Upload 10 files
Browse files- config.json +65 -0
- rng_state.pth +3 -0
- runs/Jun05_09-58-51_a44b8c1d0683/1685959135.0244417/events.out.tfevents.1685959135.a44b8c1d0683.15968.1 +3 -0
- runs/Jun05_09-58-51_a44b8c1d0683/events.out.tfevents.1685959135.a44b8c1d0683.15968.0 +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer_config.json +16 -0
- trainer_state.json +292 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/content/hierarchical-transformers/data/PLMs/hi-transformer-p1-grouped/",
|
3 |
+
"architectures": [
|
4 |
+
"HATForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"encoder_layout": {
|
9 |
+
"0": {
|
10 |
+
"document_encoder": false,
|
11 |
+
"sentence_encoder": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"document_encoder": true,
|
15 |
+
"sentence_encoder": true
|
16 |
+
},
|
17 |
+
"2": {
|
18 |
+
"document_encoder": false,
|
19 |
+
"sentence_encoder": true
|
20 |
+
},
|
21 |
+
"3": {
|
22 |
+
"document_encoder": true,
|
23 |
+
"sentence_encoder": true
|
24 |
+
},
|
25 |
+
"4": {
|
26 |
+
"document_encoder": false,
|
27 |
+
"sentence_encoder": true
|
28 |
+
},
|
29 |
+
"5": {
|
30 |
+
"document_encoder": true,
|
31 |
+
"sentence_encoder": true
|
32 |
+
},
|
33 |
+
"6": {
|
34 |
+
"document_encoder": false,
|
35 |
+
"sentence_encoder": true
|
36 |
+
},
|
37 |
+
"7": {
|
38 |
+
"document_encoder": true,
|
39 |
+
"sentence_encoder": true
|
40 |
+
}
|
41 |
+
},
|
42 |
+
"hidden_act": "gelu",
|
43 |
+
"hidden_dropout_prob": 0.1,
|
44 |
+
"hidden_size": 256,
|
45 |
+
"initializer_range": 0.02,
|
46 |
+
"intermediate_size": 1024,
|
47 |
+
"layer_norm_eps": 1e-12,
|
48 |
+
"max_position_embeddings": 128,
|
49 |
+
"max_sentence_length": 128,
|
50 |
+
"max_sentence_size": 128,
|
51 |
+
"max_sentences": 5,
|
52 |
+
"model_max_length": 640,
|
53 |
+
"model_type": "hi-transformer",
|
54 |
+
"num_attention_heads": 4,
|
55 |
+
"num_hidden_layers": 8,
|
56 |
+
"output_past": true,
|
57 |
+
"pad_token_id": 0,
|
58 |
+
"parameters": 136350720,
|
59 |
+
"position_embedding_type": "absolute",
|
60 |
+
"torch_dtype": "float32",
|
61 |
+
"transformers_version": "4.29.2",
|
62 |
+
"type_vocab_size": 2,
|
63 |
+
"use_cache": true,
|
64 |
+
"vocab_size": 30522
|
65 |
+
}
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39b794c514652d0247951ef36f772c098cb42d95dcb0ab637d11adcd57bf4c8b
|
3 |
+
size 14575
|
runs/Jun05_09-58-51_a44b8c1d0683/1685959135.0244417/events.out.tfevents.1685959135.a44b8c1d0683.15968.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2fdf34f5304238b5830f85d202b4994cb9b687755e99aed46ce37e985afddd06
|
3 |
+
size 6054
|
runs/Jun05_09-58-51_a44b8c1d0683/events.out.tfevents.1685959135.a44b8c1d0683.15968.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:37ba5dc4693fd06a71bdf0fb8b7bb606a0202e29640eebdb1af1b8cc2e557aec
|
3 |
+
size 13114
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d59c7c8617e73104e711fa948abe57fd3ce7c6b1ac471ce13de0a705a209a2bf
|
3 |
+
size 627
|
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_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"clean_up_tokenization_spaces": true,
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"do_basic_tokenize": true,
|
5 |
+
"do_lower_case": true,
|
6 |
+
"mask_token": "[MASK]",
|
7 |
+
"model_max_length": 640,
|
8 |
+
"never_split": null,
|
9 |
+
"pad_token": "[PAD]",
|
10 |
+
"sep_token": "[SEP]",
|
11 |
+
"strip_accents": null,
|
12 |
+
"tokenize_chinese_chars": true,
|
13 |
+
"tokenizer_class": "BertTokenizer",
|
14 |
+
"unk_token": "[UNK]",
|
15 |
+
"use_fast": true
|
16 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,292 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0778911060510112,
|
5 |
+
