Training in progress, step 10000
Browse files- .gitignore +1 -0
- README.md +6 -0
- config.json +33 -0
- last-checkpoint/config.json +33 -0
- last-checkpoint/merges.txt +0 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/pytorch_model.bin +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scaler.pt +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +7 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +14 -0
- last-checkpoint/trainer_state.json +316 -0
- last-checkpoint/training_args.bin +3 -0
- last-checkpoint/vocab.json +0 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
README.md
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
mask_token: "[MASK]"
|
3 |
+
widget:
|
4 |
+
- text: 대한민국의 수도는 [MASK]입니다.
|
5 |
+
---
|
6 |
+
![image](https://i.imgur.com/brETMw6.jpg)
|
config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Bingsu/my_mobilebert_untrained",
|
3 |
+
"architectures": [
|
4 |
+
"MobileBertForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_activation": false,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"embedding_size": 128,
|
10 |
+
"hidden_act": "relu",
|
11 |
+
"hidden_dropout_prob": 0.0,
|
12 |
+
"hidden_size": 512,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 512,
|
15 |
+
"intra_bottleneck_size": 128,
|
16 |
+
"key_query_shared_bottleneck": true,
|
17 |
+
"layer_norm_eps": 1e-12,
|
18 |
+
"max_position_embeddings": 512,
|
19 |
+
"model_type": "mobilebert",
|
20 |
+
"normalization_type": "no_norm",
|
21 |
+
"num_attention_heads": 4,
|
22 |
+
"num_feedforward_networks": 4,
|
23 |
+
"num_hidden_layers": 24,
|
24 |
+
"pad_token_id": 1,
|
25 |
+
"torch_dtype": "float32",
|
26 |
+
"transformers_version": "4.21.2",
|
27 |
+
"trigram_input": true,
|
28 |
+
"true_hidden_size": 128,
|
29 |
+
"type_vocab_size": 2,
|
30 |
+
"use_bottleneck": true,
|
31 |
+
"use_bottleneck_attention": false,
|
32 |
+
"vocab_size": 30522
|
33 |
+
}
|
last-checkpoint/config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Bingsu/my_mobilebert_untrained",
|
3 |
+
"architectures": [
|
4 |
+
"MobileBertForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_activation": false,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"embedding_size": 128,
|
10 |
+
"hidden_act": "relu",
|
11 |
+
"hidden_dropout_prob": 0.0,
|
12 |
+
"hidden_size": 512,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 512,
|
15 |
+
"intra_bottleneck_size": 128,
|
16 |
+
"key_query_shared_bottleneck": true,
|
17 |
+
"layer_norm_eps": 1e-12,
|
18 |
+
"max_position_embeddings": 512,
|
19 |
+
"model_type": "mobilebert",
|
20 |
+
"normalization_type": "no_norm",
|
21 |
+
"num_attention_heads": 4,
|
22 |
+
"num_feedforward_networks": 4,
|
23 |
+
"num_hidden_layers": 24,
|
24 |
+
"pad_token_id": 1,
|
25 |
+
"torch_dtype": "float32",
|
26 |
+
"transformers_version": "4.21.2",
|
27 |
+
"trigram_input": true,
|
28 |
+
"true_hidden_size": 128,
|
29 |
+
"type_vocab_size": 2,
|
30 |
+
"use_bottleneck": true,
|
31 |
+
"use_bottleneck_attention": false,
|
32 |
+
"vocab_size": 30522
|
33 |
+
}
|
last-checkpoint/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
last-checkpoint/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53f86b7c1030ebca2a6c99c8f28aa18cf67208a00e08fce91eea733b0bec240d
|
3 |
+
size 586826597
|
last-checkpoint/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cf968dcd05b6ea7291b9aeb7e5ef41b98d5f49af58bdfb3e0c4d109ecd867493
|
3 |
+
size 146774203
|
last-checkpoint/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61fc86e3bcf87c025896dabb3d26ce5ca4103ec874557dd3084c738eb61ad888
|
3 |
+
size 14503
|
last-checkpoint/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e22a3cb16dddcf5c937c1bec07637df13115b6ed917ca6d1cf2116057954b0ce
|
3 |
+
size 559
|
last-checkpoint/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3821f79e5d56fda48d85057de6b8b28f1ffc595c942da877ef01bd25d14ff514
|
3 |
+
size 733553608
|
last-checkpoint/special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
last-checkpoint/tokenizer_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"do_lower_case": true,
|
4 |
+
"mask_token": "[MASK]",
|
5 |
+
"model_max_length": 512,
|
6 |
+
"name_or_path": "Bingsu/my_mobilebert_untrained",
|
7 |
+
"pad_token": "[PAD]",
|
8 |
+
"sep_token": "[SEP]",
|
9 |
+
"special_tokens_map_file": null,
|
10 |
+
"strip_accents": null,
|
11 |
+
"tokenize_chinese_chars": true,
|
12 |
+
"tokenizer_class": "MobileBertTokenizer",
|
13 |
+
"unk_token": "[UNK]"
|
14 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
