demdecuong
commited on
Commit
•
8800bcc
1
Parent(s):
bf3b39b
update weight
Browse files- config.json +24 -0
- eval_results.txt +12 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- train_results.txt +3 -0
- trainer_state.json +621 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "result/my-unsup-simcse-bert-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertModel"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"gradient_checkpointing": false,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 3072,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 512,
|
15 |
+
"model_type": "bert",
|
16 |
+
"num_attention_heads": 12,
|
17 |
+
"num_hidden_layers": 12,
|
18 |
+
"pad_token_id": 0,
|
19 |
+
"position_embedding_type": "absolute",
|
20 |
+
"transformers_version": "4.2.1",
|
21 |
+
"type_vocab_size": 2,
|
22 |
+
"use_cache": true,
|
23 |
+
"vocab_size": 30522
|
24 |
+
}
|
eval_results.txt
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
epoch = 1.0
|
2 |
+
eval_CR = 88.36
|
3 |
+
eval_MPQA = 89.04
|
4 |
+
eval_MR = 81.06
|
5 |
+
eval_MRPC = 76.06
|
6 |
+
eval_SST2 = 87.16
|
7 |
+
eval_SUBJ = 99.65
|
8 |
+
eval_TREC = 79.7
|
9 |
+
eval_avg_sts = 0.8421121351950032
|
10 |
+
eval_avg_transfer = 85.8614285714286
|
11 |
+
eval_sickr_spearman = 0.820764443190438
|
12 |
+
eval_stsb_spearman = 0.8634598271995685
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2880a41c4eb9a080da0c034f8df451798181a3ea8ee839cb33f75289df3417d8
|
3 |
+
size 875982637
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ac10a7342bcb924fa670cac5d0ce8e53792e48f3dc951922c6b5aa969d0145cb
|
3 |
+
size 438000759
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c34cd051aa6ad6fcf6928a695e010571d3c11ed8823b2adfbdc9e20bb28bc4c0
|
3 |
+
size 623
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "princeton-nlp/sup-simcse-bert-base-uncased", "special_tokens_map_file": "/home/minh/.cache/huggingface/transformers/8c406286308d13c3a53bc10c3d1a2d5113d4e46a34cb6ec5ee06e5d9762c462c.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "do_basic_tokenize": true, "never_split": null}
|
train_results.txt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
epoch = 1.0
|
2 |
+
train_runtime = 4791.076
|
3 |
+
train_samples_per_second = 1.859
|
trainer_state.json
ADDED
@@ -0,0 +1,621 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8634598271995685,
|
3 |
+
"best_model_checkpoint": "result/my-unsup-simcse-bert-base-uncased",
|
4 |
+
"epoch": 1.0,
|
5 |
+
"global_step": 8909,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"eval_avg_sts": 0.8421121351950032,
|
13 |
+
"eval_sickr_spearman": 0.820764443190438,
|
14 |
+
"eval_stsb_spearman": 0.8634598271995685,
|
15 |
+
"step": 125
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.03,
|
19 |
+
"eval_avg_sts": 0.8272876260858589,
|
20 |
+
"eval_sickr_spearman": 0.8016297649684213,
|
21 |
+
"eval_stsb_spearman": 0.8529454872032965,
|
22 |
+
"step": 250
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"epoch": 0.04,
|
26 |
+
"eval_avg_sts": 0.8349849313583235,
|
27 |
+
"eval_sickr_spearman": 0.8131009368626829,
|
28 |
+
"eval_stsb_spearman": 0.8568689258539641,
|
29 |
+
"step": 375
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.06,
|
33 |
+
"learning_rate": 2.831630935009541e-05,
|
34 |
+
"loss": 0.0001,
|
35 |
+
"step": 500
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 0.06,
|
39 |
+
"eval_avg_sts": 0.8324840345106272,
|
40 |
+
"eval_sickr_spearman": 0.8115295033169321,
|
41 |
+
"eval_stsb_spearman": 0.8534385657043223,
|
42 |
+
"step": 500
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.07,
|
46 |
+
"eval_avg_sts": 0.8342309434678776,
|
47 |
+
