<extra_id_13>",
843
+ "<extra_id_14>",
844
+ "<extra_id_15>",
845
+ "<extra_id_16>",
846
+ "<extra_id_17>",
847
+ "<extra_id_18>",
848
+ "<extra_id_19>",
849
+ "<extra_id_20>",
850
+ "<extra_id_21>",
851
+ "<extra_id_22>",
852
+ "<extra_id_23>",
853
+ "<extra_id_24>",
854
+ "<extra_id_25>",
855
+ "<extra_id_26>",
856
+ "<extra_id_27>",
857
+ "<extra_id_28>",
858
+ "<extra_id_29>",
859
+ "<extra_id_30>",
860
+ "<extra_id_31>",
861
+ "<extra_id_32>",
862
+ "<extra_id_33>",
863
+ "<extra_id_34>",
864
+ "<extra_id_35>",
865
+ "<extra_id_36>",
866
+ "<extra_id_37>",
867
+ "<extra_id_38>",
868
+ "<extra_id_39>",
869
+ "<extra_id_40>",
870
+ "<extra_id_41>",
871
+ "<extra_id_42>",
872
+ "<extra_id_43>",
873
+ "<extra_id_44>",
874
+ "<extra_id_45>",
875
+ "<extra_id_46>",
876
+ "<extra_id_47>",
877
+ "<extra_id_48>",
878
+ "<extra_id_49>",
879
+ "<extra_id_50>",
880
+ "<extra_id_51>",
881
+ "<extra_id_52>",
882
+ "<extra_id_53>",
883
+ "<extra_id_54>",
884
+ "<extra_id_55>",
885
+ "<extra_id_56>",
886
+ "<extra_id_57>",
887
+ "<extra_id_58>",
888
+ "<extra_id_59>",
889
+ "<extra_id_60>",
890
+ "<extra_id_61>",
891
+ "<extra_id_62>",
892
+ "<extra_id_63>",
893
+ "<extra_id_64>",
894
+ "<extra_id_65>",
895
+ "<extra_id_66>",
896
+ "<extra_id_67>",
897
+ "<extra_id_68>",
898
+ "<extra_id_69>",
899
+ "<extra_id_70>",
900
+ "<extra_id_71>",
901
+ "<extra_id_72>",
902
+ "<extra_id_73>",
903
+ "<extra_id_74>",
904
+ "<extra_id_75>",
905
+ "<extra_id_76>",
906
+ "<extra_id_77>",
907
+ "<extra_id_78>",
908
+ "<extra_id_79>",
909
+ "<extra_id_80>",
910
+ "<extra_id_81>",
911
+ "<extra_id_82>",
912
+ "<extra_id_83>",
913
+ "<extra_id_84>",
914
+ "<extra_id_85>",
915
+ "<extra_id_86>",
916
+ "<extra_id_87>",
917
+ "<extra_id_88>",
918
+ "<extra_id_89>",
919
+ "<extra_id_90>",
920
+ "<extra_id_91>",
921
+ "<extra_id_92>",
922
+ "<extra_id_93>",
923
+ "<extra_id_94>",
924
+ "<extra_id_95>",
925
+ "<extra_id_96>",
926
+ "<extra_id_97>",
927
+ "<extra_id_98>",
928
+ "<extra_id_99>"
929
+ ],
930
+ "clean_up_tokenization_spaces": true,
931
+ "eos_token": "</s>",
932
+ "extra_ids": 100,
933
+ "model_max_length": 1000000000000000019884624838656,
934
+ "pad_token": "<pad>",
935
+ "tokenizer_class": "T5Tokenizer",
936
+ "unk_token": "<unk>"
937
+ }
checkpoint-13407/trainer_state.json ADDED
@@ -0,0 +1,3833 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 3.4647490978240967,
3
+ "best_model_checkpoint": "autotrain-7wlo5-redoy/checkpoint-13407",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 13407,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005594092638174088,
13
+ "grad_norm": 15.946615219116211,
14
+ "learning_rate": 7.829977628635347e-07,
15
+ "loss": 5.5984,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.011188185276348177,
20
+ "grad_norm": 8.801526069641113,
21
+ "learning_rate": 1.7151379567486951e-06,
22
+ "loss": 6.0153,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.016782277914522265,
27
+ "grad_norm": 9.853492736816406,
28
+ "learning_rate": 2.6472781506338553e-06,
29
+ "loss": 5.3835,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.022376370552696354,
34
+ "grad_norm": 7.4353346824646,
35
+ "learning_rate": 3.5794183445190157e-06,
36
+ "loss": 5.6891,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.027970463190870442,
41
+ "grad_norm": 8.300207138061523,
42
+ "learning_rate": 4.511558538404176e-06,
43
+ "loss": 5.5559,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.03356455582904453,
48
+ "grad_norm": 11.705302238464355,
49
+ "learning_rate": 5.4436987322893364e-06,
50
+ "loss": 5.285,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.039158648467218615,
55
+ "grad_norm": 7.95635461807251,
56
+ "learning_rate": 6.375838926174497e-06,
57
+ "loss": 5.4299,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.04475274110539271,
62
+ "grad_norm": 10.348878860473633,
63
+ "learning_rate": 7.307979120059657e-06,
64
+ "loss": 5.2698,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.05034683374356679,
69
+ "grad_norm": 9.935009002685547,
70
+ "learning_rate": 8.240119313944818e-06,
71
+ "loss": 4.989,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.055940926381740884,
76
+ "grad_norm": 8.788864135742188,
77
+ "learning_rate": 9.172259507829977e-06,
78
+ "loss": 4.9269,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.06153501901991497,
83
+ "grad_norm": 7.211750030517578,
84
+ "learning_rate": 1.0104399701715138e-05,
85
+ "loss": 5.2554,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.06712911165808906,
90
+ "grad_norm": 5.683162689208984,
91
+ "learning_rate": 1.1036539895600299e-05,
92
+ "loss": 4.7611,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.07272320429626315,
97
+ "grad_norm": 6.615201950073242,
98
+ "learning_rate": 1.196868008948546e-05,
99
+ "loss": 4.7114,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.07831729693443723,
104
+ "grad_norm": 6.1170454025268555,
105
+ "learning_rate": 1.2900820283370619e-05,
106
+ "loss": 4.644,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.08391138957261132,
111
+ "grad_norm": 7.986518383026123,
112
+ "learning_rate": 1.383296047725578e-05,
113
+ "loss": 4.72,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.08950548221078541,
118
+ "grad_norm": 4.3889899253845215,
119
+ "learning_rate": 1.4765100671140942e-05,
120
+ "loss": 4.7098,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.09509957484895949,
125
+ "grad_norm": 5.542903900146484,
126
+ "learning_rate": 1.56972408650261e-05,
127
+ "loss": 4.6695,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.10069366748713358,
132
+ "grad_norm": 5.111086368560791,
133
+ "learning_rate": 1.6629381058911262e-05,
134
+ "loss": 4.6613,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.10628776012530768,
139
+ "grad_norm": 4.380405902862549,
140
+ "learning_rate": 1.756152125279642e-05,
141
+ "loss": 4.5501,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.11188185276348177,
146
+ "grad_norm": 4.832024574279785,
147
+ "learning_rate": 1.849366144668158e-05,
148
+ "loss": 4.351,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.11747594540165585,
153
+ "grad_norm": 5.491298675537109,
154
+ "learning_rate": 1.9425801640566744e-05,
155
+ "loss": 4.727,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.12307003803982994,
160
+ "grad_norm": 4.571116924285889,
161
+ "learning_rate": 2.03579418344519e-05,
162
+ "loss": 4.2995,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.12866413067800403,
167
+ "grad_norm": 7.897008895874023,
168
+ "learning_rate": 2.1290082028337062e-05,
169
+ "loss": 4.4362,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.13425822331617812,
174
+ "grad_norm": 5.006298542022705,
175
+ "learning_rate": 2.2222222222222223e-05,
176
+ "loss": 4.3874,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 0.1398523159543522,
181
+ "grad_norm": 4.3377838134765625,
182
+ "learning_rate": 2.3154362416107384e-05,
183
+ "loss": 4.696,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 0.1454464085925263,
188
+ "grad_norm": 4.403193950653076,
189
+ "learning_rate": 2.4086502609992545e-05,
190
+ "loss": 4.3924,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 0.15104050123070037,
195
+ "grad_norm": 5.898466110229492,
196
+ "learning_rate": 2.5018642803877706e-05,
197
+ "loss": 4.3397,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 0.15663459386887446,
202
+ "grad_norm": 5.418743133544922,
203
+ "learning_rate": 2.5950782997762863e-05,
204
+ "loss": 4.2487,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 0.16222868650704855,
209
+ "grad_norm": 5.5393548011779785,
210
+ "learning_rate": 2.6882923191648024e-05,
211
+ "loss": 4.3862,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 0.16782277914522264,
216
+ "grad_norm": 4.636393070220947,
217
+ "learning_rate": 2.7815063385533185e-05,
218
+ "loss": 4.3367,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 0.17341687178339674,
223
+ "grad_norm": 4.599682807922363,
224
+ "learning_rate": 2.8747203579418346e-05,
225
+ "loss": 4.3078,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 0.17901096442157083,
230
+ "grad_norm": 3.6336870193481445,
231
+ "learning_rate": 2.967934377330351e-05,
232
+ "loss": 4.2695,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 0.18460505705974492,
237
+ "grad_norm": 5.366154193878174,
238
+ "learning_rate": 3.061148396718867e-05,
239
+ "loss": 4.1408,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 0.19019914969791898,
244
+ "grad_norm": 4.447692394256592,
245
+ "learning_rate": 3.1543624161073825e-05,
246
+ "loss": 4.3595,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 0.19579324233609308,
251
+ "grad_norm": 5.87070894241333,
252
+ "learning_rate": 3.247576435495899e-05,
253
+ "loss": 3.9519,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 0.20138733497426717,
258
+ "grad_norm": 4.60684871673584,
259
+ "learning_rate": 3.3407904548844146e-05,
260
+ "loss": 3.9998,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 0.20698142761244126,
265
+ "grad_norm": 5.918384552001953,
266
+ "learning_rate": 3.434004474272931e-05,
267
+ "loss": 4.3183,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 0.21257552025061535,
272
+ "grad_norm": 8.382468223571777,
273
+ "learning_rate": 3.527218493661447e-05,
274
+ "loss": 4.0922,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 0.21816961288878944,
279
+ "grad_norm": 5.649718761444092,
280
+ "learning_rate": 3.6204325130499625e-05,
281
+ "loss": 4.2682,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 0.22376370552696354,
286
+ "grad_norm": 7.753997802734375,
287
+ "learning_rate": 3.713646532438479e-05,
288
+ "loss": 3.9353,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 0.22935779816513763,
293
+ "grad_norm": 5.820228576660156,
294
+ "learning_rate": 3.8068605518269954e-05,
295
+ "loss": 4.2482,
296
+ "step": 1025
297
+ },
298
+ {
299
+ "epoch": 0.2349518908033117,
300
+ "grad_norm": 5.172187805175781,
301
+ "learning_rate": 3.900074571215511e-05,
302
+ "loss": 4.1534,
303
+ "step": 1050
304
+ },
305
+ {
306
+ "epoch": 0.24054598344148578,
307
+ "grad_norm": 4.996575355529785,
308
+ "learning_rate": 3.993288590604027e-05,
309
+ "loss": 3.9996,
310
+ "step": 1075
311
+ },
312
+ {
313
+ "epoch": 0.24614007607965988,
314
+ "grad_norm": 3.569652557373047,
315
+ "learning_rate": 4.086502609992543e-05,
316
+ "loss": 4.0917,
317
+ "step": 1100
318
+ },
319
+ {
320
+ "epoch": 0.251734168717834,
321
+ "grad_norm": 5.313923358917236,
322
+ "learning_rate": 4.179716629381059e-05,
323
+ "loss": 4.0334,
324
+ "step": 1125
325
+ },
326
+ {
327
+ "epoch": 0.25732826135600806,
328
+ "grad_norm": 4.535011291503906,
329
+ "learning_rate": 4.2729306487695754e-05,
330
+ "loss": 3.9393,
331
+ "step": 1150
332
+ },
333
+ {
334
+ "epoch": 0.2629223539941821,
335
+ "grad_norm": 4.798233509063721,
336
+ "learning_rate": 4.366144668158091e-05,
337
+ "loss": 3.7549,
338
+ "step": 1175
339
+ },
340
+ {
341
+ "epoch": 0.26851644663235624,
342
+ "grad_norm": 5.069913864135742,
343
+ "learning_rate": 4.459358687546607e-05,
344
+ "loss": 4.0816,
345
+ "step": 1200
346
+ },
347
+ {
348
+ "epoch": 0.2741105392705303,
349
+ "grad_norm": 3.90407657623291,
350
+ "learning_rate": 4.552572706935123e-05,
351
+ "loss": 4.2875,
352
+ "step": 1225
353
+ },
354
+ {
355
+ "epoch": 0.2797046319087044,
356
+ "grad_norm": 6.95057487487793,
357
+ "learning_rate": 4.645786726323639e-05,
358
+ "loss": 3.9592,
359
+ "step": 1250
360
+ },
361
+ {
362
+ "epoch": 0.2852987245468785,
363
+ "grad_norm": 4.956694602966309,
364
+ "learning_rate": 4.7390007457121555e-05,
365
+ "loss": 4.0485,
366
+ "step": 1275
367
+ },
368
+ {
369
+ "epoch": 0.2908928171850526,
370
+ "grad_norm": 4.546765327453613,
371
+ "learning_rate": 4.832214765100672e-05,
372
+ "loss": 4.1542,
373
+ "step": 1300
374
+ },
375
+ {
376
+ "epoch": 0.2964869098232267,
377
+ "grad_norm": 4.129889011383057,
378
+ "learning_rate": 4.925428784489187e-05,
379
+ "loss": 3.978,
380
+ "step": 1325
381
+ },
382
+ {
383
+ "epoch": 0.30208100246140074,
384
+ "grad_norm": 5.083676815032959,
385
+ "learning_rate": 4.9979280623238856e-05,
386
+ "loss": 3.9948,
387
+ "step": 1350
388
+ },
389
+ {
390
+ "epoch": 0.30767509509957486,
391
+ "grad_norm": 6.2106781005859375,
392
+ "learning_rate": 4.987568373943312e-05,
393
+ "loss": 3.9205,
394
+ "step": 1375
395
+ },
396
+ {
397
+ "epoch": 0.3132691877377489,
398
+ "grad_norm": 5.162880897521973,
399
+ "learning_rate": 4.977208685562739e-05,
400
+ "loss": 3.9022,
401
+ "step": 1400
402
+ },
403
+ {
404
+ "epoch": 0.31886328037592304,
405
+ "grad_norm": 4.337917804718018,
406
+ "learning_rate": 4.966848997182165e-05,
407
+ "loss": 4.0203,
408
+ "step": 1425
409
+ },
410
+ {
411
+ "epoch": 0.3244573730140971,
412
+ "grad_norm": 4.488531589508057,
413
+ "learning_rate": 4.956489308801591e-05,
414
+ "loss": 3.9798,
415
+ "step": 1450
416
+ },
417
+ {
418
+ "epoch": 0.3300514656522712,
419
+ "grad_norm": 5.045831203460693,
420
+ "learning_rate": 4.946129620421018e-05,
421
+ "loss": 3.7537,
422
+ "step": 1475
423
+ },
424
+ {
425
+ "epoch": 0.3356455582904453,
426
+ "grad_norm": 4.726805686950684,
427
+ "learning_rate": 4.935769932040444e-05,
428
+ "loss": 3.7138,
429
+ "step": 1500
430
+ },
431
+ {
432
+ "epoch": 0.34123965092861935,
433
+ "grad_norm": 3.9257240295410156,
434
+ "learning_rate": 4.9254102436598706e-05,
435
+ "loss": 3.9845,
436
+ "step": 1525
437
+ },
438
+ {
439
+ "epoch": 0.3468337435667935,
440
+ "grad_norm": 3.3794500827789307,
441
+ "learning_rate": 4.9150505552792977e-05,
442
+ "loss": 3.6542,
443
+ "step": 1550
444
+ },
445
+ {
446
+ "epoch": 0.35242783620496754,
447
+ "grad_norm": 3.749429225921631,
448
+ "learning_rate": 4.904690866898724e-05,
449
+ "loss": 3.8954,
450
+ "step": 1575
451
+ },
452
+ {
453
+ "epoch": 0.35802192884314166,
454
+ "grad_norm": 4.707284927368164,
455
+ "learning_rate": 4.89433117851815e-05,
456
+ "loss": 4.0279,
457
+ "step": 1600
458
+ },
459
+ {
460
+ "epoch": 0.3636160214813157,
461
+ "grad_norm": 4.450760364532471,
462
+ "learning_rate": 4.883971490137577e-05,
463
+ "loss": 4.0055,
464
+ "step": 1625
465
+ },
466
+ {
467
+ "epoch": 0.36921011411948984,
468
+ "grad_norm": 6.624269485473633,
469
+ "learning_rate": 4.873611801757003e-05,
470
+ "loss": 3.8822,
471
+ "step": 1650
472
+ },
473
+ {
474
+ "epoch": 0.3748042067576639,
475
+ "grad_norm": 4.564643859863281,
476
+ "learning_rate": 4.86325211337643e-05,
477
+ "loss": 3.9155,
478
+ "step": 1675
479
+ },
480
+ {
481
+ "epoch": 0.38039829939583797,
482
+ "grad_norm": 4.664186000823975,
483
+ "learning_rate": 4.852892424995856e-05,
484
+ "loss": 3.9958,
485
+ "step": 1700
486
+ },
487
+ {
488
+ "epoch": 0.3859923920340121,
489
+ "grad_norm": 3.393782377243042,
490
+ "learning_rate": 4.842532736615283e-05,
491
+ "loss": 4.1503,
492
+ "step": 1725
493
+ },
494
+ {
495
+ "epoch": 0.39158648467218615,
496
+ "grad_norm": 3.4447598457336426,
497
+ "learning_rate": 4.83217304823471e-05,
498
+ "loss": 3.763,
499
+ "step": 1750
500
+ },
501
+ {
502
+ "epoch": 0.39718057731036027,
503
+ "grad_norm": 4.141654014587402,
504
+ "learning_rate": 4.821813359854136e-05,
505
+ "loss": 3.7629,
506
+ "step": 1775
507
+ },
508
+ {
509
+ "epoch": 0.40277466994853434,
510
+ "grad_norm": 4.235133171081543,
511