@@ -0,0 +1,316 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.04297249330703417,
|
5 |
+
"global_step": 10000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.0,
|
12 |
+
"learning_rate": 0.0004000105276443632,
|
13 |
+
"loss": 10.1199,
|
14 |
+
"step": 200
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.0,
|
18 |
+
"learning_rate": 0.00040004211053127486,
|
19 |
+
"loss": 9.997,
|
20 |
+
"step": 400
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.0,
|
24 |
+
"learning_rate": 0.000400094748522194,
|
25 |
+
"loss": 9.9386,
|
26 |
+
"step": 600
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.0,
|
30 |
+
"learning_rate": 0.00040016844138622554,
|
31 |
+
"loss": 9.8988,
|
32 |
+
"step": 800
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.0,
|
36 |
+
"learning_rate": 0.0004002631888001141,
|
37 |
+
"loss": 9.8579,
|
38 |
+
"step": 1000
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.01,
|
42 |
+
"learning_rate": 0.0004003789903482477,
|
43 |
+
"loss": 9.8159,
|
44 |
+
"step": 1200
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.01,
|
48 |
+
"learning_rate": 0.0004005158455226594,
|
49 |
+
"loss": 9.7867,
|
50 |
+
"step": 1400
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.01,
|
54 |
+
"learning_rate": 0.0004006737537230326,
|
55 |
+
"loss": 9.7605,
|
56 |
+
"step": 1600
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.01,
|
60 |
+
"learning_rate": 0.0004008527142566991,
|
61 |
+
"loss": 9.7357,
|
62 |
+
"step": 1800
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.01,
|
66 |
+
"learning_rate": 0.0004010527263386479,
|
67 |
+
"loss": 9.7138,
|
68 |
+
"step": 2000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.01,
|
72 |
+
"learning_rate": 0.00040127378909152016,
|
73 |
+
"loss": 9.6894,
|
74 |
+
"step": 2200
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.01,
|
78 |
+
"learning_rate": 0.000401515901545621,
|
79 |
+
"loss": 9.6634,
|
80 |
+
"step": 2400
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.01,
|
84 |
+
"learning_rate": 0.00040177906263891804,
|
85 |
+
"loss": 9.6451,
|
86 |
+
"step": 2600
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.01,
|
90 |
+
"learning_rate": 0.00040206327121705167,
|
91 |
+
"loss": 9.6279,
|
92 |
+
"step": 2800
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.01,
|
96 |
+
"learning_rate": 0.00040236852603333685,
|
97 |
+
"loss": 9.6038,
|
98 |
+
"step": 3000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.01,
|
102 |
+
"learning_rate": 0.0004026948257487631,
|
103 |
+
"loss": 9.5874,
|
104 |
+
"step": 3200
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.01,
|
108 |
+
"learning_rate": 0.00040304216893201697,
|
109 |
+
"loss": 9.5729,
|
110 |
+
"step": 3400
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.02,
|
114 |
+
"learning_rate": 0.0004034105540594666,
|
115 |
+
"loss": 9.547,
|
116 |
+
"step": 3600
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.02,
|
120 |
+
"learning_rate": 0.0004037999795151858,
|
121 |
+
"loss": 9.5348,
|
122 |
+
"step": 3800
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.02,
|
126 |
+
"learning_rate": 0.0004042104435909525,
|
127 |
+
"loss": 9.5207,
|
128 |
+
"step": 4000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.02,
|
132 |
+
"learning_rate": 0.0004046419444862573,
|
133 |
+
"loss": 9.5061,
|
134 |
+
"step": 4200
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.02,
|
138 |
+
"learning_rate": 0.0004050944803083139,
|
139 |
+
"loss": 9.493,
|
140 |
+
"step": 4400
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.02,
|
144 |
+
"learning_rate": 0.0004055680490720661,
|
145 |
+
"loss": 9.4782,
|
146 |
+
"step": 4600
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.02,
|
150 |
+
"learning_rate": 0.0004060626487001964,
|
151 |
+
"loss": 9.4636,
|
152 |
+
"step": 4800
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.02,
|
156 |
+
"learning_rate": 0.0004065782770231313,
|
157 |
+
"loss": 9.4546,
|
158 |
+
"step": 5000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.02,
|
162 |
+
"learning_rate": 0.000407114931779062,
|
163 |
+
"loss": 9.4453,
|
164 |
+
"step": 5200
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.02,
|
168 |
+
"learning_rate": 0.00040767261061393917,
|
169 |
+
"loss": 9.4174,
|
170 |
+
"step": 5400
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.02,
|
174 |
+
"learning_rate": 0.00040825131108149573,
|
175 |
+
"loss": 9.4159,
|
176 |
+
"step": 5600
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.02,
|
180 |
+
"learning_rate": 0.00040885103064325357,
