gguichard commited on
Commit
e734f3e
1 Parent(s): edc9de6

Training in progress, epoch 9, checkpoint

Browse files
checkpoint-1908/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-1908/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": 16,
18
+ "revision": null,
19
+ "target_modules": [
20
+ "query",
21
+ "key",
22
+ "value",
23
+ "dense"
24
+ ],
25
+ "task_type": "TOKEN_CLS"
26
+ }
checkpoint-1908/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68024815d1ea4c204bbc08947303f452e164f6cb95b84cb23764737ec51c1bda
3
+ size 32369117
checkpoint-1908/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fc721034e8f8d7d58f429c4890280625995b9427364fccac3fd3683198cd5fc
3
+ size 64782278
checkpoint-1908/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09dc264af7f407d231aec865f06b726a55bbd0d0118ee7c6052f8de350cbe631
3
+ size 14575
checkpoint-1908/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:643c232092163254ce86613fb97c17e3e2bd3932ce3c4c2ce7e3c96298d350cc
3
+ size 627
checkpoint-1908/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-1908/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-1908/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1908/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-1908/trainer_state.json ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1908,
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": 6.3597,
15
+ "step": 212
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_loss": 5.853069305419922,
20
+ "eval_runtime": 3.5937,
21
+ "eval_samples_per_second": 188.106,
22
+ "eval_steps_per_second": 23.652,
23
+ "step": 212
24
+ },
25
+ {
26
+ "epoch": 2.0,
27
+ "learning_rate": 9.900000000000001e-05,
28
+ "loss": 5.5629,
29
+ "step": 424
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "eval_loss": 4.915834903717041,
34
+ "eval_runtime": 3.704,
35
+ "eval_samples_per_second": 182.506,
36
+ "eval_steps_per_second": 22.948,
37
+ "step": 424
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "learning_rate": 9.850000000000001e-05,
42
+ "loss": 4.8228,
43
+ "step": 636
44
+ },
45
+ {
46
+ "epoch": 3.0,
47
+ "eval_loss": 4.177055835723877,
48
+ "eval_runtime": 3.7104,
49
+ "eval_samples_per_second": 182.193,
50
+ "eval_steps_per_second": 22.909,
51
+ "step": 636
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "learning_rate": 9.8e-05,
56
+ "loss": 4.1035,
57
+ "step": 848
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "eval_loss": 3.462527275085449,
62
+ "eval_runtime": 3.1205,
63
+ "eval_samples_per_second": 216.633,
64
+ "eval_steps_per_second": 27.239,
65
+ "step": 848
66
+ },
67
+ {
68
+ "epoch": 5.0,
69
+ "learning_rate": 9.75e-05,
70
+ "loss": 3.467,
71
+ "step": 1060
72
+ },
73
+ {
74
+ "epoch": 5.0,
75
+ "eval_loss": 2.8398332595825195,
76
+ "eval_runtime": 3.3235,
77
+ "eval_samples_per_second": 203.402,
78
+ "eval_steps_per_second": 25.576,
79
+ "step": 1060
80
+ },
81
+ {
82
+ "epoch": 6.0,
83
+ "learning_rate": 9.7e-05,
84
+ "loss": 2.8887,
85
+ "step": 1272
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "eval_loss": 2.3201420307159424,
90
+ "eval_runtime": 3.4255,
91
+ "eval_samples_per_second": 197.342,
92
+ "eval_steps_per_second": 24.814,
93
+ "step": 1272
94
+ },
95
+ {
96
+ "epoch": 7.0,
97
+ "learning_rate": 9.65e-05,
98
+ "loss": 2.4003,
99
+ "step": 1484
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "eval_loss": 1.8698177337646484,
104
+ "eval_runtime": 3.5218,
105
+ "eval_samples_per_second": 191.945,
106
+ "eval_steps_per_second": 24.135,
107
+ "step": 1484
108
+ },
109
+ {
110
+ "epoch": 8.0,
111
+ "learning_rate": 9.6e-05,
112
+ "loss": 1.9919,
113
+ "step": 1696
114
+ },
115
+ {
116
+ "epoch": 8.0,
117
+ "eval_loss": 1.532776951789856,
118
+ "eval_runtime": 3.5514,
119
+ "eval_samples_per_second": 190.345,
120
+ "eval_steps_per_second": 23.934,
121
+ "step": 1696
122
+ },
123
+ {
124
+ "epoch": 9.0,
125
+ "learning_rate": 9.55e-05,
126
+ "loss": 1.6656,
127
+ "step": 1908
128
+ },
129
+ {
130
+ "epoch": 9.0,
131
+ "eval_loss": 1.2841078042984009,
132
+ "eval_runtime": 3.622,
133
+ "eval_samples_per_second": 186.637,
134
+ "eval_steps_per_second": 23.468,
135
+ "step": 1908
136
+ }
137
+ ],
138
+ "logging_steps": 500,
139
+ "max_steps": 42400,
140
+ "num_train_epochs": 200,
141
+ "save_steps": 500,
142
+ "total_flos": 2399636558645124.0,
143
+ "trial_name": null,
144
+ "trial_params": null
145
+ }
checkpoint-1908/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bbad506d5d7c9f6e6c369455b23a5312c048f6fdad38f9f3772905289326b4f
3
+ size 4155