gguichard commited on
Commit
cfca728
1 Parent(s): 33a3c3f

Training in progress, epoch 13, checkpoint

Browse files
checkpoint-2756/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-2756/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "camembert/camembert-large",
4
+ "bias": "all",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": [
13
+ "classifier.bias",
14
+ "classifier.weight"
15
+ ],
16
+ "peft_type": "LORA",
17
+ "r": 64,
18
+ "revision": null,
19
+ "target_modules": [
20
+ "query",
21
+ "key",
22
+ "value",
23
+ "dense"
24
+ ],
25
+ "task_type": "TOKEN_CLS"
26
+ }
checkpoint-2756/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a040af46465d9f9826dcaaf587c3e890f9ffd775ae91912e24aa92043804240
3
+ size 117304221
checkpoint-2756/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e25274894be7475a14c709487d7eab6f727d198b0285856877bdfcc8c86bd663
3
+ size 234652550
checkpoint-2756/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c60933fb688cde84167ff6620e14f6124ccb62aa173f5c56ee382f39ee3c718e
3
+ size 14575
checkpoint-2756/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88b50c69157118f984fd17a66b6634ca197ec0c5968d3df000ad14452e5dfc09
3
+ size 627
checkpoint-2756/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f98f266fdc548c94216aaadc13ffaaafacf0c8793303e2195322d954549ea261
3
+ size 808767
checkpoint-2756/special_tokens_map.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "cls_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "mask_token": {
10
+ "content": "<mask>",
11
+ "lstrip": true,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "unk_token": "<unk>"
19
+ }
checkpoint-2756/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2756/tokenizer_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "clean_up_tokenization_spaces": true,
8
+ "cls_token": "<s>",
9
+ "eos_token": "</s>",
10
+ "mask_token": {
11
+ "__type": "AddedToken",
12
+ "content": "<mask>",
13
+ "lstrip": true,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "model_max_length": 1000000000000000019884624838656,
19
+ "pad_token": "<pad>",
20
+ "sep_token": "</s>",
21
+ "sp_model_kwargs": {},
22
+ "tokenizer_class": "CamembertTokenizer",
23
+ "unk_token": "<unk>"
24
+ }
checkpoint-2756/trainer_state.json ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 13.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2756,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 9.95e-05,
14
+ "loss": 6.3714,
15
+ "step": 212
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_loss": 5.850927352905273,
20
+ "eval_runtime": 3.8282,
21
+ "eval_samples_per_second": 176.584,
22
+ "eval_steps_per_second": 22.204,
23
+ "step": 212
24
+ },
25
+ {
26
+ "epoch": 2.0,
27
+ "learning_rate": 9.900000000000001e-05,
28
+ "loss": 5.5438,
29
+ "step": 424
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "eval_loss": 4.862037658691406,
34
+ "eval_runtime": 3.7425,
35
+ "eval_samples_per_second": 180.626,
36
+ "eval_steps_per_second": 22.712,
37
+ "step": 424
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "learning_rate": 9.850000000000001e-05,
42
+ "loss": 4.6993,
43
+ "step": 636
44
+ },
45
+ {
46
+ "epoch": 3.0,
47
+ "eval_loss": 4.018372058868408,
48
+ "eval_runtime": 4.0424,
49
+ "eval_samples_per_second": 167.229,
50
+ "eval_steps_per_second": 21.027,
51
+ "step": 636
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "learning_rate": 9.8e-05,
56
+ "loss": 3.967,
57
+ "step": 848