|
181 |
+
"loss": 9.3993,
|
182 |
+
"step": 5800
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.03,
|
186 |
+
"learning_rate": 0.00040947176666852707,
|
187 |
+
"loss": 9.3953,
|
188 |
+
"step": 6000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.03,
|
192 |
+
"learning_rate": 0.00041011351643444917,
|
193 |
+
"loss": 9.3854,
|
194 |
+
"step": 6200
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.03,
|
198 |
+
"learning_rate": 0.0004107762771259713,
|
199 |
+
"loss": 9.3679,
|
200 |
+
"step": 6400
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.03,
|
204 |
+
"learning_rate": 0.0004114600458358809,
|
205 |
+
"loss": 9.3595,
|
206 |
+
"step": 6600
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.03,
|
210 |
+
"learning_rate": 0.00041216481956481664,
|
211 |
+
"loss": 9.3504,
|
212 |
+
"step": 6800
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.03,
|
216 |
+
"learning_rate": 0.00041289059522127414,
|
217 |
+
"loss": 9.3417,
|
218 |
+
"step": 7000
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.03,
|
222 |
+
"learning_rate": 0.0004136373696216229,
|
223 |
+
"loss": 9.3275,
|
224 |
+
"step": 7200
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.03,
|
228 |
+
"learning_rate": 0.0004144051394901274,
|
229 |
+
"loss": 9.3201,
|
230 |
+
"step": 7400
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.03,
|
234 |
+
"learning_rate": 0.0004151939014589469,
|
235 |
+
"loss": 9.3123,
|
236 |
+
"step": 7600
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.03,
|
240 |
+
"learning_rate": 0.0004160036520681667,
|
241 |
+
"loss": 9.3084,
|
242 |
+
"step": 7800
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.03,
|
246 |
+
"learning_rate": 0.0004168343877657965,
|
247 |
+
"loss": 9.2954,
|
248 |
+
"step": 8000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.04,
|
252 |
+
"learning_rate": 0.00041768179413688954,
|
253 |
+
"loss": 9.2862,
|
254 |
+
"step": 8200
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.04,
|
258 |
+
"learning_rate": 0.00041855438410810103,
|
259 |
+
"loss": 9.283,
|
260 |
+
"step": 8400
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.04,
|
264 |
+
"learning_rate": 0.00041944794797888797,
|
265 |
+
"loss": 9.2711,
|
266 |
+
"step": 8600
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.04,
|
270 |
+
"learning_rate": 0.00042036248182962185,
|
271 |
+
"loss": 9.2726,
|
272 |
+
"step": 8800
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.04,
|
276 |
+
"learning_rate": 0.0004212979816486783,
|
277 |
+
"loss": 9.2621,
|
278 |
+
"step": 9000
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.04,
|
282 |
+
"learning_rate": 0.00042225444333247354,
|
283 |
+
"loss": 9.2527,
|
284 |
+
"step": 9200
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.04,
|
288 |
+
"learning_rate": 0.0004232318626854678,
|
289 |
+
"loss": 9.2453,
|
290 |
+
"step": 9400
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.04,
|
294 |
+
"learning_rate": 0.0004242302354201949,
|
295 |
+
"loss": 9.2314,
|
296 |
+
"step": 9600
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.04,
|
300 |
+
"learning_rate": 0.000425249557157276,
|
301 |
+
"loss": 9.2337,
|
302 |
+
"step": 9800
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.04,
|
306 |
+
"learning_rate": 0.00042628982342543184,
|
307 |
+
"loss": 9.2276,
|
308 |
+
"step": 10000
|
309 |
+
}
|
310 |
+
],
|
311 |
+
"max_steps": 1000000,
|
312 |
+
"num_train_epochs": 5,
|
313 |
+
"total_flos": 1.593829982208e+16,
|
314 |
+
"trial_name": null,
|
315 |
+
"trial_params": null
|
316 |
+
}
|
last-checkpoint/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c33b030231937c49711cbb55890f65b86f81a68638a6a5c30e4f67ed0b41b6f
|
3 |
+
size 3375
|
last-checkpoint/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cf968dcd05b6ea7291b9aeb7e5ef41b98d5f49af58bdfb3e0c4d109ecd867493
|
3 |
+
size 146774203
|
special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"do_lower_case": true,
|
4 |
+
"mask_token": "[MASK]",
|
5 |
+
"model_max_length": 512,
|
6 |
+
"name_or_path": "Bingsu/my_mobilebert_untrained",
|
7 |
+
"pad_token": "[PAD]",
|
8 |
+
"sep_token": "[SEP]",
|
9 |
+
"special_tokens_map_file": null,
|
10 |
+
"strip_accents": null,
|
11 |
+
"tokenize_chinese_chars": true,
|
12 |
+
"tokenizer_class": "MobileBertTokenizer",
|
13 |
+
"unk_token": "[UNK]"
|
14 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c33b030231937c49711cbb55890f65b86f81a68638a6a5c30e4f67ed0b41b6f
|
3 |
+
size 3375
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|