+ "learning_rate": 4.8114536714735623e-05,
512
+ "loss": 3.9352,
513
+ "step": 1800
514
+ },
515
+ {
516
+ "epoch": 0.40836876258670846,
517
+ "grad_norm": 4.619142055511475,
518
+ "learning_rate": 4.801093983092989e-05,
519
+ "loss": 3.9759,
520
+ "step": 1825
521
+ },
522
+ {
523
+ "epoch": 0.4139628552248825,
524
+ "grad_norm": 4.4145121574401855,
525
+ "learning_rate": 4.790734294712415e-05,
526
+ "loss": 3.9988,
527
+ "step": 1850
528
+ },
529
+ {
530
+ "epoch": 0.41955694786305664,
531
+ "grad_norm": 4.787710666656494,
532
+ "learning_rate": 4.780374606331841e-05,
533
+ "loss": 3.5465,
534
+ "step": 1875
535
+ },
536
+ {
537
+ "epoch": 0.4251510405012307,
538
+ "grad_norm": 3.8803603649139404,
539
+ "learning_rate": 4.7700149179512683e-05,
540
+ "loss": 3.7941,
541
+ "step": 1900
542
+ },
543
+ {
544
+ "epoch": 0.43074513313940477,
545
+ "grad_norm": 3.9608070850372314,
546
+ "learning_rate": 4.759655229570695e-05,
547
+ "loss": 4.0338,
548
+ "step": 1925
549
+ },
550
+ {
551
+ "epoch": 0.4363392257775789,
552
+ "grad_norm": 5.147342681884766,
553
+ "learning_rate": 4.749295541190121e-05,
554
+ "loss": 3.8809,
555
+ "step": 1950
556
+ },
557
+ {
558
+ "epoch": 0.44193331841575295,
559
+ "grad_norm": 4.146068096160889,
560
+ "learning_rate": 4.738935852809548e-05,
561
+ "loss": 3.8297,
562
+ "step": 1975
563
+ },
564
+ {
565
+ "epoch": 0.44752741105392707,
566
+ "grad_norm": 6.210178852081299,
567
+ "learning_rate": 4.7285761644289744e-05,
568
+ "loss": 3.8976,
569
+ "step": 2000
570
+ },
571
+ {
572
+ "epoch": 0.45312150369210114,
573
+ "grad_norm": 4.273529052734375,
574
+ "learning_rate": 4.718216476048401e-05,
575
+ "loss": 3.6413,
576
+ "step": 2025
577
+ },
578
+ {
579
+ "epoch": 0.45871559633027525,
580
+ "grad_norm": 3.2547686100006104,
581
+ "learning_rate": 4.707856787667827e-05,
582
+ "loss": 3.6965,
583
+ "step": 2050
584
+ },
585
+ {
586
+ "epoch": 0.4643096889684493,
587
+ "grad_norm": 4.045331954956055,
588
+ "learning_rate": 4.6974970992872534e-05,
589
+ "loss": 3.7538,
590
+ "step": 2075
591
+ },
592
+ {
593
+ "epoch": 0.4699037816066234,
594
+ "grad_norm": 4.639896392822266,
595
+ "learning_rate": 4.6871374109066804e-05,
596
+ "loss": 3.7776,
597
+ "step": 2100
598
+ },
599
+ {
600
+ "epoch": 0.4754978742447975,
601
+ "grad_norm": 3.88511323928833,
602
+ "learning_rate": 4.676777722526107e-05,
603
+ "loss": 3.8255,
604
+ "step": 2125
605
+ },
606
+ {
607
+ "epoch": 0.48109196688297157,
608
+ "grad_norm": 3.2007739543914795,
609
+ "learning_rate": 4.666418034145533e-05,
610
+ "loss": 3.9731,
611
+ "step": 2150
612
+ },
613
+ {
614
+ "epoch": 0.4866860595211457,
615
+ "grad_norm": 3.954047679901123,
616
+ "learning_rate": 4.65605834576496e-05,
617
+ "loss": 3.7451,
618
+ "step": 2175
619
+ },
620
+ {
621
+ "epoch": 0.49228015215931975,
622
+ "grad_norm": 5.1558427810668945,
623
+ "learning_rate": 4.6456986573843864e-05,
624
+ "loss": 3.6939,
625
+ "step": 2200
626
+ },
627
+ {
628
+ "epoch": 0.49787424479749387,
629
+ "grad_norm": 3.4334099292755127,
630
+ "learning_rate": 4.635338969003813e-05,
631
+ "loss": 3.7097,
632
+ "step": 2225
633
+ },
634
+ {
635
+ "epoch": 0.503468337435668,
636
+ "grad_norm": 5.569153785705566,
637
+ "learning_rate": 4.624979280623239e-05,
638
+ "loss": 3.6046,
639
+ "step": 2250
640
+ },
641
+ {
642
+ "epoch": 0.509062430073842,
643
+ "grad_norm": 5.0056867599487305,
644
+ "learning_rate": 4.6146195922426654e-05,
645
+ "loss": 3.6809,
646
+ "step": 2275
647
+ },
648
+ {
649
+ "epoch": 0.5146565227120161,
650
+ "grad_norm": 4.939054012298584,
651
+ "learning_rate": 4.6042599038620924e-05,
652
+ "loss": 3.6925,
653
+ "step": 2300
654
+ },
655
+ {
656
+ "epoch": 0.5202506153501902,
657
+ "grad_norm": 3.5806541442871094,
658
+ "learning_rate": 4.593900215481519e-05,
659
+ "loss": 3.9425,
660
+ "step": 2325
661
+ },
662
+ {
663
+ "epoch": 0.5258447079883642,
664
+ "grad_norm": 3.3533098697662354,
665
+ "learning_rate": 4.583540527100945e-05,
666
+ "loss": 3.879,
667
+ "step": 2350
668
+ },
669
+ {
670
+ "epoch": 0.5314388006265384,
671
+ "grad_norm": 3.3970768451690674,
672
+ "learning_rate": 4.573180838720372e-05,
673
+ "loss": 3.7817,
674
+ "step": 2375
675
+ },
676
+ {
677
+ "epoch": 0.5370328932647125,
678
+ "grad_norm": 4.292008399963379,
679
+ "learning_rate": 4.562821150339798e-05,
680
+ "loss": 3.8217,
681
+ "step": 2400
682
+ },
683
+ {
684
+ "epoch": 0.5426269859028866,
685
+ "grad_norm": 3.9546926021575928,
686
+ "learning_rate": 4.552461461959224e-05,
687
+ "loss": 3.8219,
688
+ "step": 2425
689
+ },
690
+ {
691
+ "epoch": 0.5482210785410606,
692
+ "grad_norm": 4.9536638259887695,
693
+ "learning_rate": 4.542101773578651e-05,
694
+ "loss": 3.7819,
695
+ "step": 2450
696
+ },
697
+ {
698
+ "epoch": 0.5538151711792347,
699
+ "grad_norm": 4.847626686096191,
700
+ "learning_rate": 4.5317420851980774e-05,
701
+ "loss": 3.717,
702
+ "step": 2475
703
+ },
704
+ {
705
+ "epoch": 0.5594092638174089,
706
+ "grad_norm": 3.5468544960021973,
707
+ "learning_rate": 4.521382396817504e-05,
708
+ "loss": 3.8753,
709
+ "step": 2500
710
+ },
711
+ {
712
+ "epoch": 0.5650033564555829,
713
+ "grad_norm": 3.327033042907715,
714
+ "learning_rate": 4.511022708436931e-05,
715
+ "loss": 4.059,
716
+ "step": 2525
717
+ },
718
+ {
719
+ "epoch": 0.570597449093757,
720
+ "grad_norm": 3.7893993854522705,
721
+ "learning_rate": 4.500663020056357e-05,
722
+ "loss": 3.7822,
723
+ "step": 2550
724
+ },
725
+ {
726
+ "epoch": 0.5761915417319311,
727
+ "grad_norm": 4.392114162445068,
728
+ "learning_rate": 4.4903033316757834e-05,
729
+ "loss": 3.6745,
730
+ "step": 2575
731
+ },
732
+ {
733
+ "epoch": 0.5817856343701052,
734
+ "grad_norm": 3.785527467727661,
735
+ "learning_rate": 4.47994364329521e-05,
736
+ "loss": 3.6539,
737
+ "step": 2600
738
+ },
739
+ {
740
+ "epoch": 0.5873797270082792,
741
+ "grad_norm": 3.4213197231292725,
742
+ "learning_rate": 4.469583954914636e-05,
743
+ "loss": 3.5215,
744
+ "step": 2625
745
+ },
746
+ {
747
+ "epoch": 0.5929738196464533,
748
+ "grad_norm": 5.995355606079102,
749
+ "learning_rate": 4.459224266534063e-05,
750
+ "loss": 3.7344,
751
+ "step": 2650
752
+ },
753
+ {
754
+ "epoch": 0.5985679122846275,
755
+ "grad_norm": 4.48270320892334,
756
+ "learning_rate": 4.4488645781534894e-05,
757
+ "loss": 3.6152,
758
+ "step": 2675
759
+ },
760
+ {
761
+ "epoch": 0.6041620049228015,
762
+ "grad_norm": 4.63301420211792,
763
+ "learning_rate": 4.438504889772916e-05,
764
+ "loss": 3.9936,
765
+ "step": 2700
766
+ },
767
+ {
768
+ "epoch": 0.6097560975609756,
769
+ "grad_norm": 3.6644222736358643,
770
+ "learning_rate": 4.428145201392343e-05,
771
+ "loss": 4.0202,
772
+ "step": 2725
773
+ },
774
+ {
775
+ "epoch": 0.6153501901991497,
776
+ "grad_norm": 4.178006649017334,
777
+ "learning_rate": 4.417785513011769e-05,
778
+ "loss": 3.7906,
779
+ "step": 2750
780
+ },
781
+ {
782
+ "epoch": 0.6209442828373238,
783
+ "grad_norm": 5.096932888031006,
784
+ "learning_rate": 4.4074258246311954e-05,
785
+ "loss": 3.9282,
786
+ "step": 2775
787
+ },
788
+ {
789
+ "epoch": 0.6265383754754978,
790
+ "grad_norm": 5.241128921508789,
791
+ "learning_rate": 4.397066136250622e-05,
792
+ "loss": 3.7666,
793
+ "step": 2800
794
+ },
795
+ {
796
+ "epoch": 0.632132468113672,
797
+ "grad_norm": 3.220507860183716,
798
+ "learning_rate": 4.386706447870048e-05,
799
+ "loss": 3.7387,
800
+ "step": 2825
801
+ },
802
+ {
803
+ "epoch": 0.6377265607518461,
804
+ "grad_norm": 4.53447961807251,
805
+ "learning_rate": 4.3763467594894744e-05,
806
+ "loss": 3.6855,
807
+ "step": 2850
808
+ },
809
+ {
810
+ "epoch": 0.6433206533900201,
811
+ "grad_norm": 3.06878924369812,
812
+ "learning_rate": 4.3659870711089014e-05,
813
+ "loss": 3.6931,
814
+ "step": 2875
815
+ },
816
+ {
817
+ "epoch": 0.6489147460281942,
818
+ "grad_norm": 3.6970791816711426,
819
+ "learning_rate": 4.355627382728328e-05,
820
+ "loss": 3.8291,
821
+ "step": 2900
822
+ },
823
+ {
824
+ "epoch": 0.6545088386663683,
825
+ "grad_norm": 4.492339611053467,
826
+ "learning_rate": 4.345267694347754e-05,
827
+ "loss": 3.7108,
828
+ "step": 2925
829
+ },
830
+ {
831
+ "epoch": 0.6601029313045425,
832
+ "grad_norm": 2.8937885761260986,
833
+ "learning_rate": 4.334908005967181e-05,
834
+ "loss": 3.671,
835
+ "step": 2950
836
+ },
837
+ {
838
+ "epoch": 0.6656970239427165,
839
+ "grad_norm": 4.814384937286377,
840
+ "learning_rate": 4.324548317586607e-05,
841
+ "loss": 3.7018,
842
+ "step": 2975
843
+ },
844
+ {
845
+ "epoch": 0.6712911165808906,
846
+ "grad_norm": 4.3328118324279785,
847
+ "learning_rate": 4.314188629206034e-05,
848
+ "loss": 3.8001,
849
+ "step": 3000
850
+ },
851
+ {
852
+ "epoch": 0.6768852092190647,
853
+ "grad_norm": 3.7791872024536133,
854
+ "learning_rate": 4.30382894082546e-05,
855
+ "loss": 3.9265,
856
+ "step": 3025
857
+ },
858
+ {
859
+ "epoch": 0.6824793018572387,
860
+ "grad_norm": 3.6916797161102295,
861
+ "learning_rate": 4.2934692524448864e-05,
862
+ "loss": 3.9498,
863
+ "step": 3050
864
+ },
865
+ {
866
+ "epoch": 0.6880733944954128,
867
+ "grad_norm": 4.374492168426514,
868
+ "learning_rate": 4.2831095640643134e-05,
869
+ "loss": 3.7367,
870
+ "step": 3075
871
+ },
872
+ {
873
+ "epoch": 0.693667487133587,
874
+ "grad_norm": 4.071556568145752,
875
+ "learning_rate": 4.27274987568374e-05,
876
+ "loss": 3.8503,
877
+ "step": 3100
878
+ },
879
+ {
880
+ "epoch": 0.6992615797717611,
881
+ "grad_norm": 4.232872486114502,
882
+ "learning_rate": 4.262390187303166e-05,
883
+ "loss": 3.7388,
884
+ "step": 3125
885
+ },
886
+ {
887
+ "epoch": 0.7048556724099351,
888
+ "grad_norm": 4.348662376403809,
889
+ "learning_rate": 4.252030498922593e-05,
890
+ "loss": 3.6244,
891
+ "step": 3150
892
+ },
893
+ {
894
+ "epoch": 0.7104497650481092,
895
+ "grad_norm": 3.3647055625915527,
896
+ "learning_rate": 4.241670810542019e-05,
897
+ "loss": 3.8258,
898
+ "step": 3175
899
+ },
900
+ {
901
+ "epoch": 0.7160438576862833,
902
+ "grad_norm": 4.311660289764404,
903
+ "learning_rate": 4.231311122161445e-05,
904
+ "loss": 3.9008,
905
+ "step": 3200
906
+ },
907
+ {
908
+ "epoch": 0.7216379503244573,
909
+ "grad_norm": 5.2813615798950195,
910
+ "learning_rate": 4.220951433780872e-05,
911
+ "loss": 3.6071,
912
+ "step": 3225
913
+ },
914
+ {
915
+ "epoch": 0.7272320429626314,
916
+ "grad_norm": 2.818814277648926,
917
+ "learning_rate": 4.2105917454002984e-05,
918
+ "loss": 3.6422,
919
+ "step": 3250
920
+ },
921
+ {
922
+ "epoch": 0.7328261356008056,
923
+ "grad_norm": 4.551268577575684,
924
+ "learning_rate": 4.200232057019725e-05,
925
+ "loss": 3.6364,
926
+ "step": 3275
927
+ },
928
+ {
929
+ "epoch": 0.7384202282389797,
930
+ "grad_norm": 3.7793288230895996,
931
+ "learning_rate": 4.189872368639152e-05,
932
+ "loss": 3.9139,
933
+ "step": 3300
934
+ },
935
+ {
936
+ "epoch": 0.7440143208771537,
937
+ "grad_norm": 2.9679417610168457,
938
+ "learning_rate": 4.179512680258578e-05,
939
+ "loss": 3.8005,
940
+ "step": 3325
941
+ },
942
+ {
943
+ "epoch": 0.7496084135153278,
944
+ "grad_norm": 3.486783027648926,
945
+ "learning_rate": 4.1691529918780045e-05,
946
+ "loss": 3.7324,
947
+ "step": 3350
948
+ },
949
+ {
950
+ "epoch": 0.7552025061535019,
951
+ "grad_norm": 4.470802307128906,
952
+ "learning_rate": 4.158793303497431e-05,
953
+ "loss": 3.6514,
954
+ "step": 3375
955
+ },
956
+ {
957
+ "epoch": 0.7607965987916759,
958
+ "grad_norm": 3.055450201034546,
959
+ "learning_rate": 4.148433615116857e-05,
960
+ "loss": 3.6617,
961
+ "step": 3400
962
+ },
963
+ {
964
+ "epoch": 0.7663906914298501,
965
+ "grad_norm": 6.0705413818359375,
966
+ "learning_rate": 4.138073926736284e-05,
967
+ "loss": 3.8084,
968
+ "step": 3425
969
+ },
970
+ {
971
+ "epoch": 0.7719847840680242,
972
+ "grad_norm": 2.6818230152130127,
973
+ "learning_rate": 4.1277142383557105e-05,
974
+ "loss": 3.7827,
975
+ "step": 3450
976
+ },
977
+ {
978
+ "epoch": 0.7775788767061983,
979
+ "grad_norm": 4.817789077758789,
980
+ "learning_rate": 4.117354549975137e-05,
981
+ "loss": 3.5952,
982
+ "step": 3475
983
+ },
984
+ {
985
+ "epoch": 0.7831729693443723,
986
+ "grad_norm": 5.3004560470581055,
987
+ "learning_rate": 4.106994861594564e-05,
988
+ "loss": 3.8184,
989
+ "step": 3500
990
+ },
991
+ {
992
+ "epoch": 0.7887670619825464,
993
+ "grad_norm": 4.656194686889648,
994
+ "learning_rate": 4.09663517321399e-05,
995
+ "loss": 3.6244,
996
+ "step": 3525
997
+ },
998
+ {
999
+ "epoch": 0.7943611546207205,
1000
+ "grad_norm": 2.6837692260742188,
1001
+ "learning_rate": 4.0862754848334165e-05,
1002
+ "loss": 3.727,
1003
+ "step": 3550
1004
+ },
1005
+ {
1006
+ "epoch": 0.7999552472588946,
1007
+ "grad_norm": 3.2601094245910645,
1008
+ "learning_rate": 4.075915796452843e-05,
1009
+ "loss": 3.7806,
1010
+ "step": 3575
1011
+ },
1012
+ {
1013
+ "epoch": 0.8055493398970687,
1014
+ "grad_norm": 4.067142963409424,
1015
+ "learning_rate": 4.065556108072269e-05,
1016
+ "loss": 3.8446,
1017
+ "step": 3600
1018
+ },
1019
+ {
1020
+ "epoch": 0.8111434325352428,
1021
+ "grad_norm": 2.995216131210327,
1022
+ "learning_rate": 4.055196419691696e-05,
1023
+ "loss": 3.6184,
1024
+ "step": 3625
1025
+ },
1026
+ {
1027
+ "epoch": 0.8167375251734169,
1028
+ "grad_norm": 3.2228896617889404,
1029
+ "learning_rate": 4.0448367313111225e-05,
1030
+ "loss": 3.7542,
1031
+ "step": 3650
1032
+ },
1033
+ {
1034
+ "epoch": 0.8223316178115909,
1035
+ "grad_norm": 4.756741523742676,
1036
+ "learning_rate": 4.034477042930549e-05,
1037
+ "loss": 3.6299,
1038
+ "step": 3675
1039
+ },
1040
+ {
1041
+ "epoch": 0.827925710449765,
1042
+ "grad_norm": 4.418304920196533,
1043
+ "learning_rate": 4.024117354549976e-05,
1044
+ "loss": 3.5213,
1045
+ "step": 3700
1046
+ },
1047
+ {
1048
+ "epoch": 0.8335198030879392,
1049
+ "grad_norm": 3.9183456897735596,
1050
+ "learning_rate": 4.013757666169402e-05,
1051
+ "loss": 3.8331,
1052
+ "step": 3725
1053
+ },
1054
+ {
1055
+ "epoch": 0.8391138957261133,
1056
+ "grad_norm": 3.417017698287964,
1057
+ "learning_rate": 4.003397977788828e-05,
1058
+ "loss": 3.6406,
1059
+ "step": 3750
1060
+ },
1061
+ {
1062
+ "epoch": 0.8447079883642873,
1063
+ "grad_norm": 3.881243944168091,
1064
+ "learning_rate": 3.993038289408255e-05,
1065
+ "loss": 3.678,
1066
+ "step": 3775
1067
+ },
1068
+ {
1069
+ "epoch": 0.8503020810024614,
1070
+ "grad_norm": 4.177762508392334,
1071
+ "learning_rate": 3.982678601027681e-05,
1072
+ "loss": 3.6046,
1073
+ "step": 3800
1074
+ },
1075
+ {
1076
+ "epoch": 0.8558961736406355,
1077
+ "grad_norm": 5.575811386108398,
1078
+ "learning_rate": 3.9723189126471075e-05,
1079
+ "loss": 3.7233,
1080
+ "step": 3825
1081
+ },
1082
+ {
1083
+ "epoch": 0.8614902662788095,
1084
+ "grad_norm": 3.4432308673858643,
1085
+ "learning_rate": 3.9619592242665345e-05,
1086
+ "loss": 3.6772,
1087
+ "step": 3850
1088
+ },
1089
+ {
1090
+ "epoch": 0.8670843589169837,
1091
+ "grad_norm": 5.3436970710754395,
1092
+ "learning_rate": 3.951599535885961e-05,
1093
+ "loss": 3.8025,
1094
+ "step": 3875
1095
+ },
1096
+ {
1097
+ "epoch": 0.8726784515551578,
1098
+ "grad_norm": 4.009333610534668,
1099
+ "learning_rate": 3.941239847505387e-05,
1100
+ "loss": 3.6119,
1101
+ "step": 3900
1102
+ },
1103
+ {
1104
+ "epoch": 0.8782725441933319,
1105
+ "grad_norm": 4.884556770324707,
1106
+ "learning_rate": 3.9308801591248135e-05,
1107
+ "loss": 3.626,
1108
+ "step": 3925
1109
+ },
1110