"eval_sickr_spearman": 0.8101102323005372,
|
48 |
+
"eval_stsb_spearman": 0.858351654635218,
|
49 |
+
"step": 625
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.08,
|
53 |
+
"eval_avg_sts": 0.8358817878428014,
|
54 |
+
"eval_sickr_spearman": 0.8152456216040973,
|
55 |
+
"eval_stsb_spearman": 0.8565179540815056,
|
56 |
+
"step": 750
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.1,
|
60 |
+
"eval_avg_sts": 0.8337641137899596,
|
61 |
+
"eval_sickr_spearman": 0.8073721233008548,
|
62 |
+
"eval_stsb_spearman": 0.8601561042790645,
|
63 |
+
"step": 875
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.11,
|
67 |
+
"learning_rate": 2.663261870019082e-05,
|
68 |
+
"loss": 0.0001,
|
69 |
+
"step": 1000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.11,
|
73 |
+
"eval_avg_sts": 0.8318439410449322,
|
74 |
+
"eval_sickr_spearman": 0.811522586838325,
|
75 |
+
"eval_stsb_spearman": 0.8521652952515394,
|
76 |
+
"step": 1000
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.13,
|
80 |
+
"eval_avg_sts": 0.8304909909975411,
|
81 |
+
"eval_sickr_spearman": 0.8131338381671679,
|
82 |
+
"eval_stsb_spearman": 0.8478481438279143,
|
83 |
+
"step": 1125
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.14,
|
87 |
+
"eval_avg_sts": 0.829039779377259,
|
88 |
+
"eval_sickr_spearman": 0.8045460213233376,
|
89 |
+
"eval_stsb_spearman": 0.8535335374311805,
|
90 |
+
"step": 1250
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 0.15,
|
94 |
+
"eval_avg_sts": 0.8307787364299535,
|
95 |
+
"eval_sickr_spearman": 0.8137019980660793,
|
96 |
+
"eval_stsb_spearman": 0.8478554747938275,
|
97 |
+
"step": 1375
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 0.17,
|
101 |
+
"learning_rate": 2.4948928050286226e-05,
|
102 |
+
"loss": 0.0001,
|
103 |
+
"step": 1500
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 0.17,
|
107 |
+
"eval_avg_sts": 0.831658365337354,
|
108 |
+
"eval_sickr_spearman": 0.8159185373352453,
|
109 |
+
"eval_stsb_spearman": 0.8473981933394628,
|
110 |
+
"step": 1500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.18,
|
114 |
+
"eval_avg_sts": 0.8294868187051994,
|
115 |
+
"eval_sickr_spearman": 0.8144402841262847,
|
116 |
+
"eval_stsb_spearman": 0.8445333532841142,
|
117 |
+
"step": 1625
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.2,
|
121 |
+
"eval_avg_sts": 0.8268186287541337,
|
122 |
+
"eval_sickr_spearman": 0.8082519570169981,
|
123 |
+
"eval_stsb_spearman": 0.8453853004912691,
|
124 |
+
"step": 1750
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.21,
|
128 |
+
"eval_avg_sts": 0.8313195396147719,
|
129 |
+
"eval_sickr_spearman": 0.8177095210456699,
|
130 |
+
"eval_stsb_spearman": 0.844929558183874,
|
131 |
+
"step": 1875
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.22,
|
135 |
+
"learning_rate": 2.3265237400381636e-05,
|
136 |
+
"loss": 0.0001,
|
137 |
+
"step": 2000
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.22,
|
141 |
+
"eval_avg_sts": 0.8266715243566689,
|
142 |
+
"eval_sickr_spearman": 0.8150033527284436,
|
143 |
+
"eval_stsb_spearman": 0.8383396959848942,
|
144 |
+
"step": 2000
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.24,
|
148 |
+
"eval_avg_sts": 0.8293624623711227,
|
149 |
+
"eval_sickr_spearman": 0.8175795488851783,
|
150 |
+
"eval_stsb_spearman": 0.841145375857067,
|
151 |
+
"step": 2125
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 0.25,
|
155 |
+
"eval_avg_sts": 0.8322036045569485,
|
156 |
+
"eval_sickr_spearman": 0.8175970802372032,
|
157 |
+
"eval_stsb_spearman": 0.8468101288766937,
|
158 |
+
"step": 2250
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.27,
|
162 |
+
"eval_avg_sts": 0.8314845597378098,
|
163 |
+
