versae commited on
Commit
3bab364
1 Parent(s): e8b7593

Gaussian 512 fine-tuned PAWS-X es with acc 0.8965

Browse files
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.8965,
4
+ "eval_loss": 0.5265980958938599,
5
+ "eval_runtime": 21.0936,
6
+ "eval_samples": 2000,
7
+ "eval_samples_per_second": 94.815,
8
+ "eval_steps_per_second": 5.926,
9
+ "train_loss": 0.1811776852978326,
10
+ "train_runtime": 8418.9007,
11
+ "train_samples": 49401,
12
+ "train_samples_per_second": 29.339,
13
+ "train_steps_per_second": 1.834
14
+ }
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bertin-project/bertin-base-gaussian-exp-512seqlen",
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "eos_token_id": 2,
9
+ "finetuning_task": "paws-x",
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-05,
17
+ "max_position_embeddings": 514,
18
+ "model_type": "roberta",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "problem_type": "single_label_classification",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.9.0.dev0",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 50265
29
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.8965,
4
+ "eval_loss": 0.5265980958938599,
5
+ "eval_runtime": 21.0936,
6
+ "eval_samples": 2000,
7
+ "eval_samples_per_second": 94.815,
8
+ "eval_steps_per_second": 5.926
9
+ }
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:f33890dc060a5e2908bbd3d4fe506d7ddc8194086f80f92d841e8c985f6f7b73
3
+ size 498674093
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "bertin-project/bertin-base-gaussian-exp-512seqlen", "tokenizer_class": "RobertaTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "train_loss": 0.1811776852978326,
4
+ "train_runtime": 8418.9007,
5
+ "train_samples": 49401,
6
+ "train_samples_per_second": 29.339,
7
+ "train_steps_per_second": 1.834
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,205 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.0,
5
+ "global_step": 15440,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.16,
12
+ "learning_rate": 4.8380829015544046e-05,
13
+ "loss": 0.5771,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.32,
18
+ "learning_rate": 4.676165803108808e-05,
19
+ "loss": 0.3961,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.49,
24
+ "learning_rate": 4.5142487046632126e-05,
25
+ "loss": 0.3378,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.65,
30
+ "learning_rate": 4.352331606217617e-05,
31
+ "loss": 0.3113,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.81,
36
+ "learning_rate": 4.190414507772021e-05,
37
+ "loss": 0.2978,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.97,
42
+ "learning_rate": 4.028497409326425e-05,
43
+ "loss": 0.269,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 1.13,
48
+ "learning_rate": 3.8665803108808294e-05,
49
+ "loss": 0.2396,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 1.3,
54
+ "learning_rate": 3.704663212435233e-05,
55
+ "loss": 0.215,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 1.46,
60
+ "learning_rate": 3.5427461139896374e-05,
61
+ "loss": 0.215,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 1.62,
66
+ "learning_rate": 3.380829015544041e-05,
67
+ "loss": 0.1969,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 1.78,
72
+ "learning_rate": 3.2189119170984454e-05,
73
+ "loss": 0.1983,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 1.94,
78
+ "learning_rate": 3.05699481865285e-05,
79
+ "loss": 0.2113,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 2.1,
84
+ "learning_rate": 2.8950777202072538e-05,
85
+ "loss": 0.1659,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 2.27,
90
+ "learning_rate": 2.7331606217616585e-05,
91
+ "loss": 0.1544,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 2.43,
96
+ "learning_rate": 2.5712435233160625e-05,
97
+ "loss": 0.1517,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 2.59,
102
+ "learning_rate": 2.4093264248704665e-05,
103
+ "loss": 0.156,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 2.75,
108
+ "learning_rate": 2.2474093264248706e-05,
109
+ "loss": 0.1547,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 2.91,
114
+ "learning_rate": 2.0854922279792746e-05,
115
+ "loss": 0.1331,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 3.08,
120
+ "learning_rate": 1.9235751295336786e-05,
121
+ "loss": 0.1278,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 3.24,
126
+ "learning_rate": 1.761658031088083e-05,
127
+ "loss": 0.1125,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 3.4,
132
+ "learning_rate": 1.5997409326424873e-05,
133
+ "loss": 0.1148,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 3.56,
138
+ "learning_rate": 1.4378238341968913e-05,
139
+ "loss": 0.1123,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 3.72,
144
+ "learning_rate": 1.2759067357512955e-05,
145
+ "loss": 0.106,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 3.89,
150
+ "learning_rate": 1.1139896373056995e-05,
151
+ "loss": 0.1053,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 4.05,
156
+ "learning_rate": 9.520725388601037e-06,
157
+ "loss": 0.0906,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 4.21,
162
+ "learning_rate": 7.901554404145079e-06,
163
+ "loss": 0.0792,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 4.37,
168
+ "learning_rate": 6.282383419689119e-06,
169
+ "loss": 0.0726,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 4.53,
174
+ "learning_rate": 4.663212435233161e-06,
175
+ "loss": 0.0781,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 4.7,
180
+ "learning_rate": 3.044041450777202e-06,
181
+ "loss": 0.0697,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 4.86,
186
+ "learning_rate": 1.4248704663212437e-06,
187
+ "loss": 0.075,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 5.0,
192
+ "step": 15440,
193
+ "total_flos": 6.49897462291968e+16,
194
+ "train_loss": 0.1811776852978326,
195
+ "train_runtime": 8418.9007,
196
+ "train_samples_per_second": 29.339,
197
+ "train_steps_per_second": 1.834
198
+ }
199
+ ],
200
+ "max_steps": 15440,
201
+ "num_train_epochs": 5,
202
+ "total_flos": 6.49897462291968e+16,
203
+ "trial_name": null,
204
+ "trial_params": null
205
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:576eca8f4ddd686c360a049868136f08fcaf73a619b5fc176fba347c70891c2e
3
+ size 2735
vocab.json ADDED
The diff for this file is too large to render. See raw diff