+ {
1111
+ "epoch": 0.8838666368315059,
1112
+ "grad_norm": 3.6885242462158203,
1113
+ "learning_rate": 3.92052047074424e-05,
1114
+ "loss": 3.6389,
1115
+ "step": 3950
1116
+ },
1117
+ {
1118
+ "epoch": 0.88946072946968,
1119
+ "grad_norm": 4.574316501617432,
1120
+ "learning_rate": 3.910160782363667e-05,
1121
+ "loss": 3.876,
1122
+ "step": 3975
1123
+ },
1124
+ {
1125
+ "epoch": 0.8950548221078541,
1126
+ "grad_norm": 4.40355110168457,
1127
+ "learning_rate": 3.899801093983093e-05,
1128
+ "loss": 3.8318,
1129
+ "step": 4000
1130
+ },
1131
+ {
1132
+ "epoch": 0.9006489147460282,
1133
+ "grad_norm": 2.6952688694000244,
1134
+ "learning_rate": 3.8894414056025195e-05,
1135
+ "loss": 3.7394,
1136
+ "step": 4025
1137
+ },
1138
+ {
1139
+ "epoch": 0.9062430073842023,
1140
+ "grad_norm": 3.1016721725463867,
1141
+ "learning_rate": 3.8790817172219465e-05,
1142
+ "loss": 3.6001,
1143
+ "step": 4050
1144
+ },
1145
+ {
1146
+ "epoch": 0.9118371000223764,
1147
+ "grad_norm": 3.096198797225952,
1148
+ "learning_rate": 3.868722028841373e-05,
1149
+ "loss": 3.6943,
1150
+ "step": 4075
1151
+ },
1152
+ {
1153
+ "epoch": 0.9174311926605505,
1154
+ "grad_norm": 3.518115758895874,
1155
+ "learning_rate": 3.858362340460799e-05,
1156
+ "loss": 3.5888,
1157
+ "step": 4100
1158
+ },
1159
+ {
1160
+ "epoch": 0.9230252852987245,
1161
+ "grad_norm": 3.107680320739746,
1162
+ "learning_rate": 3.8480026520802255e-05,
1163
+ "loss": 3.6696,
1164
+ "step": 4125
1165
+ },
1166
+ {
1167
+ "epoch": 0.9286193779368986,
1168
+ "grad_norm": 3.326657295227051,
1169
+ "learning_rate": 3.837642963699652e-05,
1170
+ "loss": 4.0632,
1171
+ "step": 4150
1172
+ },
1173
+ {
1174
+ "epoch": 0.9342134705750728,
1175
+ "grad_norm": 4.73927116394043,
1176
+ "learning_rate": 3.827283275319078e-05,
1177
+ "loss": 3.6437,
1178
+ "step": 4175
1179
+ },
1180
+ {
1181
+ "epoch": 0.9398075632132468,
1182
+ "grad_norm": 4.9470696449279785,
1183
+ "learning_rate": 3.816923586938505e-05,
1184
+ "loss": 3.5825,
1185
+ "step": 4200
1186
+ },
1187
+ {
1188
+ "epoch": 0.9454016558514209,
1189
+ "grad_norm": 3.3303074836730957,
1190
+ "learning_rate": 3.8065638985579315e-05,
1191
+ "loss": 3.6707,
1192
+ "step": 4225
1193
+ },
1194
+ {
1195
+ "epoch": 0.950995748489595,
1196
+ "grad_norm": 4.487314701080322,
1197
+ "learning_rate": 3.796204210177358e-05,
1198
+ "loss": 3.6683,
1199
+ "step": 4250
1200
+ },
1201
+ {
1202
+ "epoch": 0.9565898411277691,
1203
+ "grad_norm": 3.78049635887146,
1204
+ "learning_rate": 3.785844521796785e-05,
1205
+ "loss": 3.6519,
1206
+ "step": 4275
1207
+ },
1208
+ {
1209
+ "epoch": 0.9621839337659431,
1210
+ "grad_norm": 3.871126651763916,
1211
+ "learning_rate": 3.775484833416211e-05,
1212
+ "loss": 3.6496,
1213
+ "step": 4300
1214
+ },
1215
+ {
1216
+ "epoch": 0.9677780264041173,
1217
+ "grad_norm": 3.867600440979004,
1218
+ "learning_rate": 3.7651251450356375e-05,
1219
+ "loss": 3.8023,
1220
+ "step": 4325
1221
+ },
1222
+ {
1223
+ "epoch": 0.9733721190422914,
1224
+ "grad_norm": 3.31608510017395,
1225
+ "learning_rate": 3.754765456655064e-05,
1226
+ "loss": 3.6372,
1227
+ "step": 4350
1228
+ },
1229
+ {
1230
+ "epoch": 0.9789662116804654,
1231
+ "grad_norm": 3.9523160457611084,
1232
+ "learning_rate": 3.74440576827449e-05,
1233
+ "loss": 3.5951,
1234
+ "step": 4375
1235
+ },
1236
+ {
1237
+ "epoch": 0.9845603043186395,
1238
+ "grad_norm": 3.3369760513305664,
1239
+ "learning_rate": 3.734046079893917e-05,
1240
+ "loss": 3.6869,
1241
+ "step": 4400
1242
+ },
1243
+ {
1244
+ "epoch": 0.9901543969568136,
1245
+ "grad_norm": 4.632723808288574,
1246
+ "learning_rate": 3.7236863915133435e-05,
1247
+ "loss": 3.7278,
1248
+ "step": 4425
1249
+ },
1250
+ {
1251
+ "epoch": 0.9957484895949877,
1252
+ "grad_norm": 4.139824390411377,
1253
+ "learning_rate": 3.71332670313277e-05,
1254
+ "loss": 3.5373,
1255
+ "step": 4450
1256
+ },
1257
+ {
1258
+ "epoch": 1.0,
1259
+ "eval_gen_len": 17.5264,
1260
+ "eval_loss": 3.545018434524536,
1261
+ "eval_rouge1": 34.5248,
1262
+ "eval_rouge2": 14.772,
1263
+ "eval_rougeL": 33.5058,
1264
+ "eval_rougeLsum": 33.6413,
1265
+ "eval_runtime": 119.0687,
1266
+ "eval_samples_per_second": 9.381,
1267
+ "eval_steps_per_second": 2.352,
1268
+ "step": 4469
1269
+ },
1270
+ {
1271
+ "epoch": 1.0013425822331619,
1272
+ "grad_norm": 3.558868885040283,
1273
+ "learning_rate": 3.703381402287419e-05,
1274
+ "loss": 3.6502,
1275
+ "step": 4475
1276
+ },
1277
+ {
1278
+ "epoch": 1.006936674871336,
1279
+ "grad_norm": 3.240208387374878,
1280
+ "learning_rate": 3.693021713906846e-05,
1281
+ "loss": 3.5634,
1282
+ "step": 4500
1283
+ },
1284
+ {
1285
+ "epoch": 1.0125307675095099,
1286
+ "grad_norm": 3.541520118713379,
1287
+ "learning_rate": 3.682662025526272e-05,
1288
+ "loss": 3.7435,
1289
+ "step": 4525
1290
+ },
1291
+ {
1292
+ "epoch": 1.018124860147684,
1293
+ "grad_norm": 3.0394344329833984,
1294
+ "learning_rate": 3.6723023371456985e-05,
1295
+ "loss": 3.458,
1296
+ "step": 4550
1297
+ },
1298
+ {
1299
+ "epoch": 1.0237189527858581,
1300
+ "grad_norm": 5.072175025939941,
1301
+ "learning_rate": 3.6619426487651255e-05,
1302
+ "loss": 3.3515,
1303
+ "step": 4575
1304
+ },
1305
+ {
1306
+ "epoch": 1.0293130454240322,
1307
+ "grad_norm": 7.710240840911865,
1308
+ "learning_rate": 3.651582960384552e-05,
1309
+ "loss": 3.6072,
1310
+ "step": 4600
1311
+ },
1312
+ {
1313
+ "epoch": 1.0349071380622064,
1314
+ "grad_norm": 3.927171468734741,
1315
+ "learning_rate": 3.641223272003978e-05,
1316
+ "loss": 3.6378,
1317
+ "step": 4625
1318
+ },
1319
+ {
1320
+ "epoch": 1.0405012307003805,
1321
+ "grad_norm": 3.733412504196167,
1322
+ "learning_rate": 3.6308635836234045e-05,
1323
+ "loss": 3.7716,
1324
+ "step": 4650
1325
+ },
1326
+ {
1327
+ "epoch": 1.0460953233385544,
1328
+ "grad_norm": 3.23052978515625,
1329
+ "learning_rate": 3.620503895242831e-05,
1330
+ "loss": 3.5631,
1331
+ "step": 4675
1332
+ },
1333
+ {
1334
+ "epoch": 1.0516894159767285,
1335
+ "grad_norm": 3.7507712841033936,
1336
+ "learning_rate": 3.610144206862258e-05,
1337
+ "loss": 3.5822,
1338
+ "step": 4700
1339
+ },
1340
+ {
1341
+ "epoch": 1.0572835086149026,
1342
+ "grad_norm": 2.8628463745117188,
1343
+ "learning_rate": 3.599784518481684e-05,
1344
+ "loss": 3.5251,
1345
+ "step": 4725
1346
+ },
1347
+ {
1348
+ "epoch": 1.0628776012530767,
1349
+ "grad_norm": 3.47662353515625,
1350
+ "learning_rate": 3.5894248301011105e-05,
1351
+ "loss": 3.4988,
1352
+ "step": 4750
1353
+ },
1354
+ {
1355
+ "epoch": 1.0684716938912509,
1356
+ "grad_norm": 6.121307373046875,
1357
+ "learning_rate": 3.5790651417205375e-05,
1358
+ "loss": 3.3226,
1359
+ "step": 4775
1360
+ },
1361
+ {
1362
+ "epoch": 1.074065786529425,
1363
+ "grad_norm": 3.7090487480163574,
1364
+ "learning_rate": 3.568705453339964e-05,
1365
+ "loss": 3.7922,
1366
+ "step": 4800
1367
+ },
1368
+ {
1369
+ "epoch": 1.079659879167599,
1370
+ "grad_norm": 3.2307398319244385,
1371
+ "learning_rate": 3.55834576495939e-05,
1372
+ "loss": 3.4743,
1373
+ "step": 4825
1374
+ },
1375
+ {
1376
+ "epoch": 1.085253971805773,
1377
+ "grad_norm": 3.6718027591705322,
1378
+ "learning_rate": 3.5479860765788165e-05,
1379
+ "loss": 3.63,
1380
+ "step": 4850
1381
+ },
1382
+ {
1383
+ "epoch": 1.090848064443947,
1384
+ "grad_norm": 5.188362121582031,
1385
+ "learning_rate": 3.537626388198243e-05,
1386
+ "loss": 3.4701,
1387
+ "step": 4875
1388
+ },
1389
+ {
1390
+ "epoch": 1.0964421570821212,
1391
+ "grad_norm": 3.4776225090026855,
1392
+ "learning_rate": 3.52726669981767e-05,
1393
+ "loss": 3.3925,
1394
+ "step": 4900
1395
+ },
1396
+ {
1397
+ "epoch": 1.1020362497202953,
1398
+ "grad_norm": 5.021528244018555,
1399
+ "learning_rate": 3.516907011437096e-05,
1400
+ "loss": 3.837,
1401
+ "step": 4925
1402
+ },
1403
+ {
1404
+ "epoch": 1.1076303423584695,
1405
+ "grad_norm": 4.035079002380371,
1406
+ "learning_rate": 3.5065473230565226e-05,
1407
+ "loss": 3.6109,
1408
+ "step": 4950
1409
+ },
1410
+ {
1411
+ "epoch": 1.1132244349966436,
1412
+ "grad_norm": 3.063847303390503,
1413
+ "learning_rate": 3.4961876346759496e-05,
1414
+ "loss": 3.551,
1415
+ "step": 4975
1416
+ },
1417
+ {
1418
+ "epoch": 1.1188185276348177,
1419
+ "grad_norm": 3.213472843170166,
1420
+ "learning_rate": 3.485827946295376e-05,
1421
+ "loss": 3.6068,
1422
+ "step": 5000
1423
+ },
1424
+ {
1425
+ "epoch": 1.1244126202729916,
1426
+ "grad_norm": 3.4713425636291504,
1427
+ "learning_rate": 3.475468257914802e-05,
1428
+ "loss": 3.6539,
1429
+ "step": 5025
1430
+ },
1431
+ {
1432
+ "epoch": 1.1300067129111657,
1433
+ "grad_norm": 4.657286643981934,
1434
+ "learning_rate": 3.4651085695342286e-05,
1435
+ "loss": 3.5507,
1436
+ "step": 5050
1437
+ },
1438
+ {
1439
+ "epoch": 1.1356008055493398,
1440
+ "grad_norm": 7.0956501960754395,
1441
+ "learning_rate": 3.454748881153655e-05,
1442
+ "loss": 3.6902,
1443
+ "step": 5075
1444
+ },
1445
+ {
1446
+ "epoch": 1.141194898187514,
1447
+ "grad_norm": 3.1084699630737305,
1448
+ "learning_rate": 3.444389192773081e-05,
1449
+ "loss": 3.5052,
1450
+ "step": 5100
1451
+ },
1452
+ {
1453
+ "epoch": 1.146788990825688,
1454
+ "grad_norm": 3.051090955734253,
1455
+ "learning_rate": 3.434029504392508e-05,
1456
+ "loss": 3.7453,
1457
+ "step": 5125
1458
+ },
1459
+ {
1460
+ "epoch": 1.1523830834638622,
1461
+ "grad_norm": 3.2541403770446777,
1462
+ "learning_rate": 3.4236698160119346e-05,
1463
+ "loss": 3.6949,
1464
+ "step": 5150
1465
+ },
1466
+ {
1467
+ "epoch": 1.1579771761020363,
1468
+ "grad_norm": 3.9791665077209473,
1469
+ "learning_rate": 3.413310127631361e-05,
1470
+ "loss": 3.5053,
1471
+ "step": 5175
1472
+ },
1473
+ {
1474
+ "epoch": 1.1635712687402102,
1475
+ "grad_norm": 3.310248374938965,
1476
+ "learning_rate": 3.402950439250788e-05,
1477
+ "loss": 3.5689,
1478
+ "step": 5200
1479
+ },
1480
+ {
1481
+ "epoch": 1.1691653613783843,
1482
+ "grad_norm": 3.229642868041992,
1483
+ "learning_rate": 3.3925907508702136e-05,
1484
+ "loss": 3.7489,
1485
+ "step": 5225
1486
+ },
1487
+ {
1488
+ "epoch": 1.1747594540165585,
1489
+ "grad_norm": 2.984686851501465,
1490
+ "learning_rate": 3.3822310624896406e-05,
1491
+ "loss": 3.4897,
1492
+ "step": 5250
1493
+ },
1494
+ {
1495
+ "epoch": 1.1803535466547326,
1496
+ "grad_norm": 4.77843713760376,
1497
+ "learning_rate": 3.371871374109067e-05,
1498
+ "loss": 3.4946,
1499
+ "step": 5275
1500
+ },
1501
+ {
1502
+ "epoch": 1.1859476392929067,
1503
+ "grad_norm": 6.428030014038086,
1504
+ "learning_rate": 3.361511685728493e-05,
1505
+ "loss": 3.5011,
1506
+ "step": 5300
1507
+ },
1508
+ {
1509
+ "epoch": 1.1915417319310808,
1510
+ "grad_norm": 3.6012275218963623,
1511
+ "learning_rate": 3.35115199734792e-05,
1512
+ "loss": 3.637,
1513
+ "step": 5325
1514
+ },
1515
+ {
1516
+ "epoch": 1.197135824569255,
1517
+ "grad_norm": 3.7239620685577393,
1518
+ "learning_rate": 3.3407923089673466e-05,
1519
+ "loss": 3.5698,
1520
+ "step": 5350
1521
+ },
1522
+ {
1523
+ "epoch": 1.2027299172074288,
1524
+ "grad_norm": 4.721301078796387,
1525
+ "learning_rate": 3.330432620586773e-05,
1526
+ "loss": 3.5866,
1527
+ "step": 5375
1528
+ },
1529
+ {
1530
+ "epoch": 1.208324009845603,
1531
+ "grad_norm": 3.069242000579834,
1532
+ "learning_rate": 3.3200729322062e-05,
1533
+ "loss": 3.5487,
1534
+ "step": 5400
1535
+ },
1536
+ {
1537
+ "epoch": 1.213918102483777,
1538
+ "grad_norm": 4.180321216583252,
1539
+ "learning_rate": 3.3097132438256256e-05,
1540
+ "loss": 3.7694,
1541
+ "step": 5425
1542
+ },
1543
+ {
1544
+ "epoch": 1.2195121951219512,
1545
+ "grad_norm": 3.361286163330078,
1546
+ "learning_rate": 3.299353555445052e-05,
1547
+ "loss": 3.5435,
1548
+ "step": 5450
1549
+ },
1550
+ {
1551
+ "epoch": 1.2251062877601253,
1552
+ "grad_norm": 4.925527095794678,
1553
+ "learning_rate": 3.288993867064479e-05,
1554
+ "loss": 3.5824,
1555
+ "step": 5475
1556
+ },
1557
+ {
1558
+ "epoch": 1.2307003803982994,
1559
+ "grad_norm": 2.9288322925567627,
1560
+ "learning_rate": 3.278634178683905e-05,
1561
+ "loss": 3.6413,
1562
+ "step": 5500
1563
+ },
1564
+ {
1565
+ "epoch": 1.2362944730364736,
1566
+ "grad_norm": 4.207690238952637,
1567
+ "learning_rate": 3.2682744903033316e-05,
1568
+ "loss": 3.6995,
1569
+ "step": 5525
1570
+ },
1571
+ {
1572
+ "epoch": 1.2418885656746474,
1573
+ "grad_norm": 4.083844184875488,
1574
+ "learning_rate": 3.2579148019227586e-05,
1575
+ "loss": 3.841,
1576
+ "step": 5550
1577
+ },
1578
+ {
1579
+ "epoch": 1.2474826583128216,
1580
+ "grad_norm": 4.099181652069092,
1581
+ "learning_rate": 3.247555113542185e-05,
1582
+ "loss": 3.5231,
1583
+ "step": 5575
1584
+ },
1585
+ {
1586
+ "epoch": 1.2530767509509957,
1587
+ "grad_norm": 4.01384162902832,
1588
+ "learning_rate": 3.237195425161611e-05,
1589
+ "loss": 3.6922,
1590
+ "step": 5600
1591
+ },
1592
+ {
1593
+ "epoch": 1.2586708435891698,
1594
+ "grad_norm": 4.1103973388671875,
1595
+ "learning_rate": 3.2268357367810376e-05,
1596
+ "loss": 3.5254,
1597
+ "step": 5625
1598
+ },
1599
+ {
1600
+ "epoch": 1.264264936227344,
1601
+ "grad_norm": 3.1536667346954346,
1602
+ "learning_rate": 3.216476048400464e-05,
1603
+ "loss": 3.3893,
1604
+ "step": 5650
1605
+ },
1606
+ {
1607
+ "epoch": 1.269859028865518,
1608
+ "grad_norm": 2.966480255126953,
1609
+ "learning_rate": 3.206116360019891e-05,
1610
+ "loss": 3.6971,
1611
+ "step": 5675
1612
+ },
1613
+ {
1614
+ "epoch": 1.2754531215036922,
1615
+ "grad_norm": 3.709972620010376,
1616
+ "learning_rate": 3.195756671639317e-05,
1617
+ "loss": 3.5895,
1618
+ "step": 5700
1619
+ },
1620
+ {
1621
+ "epoch": 1.281047214141866,
1622
+ "grad_norm": 2.565446615219116,
1623
+ "learning_rate": 3.1853969832587436e-05,
1624
+ "loss": 3.57,
1625
+ "step": 5725
1626
+ },
1627
+ {
1628
+ "epoch": 1.2866413067800404,
1629
+ "grad_norm": 4.41102933883667,
1630
+ "learning_rate": 3.1750372948781706e-05,
1631
+ "loss": 3.549,
1632
+ "step": 5750
1633
+ },
1634
+ {
1635
+ "epoch": 1.2922353994182143,
1636
+ "grad_norm": 4.337641716003418,
1637
+ "learning_rate": 3.164677606497597e-05,
1638
+ "loss": 3.6633,
1639
+ "step": 5775
1640
+ },
1641
+ {
1642
+ "epoch": 1.2978294920563884,
1643
+ "grad_norm": 3.943817138671875,
1644
+ "learning_rate": 3.1543179181170226e-05,
1645
+ "loss": 3.4278,
1646
+ "step": 5800
1647
+ },
1648
+ {
1649
+ "epoch": 1.3034235846945625,
1650
+ "grad_norm": 4.388129711151123,
1651
+ "learning_rate": 3.1439582297364496e-05,
1652
+ "loss": 3.3916,
1653
+ "step": 5825
1654
+ },
1655
+ {
1656
+ "epoch": 1.3090176773327367,
1657
+ "grad_norm": 3.093392848968506,
1658
+ "learning_rate": 3.133598541355876e-05,
1659
+ "loss": 3.6618,
1660
+ "step": 5850
1661
+ },
1662
+ {
1663
+ "epoch": 1.3146117699709108,
1664
+ "grad_norm": 4.840465068817139,
1665
+ "learning_rate": 3.123238852975302e-05,
1666
+ "loss": 3.6986,
1667
+ "step": 5875
1668
+ },
1669
+ {
1670
+ "epoch": 1.3202058626090847,
1671
+ "grad_norm": 3.259441375732422,
1672
+ "learning_rate": 3.112879164594729e-05,
1673
+ "loss": 3.7613,
1674
+ "step": 5900
1675
+ },
1676
+ {
1677
+ "epoch": 1.325799955247259,
1678
+ "grad_norm": 3.1994192600250244,
1679
+ "learning_rate": 3.1025194762141556e-05,
1680
+ "loss": 3.3959,
1681
+ "step": 5925
1682
+ },
1683
+ {
1684
+ "epoch": 1.331394047885433,
1685
+ "grad_norm": 5.499666690826416,
1686
+ "learning_rate": 3.092159787833582e-05,
1687
+ "loss": 3.5265,
1688
+ "step": 5950
1689
+ },
1690
+ {
1691