"eval_sickr_spearman": 0.8156567678324077,
|
164 |
+
"eval_stsb_spearman": 0.8473123516432118,
|
165 |
+
"step": 2375
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.28,
|
169 |
+
"learning_rate": 2.1581546750477045e-05,
|
170 |
+
"loss": 0.0001,
|
171 |
+
"step": 2500
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.28,
|
175 |
+
"eval_avg_sts": 0.8298162063992072,
|
176 |
+
"eval_sickr_spearman": 0.8121817176433601,
|
177 |
+
"eval_stsb_spearman": 0.8474506951550543,
|
178 |
+
"step": 2500
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 0.29,
|
182 |
+
"eval_avg_sts": 0.8288600170060807,
|
183 |
+
"eval_sickr_spearman": 0.8109708535761053,
|
184 |
+
"eval_stsb_spearman": 0.8467491804360561,
|
185 |
+
"step": 2625
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.31,
|
189 |
+
"eval_avg_sts": 0.8307741487193253,
|
190 |
+
"eval_sickr_spearman": 0.8109006321058029,
|
191 |
+
"eval_stsb_spearman": 0.8506476653328477,
|
192 |
+
"step": 2750
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.32,
|
196 |
+
"eval_avg_sts": 0.8299073386983135,
|
197 |
+
"eval_sickr_spearman": 0.8106493814141802,
|
198 |
+
"eval_stsb_spearman": 0.8491652959824467,
|
199 |
+
"step": 2875
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 0.34,
|
203 |
+
"learning_rate": 1.9897856100572458e-05,
|
204 |
+
"loss": 0.0,
|
205 |
+
"step": 3000
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.34,
|
209 |
+
"eval_avg_sts": 0.82947527087573,
|
210 |
+
"eval_sickr_spearman": 0.808709645382602,
|
211 |
+
"eval_stsb_spearman": 0.8502408963688581,
|
212 |
+
"step": 3000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.35,
|
216 |
+
"eval_avg_sts": 0.8247286673217802,
|
217 |
+
"eval_sickr_spearman": 0.8044403048690724,
|
218 |
+
"eval_stsb_spearman": 0.8450170297744881,
|
219 |
+
"step": 3125
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.36,
|
223 |
+
"eval_avg_sts": 0.8219332102471997,
|
224 |
+
"eval_sickr_spearman": 0.806001651883521,
|
225 |
+
"eval_stsb_spearman": 0.8378647686108784,
|
226 |
+
"step": 3250
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.38,
|
230 |
+
"eval_avg_sts": 0.8111430899888176,
|
231 |
+
"eval_sickr_spearman": 0.7991139439061943,
|
232 |
+
"eval_stsb_spearman": 0.8231722360714409,
|
233 |
+
"step": 3375
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.39,
|
237 |
+
"learning_rate": 1.8214165450667864e-05,
|
238 |
+
"loss": 0.0001,
|
239 |
+
"step": 3500
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.39,
|
243 |
+
"eval_avg_sts": 0.807692571896139,
|
244 |
+
"eval_sickr_spearman": 0.7927039052326695,
|
245 |
+
"eval_stsb_spearman": 0.8226812385596084,
|
246 |
+
"step": 3500
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.41,
|
250 |
+
"eval_avg_sts": 0.8135397552735281,
|
251 |
+
"eval_sickr_spearman": 0.7942567027110624,
|
252 |
+
"eval_stsb_spearman": 0.8328228078359937,
|
253 |
+
"step": 3625
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 0.42,
|
257 |
+
"eval_avg_sts": 0.8047796899452875,
|
258 |
+
"eval_sickr_spearman": 0.7930084224157872,
|
259 |
+
"eval_stsb_spearman": 0.8165509574747878,
|
260 |
+
"step": 3750
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.43,
|
264 |
+
"eval_avg_sts": 0.8132531030215364,
|
265 |
+
"eval_sickr_spearman": 0.7956294796212651,
|
266 |
+
"eval_stsb_spearman": 0.8308767264218079,
|
267 |
+
"step": 3875
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.45,
|
271 |
+
"learning_rate": 1.6530474800763274e-05,
|
272 |
+
"loss": 0.0001,
|
273 |
+
"step": 4000
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 0.45,
|
277 |
+
"eval_avg_sts": 0.8062306897971802,
|
278 |
+
"eval_sickr_spearman": 0.7966211297415562,
|
279 |
+