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "eval_loss": 3.290557384490967,
62
+ "eval_runtime": 4.3933,
63
+ "eval_samples_per_second": 153.872,
64
+ "eval_steps_per_second": 19.348,
65
+ "step": 848
66
+ },
67
+ {
68
+ "epoch": 5.0,
69
+ "learning_rate": 9.75e-05,
70
+ "loss": 3.305,
71
+ "step": 1060
72
+ },
73
+ {
74
+ "epoch": 5.0,
75
+ "eval_loss": 2.6497254371643066,
76
+ "eval_runtime": 4.3562,
77
+ "eval_samples_per_second": 155.182,
78
+ "eval_steps_per_second": 19.513,
79
+ "step": 1060
80
+ },
81
+ {
82
+ "epoch": 6.0,
83
+ "learning_rate": 9.7e-05,
84
+ "loss": 2.7185,
85
+ "step": 1272
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "eval_loss": 2.1158342361450195,
90
+ "eval_runtime": 4.5634,
91
+ "eval_samples_per_second": 148.135,
92
+ "eval_steps_per_second": 18.626,
93
+ "step": 1272
94
+ },
95
+ {
96
+ "epoch": 7.0,
97
+ "learning_rate": 9.65e-05,
98
+ "loss": 2.2181,
99
+ "step": 1484
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "eval_loss": 1.6882902383804321,
104
+ "eval_runtime": 4.3232,
105
+ "eval_samples_per_second": 156.367,
106
+ "eval_steps_per_second": 19.662,
107
+ "step": 1484
108
+ },
109
+ {
110
+ "epoch": 8.0,
111
+ "learning_rate": 9.6e-05,
112
+ "loss": 1.8106,
113
+ "step": 1696
114
+ },
115
+ {
116
+ "epoch": 8.0,
117
+ "eval_loss": 1.3730388879776,
118
+ "eval_runtime": 4.3486,
119
+ "eval_samples_per_second": 155.451,
120
+ "eval_steps_per_second": 19.546,
121
+ "step": 1696
122
+ },
123
+ {
124
+ "epoch": 9.0,
125
+ "learning_rate": 9.55e-05,
126
+ "loss": 1.497,
127
+ "step": 1908
128
+ },
129
+ {
130
+ "epoch": 9.0,
131
+ "eval_loss": 1.142008900642395,
132
+ "eval_runtime": 4.2743,
133
+ "eval_samples_per_second": 158.154,
134
+ "eval_steps_per_second": 19.886,
135
+ "step": 1908
136
+ },
137
+ {
138
+ "epoch": 10.0,
139
+ "learning_rate": 9.5e-05,
140
+ "loss": 1.2526,
141
+ "step": 2120
142
+ },
143
+ {
144
+ "epoch": 10.0,
145
+ "eval_loss": 0.9753935933113098,
146
+ "eval_runtime": 4.2134,
147
+ "eval_samples_per_second": 160.44,
148
+ "eval_steps_per_second": 20.174,
149
+ "step": 2120
150
+ },
151
+ {
152
+ "epoch": 11.0,
153
+ "learning_rate": 9.449999999999999e-05,
154
+ "loss": 1.0576,
155
+ "step": 2332
156
+ },
157
+ {
158
+ "epoch": 11.0,
159
+ "eval_loss": 0.8463678956031799,
160
+ "eval_runtime": 4.432,
161
+ "eval_samples_per_second": 152.527,
162
+ "eval_steps_per_second": 19.179,
163
+ "step": 2332
164
+ },
165
+ {
166
+ "epoch": 12.0,
167
+ "learning_rate": 9.4e-05,
168
+ "loss": 0.8965,
169
+ "step": 2544
170
+ },
171
+ {
172
+ "epoch": 12.0,
173
+ "eval_loss": 0.7349700927734375,
174
+ "eval_runtime": 4.5659,
175
+ "eval_samples_per_second": 148.054,
176
+ "eval_steps_per_second": 18.616,
177
+ "step": 2544
178
+ },
179
+ {
180
+ "epoch": 13.0,
181
+ "learning_rate": 9.350000000000001e-05,
182
+ "loss": 0.7613,
183
+ "step": 2756
184
+ },
185
+ {
186
+ "epoch": 13.0,
187
+ "eval_loss": 0.6494072675704956,
188
+ "eval_runtime": 4.499,
189
+ "eval_samples_per_second": 150.254,
190
+ "eval_steps_per_second": 18.893,
191
+ "step": 2756
192
+ }
193
+ ],
194
+ "logging_steps": 500,
195
+ "max_steps": 42400,
196
+ "num_train_epochs": 200,
197
+ "save_steps": 500,
198
+ "total_flos": 3701919838610772.0,
199
+ "trial_name": null,
200
+ "trial_params": null
201
+ }
checkpoint-2756/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e7d47d9ef4da320ee46463f8716d365eb34b4bc5488090164af1bcf3ad60b35
3
+ size 4155