+ "epoch": 1.336988140523607,
1692
+ "grad_norm": 4.240040302276611,
1693
+ "learning_rate": 3.081800099453009e-05,
1694
+ "loss": 3.5911,
1695
+ "step": 5975
1696
+ },
1697
+ {
1698
+ "epoch": 1.3425822331617812,
1699
+ "grad_norm": 4.083096981048584,
1700
+ "learning_rate": 3.0714404110724346e-05,
1701
+ "loss": 3.6357,
1702
+ "step": 6000
1703
+ },
1704
+ {
1705
+ "epoch": 1.3481763257999553,
1706
+ "grad_norm": 4.0603203773498535,
1707
+ "learning_rate": 3.0610807226918616e-05,
1708
+ "loss": 3.277,
1709
+ "step": 6025
1710
+ },
1711
+ {
1712
+ "epoch": 1.3537704184381294,
1713
+ "grad_norm": 3.810168743133545,
1714
+ "learning_rate": 3.050721034311288e-05,
1715
+ "loss": 3.3425,
1716
+ "step": 6050
1717
+ },
1718
+ {
1719
+ "epoch": 1.3593645110763033,
1720
+ "grad_norm": 4.005222797393799,
1721
+ "learning_rate": 3.0403613459307146e-05,
1722
+ "loss": 3.4757,
1723
+ "step": 6075
1724
+ },
1725
+ {
1726
+ "epoch": 1.3649586037144776,
1727
+ "grad_norm": 4.205389499664307,
1728
+ "learning_rate": 3.030001657550141e-05,
1729
+ "loss": 3.5328,
1730
+ "step": 6100
1731
+ },
1732
+ {
1733
+ "epoch": 1.3705526963526515,
1734
+ "grad_norm": 3.154475450515747,
1735
+ "learning_rate": 3.0196419691695676e-05,
1736
+ "loss": 3.6856,
1737
+ "step": 6125
1738
+ },
1739
+ {
1740
+ "epoch": 1.3761467889908257,
1741
+ "grad_norm": 3.6082093715667725,
1742
+ "learning_rate": 3.0092822807889943e-05,
1743
+ "loss": 3.51,
1744
+ "step": 6150
1745
+ },
1746
+ {
1747
+ "epoch": 1.3817408816289998,
1748
+ "grad_norm": 4.1572651863098145,
1749
+ "learning_rate": 2.9989225924084206e-05,
1750
+ "loss": 3.3692,
1751
+ "step": 6175
1752
+ },
1753
+ {
1754
+ "epoch": 1.387334974267174,
1755
+ "grad_norm": 4.0159454345703125,
1756
+ "learning_rate": 2.9885629040278466e-05,
1757
+ "loss": 3.407,
1758
+ "step": 6200
1759
+ },
1760
+ {
1761
+ "epoch": 1.392929066905348,
1762
+ "grad_norm": 3.2002291679382324,
1763
+ "learning_rate": 2.9782032156472733e-05,
1764
+ "loss": 3.4633,
1765
+ "step": 6225
1766
+ },
1767
+ {
1768
+ "epoch": 1.398523159543522,
1769
+ "grad_norm": 3.884312629699707,
1770
+ "learning_rate": 2.9678435272667e-05,
1771
+ "loss": 3.5748,
1772
+ "step": 6250
1773
+ },
1774
+ {
1775
+ "epoch": 1.4041172521816963,
1776
+ "grad_norm": 3.8325295448303223,
1777
+ "learning_rate": 2.9574838388861263e-05,
1778
+ "loss": 3.4883,
1779
+ "step": 6275
1780
+ },
1781
+ {
1782
+ "epoch": 1.4097113448198701,
1783
+ "grad_norm": 3.0150301456451416,
1784
+ "learning_rate": 2.947124150505553e-05,
1785
+ "loss": 3.4176,
1786
+ "step": 6300
1787
+ },
1788
+ {
1789
+ "epoch": 1.4153054374580443,
1790
+ "grad_norm": 3.5329902172088623,
1791
+ "learning_rate": 2.9367644621249797e-05,
1792
+ "loss": 3.4793,
1793
+ "step": 6325
1794
+ },
1795
+ {
1796
+ "epoch": 1.4208995300962184,
1797
+ "grad_norm": 2.890631914138794,
1798
+ "learning_rate": 2.926404773744406e-05,
1799
+ "loss": 3.6086,
1800
+ "step": 6350
1801
+ },
1802
+ {
1803
+ "epoch": 1.4264936227343925,
1804
+ "grad_norm": 4.880382537841797,
1805
+ "learning_rate": 2.916045085363832e-05,
1806
+ "loss": 3.3194,
1807
+ "step": 6375
1808
+ },
1809
+ {
1810
+ "epoch": 1.4320877153725666,
1811
+ "grad_norm": 5.451873779296875,
1812
+ "learning_rate": 2.9056853969832587e-05,
1813
+ "loss": 3.4783,
1814
+ "step": 6400
1815
+ },
1816
+ {
1817
+ "epoch": 1.4376818080107405,
1818
+ "grad_norm": 3.3253021240234375,
1819
+ "learning_rate": 2.8953257086026853e-05,
1820
+ "loss": 3.6597,
1821
+ "step": 6425
1822
+ },
1823
+ {
1824
+ "epoch": 1.4432759006489149,
1825
+ "grad_norm": 4.4339776039123535,
1826
+ "learning_rate": 2.8849660202221117e-05,
1827
+ "loss": 3.3595,
1828
+ "step": 6450
1829
+ },
1830
+ {
1831
+ "epoch": 1.4488699932870888,
1832
+ "grad_norm": 4.412319183349609,
1833
+ "learning_rate": 2.8746063318415383e-05,
1834
+ "loss": 3.4645,
1835
+ "step": 6475
1836
+ },
1837
+ {
1838
+ "epoch": 1.4544640859252629,
1839
+ "grad_norm": 4.133786201477051,
1840
+ "learning_rate": 2.864246643460965e-05,
1841
+ "loss": 3.7099,
1842
+ "step": 6500
1843
+ },
1844
+ {
1845
+ "epoch": 1.460058178563437,
1846
+ "grad_norm": 2.908602476119995,
1847
+ "learning_rate": 2.8538869550803913e-05,
1848
+ "loss": 3.683,
1849
+ "step": 6525
1850
+ },
1851
+ {
1852
+ "epoch": 1.4656522712016111,
1853
+ "grad_norm": 3.2811191082000732,
1854
+ "learning_rate": 2.843527266699818e-05,
1855
+ "loss": 3.5035,
1856
+ "step": 6550
1857
+ },
1858
+ {
1859
+ "epoch": 1.4712463638397852,
1860
+ "grad_norm": 4.293539524078369,
1861
+ "learning_rate": 2.833167578319244e-05,
1862
+ "loss": 3.5384,
1863
+ "step": 6575
1864
+ },
1865
+ {
1866
+ "epoch": 1.4768404564779591,
1867
+ "grad_norm": 2.9331464767456055,
1868
+ "learning_rate": 2.8228078899386707e-05,
1869
+ "loss": 3.6,
1870
+ "step": 6600
1871
+ },
1872
+ {
1873
+ "epoch": 1.4824345491161335,
1874
+ "grad_norm": 3.5864765644073486,
1875
+ "learning_rate": 2.8124482015580973e-05,
1876
+ "loss": 3.6206,
1877
+ "step": 6625
1878
+ },
1879
+ {
1880
+ "epoch": 1.4880286417543074,
1881
+ "grad_norm": 4.750547885894775,
1882
+ "learning_rate": 2.8020885131775237e-05,
1883
+ "loss": 3.5364,
1884
+ "step": 6650
1885
+ },
1886
+ {
1887
+ "epoch": 1.4936227343924815,
1888
+ "grad_norm": 2.7902162075042725,
1889
+ "learning_rate": 2.7917288247969504e-05,
1890
+ "loss": 3.4523,
1891
+ "step": 6675
1892
+ },
1893
+ {
1894
+ "epoch": 1.4992168270306556,
1895
+ "grad_norm": 4.09032678604126,
1896
+ "learning_rate": 2.781369136416377e-05,
1897
+ "loss": 3.4141,
1898
+ "step": 6700
1899
+ },
1900
+ {
1901
+ "epoch": 1.5048109196688297,
1902
+ "grad_norm": 2.7837555408477783,
1903
+ "learning_rate": 2.7710094480358034e-05,
1904
+ "loss": 3.5636,
1905
+ "step": 6725
1906
+ },
1907
+ {
1908
+ "epoch": 1.5104050123070039,
1909
+ "grad_norm": 4.959596633911133,
1910
+ "learning_rate": 2.7606497596552293e-05,
1911
+ "loss": 3.5464,
1912
+ "step": 6750
1913
+ },
1914
+ {
1915
+ "epoch": 1.5159991049451778,
1916
+ "grad_norm": 4.016523838043213,
1917
+ "learning_rate": 2.750290071274656e-05,
1918
+ "loss": 3.6429,
1919
+ "step": 6775
1920
+ },
1921
+ {
1922
+ "epoch": 1.521593197583352,
1923
+ "grad_norm": 2.876103639602661,
1924
+ "learning_rate": 2.7399303828940827e-05,
1925
+ "loss": 3.6272,
1926
+ "step": 6800
1927
+ },
1928
+ {
1929
+ "epoch": 1.527187290221526,
1930
+ "grad_norm": 3.660968780517578,
1931
+ "learning_rate": 2.729570694513509e-05,
1932
+ "loss": 3.5362,
1933
+ "step": 6825
1934
+ },
1935
+ {
1936
+ "epoch": 1.5327813828597001,
1937
+ "grad_norm": 3.6049230098724365,
1938
+ "learning_rate": 2.7192110061329357e-05,
1939
+ "loss": 3.5556,
1940
+ "step": 6850
1941
+ },
1942
+ {
1943
+ "epoch": 1.5383754754978742,
1944
+ "grad_norm": 4.5552215576171875,
1945
+ "learning_rate": 2.7088513177523624e-05,
1946
+ "loss": 3.447,
1947
+ "step": 6875
1948
+ },
1949
+ {
1950
+ "epoch": 1.5439695681360484,
1951
+ "grad_norm": 3.871140241622925,
1952
+ "learning_rate": 2.6989060169070117e-05,
1953
+ "loss": 3.6483,
1954
+ "step": 6900
1955
+ },
1956
+ {
1957
+ "epoch": 1.5495636607742225,
1958
+ "grad_norm": 3.9797022342681885,
1959
+ "learning_rate": 2.6885463285264377e-05,
1960
+ "loss": 3.5776,
1961
+ "step": 6925
1962
+ },
1963
+ {
1964
+ "epoch": 1.5551577534123964,
1965
+ "grad_norm": 3.6146976947784424,
1966
+ "learning_rate": 2.6781866401458644e-05,
1967
+ "loss": 3.4178,
1968
+ "step": 6950
1969
+ },
1970
+ {
1971
+ "epoch": 1.5607518460505707,
1972
+ "grad_norm": 3.5435166358947754,
1973
+ "learning_rate": 2.667826951765291e-05,
1974
+ "loss": 3.4008,
1975
+ "step": 6975
1976
+ },
1977
+ {
1978
+ "epoch": 1.5663459386887446,
1979
+ "grad_norm": 3.3893115520477295,
1980
+ "learning_rate": 2.6574672633847174e-05,
1981
+ "loss": 3.5622,
1982
+ "step": 7000
1983
+ },
1984
+ {
1985
+ "epoch": 1.5719400313269187,
1986
+ "grad_norm": 4.91843843460083,
1987
+ "learning_rate": 2.647107575004144e-05,
1988
+ "loss": 3.5347,
1989
+ "step": 7025
1990
+ },
1991
+ {
1992
+ "epoch": 1.5775341239650928,
1993
+ "grad_norm": 3.9048681259155273,
1994
+ "learning_rate": 2.6367478866235707e-05,
1995
+ "loss": 3.5438,
1996
+ "step": 7050
1997
+ },
1998
+ {
1999
+ "epoch": 1.583128216603267,
2000
+ "grad_norm": 4.051578998565674,
2001
+ "learning_rate": 2.626388198242997e-05,
2002
+ "loss": 3.5693,
2003
+ "step": 7075
2004
+ },
2005
+ {
2006
+ "epoch": 1.588722309241441,
2007
+ "grad_norm": 4.973071098327637,
2008
+ "learning_rate": 2.6160285098624234e-05,
2009
+ "loss": 3.5851,
2010
+ "step": 7100
2011
+ },
2012
+ {
2013
+ "epoch": 1.594316401879615,
2014
+ "grad_norm": 3.8136496543884277,
2015
+ "learning_rate": 2.6056688214818497e-05,
2016
+ "loss": 3.5712,
2017
+ "step": 7125
2018
+ },
2019
+ {
2020
+ "epoch": 1.5999104945177893,
2021
+ "grad_norm": 5.140013694763184,
2022
+ "learning_rate": 2.5953091331012764e-05,
2023
+ "loss": 3.4751,
2024
+ "step": 7150
2025
+ },
2026
+ {
2027
+ "epoch": 1.6055045871559632,
2028
+ "grad_norm": 3.7872066497802734,
2029
+ "learning_rate": 2.584949444720703e-05,
2030
+ "loss": 3.5062,
2031
+ "step": 7175
2032
+ },
2033
+ {
2034
+ "epoch": 1.6110986797941373,
2035
+ "grad_norm": 4.849021911621094,
2036
+ "learning_rate": 2.5745897563401294e-05,
2037
+ "loss": 3.7954,
2038
+ "step": 7200
2039
+ },
2040
+ {
2041
+ "epoch": 1.6166927724323115,
2042
+ "grad_norm": 3.8170666694641113,
2043
+ "learning_rate": 2.564230067959556e-05,
2044
+ "loss": 3.5713,
2045
+ "step": 7225
2046
+ },
2047
+ {
2048
+ "epoch": 1.6222868650704856,
2049
+ "grad_norm": 3.0287299156188965,
2050
+ "learning_rate": 2.5538703795789827e-05,
2051
+ "loss": 3.5479,
2052
+ "step": 7250
2053
+ },
2054
+ {
2055
+ "epoch": 1.6278809577086597,
2056
+ "grad_norm": 3.0226757526397705,
2057
+ "learning_rate": 2.543510691198409e-05,
2058
+ "loss": 3.5507,
2059
+ "step": 7275
2060
+ },
2061
+ {
2062
+ "epoch": 1.6334750503468336,
2063
+ "grad_norm": 4.079787731170654,
2064
+ "learning_rate": 2.533151002817835e-05,
2065
+ "loss": 3.5487,
2066
+ "step": 7300
2067
+ },
2068
+ {
2069
+ "epoch": 1.639069142985008,
2070
+ "grad_norm": 3.4698259830474854,
2071
+ "learning_rate": 2.5227913144372617e-05,
2072
+ "loss": 3.3525,
2073
+ "step": 7325
2074
+ },
2075
+ {
2076
+ "epoch": 1.6446632356231818,
2077
+ "grad_norm": 4.5240864753723145,
2078
+ "learning_rate": 2.5124316260566884e-05,
2079
+ "loss": 3.4664,
2080
+ "step": 7350
2081
+ },
2082
+ {
2083
+ "epoch": 1.650257328261356,
2084
+ "grad_norm": 4.2261457443237305,
2085
+ "learning_rate": 2.5020719376761147e-05,
2086
+ "loss": 3.6773,
2087
+ "step": 7375
2088
+ },
2089
+ {
2090
+ "epoch": 1.65585142089953,
2091
+ "grad_norm": 3.5485663414001465,
2092
+ "learning_rate": 2.4917122492955414e-05,
2093
+ "loss": 3.5225,
2094
+ "step": 7400
2095
+ },
2096
+ {
2097
+ "epoch": 1.6614455135377042,
2098
+ "grad_norm": 3.4678995609283447,
2099
+ "learning_rate": 2.481352560914968e-05,
2100
+ "loss": 3.4339,
2101
+ "step": 7425
2102
+ },
2103
+ {
2104
+ "epoch": 1.6670396061758783,
2105
+ "grad_norm": 4.038591384887695,
2106
+ "learning_rate": 2.470992872534394e-05,
2107
+ "loss": 3.2773,
2108
+ "step": 7450
2109
+ },
2110
+ {
2111
+ "epoch": 1.6726336988140522,
2112
+ "grad_norm": 4.297801494598389,
2113
+ "learning_rate": 2.4606331841538207e-05,
2114
+ "loss": 3.3887,
2115
+ "step": 7475
2116
+ },
2117
+ {
2118
+ "epoch": 1.6782277914522266,
2119
+ "grad_norm": 3.6633341312408447,
2120
+ "learning_rate": 2.4502734957732474e-05,
2121
+ "loss": 3.5098,
2122
+ "step": 7500
2123
+ },
2124
+ {
2125
+ "epoch": 1.6838218840904005,
2126
+ "grad_norm": 4.264350414276123,
2127
+ "learning_rate": 2.4399138073926737e-05,
2128
+ "loss": 3.4985,
2129
+ "step": 7525
2130
+ },
2131
+ {
2132
+ "epoch": 1.6894159767285746,
2133
+ "grad_norm": 4.352592945098877,
2134
+ "learning_rate": 2.4295541190121e-05,
2135
+ "loss": 3.3587,
2136
+ "step": 7550
2137
+ },
2138
+ {
2139
+ "epoch": 1.6950100693667487,
2140
+ "grad_norm": 3.8767361640930176,
2141
+ "learning_rate": 2.4191944306315267e-05,
2142
+ "loss": 3.3361,
2143
+ "step": 7575
2144
+ },
2145
+ {
2146
+ "epoch": 1.7006041620049228,
2147
+ "grad_norm": 3.0718719959259033,
2148
+ "learning_rate": 2.4088347422509534e-05,
2149
+ "loss": 3.6612,
2150
+ "step": 7600
2151
+ },
2152
+ {
2153
+ "epoch": 1.706198254643097,
2154
+ "grad_norm": 4.193331241607666,
2155
+ "learning_rate": 2.3984750538703797e-05,
2156
+ "loss": 3.5357,
2157
+ "step": 7625
2158
+ },
2159
+ {
2160
+ "epoch": 1.7117923472812708,
2161
+ "grad_norm": 2.363382339477539,
2162
+ "learning_rate": 2.388115365489806e-05,
2163
+ "loss": 3.4306,
2164
+ "step": 7650
2165
+ },
2166
+ {
2167
+ "epoch": 1.7173864399194452,
2168
+ "grad_norm": 11.161027908325195,
2169
+ "learning_rate": 2.3777556771092327e-05,
2170
+ "loss": 3.5064,
2171
+ "step": 7675
2172
+ },
2173
+ {
2174
+ "epoch": 1.722980532557619,
2175
+ "grad_norm": 5.971621036529541,
2176
+ "learning_rate": 2.3673959887286594e-05,
2177
+ "loss": 3.4386,
2178
+ "step": 7700
2179
+ },
2180
+ {
2181
+ "epoch": 1.7285746251957932,
2182
+ "grad_norm": 4.31216287612915,
2183
+ "learning_rate": 2.3570363003480854e-05,
2184
+ "loss": 3.6187,
2185
+ "step": 7725
2186
+ },
2187
+ {
2188
+ "epoch": 1.7341687178339673,
2189
+ "grad_norm": 3.554739475250244,
2190
+ "learning_rate": 2.346676611967512e-05,
2191
+ "loss": 3.525,
2192
+ "step": 7750
2193
+ },
2194
+ {
2195
+ "epoch": 1.7397628104721414,
2196
+ "grad_norm": 4.137548446655273,
2197
+ "learning_rate": 2.3363169235869388e-05,
2198
+ "loss": 3.669,
2199
+ "step": 7775
2200
+ },
2201
+ {
2202
+ "epoch": 1.7453569031103155,
2203
+ "grad_norm": 3.0854902267456055,
2204
+ "learning_rate": 2.325957235206365e-05,
2205
+ "loss": 3.499,
2206
+ "step": 7800
2207
+ },
2208
+ {
2209
+ "epoch": 1.7509509957484894,
2210
+ "grad_norm": 3.3931126594543457,
2211
+ "learning_rate": 2.3155975468257914e-05,
2212
+ "loss": 3.5291,
2213
+ "step": 7825
2214
+ },
2215
+ {
2216
+ "epoch": 1.7565450883866638,
2217
+ "grad_norm": 3.637080192565918,
2218
+ "learning_rate": 2.305237858445218e-05,
2219
+ "loss": 3.3685,
2220
+ "step": 7850
2221
+ },
2222
+ {
2223
+ "epoch": 1.7621391810248377,
2224
+ "grad_norm": 3.0859596729278564,
2225
+ "learning_rate": 2.2948781700646448e-05,
2226
+ "loss": 3.4383,
2227
+ "step": 7875
2228
+ },
2229
+ {
2230
+ "epoch": 1.7677332736630118,
2231
+ "grad_norm": 4.485119342803955,
2232
+ "learning_rate": 2.284518481684071e-05,
2233
+ "loss": 3.4679,
2234
+ "step": 7900
2235
+ },
2236
+ {
2237
+ "epoch": 1.773327366301186,
2238
+ "grad_norm": 3.628810167312622,
2239
+ "learning_rate": 2.2741587933034974e-05,
2240
+ "loss": 3.6006,
2241
+ "step": 7925
2242
+ },
2243
+ {
2244
+ "epoch": 1.77892145893936,
2245
+ "grad_norm": 4.076844215393066,
2246
+ "learning_rate": 2.263799104922924e-05,
2247
+ "loss": 3.6363,
2248
+ "step": 7950
2249
+ },
2250
+ {
2251
+ "epoch": 1.7845155515775342,
2252
+ "grad_norm": 2.8478996753692627,
2253
+ "learning_rate": 2.2534394165423504e-05,
2254
+ "loss": 3.5581,
2255
+ "step": 7975
2256
+ },
2257
+ {
2258
+ "epoch": 1.790109644215708,
2259
+ "grad_norm": 3.94975209236145,
2260
+ "learning_rate": 2.243079728161777e-05,
2261
+ "loss": 3.5994,
2262
+ "step": 8000
2263
+ },
2264
+ {
2265
+ "epoch": 1.7957037368538824,
2266
+ "grad_norm": 4.70791482925415,
2267
+ "learning_rate": 2.2327200397812034e-05,
2268
+ "loss": 3.6401,
2269