"eval_stsb_spearman": 0.8158402498528041,
|
280 |
+
"step": 4000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 0.46,
|
284 |
+
"eval_avg_sts": 0.8163206710413518,
|
285 |
+
"eval_sickr_spearman": 0.80029277122879,
|
286 |
+
"eval_stsb_spearman": 0.8323485708539136,
|
287 |
+
"step": 4125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 0.48,
|
291 |
+
"eval_avg_sts": 0.8165420284081878,
|
292 |
+
"eval_sickr_spearman": 0.8011630467557471,
|
293 |
+
"eval_stsb_spearman": 0.8319210100606285,
|
294 |
+
"step": 4250
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 0.49,
|
298 |
+
"eval_avg_sts": 0.8187419551614288,
|
299 |
+
"eval_sickr_spearman": 0.8018022926847871,
|
300 |
+
"eval_stsb_spearman": 0.8356816176380705,
|
301 |
+
"step": 4375
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 0.51,
|
305 |
+
"learning_rate": 1.4846784150858683e-05,
|
306 |
+
"loss": 0.0,
|
307 |
+
"step": 4500
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 0.51,
|
311 |
+
"eval_avg_sts": 0.8195743034210232,
|
312 |
+
"eval_sickr_spearman": 0.8021342836579273,
|
313 |
+
"eval_stsb_spearman": 0.8370143231841193,
|
314 |
+
"step": 4500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.52,
|
318 |
+
"eval_avg_sts": 0.8193325419945161,
|
319 |
+
"eval_sickr_spearman": 0.8013091573663218,
|
320 |
+
"eval_stsb_spearman": 0.8373559266227103,
|
321 |
+
"step": 4625
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 0.53,
|
325 |
+
"eval_avg_sts": 0.8183510820774517,
|
326 |
+
"eval_sickr_spearman": 0.7999737966841394,
|
327 |
+
"eval_stsb_spearman": 0.836728367470764,
|
328 |
+
"step": 4750
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.55,
|
332 |
+
"eval_avg_sts": 0.8168697079122389,
|
333 |
+
"eval_sickr_spearman": 0.7952836076598091,
|
334 |
+
"eval_stsb_spearman": 0.8384558081646686,
|
335 |
+
"step": 4875
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 0.56,
|
339 |
+
"learning_rate": 1.3163093500954093e-05,
|
340 |
+
"loss": 0.0001,
|
341 |
+
"step": 5000
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 0.56,
|
345 |
+
"eval_avg_sts": 0.8170635469307033,
|
346 |
+
"eval_sickr_spearman": 0.796297256024559,
|
347 |
+
"eval_stsb_spearman": 0.8378298378368476,
|
348 |
+
"step": 5000
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 0.58,
|
352 |
+
"eval_avg_sts": 0.8188600980624421,
|
353 |
+
"eval_sickr_spearman": 0.7980452999302978,
|
354 |
+
"eval_stsb_spearman": 0.8396748961945864,
|
355 |
+
"step": 5125
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 0.59,
|
359 |
+
"eval_avg_sts": 0.8179261372960408,
|
360 |
+
"eval_sickr_spearman": 0.7982228709123145,
|
361 |
+
"eval_stsb_spearman": 0.8376294036797671,
|
362 |
+
"step": 5250
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 0.6,
|
366 |
+
"eval_avg_sts": 0.8152302038847802,
|
367 |
+
"eval_sickr_spearman": 0.8007572319796964,
|
368 |
+
"eval_stsb_spearman": 0.829703175789864,
|
369 |
+
"step": 5375
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.62,
|
373 |
+
"learning_rate": 1.14794028510495e-05,
|
374 |
+
"loss": 0.0,
|
375 |
+
"step": 5500
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 0.62,
|
379 |
+
"eval_avg_sts": 0.8173976618782237,
|
380 |
+
"eval_sickr_spearman": 0.8015423483638039,
|
381 |
+
"eval_stsb_spearman": 0.8332529753926434,
|
382 |
+
"step": 5500
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 0.63,
|
386 |
+
"eval_avg_sts": 0.8175197579562594,
|
387 |
+
"eval_sickr_spearman": 0.8017781330407636,
|
388 |
+
"eval_stsb_spearman": 0.8332613828717551,
|
389 |
+
"step": 5625
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.65,
|
393 |
+
"eval_avg_sts": 0.8155439357729999,
|
394 |
+