"global_step": 20000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 2e-05,
|
13 |
+
"loss": 3.1471,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 4e-05,
|
19 |
+
"loss": 2.6593,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.08,
|
24 |
+
"learning_rate": 6e-05,
|
25 |
+
"loss": 2.6154,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.11,
|
30 |
+
"learning_rate": 8e-05,
|
31 |
+
"loss": 2.5936,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.13,
|
36 |
+
"learning_rate": 0.0001,
|
37 |
+
"loss": 2.5892,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.16,
|
42 |
+
"learning_rate": 9.777777777777778e-05,
|
43 |
+
"loss": 2.5745,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.19,
|
48 |
+
"learning_rate": 9.555555555555557e-05,
|
49 |
+
"loss": 2.5523,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.22,
|
54 |
+
"learning_rate": 9.333333333333334e-05,
|
55 |
+
"loss": 2.5062,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.24,
|
60 |
+
"learning_rate": 9.111111111111112e-05,
|
61 |
+
"loss": 2.5098,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.27,
|
66 |
+
"learning_rate": 8.888888888888889e-05,
|
67 |
+
"loss": 2.4883,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.27,
|
72 |
+
"eval_accuracy": 0.524201837373072,
|
73 |
+
"eval_loss": 2.5380380153656006,
|
74 |
+
"eval_runtime": 2042.925,
|
75 |
+
"eval_samples_per_second": 122.374,
|
76 |
+
"eval_steps_per_second": 7.648,
|
77 |
+
"step": 5000
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.3,
|
81 |
+
"learning_rate": 8.666666666666667e-05,
|
82 |
+
"loss": 2.4751,
|
83 |
+
"step": 5500
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.32,
|
87 |
+
"learning_rate": 8.444444444444444e-05,
|
88 |
+
"loss": 2.4578,
|
89 |
+
"step": 6000
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.35,
|
93 |
+
"learning_rate": 8.222222222222222e-05,
|
94 |
+
"loss": 2.4484,
|
95 |
+
"step": 6500
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.38,
|
99 |
+
"learning_rate": 8e-05,
|
100 |
+
"loss": 2.4408,
|
101 |
+
"step": 7000
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.4,
|
105 |
+
"learning_rate": 7.777777777777778e-05,
|
106 |
+
"loss": 2.418,
|
107 |
+
"step": 7500
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.43,
|
111 |
+
"learning_rate": 7.555555555555556e-05,
|
112 |
+
"loss": 2.4162,
|
113 |
+
"step": 8000
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.46,
|
117 |
+
"learning_rate": 7.333333333333333e-05,
|
118 |
+
"loss": 2.4067,
|
119 |
+
"step": 8500
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.49,
|
123 |
+
"learning_rate": 7.111111111111112e-05,
|
124 |
+
"loss": 2.3932,
|
125 |
+
"step": 9000
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.51,
|
129 |
+
"learning_rate": 6.88888888888889e-05,
|
130 |
+
"loss": 2.3785,
|
131 |
+
"step": 9500
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.54,
|
135 |
+
"learning_rate": 6.666666666666667e-05,
|
136 |
+
"loss": 2.3628,
|
137 |
+
"step": 10000
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.54,
|
141 |
+
"eval_accuracy": 0.530768888210296,
|
142 |
+
"eval_loss": 2.4960896968841553,
|
143 |
+
"eval_runtime": 2067.1205,
|
144 |
+
"eval_samples_per_second": 120.941,
|
145 |
+
"eval_steps_per_second": 7.559,
|
146 |
+
"step": 10000
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.57,
|
150 |
+
"learning_rate": 6.444444444444446e-05,
|
151 |
+
"loss": 2.3703,
|
152 |
+
"step": 10500
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.59,
|
156 |
+
"learning_rate": 6.222222222222222e-05,
|
157 |
+
"loss": 2.3593,
|
158 |
+
"step": 11000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.62,
|
162 |
+
"learning_rate": 6e-05,
|