+ "step": 8025
2270
+ },
2271
+ {
2272
+ "epoch": 1.8012978294920563,
2273
+ "grad_norm": 4.756099224090576,
2274
+ "learning_rate": 2.22236035140063e-05,
2275
+ "loss": 3.6995,
2276
+ "step": 8050
2277
+ },
2278
+ {
2279
+ "epoch": 1.8068919221302304,
2280
+ "grad_norm": 4.0641255378723145,
2281
+ "learning_rate": 2.2120006630200564e-05,
2282
+ "loss": 3.3503,
2283
+ "step": 8075
2284
+ },
2285
+ {
2286
+ "epoch": 1.8124860147684045,
2287
+ "grad_norm": 3.7078256607055664,
2288
+ "learning_rate": 2.2016409746394828e-05,
2289
+ "loss": 3.6939,
2290
+ "step": 8100
2291
+ },
2292
+ {
2293
+ "epoch": 1.8180801074065787,
2294
+ "grad_norm": 3.761341094970703,
2295
+ "learning_rate": 2.1912812862589094e-05,
2296
+ "loss": 3.3038,
2297
+ "step": 8125
2298
+ },
2299
+ {
2300
+ "epoch": 1.8236742000447528,
2301
+ "grad_norm": 3.6664912700653076,
2302
+ "learning_rate": 2.1809215978783358e-05,
2303
+ "loss": 3.6525,
2304
+ "step": 8150
2305
+ },
2306
+ {
2307
+ "epoch": 1.8292682926829267,
2308
+ "grad_norm": 9.494356155395508,
2309
+ "learning_rate": 2.1705619094977624e-05,
2310
+ "loss": 3.5386,
2311
+ "step": 8175
2312
+ },
2313
+ {
2314
+ "epoch": 1.834862385321101,
2315
+ "grad_norm": 5.5311431884765625,
2316
+ "learning_rate": 2.1602022211171888e-05,
2317
+ "loss": 3.5256,
2318
+ "step": 8200
2319
+ },
2320
+ {
2321
+ "epoch": 1.840456477959275,
2322
+ "grad_norm": 3.7495036125183105,
2323
+ "learning_rate": 2.1498425327366154e-05,
2324
+ "loss": 3.4281,
2325
+ "step": 8225
2326
+ },
2327
+ {
2328
+ "epoch": 1.8460505705974493,
2329
+ "grad_norm": 3.225494384765625,
2330
+ "learning_rate": 2.1394828443560418e-05,
2331
+ "loss": 3.4695,
2332
+ "step": 8250
2333
+ },
2334
+ {
2335
+ "epoch": 1.8516446632356232,
2336
+ "grad_norm": 5.200911521911621,
2337
+ "learning_rate": 2.1291231559754685e-05,
2338
+ "loss": 3.5364,
2339
+ "step": 8275
2340
+ },
2341
+ {
2342
+ "epoch": 1.8572387558737973,
2343
+ "grad_norm": 4.8887619972229,
2344
+ "learning_rate": 2.1187634675948948e-05,
2345
+ "loss": 3.5523,
2346
+ "step": 8300
2347
+ },
2348
+ {
2349
+ "epoch": 1.8628328485119714,
2350
+ "grad_norm": 4.1215033531188965,
2351
+ "learning_rate": 2.1084037792143215e-05,
2352
+ "loss": 3.5867,
2353
+ "step": 8325
2354
+ },
2355
+ {
2356
+ "epoch": 1.8684269411501453,
2357
+ "grad_norm": 5.489912509918213,
2358
+ "learning_rate": 2.0980440908337478e-05,
2359
+ "loss": 3.5744,
2360
+ "step": 8350
2361
+ },
2362
+ {
2363
+ "epoch": 1.8740210337883196,
2364
+ "grad_norm": 4.38899040222168,
2365
+ "learning_rate": 2.0876844024531745e-05,
2366
+ "loss": 3.8165,
2367
+ "step": 8375
2368
+ },
2369
+ {
2370
+ "epoch": 1.8796151264264935,
2371
+ "grad_norm": 3.097547769546509,
2372
+ "learning_rate": 2.0773247140726008e-05,
2373
+ "loss": 3.2993,
2374
+ "step": 8400
2375
+ },
2376
+ {
2377
+ "epoch": 1.8852092190646679,
2378
+ "grad_norm": 3.8072097301483154,
2379
+ "learning_rate": 2.066965025692027e-05,
2380
+ "loss": 3.3597,
2381
+ "step": 8425
2382
+ },
2383
+ {
2384
+ "epoch": 1.8908033117028418,
2385
+ "grad_norm": 3.7643051147460938,
2386
+ "learning_rate": 2.0566053373114538e-05,
2387
+ "loss": 3.42,
2388
+ "step": 8450
2389
+ },
2390
+ {
2391
+ "epoch": 1.896397404341016,
2392
+ "grad_norm": 3.058461904525757,
2393
+ "learning_rate": 2.0462456489308805e-05,
2394
+ "loss": 3.6001,
2395
+ "step": 8475
2396
+ },
2397
+ {
2398
+ "epoch": 1.90199149697919,
2399
+ "grad_norm": 5.52889347076416,
2400
+ "learning_rate": 2.0358859605503068e-05,
2401
+ "loss": 3.4691,
2402
+ "step": 8500
2403
+ },
2404
+ {
2405
+ "epoch": 1.907585589617364,
2406
+ "grad_norm": 3.3969428539276123,
2407
+ "learning_rate": 2.025526272169733e-05,
2408
+ "loss": 3.5784,
2409
+ "step": 8525
2410
+ },
2411
+ {
2412
+ "epoch": 1.9131796822555382,
2413
+ "grad_norm": 4.519889831542969,
2414
+ "learning_rate": 2.0151665837891598e-05,
2415
+ "loss": 3.4325,
2416
+ "step": 8550
2417
+ },
2418
+ {
2419
+ "epoch": 1.9187737748937121,
2420
+ "grad_norm": 3.1045982837677,
2421
+ "learning_rate": 2.004806895408586e-05,
2422
+ "loss": 3.4577,
2423
+ "step": 8575
2424
+ },
2425
+ {
2426
+ "epoch": 1.9243678675318865,
2427
+ "grad_norm": 3.8332388401031494,
2428
+ "learning_rate": 1.9944472070280125e-05,
2429
+ "loss": 3.4851,
2430
+ "step": 8600
2431
+ },
2432
+ {
2433
+ "epoch": 1.9299619601700604,
2434
+ "grad_norm": 3.1713857650756836,
2435
+ "learning_rate": 1.984087518647439e-05,
2436
+ "loss": 3.3116,
2437
+ "step": 8625
2438
+ },
2439
+ {
2440
+ "epoch": 1.9355560528082345,
2441
+ "grad_norm": 2.7377285957336426,
2442
+ "learning_rate": 1.9737278302668658e-05,
2443
+ "loss": 3.5988,
2444
+ "step": 8650
2445
+ },
2446
+ {
2447
+ "epoch": 1.9411501454464086,
2448
+ "grad_norm": 3.1266770362854004,
2449
+ "learning_rate": 1.963368141886292e-05,
2450
+ "loss": 3.4971,
2451
+ "step": 8675
2452
+ },
2453
+ {
2454
+ "epoch": 1.9467442380845825,
2455
+ "grad_norm": 2.204955816268921,
2456
+ "learning_rate": 1.9530084535057185e-05,
2457
+ "loss": 3.452,
2458
+ "step": 8700
2459
+ },
2460
+ {
2461
+ "epoch": 1.9523383307227569,
2462
+ "grad_norm": 3.3418877124786377,
2463
+ "learning_rate": 1.942648765125145e-05,
2464
+ "loss": 3.6088,
2465
+ "step": 8725
2466
+ },
2467
+ {
2468
+ "epoch": 1.9579324233609308,
2469
+ "grad_norm": 4.350618839263916,
2470
+ "learning_rate": 1.9322890767445718e-05,
2471
+ "loss": 3.4842,
2472
+ "step": 8750
2473
+ },
2474
+ {
2475
+ "epoch": 1.963526515999105,
2476
+ "grad_norm": 3.0263142585754395,
2477
+ "learning_rate": 1.921929388363998e-05,
2478
+ "loss": 3.5175,
2479
+ "step": 8775
2480
+ },
2481
+ {
2482
+ "epoch": 1.969120608637279,
2483
+ "grad_norm": 3.8331007957458496,
2484
+ "learning_rate": 1.9115696999834245e-05,
2485
+ "loss": 3.5241,
2486
+ "step": 8800
2487
+ },
2488
+ {
2489
+ "epoch": 1.9747147012754531,
2490
+ "grad_norm": 2.886181116104126,
2491
+ "learning_rate": 1.901210011602851e-05,
2492
+ "loss": 3.5752,
2493
+ "step": 8825
2494
+ },
2495
+ {
2496
+ "epoch": 1.9803087939136272,
2497
+ "grad_norm": 2.8387129306793213,
2498
+ "learning_rate": 1.890850323222278e-05,
2499
+ "loss": 3.7299,
2500
+ "step": 8850
2501
+ },
2502
+ {
2503
+ "epoch": 1.9859028865518011,
2504
+ "grad_norm": 3.225780963897705,
2505
+ "learning_rate": 1.8804906348417038e-05,
2506
+ "loss": 3.4075,
2507
+ "step": 8875
2508
+ },
2509
+ {
2510
+ "epoch": 1.9914969791899755,
2511
+ "grad_norm": 3.7564287185668945,
2512
+ "learning_rate": 1.8701309464611305e-05,
2513
+ "loss": 3.4456,
2514
+ "step": 8900
2515
+ },
2516
+ {
2517
+ "epoch": 1.9970910718281494,
2518
+ "grad_norm": 4.473494529724121,
2519
+ "learning_rate": 1.859771258080557e-05,
2520
+ "loss": 3.4203,
2521
+ "step": 8925
2522
+ },
2523
+ {
2524
+ "epoch": 2.0,
2525
+ "eval_gen_len": 17.521,
2526
+ "eval_loss": 3.4858927726745605,
2527
+ "eval_rouge1": 34.5787,
2528
+ "eval_rouge2": 14.7352,
2529
+ "eval_rougeL": 33.5549,
2530
+ "eval_rougeLsum": 33.7034,
2531
+ "eval_runtime": 118.773,
2532
+ "eval_samples_per_second": 9.404,
2533
+ "eval_steps_per_second": 2.357,
2534
+ "step": 8938
2535
+ },
2536
+ {
2537
+ "epoch": 2.0026851644663237,
2538
+ "grad_norm": 3.3720297813415527,
2539
+ "learning_rate": 1.8494115696999835e-05,
2540
+ "loss": 3.4454,
2541
+ "step": 8950
2542
+ },
2543
+ {
2544
+ "epoch": 2.0082792571044976,
2545
+ "grad_norm": 3.2836527824401855,
2546
+ "learning_rate": 1.83905188131941e-05,
2547
+ "loss": 3.6566,
2548
+ "step": 8975
2549
+ },
2550
+ {
2551
+ "epoch": 2.013873349742672,
2552
+ "grad_norm": 3.623849868774414,
2553
+ "learning_rate": 1.8286921929388365e-05,
2554
+ "loss": 3.2928,
2555
+ "step": 9000
2556
+ },
2557
+ {
2558
+ "epoch": 2.019467442380846,
2559
+ "grad_norm": 2.457754373550415,
2560
+ "learning_rate": 1.8183325045582632e-05,
2561
+ "loss": 3.2728,
2562
+ "step": 9025
2563
+ },
2564
+ {
2565
+ "epoch": 2.0250615350190198,
2566
+ "grad_norm": 3.352640151977539,
2567
+ "learning_rate": 1.8079728161776895e-05,
2568
+ "loss": 3.3607,
2569
+ "step": 9050
2570
+ },
2571
+ {
2572
+ "epoch": 2.030655627657194,
2573
+ "grad_norm": 2.8351364135742188,
2574
+ "learning_rate": 1.797613127797116e-05,
2575
+ "loss": 3.283,
2576
+ "step": 9075
2577
+ },
2578
+ {
2579
+ "epoch": 2.036249720295368,
2580
+ "grad_norm": 3.3237016201019287,
2581
+ "learning_rate": 1.7872534394165425e-05,
2582
+ "loss": 3.2595,
2583
+ "step": 9100
2584
+ },
2585
+ {
2586
+ "epoch": 2.0418438129335423,
2587
+ "grad_norm": 4.141500949859619,
2588
+ "learning_rate": 1.776893751035969e-05,
2589
+ "loss": 3.3706,
2590
+ "step": 9125
2591
+ },
2592
+ {
2593
+ "epoch": 2.0474379055717162,
2594
+ "grad_norm": 3.100390672683716,
2595
+ "learning_rate": 1.7665340626553952e-05,
2596
+ "loss": 3.3923,
2597
+ "step": 9150
2598
+ },
2599
+ {
2600
+ "epoch": 2.0530319982098906,
2601
+ "grad_norm": 18.937232971191406,
2602
+ "learning_rate": 1.756174374274822e-05,
2603
+ "loss": 3.6783,
2604
+ "step": 9175
2605
+ },
2606
+ {
2607
+ "epoch": 2.0586260908480645,
2608
+ "grad_norm": 3.0570342540740967,
2609
+ "learning_rate": 1.7458146858942485e-05,
2610
+ "loss": 3.2842,
2611
+ "step": 9200
2612
+ },
2613
+ {
2614
+ "epoch": 2.0642201834862384,
2615
+ "grad_norm": 3.776122570037842,
2616
+ "learning_rate": 1.735454997513675e-05,
2617
+ "loss": 3.1484,
2618
+ "step": 9225
2619
+ },
2620
+ {
2621
+ "epoch": 2.0698142761244127,
2622
+ "grad_norm": 3.5520737171173096,
2623
+ "learning_rate": 1.7250953091331012e-05,
2624
+ "loss": 3.3581,
2625
+ "step": 9250
2626
+ },
2627
+ {
2628
+ "epoch": 2.0754083687625866,
2629
+ "grad_norm": 3.331061363220215,
2630
+ "learning_rate": 1.714735620752528e-05,
2631
+ "loss": 3.5026,
2632
+ "step": 9275
2633
+ },
2634
+ {
2635
+ "epoch": 2.081002461400761,
2636
+ "grad_norm": 3.8635380268096924,
2637
+ "learning_rate": 1.7043759323719545e-05,
2638
+ "loss": 3.5856,
2639
+ "step": 9300
2640
+ },
2641
+ {
2642
+ "epoch": 2.086596554038935,
2643
+ "grad_norm": 3.710649251937866,
2644
+ "learning_rate": 1.694016243991381e-05,
2645
+ "loss": 3.5301,
2646
+ "step": 9325
2647
+ },
2648
+ {
2649
+ "epoch": 2.0921906466771087,
2650
+ "grad_norm": 3.2452878952026367,
2651
+ "learning_rate": 1.6836565556108072e-05,
2652
+ "loss": 3.5673,
2653
+ "step": 9350
2654
+ },
2655
+ {
2656
+ "epoch": 2.097784739315283,
2657
+ "grad_norm": 4.901977062225342,
2658
+ "learning_rate": 1.673296867230234e-05,
2659
+ "loss": 3.3774,
2660
+ "step": 9375
2661
+ },
2662
+ {
2663
+ "epoch": 2.103378831953457,
2664
+ "grad_norm": 2.7200512886047363,
2665
+ "learning_rate": 1.6629371788496602e-05,
2666
+ "loss": 3.366,
2667
+ "step": 9400
2668
+ },
2669
+ {
2670
+ "epoch": 2.1089729245916313,
2671
+ "grad_norm": 4.317224979400635,
2672
+ "learning_rate": 1.652577490469087e-05,
2673
+ "loss": 3.2699,
2674
+ "step": 9425
2675
+ },
2676
+ {
2677
+ "epoch": 2.1145670172298052,
2678
+ "grad_norm": 3.473952293395996,
2679
+ "learning_rate": 1.6422178020885132e-05,
2680
+ "loss": 3.5787,
2681
+ "step": 9450
2682
+ },
2683
+ {
2684
+ "epoch": 2.1201611098679796,
2685
+ "grad_norm": 3.5654425621032715,
2686
+ "learning_rate": 1.63185811370794e-05,
2687
+ "loss": 3.2704,
2688
+ "step": 9475
2689
+ },
2690
+ {
2691
+ "epoch": 2.1257552025061535,
2692
+ "grad_norm": 4.178590774536133,
2693
+ "learning_rate": 1.6214984253273662e-05,
2694
+ "loss": 3.4346,
2695
+ "step": 9500
2696
+ },
2697
+ {
2698
+ "epoch": 2.131349295144328,
2699
+ "grad_norm": 2.804097890853882,
2700
+ "learning_rate": 1.611138736946793e-05,
2701
+ "loss": 3.3003,
2702
+ "step": 9525
2703
+ },
2704
+ {
2705
+ "epoch": 2.1369433877825017,
2706
+ "grad_norm": 4.665280342102051,
2707
+ "learning_rate": 1.6007790485662192e-05,
2708
+ "loss": 3.4827,
2709
+ "step": 9550
2710
+ },
2711
+ {
2712
+ "epoch": 2.1425374804206756,
2713
+ "grad_norm": 2.7570557594299316,
2714
+ "learning_rate": 1.5904193601856455e-05,
2715
+ "loss": 3.4484,
2716
+ "step": 9575
2717
+ },
2718
+ {
2719
+ "epoch": 2.14813157305885,
2720
+ "grad_norm": 3.2325522899627686,
2721
+ "learning_rate": 1.5800596718050722e-05,
2722
+ "loss": 3.6098,
2723
+ "step": 9600
2724
+ },
2725
+ {
2726
+ "epoch": 2.153725665697024,
2727
+ "grad_norm": 4.1614603996276855,
2728
+ "learning_rate": 1.5696999834244985e-05,
2729
+ "loss": 3.6194,
2730
+ "step": 9625
2731
+ },
2732
+ {
2733
+ "epoch": 2.159319758335198,
2734
+ "grad_norm": 3.4777424335479736,
2735
+ "learning_rate": 1.5593402950439252e-05,
2736
+ "loss": 3.4299,
2737
+ "step": 9650
2738
+ },
2739
+ {
2740
+ "epoch": 2.164913850973372,
2741
+ "grad_norm": 3.0564773082733154,
2742
+ "learning_rate": 1.5489806066633516e-05,
2743
+ "loss": 3.4714,
2744
+ "step": 9675
2745
+ },
2746
+ {
2747
+ "epoch": 2.170507943611546,
2748
+ "grad_norm": 4.6712870597839355,
2749
+ "learning_rate": 1.539035305818001e-05,
2750
+ "loss": 3.1544,
2751
+ "step": 9700
2752
+ },
2753
+ {
2754
+ "epoch": 2.1761020362497203,
2755
+ "grad_norm": 3.942474365234375,
2756
+ "learning_rate": 1.5286756174374275e-05,
2757
+ "loss": 3.5576,
2758
+ "step": 9725
2759
+ },
2760
+ {
2761
+ "epoch": 2.181696128887894,
2762
+ "grad_norm": 4.431474208831787,
2763
+ "learning_rate": 1.518315929056854e-05,
2764
+ "loss": 3.5018,
2765
+ "step": 9750
2766
+ },
2767
+ {
2768
+ "epoch": 2.1872902215260686,
2769
+ "grad_norm": 3.981989860534668,
2770
+ "learning_rate": 1.5079562406762807e-05,
2771
+ "loss": 3.3641,
2772
+ "step": 9775
2773
+ },
2774
+ {
2775
+ "epoch": 2.1928843141642425,
2776
+ "grad_norm": 4.154784202575684,
2777
+ "learning_rate": 1.4975965522957069e-05,
2778
+ "loss": 3.4361,
2779
+ "step": 9800
2780
+ },
2781
+ {
2782
+ "epoch": 2.198478406802417,
2783
+ "grad_norm": 4.097687721252441,
2784
+ "learning_rate": 1.4872368639151336e-05,
2785
+ "loss": 3.3835,
2786
+ "step": 9825
2787
+ },
2788
+ {
2789
+ "epoch": 2.2040724994405907,
2790
+ "grad_norm": 3.584115743637085,
2791
+ "learning_rate": 1.47687717553456e-05,
2792
+ "loss": 3.3509,
2793
+ "step": 9850
2794
+ },
2795
+ {
2796
+ "epoch": 2.209666592078765,
2797
+ "grad_norm": 3.2317144870758057,
2798
+ "learning_rate": 1.4665174871539864e-05,
2799
+ "loss": 3.5553,
2800
+ "step": 9875
2801
+ },
2802
+ {
2803
+ "epoch": 2.215260684716939,
2804
+ "grad_norm": 3.4590249061584473,
2805
+ "learning_rate": 1.4561577987734129e-05,
2806
+ "loss": 3.4574,
2807
+ "step": 9900
2808
+ },
2809
+ {
2810
+ "epoch": 2.220854777355113,
2811
+ "grad_norm": 3.1115450859069824,
2812
+ "learning_rate": 1.4457981103928394e-05,
2813
+ "loss": 3.3254,
2814
+ "step": 9925
2815
+ },
2816
+ {
2817
+ "epoch": 2.226448869993287,
2818
+ "grad_norm": 4.1351542472839355,
2819
+ "learning_rate": 1.435438422012266e-05,
2820
+ "loss": 3.3671,
2821
+ "step": 9950
2822
+ },
2823
+ {
2824
+ "epoch": 2.232042962631461,
2825
+ "grad_norm": 3.3375813961029053,
2826
+ "learning_rate": 1.4250787336316922e-05,
2827
+ "loss": 3.1883,
2828
+ "step": 9975
2829
+ },
2830
+ {
2831
+ "epoch": 2.2376370552696354,
2832
+ "grad_norm": 3.7801668643951416,
2833
+ "learning_rate": 1.4147190452511189e-05,
2834
+ "loss": 3.3476,
2835
+ "step": 10000
2836
+ },
2837
+ {
2838
+ "epoch": 2.2432311479078093,
2839
+ "grad_norm": 2.931952953338623,
2840
+ "learning_rate": 1.4043593568705454e-05,
2841
+ "loss": 3.6136,
2842
+ "step": 10025
2843
+ },
2844
+ {
2845
+ "epoch": 2.248825240545983,
2846
+ "grad_norm": 3.703629732131958,
2847
+ "learning_rate": 1.393999668489972e-05,