"eval_sickr_spearman": 0.8033175778729565,
|
395 |
+
"eval_stsb_spearman": 0.8277702936730432,
|
396 |
+
"step": 5750
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.66,
|
400 |
+
"eval_avg_sts": 0.816245667230588,
|
401 |
+
"eval_sickr_spearman": 0.8031514863241836,
|
402 |
+
"eval_stsb_spearman": 0.8293398481369924,
|
403 |
+
"step": 5875
|
404 |
+
},
|
405 |
+
{
|
406 |
+
"epoch": 0.67,
|
407 |
+
"learning_rate": 9.79571220114491e-06,
|
408 |
+
"loss": 0.0,
|
409 |
+
"step": 6000
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.67,
|
413 |
+
"eval_avg_sts": 0.8173247407144557,
|
414 |
+
"eval_sickr_spearman": 0.802800715190381,
|
415 |
+
"eval_stsb_spearman": 0.8318487662385303,
|
416 |
+
"step": 6000
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.69,
|
420 |
+
"eval_avg_sts": 0.8175691152062283,
|
421 |
+
"eval_sickr_spearman": 0.8076647768019173,
|
422 |
+
"eval_stsb_spearman": 0.8274734536105394,
|
423 |
+
"step": 6125
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 0.7,
|
427 |
+
"eval_avg_sts": 0.8185213107446343,
|
428 |
+
"eval_sickr_spearman": 0.805431666802755,
|
429 |
+
"eval_stsb_spearman": 0.8316109546865137,
|
430 |
+
"step": 6250
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 0.72,
|
434 |
+
"eval_avg_sts": 0.8186800157417145,
|
435 |
+
"eval_sickr_spearman": 0.8050108663230556,
|
436 |
+
"eval_stsb_spearman": 0.8323491651603734,
|
437 |
+
"step": 6375
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 0.73,
|
441 |
+
"learning_rate": 8.112021551240318e-06,
|
442 |
+
"loss": 0.0001,
|
443 |
+
"step": 6500
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.73,
|
447 |
+
"eval_avg_sts": 0.8188678392487967,
|
448 |
+
"eval_sickr_spearman": 0.8050220575696909,
|
449 |
+
"eval_stsb_spearman": 0.8327136209279026,
|
450 |
+
"step": 6500
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.74,
|
454 |
+
"eval_avg_sts": 0.8176842353525476,
|
455 |
+
"eval_sickr_spearman": 0.805601264621933,
|
456 |
+
"eval_stsb_spearman": 0.8297672060831622,
|
457 |
+
"step": 6625
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.76,
|
461 |
+
"eval_avg_sts": 0.8176889108605423,
|
462 |
+
"eval_sickr_spearman": 0.8056947811764331,
|
463 |
+
"eval_stsb_spearman": 0.8296830405446516,
|
464 |
+
"step": 6750
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.77,
|
468 |
+
"eval_avg_sts": 0.8196884531122979,
|
469 |
+
"eval_sickr_spearman": 0.8050268606798346,
|
470 |
+
"eval_stsb_spearman": 0.8343500455447611,
|
471 |
+
"step": 6875
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.79,
|
475 |
+
"learning_rate": 6.428330901335728e-06,
|
476 |
+
"loss": 0.0001,
|
477 |
+
"step": 7000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 0.79,
|
481 |
+
"eval_avg_sts": 0.8202633038048985,
|
482 |
+
"eval_sickr_spearman": 0.8042502938317833,
|
483 |
+
"eval_stsb_spearman": 0.8362763137780139,
|
484 |
+
"step": 7000
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 0.8,
|
488 |
+
"eval_avg_sts": 0.819155219728974,
|
489 |
+
"eval_sickr_spearman": 0.8032414485771772,
|
490 |
+
"eval_stsb_spearman": 0.8350689908807707,
|
491 |
+
"step": 7125
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.81,
|
495 |
+
"eval_avg_sts": 0.8185946178321757,
|
496 |
+
"eval_sickr_spearman": 0.8011783206460045,
|
497 |
+
"eval_stsb_spearman": 0.836010915018347,
|
498 |
+
"step": 7250
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 0.83,
|
502 |
+
"eval_avg_sts": 0.8201412949742906,
|
503 |
+
"eval_sickr_spearman": 0.8025524424270474,
|
504 |
+
"eval_stsb_spearman": 0.8377301475215336,
|
505 |
+
"step": 7375
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 0.84,
|
509 |
+
"learning_rate": 4.744640251431137e-06,