163 |
+
"loss": 2.3603,
|
164 |
+
"step": 11500
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.65,
|
168 |
+
"learning_rate": 5.7777777777777776e-05,
|
169 |
+
"loss": 2.3447,
|
170 |
+
"step": 12000
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.67,
|
174 |
+
"learning_rate": 5.555555555555556e-05,
|
175 |
+
"loss": 2.3431,
|
176 |
+
"step": 12500
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.7,
|
180 |
+
"learning_rate": 5.333333333333333e-05,
|
181 |
+
"loss": 2.332,
|
182 |
+
"step": 13000
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.73,
|
186 |
+
"learning_rate": 5.111111111111111e-05,
|
187 |
+
"loss": 2.3199,
|
188 |
+
"step": 13500
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.75,
|
192 |
+
"learning_rate": 4.888888888888889e-05,
|
193 |
+
"loss": 2.3178,
|
194 |
+
"step": 14000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.78,
|
198 |
+
"learning_rate": 4.666666666666667e-05,
|
199 |
+
"loss": 2.3043,
|
200 |
+
"step": 14500
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.81,
|
204 |
+
"learning_rate": 4.4444444444444447e-05,
|
205 |
+
"loss": 2.3087,
|
206 |
+
"step": 15000
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.81,
|
210 |
+
"eval_accuracy": 0.5384137066763606,
|
211 |
+
"eval_loss": 2.4490230083465576,
|
212 |
+
"eval_runtime": 2059.234,
|
213 |
+
"eval_samples_per_second": 121.404,
|
214 |
+
"eval_steps_per_second": 7.588,
|
215 |
+
"step": 15000
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.84,
|
219 |
+
"learning_rate": 4.222222222222222e-05,
|
220 |
+
"loss": 2.2956,
|
221 |
+
"step": 15500
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.86,
|
225 |
+
"learning_rate": 4e-05,
|
226 |
+
"loss": 2.2908,
|
227 |
+
"step": 16000
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.89,
|
231 |
+
"learning_rate": 3.777777777777778e-05,
|
232 |
+
"loss": 2.3026,
|
233 |
+
"step": 16500
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.92,
|
237 |
+
"learning_rate": 3.555555555555556e-05,
|
238 |
+
"loss": 2.2831,
|
239 |
+
"step": 17000
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.94,
|
243 |
+
"learning_rate": 3.3333333333333335e-05,
|
244 |
+
"loss": 2.2937,
|
245 |
+
"step": 17500
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.97,
|
249 |
+
"learning_rate": 3.111111111111111e-05,
|
250 |
+
"loss": 2.2769,
|
251 |
+
"step": 18000
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 1.0,
|
255 |
+
"learning_rate": 2.8888888888888888e-05,
|
256 |
+
"loss": 2.2593,
|
257 |
+
"step": 18500
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.02,
|
261 |
+
"learning_rate": 2.6666666666666667e-05,
|
262 |
+
"loss": 2.245,
|
263 |
+
"step": 19000
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.05,
|
267 |
+
"learning_rate": 2.4444444444444445e-05,
|
268 |
+
"loss": 2.2561,
|
269 |
+
"step": 19500
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.08,
|
273 |
+
"learning_rate": 2.2222222222222223e-05,
|
274 |
+
"loss": 2.2413,
|
275 |
+
"step": 20000
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.08,
|
279 |
+
"eval_accuracy": 0.5411358684415605,
|
280 |
+
"eval_loss": 2.435114860534668,
|
281 |
+
"eval_runtime": 2054.406,
|
282 |
+
"eval_samples_per_second": 121.69,
|
283 |
+
"eval_steps_per_second": 7.606,
|
284 |
+
"step": 20000
|
285 |
+
}
|
286 |
+
],
|
287 |
+
"max_steps": 25000,
|
288 |
+
"num_train_epochs": 2,
|
289 |
+
"total_flos": 1.505927324733604e+17,
|
290 |
+
"trial_name": null,
|
291 |
+
"trial_params": null
|
292 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e778a06bf77c3d21f2666e794a2f9d05b629a3142f883b1b389ce32b9e550f08
|
3 |
+
size 3963
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|