2848
+ "loss": 3.4948,
2849
+ "step": 10050
2850
+ },
2851
+ {
2852
+ "epoch": 2.2544193331841575,
2853
+ "grad_norm": 4.5622878074646,
2854
+ "learning_rate": 1.3836399801093982e-05,
2855
+ "loss": 3.6212,
2856
+ "step": 10075
2857
+ },
2858
+ {
2859
+ "epoch": 2.2600134258223314,
2860
+ "grad_norm": 3.5357351303100586,
2861
+ "learning_rate": 1.3732802917288249e-05,
2862
+ "loss": 3.3934,
2863
+ "step": 10100
2864
+ },
2865
+ {
2866
+ "epoch": 2.265607518460506,
2867
+ "grad_norm": 2.999955177307129,
2868
+ "learning_rate": 1.3629206033482514e-05,
2869
+ "loss": 3.3663,
2870
+ "step": 10125
2871
+ },
2872
+ {
2873
+ "epoch": 2.2712016110986797,
2874
+ "grad_norm": 3.6298716068267822,
2875
+ "learning_rate": 1.3525609149676779e-05,
2876
+ "loss": 3.3121,
2877
+ "step": 10150
2878
+ },
2879
+ {
2880
+ "epoch": 2.276795703736854,
2881
+ "grad_norm": 3.9620649814605713,
2882
+ "learning_rate": 1.3422012265871042e-05,
2883
+ "loss": 3.5245,
2884
+ "step": 10175
2885
+ },
2886
+ {
2887
+ "epoch": 2.282389796375028,
2888
+ "grad_norm": 3.536635160446167,
2889
+ "learning_rate": 1.3318415382065307e-05,
2890
+ "loss": 3.3576,
2891
+ "step": 10200
2892
+ },
2893
+ {
2894
+ "epoch": 2.2879838890132023,
2895
+ "grad_norm": 4.903816223144531,
2896
+ "learning_rate": 1.3214818498259574e-05,
2897
+ "loss": 3.3997,
2898
+ "step": 10225
2899
+ },
2900
+ {
2901
+ "epoch": 2.293577981651376,
2902
+ "grad_norm": 4.816688060760498,
2903
+ "learning_rate": 1.311122161445384e-05,
2904
+ "loss": 3.4524,
2905
+ "step": 10250
2906
+ },
2907
+ {
2908
+ "epoch": 2.29917207428955,
2909
+ "grad_norm": 3.090827465057373,
2910
+ "learning_rate": 1.3007624730648103e-05,
2911
+ "loss": 3.3922,
2912
+ "step": 10275
2913
+ },
2914
+ {
2915
+ "epoch": 2.3047661669277244,
2916
+ "grad_norm": 3.400076150894165,
2917
+ "learning_rate": 1.2904027846842368e-05,
2918
+ "loss": 3.3607,
2919
+ "step": 10300
2920
+ },
2921
+ {
2922
+ "epoch": 2.3103602595658983,
2923
+ "grad_norm": 4.252440452575684,
2924
+ "learning_rate": 1.2800430963036633e-05,
2925
+ "loss": 3.6075,
2926
+ "step": 10325
2927
+ },
2928
+ {
2929
+ "epoch": 2.3159543522040726,
2930
+ "grad_norm": 3.622704029083252,
2931
+ "learning_rate": 1.2696834079230896e-05,
2932
+ "loss": 3.3601,
2933
+ "step": 10350
2934
+ },
2935
+ {
2936
+ "epoch": 2.3215484448422465,
2937
+ "grad_norm": 2.2757890224456787,
2938
+ "learning_rate": 1.2593237195425161e-05,
2939
+ "loss": 3.1384,
2940
+ "step": 10375
2941
+ },
2942
+ {
2943
+ "epoch": 2.3271425374804204,
2944
+ "grad_norm": 4.479572772979736,
2945
+ "learning_rate": 1.2489640311619428e-05,
2946
+ "loss": 3.525,
2947
+ "step": 10400
2948
+ },
2949
+ {
2950
+ "epoch": 2.332736630118595,
2951
+ "grad_norm": 3.390577793121338,
2952
+ "learning_rate": 1.2386043427813691e-05,
2953
+ "loss": 3.3901,
2954
+ "step": 10425
2955
+ },
2956
+ {
2957
+ "epoch": 2.3383307227567687,
2958
+ "grad_norm": 4.7925920486450195,
2959
+ "learning_rate": 1.2282446544007958e-05,
2960
+ "loss": 3.5471,
2961
+ "step": 10450
2962
+ },
2963
+ {
2964
+ "epoch": 2.343924815394943,
2965
+ "grad_norm": 4.597250461578369,
2966
+ "learning_rate": 1.2178849660202221e-05,
2967
+ "loss": 3.4124,
2968
+ "step": 10475
2969
+ },
2970
+ {
2971
+ "epoch": 2.349518908033117,
2972
+ "grad_norm": 5.611468315124512,
2973
+ "learning_rate": 1.2075252776396486e-05,
2974
+ "loss": 3.44,
2975
+ "step": 10500
2976
+ },
2977
+ {
2978
+ "epoch": 2.3551130006712913,
2979
+ "grad_norm": 4.439332962036133,
2980
+ "learning_rate": 1.1971655892590751e-05,
2981
+ "loss": 3.5714,
2982
+ "step": 10525
2983
+ },
2984
+ {
2985
+ "epoch": 2.360707093309465,
2986
+ "grad_norm": 3.1640384197235107,
2987
+ "learning_rate": 1.1868059008785016e-05,
2988
+ "loss": 3.3725,
2989
+ "step": 10550
2990
+ },
2991
+ {
2992
+ "epoch": 2.3663011859476395,
2993
+ "grad_norm": 3.673372745513916,
2994
+ "learning_rate": 1.1764462124979281e-05,
2995
+ "loss": 3.6341,
2996
+ "step": 10575
2997
+ },
2998
+ {
2999
+ "epoch": 2.3718952785858134,
3000
+ "grad_norm": 3.6749773025512695,
3001
+ "learning_rate": 1.1660865241173546e-05,
3002
+ "loss": 3.3728,
3003
+ "step": 10600
3004
+ },
3005
+ {
3006
+ "epoch": 2.3774893712239873,
3007
+ "grad_norm": 3.5209028720855713,
3008
+ "learning_rate": 1.1557268357367811e-05,
3009
+ "loss": 3.4008,
3010
+ "step": 10625
3011
+ },
3012
+ {
3013
+ "epoch": 2.3830834638621616,
3014
+ "grad_norm": 4.668726921081543,
3015
+ "learning_rate": 1.1453671473562076e-05,
3016
+ "loss": 3.3669,
3017
+ "step": 10650
3018
+ },
3019
+ {
3020
+ "epoch": 2.3886775565003355,
3021
+ "grad_norm": 3.522364377975464,
3022
+ "learning_rate": 1.1350074589756341e-05,
3023
+ "loss": 3.5714,
3024
+ "step": 10675
3025
+ },
3026
+ {
3027
+ "epoch": 2.39427164913851,
3028
+ "grad_norm": 3.3226089477539062,
3029
+ "learning_rate": 1.1246477705950606e-05,
3030
+ "loss": 3.3249,
3031
+ "step": 10700
3032
+ },
3033
+ {
3034
+ "epoch": 2.3998657417766838,
3035
+ "grad_norm": 6.041501522064209,
3036
+ "learning_rate": 1.114288082214487e-05,
3037
+ "loss": 3.5539,
3038
+ "step": 10725
3039
+ },
3040
+ {
3041
+ "epoch": 2.4054598344148577,
3042
+ "grad_norm": 2.5041422843933105,
3043
+ "learning_rate": 1.1039283938339136e-05,
3044
+ "loss": 3.4264,
3045
+ "step": 10750
3046
+ },
3047
+ {
3048
+ "epoch": 2.411053927053032,
3049
+ "grad_norm": 3.830946922302246,
3050
+ "learning_rate": 1.09356870545334e-05,
3051
+ "loss": 3.2978,
3052
+ "step": 10775
3053
+ },
3054
+ {
3055
+ "epoch": 2.416648019691206,
3056
+ "grad_norm": 5.300613880157471,
3057
+ "learning_rate": 1.0832090170727666e-05,
3058
+ "loss": 3.3027,
3059
+ "step": 10800
3060
+ },
3061
+ {
3062
+ "epoch": 2.4222421123293802,
3063
+ "grad_norm": 3.533928394317627,
3064
+ "learning_rate": 1.072849328692193e-05,
3065
+ "loss": 3.3647,
3066
+ "step": 10825
3067
+ },
3068
+ {
3069
+ "epoch": 2.427836204967554,
3070
+ "grad_norm": 3.791598081588745,
3071
+ "learning_rate": 1.0624896403116196e-05,
3072
+ "loss": 3.4624,
3073
+ "step": 10850
3074
+ },
3075
+ {
3076
+ "epoch": 2.4334302976057285,
3077
+ "grad_norm": 3.112946033477783,
3078
+ "learning_rate": 1.052129951931046e-05,
3079
+ "loss": 3.4873,
3080
+ "step": 10875
3081
+ },
3082
+ {
3083
+ "epoch": 2.4390243902439024,
3084
+ "grad_norm": 3.300663709640503,
3085
+ "learning_rate": 1.0417702635504725e-05,
3086
+ "loss": 3.3407,
3087
+ "step": 10900
3088
+ },
3089
+ {
3090
+ "epoch": 2.4446184828820767,
3091
+ "grad_norm": 3.0511510372161865,
3092
+ "learning_rate": 1.031410575169899e-05,
3093
+ "loss": 3.0996,
3094
+ "step": 10925
3095
+ },
3096
+ {
3097
+ "epoch": 2.4502125755202506,
3098
+ "grad_norm": 3.307197093963623,
3099
+ "learning_rate": 1.0210508867893253e-05,
3100
+ "loss": 3.3536,
3101
+ "step": 10950
3102
+ },
3103
+ {
3104
+ "epoch": 2.4558066681584245,
3105
+ "grad_norm": 3.473043918609619,
3106
+ "learning_rate": 1.010691198408752e-05,
3107
+ "loss": 3.3556,
3108
+ "step": 10975
3109
+ },
3110
+ {
3111
+ "epoch": 2.461400760796599,
3112
+ "grad_norm": 3.5455174446105957,
3113
+ "learning_rate": 1.0003315100281783e-05,
3114
+ "loss": 3.3864,
3115
+ "step": 11000
3116
+ },
3117
+ {
3118
+ "epoch": 2.4669948534347728,
3119
+ "grad_norm": 2.930676221847534,
3120
+ "learning_rate": 9.89971821647605e-06,
3121
+ "loss": 3.4506,
3122
+ "step": 11025
3123
+ },
3124
+ {
3125
+ "epoch": 2.472588946072947,
3126
+ "grad_norm": 3.216984987258911,
3127
+ "learning_rate": 9.796121332670313e-06,
3128
+ "loss": 3.2453,
3129
+ "step": 11050
3130
+ },
3131
+ {
3132
+ "epoch": 2.478183038711121,
3133
+ "grad_norm": 3.2741010189056396,
3134
+ "learning_rate": 9.692524448864578e-06,
3135
+ "loss": 3.3429,
3136
+ "step": 11075
3137
+ },
3138
+ {
3139
+ "epoch": 2.483777131349295,
3140
+ "grad_norm": 4.5768046379089355,
3141
+ "learning_rate": 9.588927565058843e-06,
3142
+ "loss": 3.0766,
3143
+ "step": 11100
3144
+ },
3145
+ {
3146
+ "epoch": 2.4893712239874692,
3147
+ "grad_norm": 3.7554574012756348,
3148
+ "learning_rate": 9.485330681253108e-06,
3149
+ "loss": 3.6496,
3150
+ "step": 11125
3151
+ },
3152
+ {
3153
+ "epoch": 2.494965316625643,
3154
+ "grad_norm": 3.8196182250976562,
3155
+ "learning_rate": 9.381733797447373e-06,
3156
+ "loss": 3.3505,
3157
+ "step": 11150
3158
+ },
3159
+ {
3160
+ "epoch": 2.5005594092638175,
3161
+ "grad_norm": 2.859930992126465,
3162
+ "learning_rate": 9.278136913641638e-06,
3163
+ "loss": 3.5258,
3164
+ "step": 11175
3165
+ },
3166
+ {
3167
+ "epoch": 2.5061535019019914,
3168
+ "grad_norm": 3.05218768119812,
3169
+ "learning_rate": 9.174540029835903e-06,
3170
+ "loss": 3.4866,
3171
+ "step": 11200
3172
+ },
3173
+ {
3174
+ "epoch": 2.5117475945401657,
3175
+ "grad_norm": 8.317590713500977,
3176
+ "learning_rate": 9.070943146030168e-06,
3177
+ "loss": 3.5313,
3178
+ "step": 11225
3179
+ },
3180
+ {
3181
+ "epoch": 2.5173416871783396,
3182
+ "grad_norm": 5.521259307861328,
3183
+ "learning_rate": 8.967346262224433e-06,
3184
+ "loss": 3.3587,
3185
+ "step": 11250
3186
+ },
3187
+ {
3188
+ "epoch": 2.522935779816514,
3189
+ "grad_norm": 3.744612693786621,
3190
+ "learning_rate": 8.863749378418698e-06,
3191
+ "loss": 3.6261,
3192
+ "step": 11275
3193
+ },
3194
+ {
3195
+ "epoch": 2.528529872454688,
3196
+ "grad_norm": 3.0806972980499268,
3197
+ "learning_rate": 8.760152494612962e-06,
3198
+ "loss": 3.4882,
3199
+ "step": 11300
3200
+ },
3201
+ {
3202
+ "epoch": 2.5341239650928618,
3203
+ "grad_norm": 2.8486557006835938,
3204
+ "learning_rate": 8.656555610807228e-06,
3205
+ "loss": 3.4368,
3206
+ "step": 11325
3207
+ },
3208
+ {
3209
+ "epoch": 2.539718057731036,
3210
+ "grad_norm": 4.572649002075195,
3211
+ "learning_rate": 8.552958727001492e-06,
3212
+ "loss": 3.2185,
3213
+ "step": 11350
3214
+ },
3215
+ {
3216
+ "epoch": 2.54531215036921,
3217
+ "grad_norm": 4.267202854156494,
3218
+ "learning_rate": 8.449361843195758e-06,
3219
+ "loss": 3.3354,
3220
+ "step": 11375
3221
+ },
3222
+ {
3223
+ "epoch": 2.5509062430073843,
3224
+ "grad_norm": 4.093388557434082,
3225
+ "learning_rate": 8.345764959390022e-06,
3226
+ "loss": 3.3239,
3227
+ "step": 11400
3228
+ },
3229
+ {
3230
+ "epoch": 2.5565003356455582,
3231
+ "grad_norm": 2.5423085689544678,
3232
+ "learning_rate": 8.242168075584287e-06,
3233
+ "loss": 3.4522,
3234
+ "step": 11425
3235
+ },
3236
+ {
3237
+ "epoch": 2.562094428283732,
3238
+ "grad_norm": 3.1850597858428955,
3239
+ "learning_rate": 8.138571191778552e-06,
3240
+ "loss": 3.3388,
3241
+ "step": 11450
3242
+ },
3243
+ {
3244
+ "epoch": 2.5676885209219065,
3245
+ "grad_norm": 4.956600189208984,
3246
+ "learning_rate": 8.034974307972817e-06,
3247
+ "loss": 3.3726,
3248
+ "step": 11475
3249
+ },
3250
+ {
3251
+ "epoch": 2.573282613560081,
3252
+ "grad_norm": 5.410662651062012,
3253
+ "learning_rate": 7.931377424167082e-06,
3254
+ "loss": 3.5072,
3255
+ "step": 11500
3256
+ },
3257
+ {
3258
+ "epoch": 2.5788767061982547,
3259
+ "grad_norm": 3.2471835613250732,
3260
+ "learning_rate": 7.827780540361345e-06,
3261
+ "loss": 3.3801,
3262
+ "step": 11525
3263
+ },
3264
+ {
3265
+ "epoch": 2.5844707988364286,
3266
+ "grad_norm": 2.7254815101623535,
3267
+ "learning_rate": 7.724183656555612e-06,
3268
+ "loss": 3.4124,
3269
+ "step": 11550
3270
+ },
3271
+ {
3272
+ "epoch": 2.590064891474603,
3273
+ "grad_norm": 4.180164337158203,
3274
+ "learning_rate": 7.620586772749876e-06,
3275
+ "loss": 3.4543,
3276
+ "step": 11575
3277
+ },
3278
+ {
3279
+ "epoch": 2.595658984112777,
3280
+ "grad_norm": 4.883584499359131,
3281
+ "learning_rate": 7.516989888944141e-06,
3282
+ "loss": 3.1982,
3283
+ "step": 11600
3284
+ },
3285
+ {
3286
+ "epoch": 2.601253076750951,
3287
+ "grad_norm": 4.658567428588867,
3288
+ "learning_rate": 7.413393005138405e-06,
3289
+ "loss": 3.4653,
3290
+ "step": 11625
3291
+ },
3292
+ {
3293
+ "epoch": 2.606847169389125,
3294
+ "grad_norm": 4.22096586227417,
3295
+ "learning_rate": 7.309796121332671e-06,
3296
+ "loss": 3.3742,
3297
+ "step": 11650
3298
+ },
3299
+ {
3300
+ "epoch": 2.612441262027299,
3301
+ "grad_norm": 3.328474760055542,
3302
+ "learning_rate": 7.206199237526935e-06,
3303
+ "loss": 3.2767,
3304
+ "step": 11675
3305
+ },
3306
+ {
3307
+ "epoch": 2.6180353546654733,
3308
+ "grad_norm": 4.015954494476318,
3309
+ "learning_rate": 7.102602353721201e-06,
3310
+ "loss": 3.1757,
3311
+ "step": 11700
3312
+ },
3313
+ {
3314
+ "epoch": 2.623629447303647,
3315
+ "grad_norm": 3.448801279067993,
3316
+ "learning_rate": 6.999005469915465e-06,
3317
+ "loss": 3.3787,
3318
+ "step": 11725
3319
+ },
3320
+ {
3321
+ "epoch": 2.6292235399418216,
3322
+ "grad_norm": 5.924972057342529,
3323
+ "learning_rate": 6.89540858610973e-06,
3324
+ "loss": 3.3822,
3325
+ "step": 11750
3326
+ },
3327
+ {
3328
+ "epoch": 2.6348176325799955,
3329
+ "grad_norm": 3.3082499504089355,
3330
+ "learning_rate": 6.7918117023039944e-06,
3331
+ "loss": 3.5094,
3332
+ "step": 11775
3333
+ },
3334
+ {
3335
+ "epoch": 2.6404117252181694,
3336
+ "grad_norm": 3.532564640045166,
3337
+ "learning_rate": 6.68821481849826e-06,
3338
+ "loss": 3.4982,
3339
+ "step": 11800
3340
+ },
3341
+ {
3342
+ "epoch": 2.6460058178563437,
3343
+ "grad_norm": 3.7091224193573,
3344
+ "learning_rate": 6.5846179346925245e-06,
3345
+ "loss": 3.4677,
3346
+ "step": 11825
3347
+ },
3348
+ {
3349
+ "epoch": 2.651599910494518,
3350
+ "grad_norm": 4.424626350402832,
3351
+ "learning_rate": 6.48102105088679e-06,
3352
+ "loss": 3.3444,
3353
+ "step": 11850
3354
+ },
3355
+ {
3356
+ "epoch": 2.657194003132692,
3357
+ "grad_norm": 3.832063913345337,
3358
+ "learning_rate": 6.3774241670810545e-06,
3359
+ "loss": 3.5021,
3360
+ "step": 11875
3361
+ },
3362
+ {
3363
+ "epoch": 2.662788095770866,
3364
+ "grad_norm": 4.093804836273193,
3365
+ "learning_rate": 6.277971158627549e-06,
3366
+ "loss": 3.1801,
3367
+ "step": 11900
3368
+ },
3369
+ {
3370
+ "epoch": 2.66838218840904,
3371
+ "grad_norm": 3.710360050201416,
3372
+ "learning_rate": 6.174374274821814e-06,
3373
+ "loss": 3.6768,
3374
+ "step": 11925
3375
+ },
3376
+ {
3377
+ "epoch": 2.673976281047214,
3378
+ "grad_norm": 3.7162835597991943,
3379
+ "learning_rate": 6.070777391016079e-06,
3380
+ "loss": 3.4159,
3381
+ "step": 11950
3382
+ },
3383
+ {
3384
+ "epoch": 2.6795703736853884,
3385
+ "grad_norm": 5.361623764038086,
3386
+ "learning_rate": 5.967180507210343e-06,
3387
+ "loss": 3.4015,
3388
+ "step": 11975
3389
+ },
3390
+ {
3391
+ "epoch": 2.6851644663235623,
3392
+ "grad_norm": 4.155981540679932,
3393
+ "learning_rate": 5.863583623404608e-06,
3394
+ "loss": 3.5182,
3395
+ "step": 12000
3396
+ },
3397
+ {
3398
+ "epoch": 2.690758558961736,
3399
+ "grad_norm": 2.906383991241455,
3400
+ "learning_rate": 5.759986739598873e-06,
3401
+ "loss": 3.4939,
3402
+ "step": 12025
3403
+ },
3404
+ {
3405
+ "epoch": 2.6963526515999106,
3406
+ "grad_norm": 3.313352584838867,
3407
+ "learning_rate": 5.656389855793138e-06,
3408
+ "loss": 3.3542,
3409
+ "step": 12050
3410
+ },
3411
+ {
3412
+ "epoch": 2.7019467442380845,
3413
+ "grad_norm": 3.462233543395996,
3414
+ "learning_rate": 5.552792971987403e-06,
3415
+ "loss": 3.7296,
3416
+ "step": 12075
3417
+ },
3418
+ {
3419
+ "epoch": 2.707540836876259,
3420
+ "grad_norm": 4.586158752441406,
3421
+ "learning_rate": 5.449196088181668e-06,
3422
+ "loss": 3.4957,
3423
+ "step": 12100
3424
+ },
3425
+ {
3426
+ "epoch": 2.7131349295144327,
3427