|
510 |
+
"loss": 0.0,
|
511 |
+
"step": 7500
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 0.84,
|
515 |
+
"eval_avg_sts": 0.8212551171670207,
|
516 |
+
"eval_sickr_spearman": 0.8034978385966538,
|
517 |
+
"eval_stsb_spearman": 0.8390123957373876,
|
518 |
+
"step": 7500
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 0.86,
|
522 |
+
"eval_avg_sts": 0.818850668226643,
|
523 |
+
"eval_sickr_spearman": 0.8027982175731062,
|
524 |
+
"eval_stsb_spearman": 0.8349031188801797,
|
525 |
+
"step": 7625
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 0.87,
|
529 |
+
"eval_avg_sts": 0.8184223281084055,
|
530 |
+
"eval_sickr_spearman": 0.8041640299736004,
|
531 |
+
"eval_stsb_spearman": 0.8326806262432108,
|
532 |
+
"step": 7750
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.88,
|
536 |
+
"eval_avg_sts": 0.818812249944629,
|
537 |
+
"eval_sickr_spearman": 0.8036731040858015,
|
538 |
+
"eval_stsb_spearman": 0.8339513958034565,
|
539 |
+
"step": 7875
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.9,
|
543 |
+
"learning_rate": 3.0609496015265466e-06,
|
544 |
+
"loss": 0.0,
|
545 |
+
"step": 8000
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.9,
|
549 |
+
"eval_avg_sts": 0.8204302396512207,
|
550 |
+
"eval_sickr_spearman": 0.8052288794924833,
|
551 |
+
"eval_stsb_spearman": 0.8356315998099582,
|
552 |
+
"step": 8000
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 0.91,
|
556 |
+
"eval_avg_sts": 0.8209129013628449,
|
557 |
+
"eval_sickr_spearman": 0.8057768663287909,
|
558 |
+
"eval_stsb_spearman": 0.8360489363968989,
|
559 |
+
"step": 8125
|
560 |
+
},
|
561 |
+
{
|
562 |
+
"epoch": 0.93,
|
563 |
+
"eval_avg_sts": 0.8191227480526566,
|
564 |
+
"eval_sickr_spearman": 0.8051447290027637,
|
565 |
+
"eval_stsb_spearman": 0.8331007671025495,
|
566 |
+
"step": 8250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.94,
|
570 |
+
"eval_avg_sts": 0.8188352393246607,
|
571 |
+
"eval_sickr_spearman": 0.804106104465266,
|
572 |
+
"eval_stsb_spearman": 0.8335643741840555,
|
573 |
+
"step": 8375
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 0.95,
|
577 |
+
"learning_rate": 1.3772589516219555e-06,
|
578 |
+
"loss": 0.0001,
|
579 |
+
"step": 8500
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 0.95,
|
583 |
+
"eval_avg_sts": 0.8199749779902297,
|
584 |
+
"eval_sickr_spearman": 0.8036332863027092,
|
585 |
+
"eval_stsb_spearman": 0.8363166696777501,
|
586 |
+
"step": 8500
|
587 |
+
},
|
588 |
+
{
|
589 |
+
"epoch": 0.97,
|
590 |
+
"eval_avg_sts": 0.8199803565006372,
|
591 |
+
"eval_sickr_spearman": 0.8035859756677927,
|
592 |
+
"eval_stsb_spearman": 0.8363747373334816,
|
593 |
+
"step": 8625
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 0.98,
|
597 |
+
"eval_avg_sts": 0.8201168764945796,
|
598 |
+
"eval_sickr_spearman": 0.803825650863969,
|
599 |
+
"eval_stsb_spearman": 0.8364081021251903,
|
600 |
+
"step": 8750
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 1.0,
|
604 |
+
"eval_avg_sts": 0.820079455438296,
|
605 |
+
"eval_sickr_spearman": 0.8038641237762209,
|
606 |
+
"eval_stsb_spearman": 0.8362947871003711,
|
607 |
+
"step": 8875
|
608 |
+
},
|
609 |
+
{
|
610 |
+
"epoch": 1.0,
|
611 |
+
"step": 8909,
|
612 |
+
"train_runtime": 4791.076,
|
613 |
+
"train_samples_per_second": 1.859
|
614 |
+
}
|
615 |
+
],
|
616 |
+
"max_steps": 8909,
|
617 |
+
"num_train_epochs": 1,
|
618 |
+
"total_flos": 24044429181321216,
|
619 |
+
"trial_name": null,
|
620 |
+
"trial_params": null
|
621 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0e37bdcc4b6228191acee7df6a97d6fd66af2b815834823cdf34bfad2a2ba0e
|
3 |
+
size 2095
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|