+ "grad_norm": 3.1800856590270996,
3428
+ "learning_rate": 5.345599204375933e-06,
3429
+ "loss": 3.1881,
3430
+ "step": 12125
3431
+ },
3432
+ {
3433
+ "epoch": 2.7187290221526066,
3434
+ "grad_norm": 4.854430675506592,
3435
+ "learning_rate": 5.242002320570198e-06,
3436
+ "loss": 3.4447,
3437
+ "step": 12150
3438
+ },
3439
+ {
3440
+ "epoch": 2.724323114790781,
3441
+ "grad_norm": 3.673379421234131,
3442
+ "learning_rate": 5.138405436764462e-06,
3443
+ "loss": 3.5881,
3444
+ "step": 12175
3445
+ },
3446
+ {
3447
+ "epoch": 2.7299172074289553,
3448
+ "grad_norm": 4.523945331573486,
3449
+ "learning_rate": 5.034808552958727e-06,
3450
+ "loss": 3.517,
3451
+ "step": 12200
3452
+ },
3453
+ {
3454
+ "epoch": 2.735511300067129,
3455
+ "grad_norm": 3.587252140045166,
3456
+ "learning_rate": 4.931211669152992e-06,
3457
+ "loss": 3.5721,
3458
+ "step": 12225
3459
+ },
3460
+ {
3461
+ "epoch": 2.741105392705303,
3462
+ "grad_norm": 5.535435676574707,
3463
+ "learning_rate": 4.827614785347257e-06,
3464
+ "loss": 3.3456,
3465
+ "step": 12250
3466
+ },
3467
+ {
3468
+ "epoch": 2.7466994853434774,
3469
+ "grad_norm": 3.8339381217956543,
3470
+ "learning_rate": 4.724017901541522e-06,
3471
+ "loss": 3.2803,
3472
+ "step": 12275
3473
+ },
3474
+ {
3475
+ "epoch": 2.7522935779816513,
3476
+ "grad_norm": 4.130194664001465,
3477
+ "learning_rate": 4.620421017735787e-06,
3478
+ "loss": 3.4254,
3479
+ "step": 12300
3480
+ },
3481
+ {
3482
+ "epoch": 2.7578876706198256,
3483
+ "grad_norm": 4.2525315284729,
3484
+ "learning_rate": 4.516824133930052e-06,
3485
+ "loss": 3.4911,
3486
+ "step": 12325
3487
+ },
3488
+ {
3489
+ "epoch": 2.7634817632579995,
3490
+ "grad_norm": 4.41726541519165,
3491
+ "learning_rate": 4.413227250124316e-06,
3492
+ "loss": 3.2429,
3493
+ "step": 12350
3494
+ },
3495
+ {
3496
+ "epoch": 2.7690758558961734,
3497
+ "grad_norm": 5.655388355255127,
3498
+ "learning_rate": 4.309630366318581e-06,
3499
+ "loss": 3.2737,
3500
+ "step": 12375
3501
+ },
3502
+ {
3503
+ "epoch": 2.774669948534348,
3504
+ "grad_norm": 3.538651704788208,
3505
+ "learning_rate": 4.2060334825128464e-06,
3506
+ "loss": 3.4575,
3507
+ "step": 12400
3508
+ },
3509
+ {
3510
+ "epoch": 2.7802640411725217,
3511
+ "grad_norm": 4.212822437286377,
3512
+ "learning_rate": 4.102436598707111e-06,
3513
+ "loss": 3.4099,
3514
+ "step": 12425
3515
+ },
3516
+ {
3517
+ "epoch": 2.785858133810696,
3518
+ "grad_norm": 6.0278639793396,
3519
+ "learning_rate": 3.998839714901376e-06,
3520
+ "loss": 3.3082,
3521
+ "step": 12450
3522
+ },
3523
+ {
3524
+ "epoch": 2.79145222644887,
3525
+ "grad_norm": 3.1115334033966064,
3526
+ "learning_rate": 3.895242831095641e-06,
3527
+ "loss": 3.5152,
3528
+ "step": 12475
3529
+ },
3530
+ {
3531
+ "epoch": 2.797046319087044,
3532
+ "grad_norm": 3.4043774604797363,
3533
+ "learning_rate": 3.7916459472899057e-06,
3534
+ "loss": 3.4762,
3535
+ "step": 12500
3536
+ },
3537
+ {
3538
+ "epoch": 2.802640411725218,
3539
+ "grad_norm": 2.902223587036133,
3540
+ "learning_rate": 3.6880490634841703e-06,
3541
+ "loss": 3.3443,
3542
+ "step": 12525
3543
+ },
3544
+ {
3545
+ "epoch": 2.8082345043633925,
3546
+ "grad_norm": 6.004910469055176,
3547
+ "learning_rate": 3.5844521796784353e-06,
3548
+ "loss": 3.3289,
3549
+ "step": 12550
3550
+ },
3551
+ {
3552
+ "epoch": 2.8138285970015664,
3553
+ "grad_norm": 4.121148109436035,
3554
+ "learning_rate": 3.4808552958727003e-06,
3555
+ "loss": 3.5039,
3556
+ "step": 12575
3557
+ },
3558
+ {
3559
+ "epoch": 2.8194226896397403,
3560
+ "grad_norm": 3.410569429397583,
3561
+ "learning_rate": 3.377258412066965e-06,
3562
+ "loss": 3.3976,
3563
+ "step": 12600
3564
+ },
3565
+ {
3566
+ "epoch": 2.8250167822779146,
3567
+ "grad_norm": 5.479353904724121,
3568
+ "learning_rate": 3.27366152826123e-06,
3569
+ "loss": 3.4147,
3570
+ "step": 12625
3571
+ },
3572
+ {
3573
+ "epoch": 2.8306108749160885,
3574
+ "grad_norm": 3.4429993629455566,
3575
+ "learning_rate": 3.170064644455495e-06,
3576
+ "loss": 3.434,
3577
+ "step": 12650
3578
+ },
3579
+ {
3580
+ "epoch": 2.836204967554263,
3581
+ "grad_norm": 2.758530855178833,
3582
+ "learning_rate": 3.06646776064976e-06,
3583
+ "loss": 3.5112,
3584
+ "step": 12675
3585
+ },
3586
+ {
3587
+ "epoch": 2.8417990601924368,
3588
+ "grad_norm": 3.5778067111968994,
3589
+ "learning_rate": 2.9628708768440246e-06,
3590
+ "loss": 3.473,
3591
+ "step": 12700
3592
+ },
3593
+ {
3594
+ "epoch": 2.8473931528306107,
3595
+ "grad_norm": 4.725358486175537,
3596
+ "learning_rate": 2.8592739930382896e-06,
3597
+ "loss": 3.578,
3598
+ "step": 12725
3599
+ },
3600
+ {
3601
+ "epoch": 2.852987245468785,
3602
+ "grad_norm": 3.3768036365509033,
3603
+ "learning_rate": 2.7556771092325546e-06,
3604
+ "loss": 3.2679,
3605
+ "step": 12750
3606
+ },
3607
+ {
3608
+ "epoch": 2.858581338106959,
3609
+ "grad_norm": 3.494004726409912,
3610
+ "learning_rate": 2.6520802254268196e-06,
3611
+ "loss": 3.3649,
3612
+ "step": 12775
3613
+ },
3614
+ {
3615
+ "epoch": 2.8641754307451333,
3616
+ "grad_norm": 4.525993347167969,
3617
+ "learning_rate": 2.5484833416210842e-06,
3618
+ "loss": 3.2952,
3619
+ "step": 12800
3620
+ },
3621
+ {
3622
+ "epoch": 2.869769523383307,
3623
+ "grad_norm": 4.161382675170898,
3624
+ "learning_rate": 2.444886457815349e-06,
3625
+ "loss": 3.0342,
3626
+ "step": 12825
3627
+ },
3628
+ {
3629
+ "epoch": 2.875363616021481,
3630
+ "grad_norm": 3.783557891845703,
3631
+ "learning_rate": 2.341289574009614e-06,
3632
+ "loss": 3.4595,
3633
+ "step": 12850
3634
+ },
3635
+ {
3636
+ "epoch": 2.8809577086596554,
3637
+ "grad_norm": 3.859001636505127,
3638
+ "learning_rate": 2.237692690203879e-06,
3639
+ "loss": 3.2808,
3640
+ "step": 12875
3641
+ },
3642
+ {
3643
+ "epoch": 2.8865518012978297,
3644
+ "grad_norm": 3.365854263305664,
3645
+ "learning_rate": 2.1340958063981435e-06,
3646
+ "loss": 3.3992,
3647
+ "step": 12900
3648
+ },
3649
+ {
3650
+ "epoch": 2.8921458939360036,
3651
+ "grad_norm": 2.6864302158355713,
3652
+ "learning_rate": 2.0304989225924085e-06,
3653
+ "loss": 3.3891,
3654
+ "step": 12925
3655
+ },
3656
+ {
3657
+ "epoch": 2.8977399865741775,
3658
+ "grad_norm": 3.899183511734009,
3659
+ "learning_rate": 1.9269020387866735e-06,
3660
+ "loss": 3.4347,
3661
+ "step": 12950
3662
+ },
3663
+ {
3664
+ "epoch": 2.903334079212352,
3665
+ "grad_norm": 3.851257801055908,
3666
+ "learning_rate": 1.8233051549809383e-06,
3667
+ "loss": 3.4567,
3668
+ "step": 12975
3669
+ },
3670
+ {
3671
+ "epoch": 2.9089281718505258,
3672
+ "grad_norm": 6.478551387786865,
3673
+ "learning_rate": 1.7197082711752031e-06,
3674
+ "loss": 3.5104,
3675
+ "step": 13000
3676
+ },
3677
+ {
3678
+ "epoch": 2.9145222644887,
3679
+ "grad_norm": 3.8405916690826416,
3680
+ "learning_rate": 1.6161113873694681e-06,
3681
+ "loss": 3.4155,
3682
+ "step": 13025
3683
+ },
3684
+ {
3685
+ "epoch": 2.920116357126874,
3686
+ "grad_norm": 4.361378192901611,
3687
+ "learning_rate": 1.512514503563733e-06,
3688
+ "loss": 3.4019,
3689
+ "step": 13050
3690
+ },
3691
+ {
3692
+ "epoch": 2.925710449765048,
3693
+ "grad_norm": 3.0873847007751465,
3694
+ "learning_rate": 1.4089176197579978e-06,
3695
+ "loss": 3.3706,
3696
+ "step": 13075
3697
+ },
3698
+ {
3699
+ "epoch": 2.9313045424032222,
3700
+ "grad_norm": 3.932100534439087,
3701
+ "learning_rate": 1.3053207359522626e-06,
3702
+ "loss": 3.47,
3703
+ "step": 13100
3704
+ },
3705
+ {
3706
+ "epoch": 2.936898635041396,
3707
+ "grad_norm": 2.9036309719085693,
3708
+ "learning_rate": 1.2017238521465276e-06,
3709
+ "loss": 3.4773,
3710
+ "step": 13125
3711
+ },
3712
+ {
3713
+ "epoch": 2.9424927276795705,
3714
+ "grad_norm": 3.846344232559204,
3715
+ "learning_rate": 1.0981269683407924e-06,
3716
+ "loss": 3.4122,
3717
+ "step": 13150
3718
+ },
3719
+ {
3720
+ "epoch": 2.9480868203177444,
3721
+ "grad_norm": 5.015985012054443,
3722
+ "learning_rate": 9.945300845350572e-07,
3723
+ "loss": 3.4759,
3724
+ "step": 13175
3725
+ },
3726
+ {
3727
+ "epoch": 2.9536809129559183,
3728
+ "grad_norm": 4.375097751617432,
3729
+ "learning_rate": 8.909332007293221e-07,
3730
+ "loss": 3.5986,
3731
+ "step": 13200
3732
+ },
3733
+ {
3734
+ "epoch": 2.9592750055940926,
3735
+ "grad_norm": 5.92262077331543,
3736
+ "learning_rate": 7.87336316923587e-07,
3737
+ "loss": 3.3251,
3738
+ "step": 13225
3739
+ },
3740
+ {
3741
+ "epoch": 2.964869098232267,
3742
+ "grad_norm": 3.339240312576294,
3743
+ "learning_rate": 6.837394331178518e-07,
3744
+ "loss": 3.4548,
3745
+ "step": 13250
3746
+ },
3747
+ {
3748
+ "epoch": 2.970463190870441,
3749
+ "grad_norm": 3.378695011138916,
3750
+ "learning_rate": 5.801425493121168e-07,
3751
+ "loss": 3.318,
3752
+ "step": 13275
3753
+ },
3754
+ {
3755
+ "epoch": 2.9760572835086148,
3756
+ "grad_norm": 3.89557147026062,
3757
+ "learning_rate": 4.765456655063816e-07,
3758
+ "loss": 3.3749,
3759
+ "step": 13300
3760
+ },
3761
+ {
3762
+ "epoch": 2.981651376146789,
3763
+ "grad_norm": 2.601956367492676,
3764
+ "learning_rate": 3.729487817006464e-07,
3765
+ "loss": 3.5431,
3766
+ "step": 13325
3767
+ },
3768
+ {
3769
+ "epoch": 2.987245468784963,
3770
+ "grad_norm": 3.6156444549560547,
3771
+ "learning_rate": 2.6935189789491134e-07,
3772
+ "loss": 3.5365,
3773
+ "step": 13350
3774
+ },
3775
+ {
3776
+ "epoch": 2.9928395614231373,
3777
+ "grad_norm": 3.4102349281311035,
3778
+ "learning_rate": 1.6575501408917623e-07,
3779
+ "loss": 3.6213,
3780
+ "step": 13375
3781
+ },
3782
+ {
3783
+ "epoch": 2.9984336540613112,
3784
+ "grad_norm": 5.817409038543701,
3785
+ "learning_rate": 6.215813028344107e-08,
3786
+ "loss": 3.3783,
3787
+ "step": 13400
3788
+ },
3789
+ {
3790
+ "epoch": 3.0,
3791
+ "eval_gen_len": 17.5112,
3792
+ "eval_loss": 3.4647490978240967,
3793
+ "eval_rouge1": 34.5915,
3794
+ "eval_rouge2": 14.7399,
3795
+ "eval_rougeL": 33.5742,
3796
+ "eval_rougeLsum": 33.7167,
3797
+ "eval_runtime": 118.5668,
3798
+ "eval_samples_per_second": 9.421,
3799
+ "eval_steps_per_second": 2.362,
3800
+ "step": 13407
3801
+ }
3802
+ ],
3803
+ "logging_steps": 25,
3804
+ "max_steps": 13407,
3805
+ "num_input_tokens_seen": 0,
3806
+ "num_train_epochs": 3,
3807
+ "save_steps": 500,
3808
+ "stateful_callbacks": {
3809
+ "EarlyStoppingCallback": {
3810
+ "args": {
3811
+ "early_stopping_patience": 5,
3812
+ "early_stopping_threshold": 0.01
3813
+ },
3814
+ "attributes": {
3815
+ "early_stopping_patience_counter": 0
3816
+ }
3817
+ },
3818
+ "TrainerControl": {
3819
+ "args": {
3820
+ "should_epoch_stop": false,
3821
+ "should_evaluate": false,
3822
+ "should_log": false,
3823
+ "should_save": true,
3824
+ "should_training_stop": true
3825
+ },
3826
+ "attributes": {}
3827
+ }
3828
+ },
3829
+ "total_flos": 1158566174208000.0,
3830
+ "train_batch_size": 2,
3831
+ "trial_name": null,
3832
+ "trial_params": null
3833
+ }
checkpoint-13407/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:037b5061a4768998e62bf58288beb11ae2f9a3594c44957cb5b4481c7508b626
3
+ size 5304
config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google-t5/t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "relu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": false,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "n_positions": 512,
21
+ "num_decoder_layers": 12,
22
+ "num_heads": 12,
23
+ "num_layers": 12,
24
+ "output_past": true,
25
+ "pad_token_id": 0,
26
+ "relative_attention_max_distance": 128,
27
+ "relative_attention_num_buckets": 32,
28
+ "task_specific_params": {
29
+ "summarization": {
30
+ "early_stopping": true,
31
+ "length_penalty": 2.0,
32
+ "max_length": 200,
33
+ "min_length": 30,
34
+ "no_repeat_ngram_size": 3,
35
+ "num_beams": 4,
36
+ "prefix": "summarize: "
37
+ },
38
+ "translation_en_to_de": {
39
+ "early_stopping": true,
40
+ "max_length": 300,
41
+ "num_beams": 4,
42
+ "prefix": "translate English to German: "
43
+ },
44
+ "translation_en_to_fr": {
45
+ "early_stopping": true,
46
+ "max_length": 300,
47
+ "num_beams": 4,
48
+ "prefix": "translate English to French: "
49
+ },
50
+ "translation_en_to_ro": {
51
+ "early_stopping": true,
52
+ "max_length": 300,
53
+ "num_beams": 4,
54
+ "prefix": "translate English to Romanian: "
55
+ }
56
+ },
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.41.0",
59
+ "use_cache": true,
60
+ "vocab_size": 32128
61
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "decoder_start_token_id": 0,
3
+ "eos_token_id": 1,
4
+ "pad_token_id": 0,
5
+ "transformers_version": "4.41.0"
6
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1451374ea6f9efea6b2c3934a2252d4458f9545ad6ad102a223e0c2e12e8a029
3
+ size 891644712
runs/Jun11_07-12-40_r-riken01-flan-t5-base-i1nyzxit-5e549-q7pdx/events.out.tfevents.1718089964.r-riken01-flan-t5-base-i1nyzxit-5e549-q7pdx.81.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8cb791a55411b9649cd6bd3bfc67e7817096a8a8c9a668ad81ca9beb515d8759
3
- size 105838
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf473a6e4091d95aef5848b5a5e00c40b4dbe4b5a5a9fad439ebb98937efe078
3
+ size 120854
runs/Jun11_07-12-40_r-riken01-flan-t5-base-i1nyzxit-5e549-q7pdx/events.out.tfevents.1718093622.r-riken01-flan-t5-base-i1nyzxit-5e549-q7pdx.81.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98e69a3f0bfee8ddd9ee184b447e9f370c2b999b522fecef6103e30f608779e0
3
+ size 613
special_tokens_map.json ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "eos_token": "</s>",
105
+ "pad_token": "<pad>",
106
+ "unk_token": "<unk>"
107
+ }
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,937 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<pad>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "</s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<unk>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "32000": {
28
+ "content": "<extra_id_99>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "32001": {
36
+ "content": "<extra_id_98>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "32002": {
44
+ "content": "<extra_id_97>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "32003": {
52
+ "content": "<extra_id_96>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "32004": {
60
+ "content": "<extra_id_95>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "32005": {
68
+ "content": "<extra_id_94>",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "32006": {
76
+ "content": "<extra_id_93>",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "32007": {
84
+ "content": "<extra_id_92>",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "32008": {
92
+ "content": "<extra_id_91>",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "32009": {
100
+ "content": "<extra_id_90>",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "32010": {
108
+ "content": "<extra_id_89>",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "32011": {
116
+ "content": "<extra_id_88>",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "32012": {
124
+ "content": "<extra_id_87>",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "32013": {
132
+ "content": "<extra_id_86>",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "32014": {
140
+ "content": "<extra_id_85>",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "32015": {
148
+ "content": "<extra_id_84>",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "32016": {
156
+ "content": "<extra_id_83>",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "32017": {
164
+ "content": "<extra_id_82>",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "32018": {
172
+ "content": "<extra_id_81>",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "32019": {
180
+ "content": "<extra_id_80>",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "32020": {
188
+ "content": "<extra_id_79>",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "32021": {
196
+ "content": "<extra_id_78>",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "32022": {
204
+ "content": "<extra_id_77>",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "32023": {
212
+ "content": "<extra_id_76>",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "32024": {
220
+ "content": "<extra_id_75>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "32025": {
228
+ "content": "<extra_id_74>",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "32026": {
236
+ "content": "<extra_id_73>",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "32027": {
244
+ "content": "<extra_id_72>",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "32028": {
252
+ "content": "<extra_id_71>",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "32029": {
260
+ "content": "<extra_id_70>",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "32030": {
268
+ "content": "<extra_id_69>",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "32031": {
276
+ "content": "<extra_id_68>",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "32032": {
284
+ "content": "<extra_id_67>",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "32033": {
292
+ "content": "<extra_id_66>",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "32034": {
300
+ "content": "<extra_id_65>",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "32035": {
308
+ "content": "<extra_id_64>",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "32036": {
316
+ "content": "<extra_id_63>",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "32037": {
324
+ "content": "<extra_id_62>",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "32038": {
332
+ "content": "<extra_id_61>",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "32039": {
340
+ "content": "<extra_id_60>",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "32040": {
348
+ "content": "<extra_id_59>",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "32041": {
356
+ "content": "<extra_id_58>",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "32042": {
364
+ "content": "<extra_id_57>",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "32043": {
372
+ "content": "<extra_id_56>",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "32044": {
380
+ "content": "<extra_id_55>",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "32045": {
388
+ "content": "<extra_id_54>",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "32046": {
396
+ "content": "<extra_id_53>",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "32047": {
404
+ "content": "<extra_id_52>",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "32048": {
412
+ "content": "<extra_id_51>",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "32049": {
420
+ "content": "<extra_id_50>",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "32050": {
428
+ "content": "<extra_id_49>",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "32051": {
436
+ "content": "<extra_id_48>",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "32052": {
444
+ "content": "<extra_id_47>",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "32053": {
452
+ "content": "<extra_id_46>",
453
+ "lstrip": false,
454
+ "normalized": false,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ },
459
+ "32054": {
460
+ "content": "<extra_id_45>",
461
+ "lstrip": false,
462
+ "normalized": false,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": true
466
+ },
467
+ "32055": {
468
+ "content": "<extra_id_44>",
469
+ "lstrip": false,
470
+ "normalized": false,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": true
474
+ },
475
+ "32056": {
476
+ "content": "<extra_id_43>",
477
+ "lstrip": false,
478
+ "normalized": false,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": true
482
+ },
483
+ "32057": {
484
+ "content": "<extra_id_42>",
485
+ "lstrip": false,
486
+ "normalized": false,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": true
490
+ },
491
+ "32058": {
492
+ "content": "<extra_id_41>",
493
+ "lstrip": false,
494
+ "normalized": false,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": true
498
+ },
499
+ "32059": {
500
+ "content": "<extra_id_40>",
501
+ "lstrip": false,
502
+ "normalized": false,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": true
506
+ },
507
+ "32060": {
508
+ "content": "<extra_id_39>",
509
+ "lstrip": false,
510
+ "normalized": false,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": true
514
+ },
515
+ "32061": {
516
+ "content": "<extra_id_38>",
517
+ "lstrip": false,
518
+ "normalized": false,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": true
522
+ },
523
+ "32062": {
524
+ "content": "<extra_id_37>",
525
+ "lstrip": false,
526
+ "normalized": false,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": true
530
+ },
531
+ "32063": {
532
+ "content": "<extra_id_36>",
533
+ "lstrip": false,
534
+ "normalized": false,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": true
538
+ },
539
+ "32064": {
540
+ "content": "<extra_id_35>",
541
+ "lstrip": false,
542
+ "normalized": false,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": true
546
+ },
547
+ "32065": {
548
+ "content": "<extra_id_34>",
549
+ "lstrip": false,
550
+ "normalized": false,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": true
554
+ },
555
+ "32066": {
556
+ "content": "<extra_id_33>",
557
+ "lstrip": false,
558
+ "normalized": false,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": true
562
+ },
563
+ "32067": {
564
+ "content": "<extra_id_32>",
565
+ "lstrip": false,
566
+ "normalized": false,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": true
570
+ },
571
+ "32068": {
572
+ "content": "<extra_id_31>",
573
+ "lstrip": false,
574
+ "normalized": false,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": true
578
+ },
579
+ "32069": {
580
+ "content": "<extra_id_30>",
581
+ "lstrip": false,
582
+ "normalized": false,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": true
586
+ },
587
+ "32070": {
588
+ "content": "<extra_id_29>",
589
+ "lstrip": false,
590
+ "normalized": false,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": true
594
+ },
595
+ "32071": {
596
+ "content": "<extra_id_28>",
597
+ "lstrip": false,
598
+ "normalized": false,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": true
602
+ },
603
+ "32072": {
604
+ "content": "<extra_id_27>",
605
+ "lstrip": false,
606
+ "normalized": false,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": true
610
+ },
611
+ "32073": {
612
+ "content": "<extra_id_26>",
613
+ "lstrip": false,
614
+ "normalized": false,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": true
618
+ },
619
+ "32074": {
620
+ "content": "<extra_id_25>",
621
+ "lstrip": false,
622
+ "normalized": false,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": true
626
+ },
627
+ "32075": {
628
+ "content": "<extra_id_24>",
629
+ "lstrip": false,
630
+ "normalized": false,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": true
634
+ },
635
+ "32076": {
636
+ "content": "<extra_id_23>",
637
+ "lstrip": false,
638
+ "normalized": false,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": true
642
+ },
643
+ "32077": {
644
+ "content": "<extra_id_22>",
645
+ "lstrip": false,
646
+ "normalized": false,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": true
650
+ },
651
+ "32078": {
652
+ "content": "<extra_id_21>",
653
+ "lstrip": false,
654
+ "normalized": false,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": true
658
+ },
659
+ "32079": {
660
+ "content": "<extra_id_20>",
661
+ "lstrip": false,
662
+ "normalized": false,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": true
666
+ },
667
+ "32080": {
668
+ "content": "<extra_id_19>",
669
+ "lstrip": false,
670
+ "normalized": false,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": true
674
+ },
675
+ "32081": {
676
+ "content": "<extra_id_18>",
677
+ "lstrip": false,
678
+ "normalized": false,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": true
682
+ },
683
+ "32082": {
684
+ "content": "<extra_id_17>",
685
+ "lstrip": false,
686
+ "normalized": false,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": true
690
+ },
691
+ "32083": {
692
+ "content": "<extra_id_16>",
693
+ "lstrip": false,
694
+ "normalized": false,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": true
698
+ },
699
+ "32084": {
700
+ "content": "<extra_id_15>",
701
+ "lstrip": false,
702
+ "normalized": false,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": true
706
+ },
707
+ "32085": {
708
+ "content": "<extra_id_14>",
709
+ "lstrip": false,
710
+ "normalized": false,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": true
714
+ },
715
+ "32086": {
716
+ "content": "<extra_id_13>",
717
+ "lstrip": false,
718
+ "normalized": false,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": true
722
+ },
723
+ "32087": {
724
+ "content": "<extra_id_12>",
725
+ "lstrip": false,
726
+ "normalized": false,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": true
730
+ },
731
+ "32088": {
732
+ "content": "<extra_id_11>",
733
+ "lstrip": false,
734
+ "normalized": false,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": true
738
+ },
739
+ "32089": {
740
+ "content": "<extra_id_10>",
741
+ "lstrip": false,
742
+ "normalized": false,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": true
746
+ },
747
+ "32090": {
748
+ "content": "<extra_id_9>",
749
+ "lstrip": false,
750
+ "normalized": false,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": true
754
+ },
755
+ "32091": {
756
+ "content": "<extra_id_8>",
757
+ "lstrip": false,
758
+ "normalized": false,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": true
762
+ },
763
+ "32092": {
764
+ "content": "<extra_id_7>",
765
+ "lstrip": false,
766
+ "normalized": false,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": true
770
+ },
771
+ "32093": {
772
+ "content": "<extra_id_6>",
773
+ "lstrip": false,
774
+ "normalized": false,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": true
778
+ },
779
+ "32094": {
780
+ "content": "<extra_id_5>",
781
+ "lstrip": false,
782
+ "normalized": false,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": true
786
+ },
787
+ "32095": {
788
+ "content": "<extra_id_4>",
789
+ "lstrip": false,
790
+ "normalized": false,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": true
794
+ },
795
+ "32096": {
796
+ "content": "<extra_id_3>",
797
+ "lstrip": false,
798
+ "normalized": false,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": true
802
+ },
803
+ "32097": {
804
+ "content": "<extra_id_2>",
805
+ "lstrip": false,
806
+ "normalized": false,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": true
810
+ },
811
+ "32098": {
812
+ "content": "<extra_id_1>",
813
+ "lstrip": false,
814
+ "normalized": false,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": true
818
+ },
819
+ "32099": {
820
+ "content": "<extra_id_0>",
821
+ "lstrip": false,
822
+ "normalized": false,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": true
826
+ }
827
+ },
828
+ "additional_special_tokens": [
829
+ "<extra_id_0>",
830
+ "<extra_id_1>",
831
+ "<extra_id_2>",
832
+ "<extra_id_3>",
833
+ "<extra_id_4>",
834
+ "<extra_id_5>",
835
+ "<extra_id_6>",
836
+ "<extra_id_7>",
837
+ "<extra_id_8>",
838
+ "<extra_id_9>",
839
+ "<extra_id_10>",
840
+ "<extra_id_11>",
841
+ "<extra_id_12>",
842
+ "<extra_id_13>",
843
+ "<extra_id_14>",
844
+ "<extra_id_15>",
845
+ "<extra_id_16>",
846
+ "<extra_id_17>",
847
+ "<extra_id_18>",
848
+ "<extra_id_19>",
849
+ "<extra_id_20>",
850
+ "<extra_id_21>",
851
+ "<extra_id_22>",
852
+ "<extra_id_23>",
853
+ "<extra_id_24>",
854
+ "<extra_id_25>",
855
+ "<extra_id_26>",
856
+ "<extra_id_27>",
857
+ "<extra_id_28>",
858
+ "<extra_id_29>",
859
+ "<extra_id_30>",
860
+ "<extra_id_31>",
861
+ "<extra_id_32>",
862
+ "<extra_id_33>",
863
+ "<extra_id_34>",
864
+ "<extra_id_35>",
865
+ "<extra_id_36>",
866
+ "<extra_id_37>",
867
+ "<extra_id_38>",
868
+ "<extra_id_39>",
869
+ "<extra_id_40>",
870
+ "<extra_id_41>",
871
+ "<extra_id_42>",
872
+ "<extra_id_43>",
873
+ "<extra_id_44>",
874
+ "<extra_id_45>",
875
+ "<extra_id_46>",
876
+ "<extra_id_47>",
877
+ "<extra_id_48>",
878
+ "<extra_id_49>",
879
+ "<extra_id_50>",
880
+ "<extra_id_51>",
881
+ "<extra_id_52>",
882
+ "<extra_id_53>",
883
+ "<extra_id_54>",
884
+ "<extra_id_55>",
885
+ "<extra_id_56>",
886
+ "<extra_id_57>",
887
+ "<extra_id_58>",
888
+ "<extra_id_59>",
889
+ "<extra_id_60>",
890
+ "<extra_id_61>",
891
+ "<extra_id_62>",
892
+ "<extra_id_63>",
893
+ "<extra_id_64>",
894
+ "<extra_id_65>",
895
+ "<extra_id_66>",
896
+ "<extra_id_67>",
897
+ "<extra_id_68>",
898
+ "<extra_id_69>",
899
+ "<extra_id_70>",
900
+ "<extra_id_71>",
901
+ "<extra_id_72>",
902
+ "<extra_id_73>",
903
+ "<extra_id_74>",
904
+ "<extra_id_75>",
905
+ "<extra_id_76>",
906
+ "<extra_id_77>",
907
+ "<extra_id_78>",
908
+ "<extra_id_79>",
909
+ "<extra_id_80>",
910
+ "<extra_id_81>",
911
+ "<extra_id_82>",
912
+ "<extra_id_83>",
913
+ "<extra_id_84>",
914
+ "<extra_id_85>",
915
+ "<extra_id_86>",
916
+ "<extra_id_87>",
917
+ "<extra_id_88>",
918
+ "<extra_id_89>",
919
+ "<extra_id_90>",
920
+ "<extra_id_91>",
921
+ "<extra_id_92>",
922
+ "<extra_id_93>",
923
+ "<extra_id_94>",
924
+ "<extra_id_95>",
925
+ "<extra_id_96>",
926
+ "<extra_id_97>",
927
+ "<extra_id_98>",
928
+ "<extra_id_99>"
929
+ ],
930
+ "clean_up_tokenization_spaces": true,
931
+ "eos_token": "</s>",
932
+ "extra_ids": 100,
933
+ "model_max_length": 1000000000000000019884624838656,
934
+ "pad_token": "<pad>",
935
+ "tokenizer_class": "T5Tokenizer",
936
+ "unk_token": "<unk>"
937
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:037b5061a4768998e62bf58288beb11ae2f9a3594c44957cb5b4481c7508b626
3
+ size 5304
training_params.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "data_path": "autotrain-7wlo5-redoy/autotrain-data",
3
+ "model": "google-t5/t5-base",
4
+ "username": "riken01",
5
+ "seed": 42,
6
+ "train_split": "train",
7
+ "valid_split": "validation",
8
+ "project_name": "autotrain-7wlo5-redoy",
9
+ "push_to_hub": true,
10
+ "text_column": "autotrain_text",
11
+ "target_column": "autotrain_label",
12
+ "lr": 5e-05,
13
+ "epochs": 3,
14
+ "max_seq_length": 128,
15
+ "max_target_length": 128,
16
+ "batch_size": 2,
17
+ "warmup_ratio": 0.1,
18
+ "gradient_accumulation": 1,
19
+ "optimizer": "adamw_torch",
20
+ "scheduler": "linear",
21
+ "weight_decay": 0.0,
22
+ "max_grad_norm": 1.0,
23
+ "logging_steps": -1,
24
+ "evaluation_strategy": "epoch",
25
+ "auto_find_batch_size": false,
26
+ "mixed_precision": "fp16",
27
+ "save_total_limit": 1,
28
+ "peft": false,
29
+ "quantization": "int4",
30
+ "lora_r": 16,
31
+ "lora_alpha": 32,
32
+ "lora_dropout": 0.05,
33
+ "target_modules": "all-linear",
34
+ "log": "tensorboard",
35
+ "early_stopping_patience": 5,
36
+ "early_stopping_threshold": 0.01
37
+ }